code
stringlengths 87
55.2k
| code_codestyle
int64 0
349
| style_context
stringlengths 135
49.1k
| style_context_codestyle
int64 0
349
| label
int64 0
1
|
|---|---|---|---|---|
"""simple docstring"""
import math
def lowercase ( A_ , A_ = 0 , A_ = 0 )-> list:
'''simple docstring'''
a : Optional[Any] = end or len(A_ )
for i in range(A_ , A_ ):
a : Optional[Any] = i
a : Dict = array[i]
while temp_index != start and temp_index_value < array[temp_index - 1]:
a : str = array[temp_index - 1]
temp_index -= 1
a : Optional[Any] = temp_index_value
return array
def lowercase ( A_ , A_ , A_ )-> None: # Max Heap
'''simple docstring'''
a : Union[str, Any] = index
a : str = 2 * index + 1 # Left Node
a : Optional[int] = 2 * index + 2 # Right Node
if left_index < heap_size and array[largest] < array[left_index]:
a : List[Any] = left_index
if right_index < heap_size and array[largest] < array[right_index]:
a : Optional[int] = right_index
if largest != index:
a , a : List[Any] = array[largest], array[index]
heapify(A_ , A_ , A_ )
def lowercase ( A_ )-> list:
'''simple docstring'''
a : Optional[int] = len(A_ )
for i in range(n // 2 , -1 , -1 ):
heapify(A_ , A_ , A_ )
for i in range(n - 1 , 0 , -1 ):
a , a : Any = array[0], array[i]
heapify(A_ , 0 , A_ )
return array
def lowercase ( A_ , A_ , A_ , A_ )-> int:
'''simple docstring'''
if (array[first_index] > array[middle_index]) != (
array[first_index] > array[last_index]
):
return array[first_index]
elif (array[middle_index] > array[first_index]) != (
array[middle_index] > array[last_index]
):
return array[middle_index]
else:
return array[last_index]
def lowercase ( A_ , A_ , A_ , A_ )-> int:
'''simple docstring'''
a : str = low
a : int = high
while True:
while array[i] < pivot:
i += 1
j -= 1
while pivot < array[j]:
j -= 1
if i >= j:
return i
a , a : int = array[j], array[i]
i += 1
def lowercase ( A_ )-> list:
'''simple docstring'''
if len(A_ ) == 0:
return array
a : Union[str, Any] = 2 * math.ceil(math.loga(len(A_ ) ) )
a : Optional[int] = 16
return intro_sort(A_ , 0 , len(A_ ) , A_ , A_ )
def lowercase ( A_ , A_ , A_ , A_ , A_ )-> list:
'''simple docstring'''
while end - start > size_threshold:
if max_depth == 0:
return heap_sort(A_ )
max_depth -= 1
a : Tuple = median_of_a(A_ , A_ , start + ((end - start) // 2) + 1 , end - 1 )
a : Optional[Any] = partition(A_ , A_ , A_ , A_ )
intro_sort(A_ , A_ , A_ , A_ , A_ )
a : Dict = p
return insertion_sort(A_ , A_ , A_ )
if __name__ == "__main__":
import doctest
doctest.testmod()
__lowercase = input("""Enter numbers separated by a comma : """).strip()
__lowercase = [float(item) for item in user_input.split(""",""")]
print(sort(unsorted))
| 40
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : int = logging.get_logger(__name__)
snake_case_ : str = {}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''llama'''
UpperCAmelCase__ : Dict = ['''past_key_values''']
def __init__( self : str , _snake_case : List[str]=32000 , _snake_case : int=4096 , _snake_case : List[str]=11008 , _snake_case : Optional[int]=32 , _snake_case : List[Any]=32 , _snake_case : Tuple=None , _snake_case : int="silu" , _snake_case : List[Any]=2048 , _snake_case : List[str]=0.0_2 , _snake_case : Any=1e-6 , _snake_case : List[str]=True , _snake_case : Optional[Any]=0 , _snake_case : Dict=1 , _snake_case : List[Any]=2 , _snake_case : str=1 , _snake_case : Union[str, Any]=False , _snake_case : str=None , **_snake_case : List[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = max_position_embeddings
UpperCAmelCase_ = hidden_size
UpperCAmelCase_ = intermediate_size
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
# for backward compatibility
if num_key_value_heads is None:
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = num_key_value_heads
UpperCAmelCase_ = hidden_act
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = rms_norm_eps
UpperCAmelCase_ = pretraining_tp
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = rope_scaling
self._rope_scaling_validation()
super().__init__(
pad_token_id=_snake_case , bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case , )
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
if self.rope_scaling is None:
return
if not isinstance(self.rope_scaling , _snake_case) or len(self.rope_scaling) != 2:
raise ValueError(
'''`rope_scaling` must be a dictionary with with two fields, `name` and `factor`, '''
F"""got {self.rope_scaling}""")
UpperCAmelCase_ = self.rope_scaling.get('''type''' , _snake_case)
UpperCAmelCase_ = self.rope_scaling.get('''factor''' , _snake_case)
if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic"]:
raise ValueError(
F"""`rope_scaling`'s name field must be one of ['linear', 'dynamic'], got {rope_scaling_type}""")
if rope_scaling_factor is None or not isinstance(_snake_case , _snake_case) or rope_scaling_factor <= 1.0:
raise ValueError(F"""`rope_scaling`'s factor field must be an float > 1, got {rope_scaling_factor}""")
| 51
| 0
|
'''simple docstring'''
from typing import TYPE_CHECKING
from ...file_utils import _LazyModule, is_torch_available
from ...utils import OptionalDependencyNotAvailable
_A : str ={
'''configuration_gpt_neox_japanese''': ['''GPT_NEOX_JAPANESE_PRETRAINED_CONFIG_ARCHIVE_MAP''', '''GPTNeoXJapaneseConfig'''],
'''tokenization_gpt_neox_japanese''': ['''GPTNeoXJapaneseTokenizer'''],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_A : Optional[Any] =[
'''GPT_NEOX_JAPANESE_PRETRAINED_MODEL_ARCHIVE_LIST''',
'''GPTNeoXJapaneseForCausalLM''',
'''GPTNeoXJapaneseLayer''',
'''GPTNeoXJapaneseModel''',
'''GPTNeoXJapanesePreTrainedModel''',
]
if TYPE_CHECKING:
from .configuration_gpt_neox_japanese import GPT_NEOX_JAPANESE_PRETRAINED_CONFIG_ARCHIVE_MAP, GPTNeoXJapaneseConfig
from .tokenization_gpt_neox_japanese import GPTNeoXJapaneseTokenizer
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_gpt_neox_japanese import (
GPT_NEOX_JAPANESE_PRETRAINED_MODEL_ARCHIVE_LIST,
GPTNeoXJapaneseForCausalLM,
GPTNeoXJapaneseLayer,
GPTNeoXJapaneseModel,
GPTNeoXJapanesePreTrainedModel,
)
else:
import sys
_A : Optional[Any] =_LazyModule(__name__, globals()['''__file__'''], _import_structure, module_spec=__spec__)
| 41
|
from collections import OrderedDict
from typing import Any, List, Mapping, Optional
from ... import PreTrainedTokenizer, TensorType, is_torch_available
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfigWithPast, PatchingSpec
from ...utils import logging
snake_case_ : List[str] = logging.get_logger(__name__)
snake_case_ : Tuple = {
"Salesforce/codegen-350M-nl": "https://huggingface.co/Salesforce/codegen-350M-nl/resolve/main/config.json",
"Salesforce/codegen-350M-multi": "https://huggingface.co/Salesforce/codegen-350M-multi/resolve/main/config.json",
"Salesforce/codegen-350M-mono": "https://huggingface.co/Salesforce/codegen-350M-mono/resolve/main/config.json",
"Salesforce/codegen-2B-nl": "https://huggingface.co/Salesforce/codegen-2B-nl/resolve/main/config.json",
"Salesforce/codegen-2B-multi": "https://huggingface.co/Salesforce/codegen-2B-multi/resolve/main/config.json",
"Salesforce/codegen-2B-mono": "https://huggingface.co/Salesforce/codegen-2B-mono/resolve/main/config.json",
"Salesforce/codegen-6B-nl": "https://huggingface.co/Salesforce/codegen-6B-nl/resolve/main/config.json",
"Salesforce/codegen-6B-multi": "https://huggingface.co/Salesforce/codegen-6B-multi/resolve/main/config.json",
"Salesforce/codegen-6B-mono": "https://huggingface.co/Salesforce/codegen-6B-mono/resolve/main/config.json",
"Salesforce/codegen-16B-nl": "https://huggingface.co/Salesforce/codegen-16B-nl/resolve/main/config.json",
"Salesforce/codegen-16B-multi": "https://huggingface.co/Salesforce/codegen-16B-multi/resolve/main/config.json",
"Salesforce/codegen-16B-mono": "https://huggingface.co/Salesforce/codegen-16B-mono/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''codegen'''
UpperCAmelCase__ : int = {
'''max_position_embeddings''': '''n_positions''',
'''hidden_size''': '''n_embd''',
'''num_attention_heads''': '''n_head''',
'''num_hidden_layers''': '''n_layer''',
}
def __init__( self : Union[str, Any] , _snake_case : Union[str, Any]=50400 , _snake_case : Optional[int]=2048 , _snake_case : Union[str, Any]=2048 , _snake_case : List[str]=4096 , _snake_case : Any=28 , _snake_case : List[str]=16 , _snake_case : int=64 , _snake_case : Tuple=None , _snake_case : Dict="gelu_new" , _snake_case : Union[str, Any]=0.0 , _snake_case : Optional[Any]=0.0 , _snake_case : List[Any]=0.0 , _snake_case : List[Any]=1e-5 , _snake_case : List[str]=0.0_2 , _snake_case : Optional[Any]=True , _snake_case : int=50256 , _snake_case : Tuple=50256 , _snake_case : int=False , **_snake_case : Any , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = n_ctx
UpperCAmelCase_ = n_positions
UpperCAmelCase_ = n_embd
UpperCAmelCase_ = n_layer
UpperCAmelCase_ = n_head
UpperCAmelCase_ = n_inner
UpperCAmelCase_ = rotary_dim
UpperCAmelCase_ = activation_function
UpperCAmelCase_ = resid_pdrop
UpperCAmelCase_ = embd_pdrop
UpperCAmelCase_ = attn_pdrop
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
super().__init__(
bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case)
class __snake_case ( a ):
def __init__( self : Tuple , _snake_case : PretrainedConfig , _snake_case : str = "default" , _snake_case : List[PatchingSpec] = None , _snake_case : bool = False , ):
"""simple docstring"""
super().__init__(_snake_case , task=_snake_case , patching_specs=_snake_case , use_past=_snake_case)
if not getattr(self._config , '''pad_token_id''' , _snake_case):
# TODO: how to do that better?
UpperCAmelCase_ = 0
@property
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = OrderedDict({'''input_ids''': {0: '''batch''', 1: '''sequence'''}})
if self.use_past:
self.fill_with_past_key_values_(_snake_case , direction='''inputs''')
UpperCAmelCase_ = {0: '''batch''', 1: '''past_sequence + sequence'''}
else:
UpperCAmelCase_ = {0: '''batch''', 1: '''sequence'''}
return common_inputs
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return self._config.n_layer
@property
def lowerCamelCase ( self : int):
"""simple docstring"""
return self._config.n_head
def lowerCamelCase ( self : Optional[int] , _snake_case : PreTrainedTokenizer , _snake_case : int = -1 , _snake_case : int = -1 , _snake_case : bool = False , _snake_case : Optional[TensorType] = None , ):
"""simple docstring"""
UpperCAmelCase_ = super(_snake_case , self).generate_dummy_inputs(
_snake_case , batch_size=_snake_case , seq_length=_snake_case , is_pair=_snake_case , framework=_snake_case)
# We need to order the input in the way they appears in the forward()
UpperCAmelCase_ = OrderedDict({'''input_ids''': common_inputs['''input_ids''']})
# Need to add the past_keys
if self.use_past:
if not is_torch_available():
raise ValueError('''Cannot generate dummy past_keys inputs without PyTorch installed.''')
else:
import torch
UpperCAmelCase_ , UpperCAmelCase_ = common_inputs['''input_ids'''].shape
# Not using the same length for past_key_values
UpperCAmelCase_ = seqlen + 2
UpperCAmelCase_ = (
batch,
self.num_attention_heads,
past_key_values_length,
self._config.hidden_size // self.num_attention_heads,
)
UpperCAmelCase_ = [
(torch.zeros(_snake_case), torch.zeros(_snake_case)) for _ in range(self.num_layers)
]
UpperCAmelCase_ = common_inputs['''attention_mask''']
if self.use_past:
UpperCAmelCase_ = ordered_inputs['''attention_mask'''].dtype
UpperCAmelCase_ = torch.cat(
[ordered_inputs['''attention_mask'''], torch.ones(_snake_case , _snake_case , dtype=_snake_case)] , dim=1)
return ordered_inputs
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return 13
| 51
| 0
|
'''simple docstring'''
import argparse
import json
import os
from tensorflow.core.protobuf.saved_model_pba import SavedModel
# All paths are set with the intent you should run this script from the root of the repo with the command
# python utils/check_copies.py
lowercase : Dict = "."
# Internal TensorFlow ops that can be safely ignored (mostly specific to a saved model)
lowercase : Optional[Any] = [
"Assert",
"AssignVariableOp",
"EmptyTensorList",
"MergeV2Checkpoints",
"ReadVariableOp",
"ResourceGather",
"RestoreV2",
"SaveV2",
"ShardedFilename",
"StatefulPartitionedCall",
"StaticRegexFullMatch",
"VarHandleOp",
]
def SCREAMING_SNAKE_CASE__ ( __A , __A , __A ) -> List[Any]:
_snake_case = SavedModel()
_snake_case = []
with open(os.path.join(__A , 'utils' , 'tf_ops' , 'onnx.json' ) ) as f:
_snake_case = json.load(__A )['opsets']
for i in range(1 , opset + 1 ):
onnx_ops.extend(onnx_opsets[str(__A )] )
with open(__A , 'rb' ) as f:
saved_model.ParseFromString(f.read() )
_snake_case = set()
# Iterate over every metagraph in case there is more than one (a saved model can contain multiple graphs)
for meta_graph in saved_model.meta_graphs:
# Add operations in the graph definition
model_op_names.update(node.op for node in meta_graph.graph_def.node )
# Go through the functions in the graph definition
for func in meta_graph.graph_def.library.function:
# Add operations in each function
model_op_names.update(node.op for node in func.node_def )
# Convert to list, sorted if you want
_snake_case = sorted(__A )
_snake_case = []
for op in model_op_names:
if op not in onnx_ops and op not in INTERNAL_OPS:
incompatible_ops.append(__A )
if strict and len(__A ) > 0:
raise Exception(F'Found the following incompatible ops for the opset {opset}:\n' + incompatible_ops )
elif len(__A ) > 0:
print(F'Found the following incompatible ops for the opset {opset}:' )
print(*__A , sep='\n' )
else:
print(F'The saved model {saved_model_path} can properly be converted with ONNX.' )
if __name__ == "__main__":
lowercase : str = argparse.ArgumentParser()
parser.add_argument("--saved_model_path", help="Path of the saved model to check (the .pb file).")
parser.add_argument(
"--opset", default=12, type=int, help="The ONNX opset against which the model has to be tested."
)
parser.add_argument(
"--framework", choices=["onnx"], default="onnx", help="Frameworks against which to test the saved model."
)
parser.add_argument(
"--strict", action="store_true", help="Whether make the checking strict (raise errors) or not (raise warnings)"
)
lowercase : List[Any] = parser.parse_args()
if args.framework == "onnx":
onnx_compliancy(args.saved_model_path, args.strict, args.opset)
| 42
|
import os
import unittest
from transformers.models.phobert.tokenization_phobert import VOCAB_FILES_NAMES, PhobertTokenizer
from ...test_tokenization_common import TokenizerTesterMixin
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : Any = PhobertTokenizer
UpperCAmelCase__ : List[str] = False
def lowerCamelCase ( self : str):
"""simple docstring"""
super().setUp()
# Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt
UpperCAmelCase_ = ['''T@@''', '''i''', '''I''', '''R@@''', '''r''', '''e@@''']
UpperCAmelCase_ = dict(zip(_snake_case , range(len(_snake_case))))
UpperCAmelCase_ = ['''#version: 0.2''', '''l à</w>''']
UpperCAmelCase_ = {'''unk_token''': '''<unk>'''}
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''vocab_file'''])
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''merges_file'''])
with open(self.vocab_file , '''w''' , encoding='''utf-8''') as fp:
for token in vocab_tokens:
fp.write(F"""{token} {vocab_tokens[token]}\n""")
with open(self.merges_file , '''w''' , encoding='''utf-8''') as fp:
fp.write('''\n'''.join(_snake_case))
def lowerCamelCase ( self : int , **_snake_case : Any):
"""simple docstring"""
kwargs.update(self.special_tokens_map)
return PhobertTokenizer.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Optional[Any] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T<unk> i <unk> <unk> <unk> <unk> <unk> <unk> I Re<unk> e<unk> <unk> <unk> <unk>'''
return input_text, output_text
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = PhobertTokenizer(self.vocab_file , self.merges_file , **self.special_tokens_map)
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T@@ ô@@ i l@@ à V@@ i@@ n@@ A@@ I R@@ e@@ s@@ e@@ a@@ r@@ c@@ h'''.split()
UpperCAmelCase_ = tokenizer.tokenize(_snake_case)
print(_snake_case)
self.assertListEqual(_snake_case , _snake_case)
UpperCAmelCase_ = tokens + [tokenizer.unk_token]
UpperCAmelCase_ = [4, 3, 5, 3, 3, 3, 3, 3, 3, 6, 7, 9, 3, 9, 3, 3, 3, 3, 3]
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case) , _snake_case)
| 51
| 0
|
import unittest
from parameterized import parameterized
from transformers import LlamaConfig, is_torch_available, set_seed
from transformers.testing_utils import require_torch, slow, torch_device
from ...generation.test_utils import GenerationTesterMixin
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import LlamaForCausalLM, LlamaForSequenceClassification, LlamaModel, LlamaTokenizer
class lowerCamelCase_ :
'''simple docstring'''
def __init__( self , __lowercase , __lowercase=13 , __lowercase=7 , __lowercase=True , __lowercase=True , __lowercase=False , __lowercase=True , __lowercase=99 , __lowercase=32 , __lowercase=5 , __lowercase=4 , __lowercase=37 , __lowercase="gelu" , __lowercase=0.1 , __lowercase=0.1 , __lowercase=512 , __lowercase=16 , __lowercase=2 , __lowercase=0.02 , __lowercase=3 , __lowercase=4 , __lowercase=None , ) -> Dict:
__UpperCamelCase :Union[str, Any] = parent
__UpperCamelCase :List[str] = batch_size
__UpperCamelCase :int = seq_length
__UpperCamelCase :str = is_training
__UpperCamelCase :Optional[Any] = use_input_mask
__UpperCamelCase :List[Any] = use_token_type_ids
__UpperCamelCase :Optional[Any] = use_labels
__UpperCamelCase :List[Any] = vocab_size
__UpperCamelCase :Optional[int] = hidden_size
__UpperCamelCase :Tuple = num_hidden_layers
__UpperCamelCase :Tuple = num_attention_heads
__UpperCamelCase :Optional[int] = intermediate_size
__UpperCamelCase :Union[str, Any] = hidden_act
__UpperCamelCase :int = hidden_dropout_prob
__UpperCamelCase :Optional[int] = attention_probs_dropout_prob
__UpperCamelCase :Tuple = max_position_embeddings
__UpperCamelCase :str = type_vocab_size
__UpperCamelCase :List[Any] = type_sequence_label_size
__UpperCamelCase :Optional[int] = initializer_range
__UpperCamelCase :str = num_labels
__UpperCamelCase :Any = num_choices
__UpperCamelCase :Any = scope
def UpperCamelCase__ ( self) -> List[Any]:
__UpperCamelCase :List[str] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size)
__UpperCamelCase :int = None
if self.use_input_mask:
__UpperCamelCase :Union[str, Any] = random_attention_mask([self.batch_size, self.seq_length])
__UpperCamelCase :str = None
if self.use_token_type_ids:
__UpperCamelCase :Dict = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size)
__UpperCamelCase :Optional[Any] = None
__UpperCamelCase :Dict = None
__UpperCamelCase :Any = None
if self.use_labels:
__UpperCamelCase :List[Any] = ids_tensor([self.batch_size] , self.type_sequence_label_size)
__UpperCamelCase :Union[str, Any] = ids_tensor([self.batch_size, self.seq_length] , self.num_labels)
__UpperCamelCase :Optional[int] = ids_tensor([self.batch_size] , self.num_choices)
__UpperCamelCase :int = self.get_config()
return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
def UpperCamelCase__ ( self) -> str:
return LlamaConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=__lowercase , initializer_range=self.initializer_range , )
def UpperCamelCase__ ( self , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase) -> Tuple:
__UpperCamelCase :str = LlamaModel(config=__lowercase)
model.to(__lowercase)
model.eval()
__UpperCamelCase :str = model(__lowercase , attention_mask=__lowercase)
__UpperCamelCase :Any = model(__lowercase)
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size))
def UpperCamelCase__ ( self , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , ) -> Tuple:
__UpperCamelCase :Optional[Any] = True
__UpperCamelCase :Dict = LlamaModel(__lowercase)
model.to(__lowercase)
model.eval()
__UpperCamelCase :int = model(
__lowercase , attention_mask=__lowercase , encoder_hidden_states=__lowercase , encoder_attention_mask=__lowercase , )
__UpperCamelCase :Any = model(
__lowercase , attention_mask=__lowercase , encoder_hidden_states=__lowercase , )
__UpperCamelCase :Tuple = model(__lowercase , attention_mask=__lowercase)
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size))
def UpperCamelCase__ ( self , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , ) -> Optional[Any]:
__UpperCamelCase :Optional[int] = LlamaForCausalLM(config=__lowercase)
model.to(__lowercase)
model.eval()
__UpperCamelCase :Any = model(__lowercase , attention_mask=__lowercase , labels=__lowercase)
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size))
def UpperCamelCase__ ( self , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , ) -> List[Any]:
__UpperCamelCase :str = True
__UpperCamelCase :Tuple = True
__UpperCamelCase :List[Any] = LlamaForCausalLM(config=__lowercase)
model.to(__lowercase)
model.eval()
# first forward pass
__UpperCamelCase :Union[str, Any] = model(
__lowercase , attention_mask=__lowercase , encoder_hidden_states=__lowercase , encoder_attention_mask=__lowercase , use_cache=__lowercase , )
__UpperCamelCase :Optional[Any] = outputs.past_key_values
# create hypothetical multiple next token and extent to next_input_ids
__UpperCamelCase :Optional[int] = ids_tensor((self.batch_size, 3) , config.vocab_size)
__UpperCamelCase :int = ids_tensor((self.batch_size, 3) , vocab_size=2)
# append to next input_ids and
__UpperCamelCase :Any = torch.cat([input_ids, next_tokens] , dim=-1)
__UpperCamelCase :Optional[int] = torch.cat([input_mask, next_mask] , dim=-1)
__UpperCamelCase :Union[str, Any] = model(
__lowercase , attention_mask=__lowercase , encoder_hidden_states=__lowercase , encoder_attention_mask=__lowercase , output_hidden_states=__lowercase , )['''hidden_states'''][0]
__UpperCamelCase :Tuple = model(
__lowercase , attention_mask=__lowercase , encoder_hidden_states=__lowercase , encoder_attention_mask=__lowercase , past_key_values=__lowercase , output_hidden_states=__lowercase , )['''hidden_states'''][0]
# select random slice
__UpperCamelCase :Any = ids_tensor((1,) , output_from_past.shape[-1]).item()
__UpperCamelCase :Optional[Any] = output_from_no_past[:, -3:, random_slice_idx].detach()
__UpperCamelCase :Union[str, Any] = output_from_past[:, :, random_slice_idx].detach()
self.parent.assertTrue(output_from_past_slice.shape[1] == next_tokens.shape[1])
# test that outputs are equal for slice
self.parent.assertTrue(torch.allclose(__lowercase , __lowercase , atol=1E-3))
def UpperCamelCase__ ( self) -> str:
__UpperCamelCase :Union[str, Any] = self.prepare_config_and_inputs()
(
(
__UpperCamelCase
) , (
__UpperCamelCase
) , (
__UpperCamelCase
) , (
__UpperCamelCase
) , (
__UpperCamelCase
) , (
__UpperCamelCase
) , (
__UpperCamelCase
) ,
) :List[Any] = config_and_inputs
__UpperCamelCase :Any = {'''input_ids''': input_ids, '''attention_mask''': input_mask}
return config, inputs_dict
@require_torch
class lowerCamelCase_ ( UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , unittest.TestCase ):
'''simple docstring'''
a__ : Union[str, Any] = (LlamaModel, LlamaForCausalLM, LlamaForSequenceClassification) if is_torch_available() else ()
a__ : Union[str, Any] = (LlamaForCausalLM,) if is_torch_available() else ()
a__ : List[Any] = (
{
"""feature-extraction""": LlamaModel,
"""text-classification""": LlamaForSequenceClassification,
"""text-generation""": LlamaForCausalLM,
"""zero-shot""": LlamaForSequenceClassification,
}
if is_torch_available()
else {}
)
a__ : Tuple = False
a__ : Any = False
def UpperCamelCase__ ( self) -> str:
__UpperCamelCase :Dict = LlamaModelTester(self)
__UpperCamelCase :Union[str, Any] = ConfigTester(self , config_class=__lowercase , hidden_size=37)
def UpperCamelCase__ ( self) -> Dict:
self.config_tester.run_common_tests()
def UpperCamelCase__ ( self) -> Tuple:
__UpperCamelCase :int = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*__lowercase)
def UpperCamelCase__ ( self) -> Union[str, Any]:
__UpperCamelCase :List[str] = self.model_tester.prepare_config_and_inputs()
for type in ["absolute", "relative_key", "relative_key_query"]:
__UpperCamelCase :Any = type
self.model_tester.create_and_check_model(*__lowercase)
def UpperCamelCase__ ( self) -> List[str]:
__UpperCamelCase , __UpperCamelCase :Union[str, Any] = self.model_tester.prepare_config_and_inputs_for_common()
__UpperCamelCase :Dict = 3
__UpperCamelCase :int = input_dict['''input_ids''']
__UpperCamelCase :Tuple = input_ids.ne(1).to(__lowercase)
__UpperCamelCase :Any = ids_tensor([self.model_tester.batch_size] , self.model_tester.type_sequence_label_size)
__UpperCamelCase :str = LlamaForSequenceClassification(__lowercase)
model.to(__lowercase)
model.eval()
__UpperCamelCase :str = model(__lowercase , attention_mask=__lowercase , labels=__lowercase)
self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels))
def UpperCamelCase__ ( self) -> int:
__UpperCamelCase , __UpperCamelCase :str = self.model_tester.prepare_config_and_inputs_for_common()
__UpperCamelCase :Optional[int] = 3
__UpperCamelCase :Optional[int] = '''single_label_classification'''
__UpperCamelCase :List[str] = input_dict['''input_ids''']
__UpperCamelCase :str = input_ids.ne(1).to(__lowercase)
__UpperCamelCase :Optional[int] = ids_tensor([self.model_tester.batch_size] , self.model_tester.type_sequence_label_size)
__UpperCamelCase :Tuple = LlamaForSequenceClassification(__lowercase)
model.to(__lowercase)
model.eval()
__UpperCamelCase :List[Any] = model(__lowercase , attention_mask=__lowercase , labels=__lowercase)
self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels))
def UpperCamelCase__ ( self) -> Dict:
__UpperCamelCase , __UpperCamelCase :Dict = self.model_tester.prepare_config_and_inputs_for_common()
__UpperCamelCase :int = 3
__UpperCamelCase :Dict = '''multi_label_classification'''
__UpperCamelCase :Optional[int] = input_dict['''input_ids''']
__UpperCamelCase :int = input_ids.ne(1).to(__lowercase)
__UpperCamelCase :Optional[int] = ids_tensor(
[self.model_tester.batch_size, config.num_labels] , self.model_tester.type_sequence_label_size).to(torch.float)
__UpperCamelCase :str = LlamaForSequenceClassification(__lowercase)
model.to(__lowercase)
model.eval()
__UpperCamelCase :Any = model(__lowercase , attention_mask=__lowercase , labels=__lowercase)
self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels))
@unittest.skip('''LLaMA buffers include complex numbers, which breaks this test''')
def UpperCamelCase__ ( self) -> Optional[Any]:
pass
@parameterized.expand([('''linear''',), ('''dynamic''',)])
def UpperCamelCase__ ( self , __lowercase) -> Optional[int]:
__UpperCamelCase , __UpperCamelCase :List[str] = self.model_tester.prepare_config_and_inputs_for_common()
__UpperCamelCase :Union[str, Any] = ids_tensor([1, 10] , config.vocab_size)
__UpperCamelCase :List[Any] = ids_tensor([1, int(config.max_position_embeddings * 1.5)] , config.vocab_size)
set_seed(42) # Fixed seed at init time so the two models get the same random weights
__UpperCamelCase :str = LlamaModel(__lowercase)
original_model.to(__lowercase)
original_model.eval()
__UpperCamelCase :Optional[Any] = original_model(__lowercase).last_hidden_state
__UpperCamelCase :Optional[int] = original_model(__lowercase).last_hidden_state
set_seed(42) # Fixed seed at init time so the two models get the same random weights
__UpperCamelCase :List[str] = {'''type''': scaling_type, '''factor''': 10.0}
__UpperCamelCase :int = LlamaModel(__lowercase)
scaled_model.to(__lowercase)
scaled_model.eval()
__UpperCamelCase :List[str] = scaled_model(__lowercase).last_hidden_state
__UpperCamelCase :List[Any] = scaled_model(__lowercase).last_hidden_state
# Dynamic scaling does not change the RoPE embeddings until it receives an input longer than the original
# maximum sequence length, so the outputs for the short input should match.
if scaling_type == "dynamic":
self.assertTrue(torch.allclose(__lowercase , __lowercase , atol=1E-5))
else:
self.assertFalse(torch.allclose(__lowercase , __lowercase , atol=1E-5))
# The output should be different for long inputs
self.assertFalse(torch.allclose(__lowercase , __lowercase , atol=1E-5))
@require_torch
class lowerCamelCase_ ( unittest.TestCase ):
'''simple docstring'''
@unittest.skip('''Logits are not exactly the same, once we fix the instabalities somehow, will update!''')
@slow
def UpperCamelCase__ ( self) -> List[str]:
__UpperCamelCase :Dict = [1, 306, 4_658, 278, 6_593, 310, 2_834, 338]
__UpperCamelCase :Optional[Any] = LlamaForCausalLM.from_pretrained('''meta-llama/Llama-2-7b-hf''' , device_map='''auto''')
__UpperCamelCase :Tuple = model(torch.tensor([input_ids]))
# Expected mean on dim = -1
__UpperCamelCase :Any = torch.tensor([[-6.65_50, -4.12_27, -4.98_59, -3.24_06, 0.82_62, -3.00_33, 1.29_64, -3.36_99]])
torch.testing.assert_close(out.mean(-1) , __lowercase , atol=1E-2 , rtol=1E-2)
# slicing logits[0, 0, 0:30]
# fmt: off
__UpperCamelCase :Any = torch.tensor([-12.82_81, -7.44_53, -0.46_39, -8.06_25, -7.25_00, -8.00_00, -6.48_83, -7.76_95, -7.84_38, -7.03_12, -6.21_88, -7.13_28, -1.84_96, 1.99_61, -8.62_50, -6.72_27, -12.82_81, -6.94_92, -7.07_42, -7.78_52, -7.58_20, -7.90_62, -6.93_75, -7.98_05, -8.34_38, -8.15_62, -8.04_69, -7.62_50, -7.74_22, -7.33_98,])
# fmt: on
torch.testing.assert_close(out[0, 0, :30] , __lowercase , atol=1E-5 , rtol=1E-5)
@unittest.skip('''Logits are not exactly the same, once we fix the instabalities somehow, will update!''')
@slow
def UpperCamelCase__ ( self) -> List[Any]:
__UpperCamelCase :str = [1, 306, 4_658, 278, 6_593, 310, 2_834, 338]
__UpperCamelCase :Any = LlamaForCausalLM.from_pretrained('''meta-llama/Llama-2-13b-hf''' , device_map='''auto''')
__UpperCamelCase :Optional[int] = model(torch.tensor(__lowercase))
# Expected mean on dim = -1
__UpperCamelCase :Any = torch.tensor([[-2.06_22, -1.27_94, -1.16_38, -0.97_88, -1.46_03, -1.02_38, -1.78_93, -1.44_11]])
torch.testing.assert_close(out.mean(-1) , __lowercase , atol=1E-2 , rtol=1E-2)
# slicing logits[0, 0, 0:30]
# fmt: off
__UpperCamelCase :Union[str, Any] = torch.tensor([-8.14_06, -8.05_47, 2.74_61, -1.23_44, -0.14_48, -1.82_62, -1.00_20, -1.81_54, -1.68_95, -1.85_16, -2.35_74, -0.92_77, 3.75_98, 6.57_42, -1.29_98, -0.11_77, -8.14_06, -2.96_88, -2.91_99, -3.16_99, -3.52_54, -2.35_55, -2.79_88, -3.41_41, -2.82_62, -4.51_95, -3.33_79, -3.31_64, -2.78_32, -3.02_73])
# fmt: on
torch.testing.assert_close(out[0, 0, :30] , __lowercase , atol=1E-5 , rtol=1E-5)
@unittest.skip('''Logits are not exactly the same, once we fix the instabalities somehow, will update!''')
@slow
def UpperCamelCase__ ( self) -> int:
__UpperCamelCase :Any = [1, 306, 4_658, 278, 6_593, 310, 2_834, 338]
__UpperCamelCase :Optional[int] = LlamaForCausalLM.from_pretrained('''meta-llama/Llama-2-13b-chat-hf''' , device_map='''auto''')
__UpperCamelCase :str = model(torch.tensor(__lowercase))
# Expected mean on dim = -1
__UpperCamelCase :List[Any] = torch.tensor([[-0.85_62, -1.85_20, -0.75_51, -0.41_62, -1.51_61, -1.20_38, -2.48_23, -2.32_54]])
torch.testing.assert_close(out.mean(-1) , __lowercase , atol=1E-2 , rtol=1E-2)
# slicing logits[0, 0, 0:30]
# fmt: off
__UpperCamelCase :Union[str, Any] = torch.tensor([-2.22_27, 4.88_28, 0.90_23, -0.45_78, -0.78_71, -0.10_33, -0.62_21, -0.57_86, -0.78_03, -1.06_74, -1.29_20, -0.15_70, 0.80_08, 2.07_23, -0.94_97, 0.27_71, -2.22_27, -0.76_12, -1.43_46, -1.20_61, -1.64_26, -0.30_00, -0.71_39, -1.19_34, -1.86_91, -1.69_73, -1.59_47, -1.27_05, -0.35_23, -0.55_13])
# fmt: on
torch.testing.assert_close(out.mean(-1) , __lowercase , atol=1E-2 , rtol=1E-2)
@unittest.skip(
'''Logits are not exactly the same, once we fix the instabalities somehow, will update! Also it is gonna be a `too_slow` test''')
@slow
def UpperCamelCase__ ( self) -> Tuple:
__UpperCamelCase :Any = [1, 306, 4_658, 278, 6_593, 310, 2_834, 338]
__UpperCamelCase :List[Any] = LlamaForCausalLM.from_pretrained('''meta-llama/Llama-2-70b-hf''' , device_map='''auto''')
__UpperCamelCase :List[str] = model(torch.tensor(__lowercase))
__UpperCamelCase :Dict = torch.tensor(
[[-4.23_27, -3.33_60, -4.66_65, -4.76_31, -1.81_80, -3.41_70, -1.42_11, -3.18_10]] , dtype=torch.floataa)
torch.testing.assert_close(out.mean(-1) , __lowercase , atol=1E-2 , rtol=1E-2)
# fmt: off
__UpperCamelCase :str = torch.tensor([-9.49_22, -3.95_51, 1.79_98, -5.67_58, -5.10_55, -5.89_84, -4.83_20, -6.80_86, -6.53_91, -5.61_72, -5.58_20, -5.53_52, 1.78_81, 3.62_89, -6.51_17, -3.47_85, -9.50_00, -6.03_52, -6.81_25, -6.01_95, -6.68_36, -5.47_27, -6.28_12, -6.03_91, -7.33_98, -7.42_97, -7.48_44, -6.58_20, -5.87_89, -5.53_12])
# fmt: on
torch.testing.assert_close(out[0, 0, :30] , __lowercase , atol=1E-5 , rtol=1E-5)
@unittest.skip('''Model is curently gated''')
@slow
def UpperCamelCase__ ( self) -> str:
__UpperCamelCase :List[Any] = '''Simply put, the theory of relativity states that 1) the laws of physics are the same everywhere in the universe and 2) the passage of time and the length of objects can vary depending on the observer\'s frame of reference.\n\nThe first part of the theory, that the laws of physics are the same everywhere, is known as the "princi'''
__UpperCamelCase :Optional[int] = '''Simply put, the theory of relativity states that '''
__UpperCamelCase :Tuple = LlamaTokenizer.from_pretrained('''meta-llama/Llama-2-13b-chat-hf''')
__UpperCamelCase :Dict = tokenizer.encode(__lowercase , return_tensors='''pt''')
__UpperCamelCase :Tuple = LlamaForCausalLM.from_pretrained(
'''meta-llama/Llama-2-13b-chat-hf''' , device_map='''sequential''' , use_safetensors=__lowercase)
# greedy generation outputs
__UpperCamelCase :List[Any] = model.generate(__lowercase , max_new_tokens=64 , top_p=__lowercase , temperature=1 , do_sample=__lowercase)
__UpperCamelCase :str = tokenizer.decode(generated_ids[0] , skip_special_tokens=__lowercase)
self.assertEqual(__lowercase , __lowercase)
| 43
|
from typing import List, Optional, TypeVar
from .arrow_dataset import Dataset, _concatenate_map_style_datasets, _interleave_map_style_datasets
from .dataset_dict import DatasetDict, IterableDatasetDict
from .info import DatasetInfo
from .iterable_dataset import IterableDataset, _concatenate_iterable_datasets, _interleave_iterable_datasets
from .splits import NamedSplit
from .utils import logging
from .utils.py_utils import Literal
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[int] = TypeVar("DatasetType", Dataset, IterableDataset)
def A (__A : List[DatasetType] , __A : Optional[List[float]] = None , __A : Optional[int] = None , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : Literal["first_exhausted", "all_exhausted"] = "first_exhausted" , ) -> DatasetType:
"""simple docstring"""
from .arrow_dataset import Dataset
from .iterable_dataset import IterableDataset
if not datasets:
raise ValueError('''Unable to interleave an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if stopping_strategy not in ["first_exhausted", "all_exhausted"]:
raise ValueError(F"""{stopping_strategy} is not supported. Please enter a valid stopping_strategy.""" )
if dataset_type is Dataset:
return _interleave_map_style_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
else:
return _interleave_iterable_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
def A (__A : List[DatasetType] , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : int = 0 , ) -> DatasetType:
"""simple docstring"""
if not dsets:
raise ValueError('''Unable to concatenate an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if dataset_type is Dataset:
return _concatenate_map_style_datasets(__A , info=__A , split=__A , axis=__A )
else:
return _concatenate_iterable_datasets(__A , info=__A , split=__A , axis=__A )
| 51
| 0
|
"""simple docstring"""
import inspect
import unittest
from transformers import ViTConfig
from transformers.testing_utils import (
require_accelerate,
require_torch,
require_torch_gpu,
require_vision,
slow,
torch_device,
)
from transformers.utils import cached_property, is_torch_available, is_vision_available
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from torch import nn
from transformers import ViTForImageClassification, ViTForMaskedImageModeling, ViTModel
from transformers.models.vit.modeling_vit import VIT_PRETRAINED_MODEL_ARCHIVE_LIST
if is_vision_available():
from PIL import Image
from transformers import ViTImageProcessor
class __A :
def __init__( self , a__ , a__=13 , a__=30 , a__=2 , a__=3 , a__=True , a__=True , a__=32 , a__=5 , a__=4 , a__=37 , a__="gelu" , a__=0.1 , a__=0.1 , a__=10 , a__=0.0_2 , a__=None , a__=2 , ):
_lowerCAmelCase : List[Any] = parent
_lowerCAmelCase : int = batch_size
_lowerCAmelCase : Tuple = image_size
_lowerCAmelCase : Optional[Any] = patch_size
_lowerCAmelCase : int = num_channels
_lowerCAmelCase : List[str] = is_training
_lowerCAmelCase : Tuple = use_labels
_lowerCAmelCase : Any = hidden_size
_lowerCAmelCase : List[Any] = num_hidden_layers
_lowerCAmelCase : List[str] = num_attention_heads
_lowerCAmelCase : Optional[Any] = intermediate_size
_lowerCAmelCase : Any = hidden_act
_lowerCAmelCase : Tuple = hidden_dropout_prob
_lowerCAmelCase : Optional[Any] = attention_probs_dropout_prob
_lowerCAmelCase : Any = type_sequence_label_size
_lowerCAmelCase : int = initializer_range
_lowerCAmelCase : Any = scope
_lowerCAmelCase : Union[str, Any] = encoder_stride
# in ViT, the seq length equals the number of patches + 1 (we add 1 for the [CLS] token)
_lowerCAmelCase : int = (image_size // patch_size) ** 2
_lowerCAmelCase : Optional[Any] = num_patches + 1
def __A ( self ):
_lowerCAmelCase : List[Any] = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] )
_lowerCAmelCase : Optional[int] = None
if self.use_labels:
_lowerCAmelCase : str = ids_tensor([self.batch_size] , self.type_sequence_label_size )
_lowerCAmelCase : Dict = self.get_config()
return config, pixel_values, labels
def __A ( self ):
return ViTConfig(
image_size=self.image_size , patch_size=self.patch_size , num_channels=self.num_channels , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , is_decoder=a__ , initializer_range=self.initializer_range , encoder_stride=self.encoder_stride , )
def __A ( self , a__ , a__ , a__ ):
_lowerCAmelCase : str = ViTModel(config=a__ )
model.to(a__ )
model.eval()
_lowerCAmelCase : Optional[int] = model(a__ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def __A ( self , a__ , a__ , a__ ):
_lowerCAmelCase : Any = ViTForMaskedImageModeling(config=a__ )
model.to(a__ )
model.eval()
_lowerCAmelCase : List[str] = model(a__ )
self.parent.assertEqual(
result.reconstruction.shape , (self.batch_size, self.num_channels, self.image_size, self.image_size) )
# test greyscale images
_lowerCAmelCase : Union[str, Any] = 1
_lowerCAmelCase : Union[str, Any] = ViTForMaskedImageModeling(a__ )
model.to(a__ )
model.eval()
_lowerCAmelCase : Any = floats_tensor([self.batch_size, 1, self.image_size, self.image_size] )
_lowerCAmelCase : Dict = model(a__ )
self.parent.assertEqual(result.reconstruction.shape , (self.batch_size, 1, self.image_size, self.image_size) )
def __A ( self , a__ , a__ , a__ ):
_lowerCAmelCase : Optional[int] = self.type_sequence_label_size
_lowerCAmelCase : Dict = ViTForImageClassification(a__ )
model.to(a__ )
model.eval()
_lowerCAmelCase : int = model(a__ , labels=a__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.type_sequence_label_size) )
# test greyscale images
_lowerCAmelCase : int = 1
_lowerCAmelCase : Dict = ViTForImageClassification(a__ )
model.to(a__ )
model.eval()
_lowerCAmelCase : List[str] = floats_tensor([self.batch_size, 1, self.image_size, self.image_size] )
_lowerCAmelCase : Tuple = model(a__ )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.type_sequence_label_size) )
def __A ( self ):
_lowerCAmelCase : Dict = self.prepare_config_and_inputs()
(
(
_lowerCAmelCase
) , (
_lowerCAmelCase
) , (
_lowerCAmelCase
) ,
) : List[str] = config_and_inputs
_lowerCAmelCase : Optional[Any] = {"""pixel_values""": pixel_values}
return config, inputs_dict
@require_torch
class __A ( SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ , unittest.TestCase ):
_UpperCamelCase : int = (
(
ViTModel,
ViTForImageClassification,
ViTForMaskedImageModeling,
)
if is_torch_available()
else ()
)
_UpperCamelCase : Tuple = (
{"feature-extraction": ViTModel, "image-classification": ViTForImageClassification}
if is_torch_available()
else {}
)
_UpperCamelCase : Dict = True
_UpperCamelCase : str = False
_UpperCamelCase : Optional[Any] = False
_UpperCamelCase : List[str] = False
def __A ( self ):
_lowerCAmelCase : Dict = ViTModelTester(self )
_lowerCAmelCase : Any = ConfigTester(self , config_class=a__ , has_text_modality=a__ , hidden_size=37 )
def __A ( self ):
self.config_tester.run_common_tests()
@unittest.skip(reason="""ViT does not use inputs_embeds""" )
def __A ( self ):
pass
def __A ( self ):
_lowerCAmelCase , _lowerCAmelCase : Optional[int] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
_lowerCAmelCase : List[str] = model_class(a__ )
self.assertIsInstance(model.get_input_embeddings() , (nn.Module) )
_lowerCAmelCase : int = model.get_output_embeddings()
self.assertTrue(x is None or isinstance(a__ , nn.Linear ) )
def __A ( self ):
_lowerCAmelCase , _lowerCAmelCase : List[str] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
_lowerCAmelCase : Optional[Any] = model_class(a__ )
_lowerCAmelCase : str = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
_lowerCAmelCase : List[str] = [*signature.parameters.keys()]
_lowerCAmelCase : List[str] = ["""pixel_values"""]
self.assertListEqual(arg_names[:1] , a__ )
def __A ( self ):
_lowerCAmelCase : Optional[int] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*a__ )
def __A ( self ):
_lowerCAmelCase : Any = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_masked_image_modeling(*a__ )
def __A ( self ):
_lowerCAmelCase : str = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_image_classification(*a__ )
@slow
def __A ( self ):
for model_name in VIT_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
_lowerCAmelCase : List[str] = ViTModel.from_pretrained(a__ )
self.assertIsNotNone(a__ )
def SCREAMING_SNAKE_CASE ( ) -> List[str]:
_lowerCAmelCase : Union[str, Any] = Image.open("""./tests/fixtures/tests_samples/COCO/000000039769.png""" )
return image
@require_torch
@require_vision
class __A ( unittest.TestCase ):
@cached_property
def __A ( self ):
return ViTImageProcessor.from_pretrained("""google/vit-base-patch16-224""" ) if is_vision_available() else None
@slow
def __A ( self ):
_lowerCAmelCase : Dict = ViTForImageClassification.from_pretrained("""google/vit-base-patch16-224""" ).to(a__ )
_lowerCAmelCase : str = self.default_image_processor
_lowerCAmelCase : str = prepare_img()
_lowerCAmelCase : Tuple = image_processor(images=a__ , return_tensors="""pt""" ).to(a__ )
# forward pass
with torch.no_grad():
_lowerCAmelCase : Optional[int] = model(**a__ )
# verify the logits
_lowerCAmelCase : str = torch.Size((1, 1000) )
self.assertEqual(outputs.logits.shape , a__ )
_lowerCAmelCase : List[Any] = torch.tensor([-0.2_7_4_4, 0.8_2_1_5, -0.0_8_3_6] ).to(a__ )
self.assertTrue(torch.allclose(outputs.logits[0, :3] , a__ , atol=1e-4 ) )
@slow
def __A ( self ):
# ViT models have an `interpolate_pos_encoding` argument in their forward method,
# allowing to interpolate the pre-trained position embeddings in order to use
# the model on higher resolutions. The DINO model by Facebook AI leverages this
# to visualize self-attention on higher resolution images.
_lowerCAmelCase : Optional[Any] = ViTModel.from_pretrained("""facebook/dino-vits8""" ).to(a__ )
_lowerCAmelCase : str = ViTImageProcessor.from_pretrained("""facebook/dino-vits8""" , size=480 )
_lowerCAmelCase : Any = prepare_img()
_lowerCAmelCase : Dict = image_processor(images=a__ , return_tensors="""pt""" )
_lowerCAmelCase : Optional[Any] = inputs.pixel_values.to(a__ )
# forward pass
with torch.no_grad():
_lowerCAmelCase : str = model(a__ , interpolate_pos_encoding=a__ )
# verify the logits
_lowerCAmelCase : str = torch.Size((1, 3601, 384) )
self.assertEqual(outputs.last_hidden_state.shape , a__ )
_lowerCAmelCase : Union[str, Any] = torch.tensor(
[[4.2_3_4_0, 4.3_9_0_6, -6.6_6_9_2], [4.5_4_6_3, 1.8_9_2_8, -6.7_2_5_7], [4.4_4_2_9, 0.8_4_9_6, -5.8_5_8_5]] ).to(a__ )
self.assertTrue(torch.allclose(outputs.last_hidden_state[0, :3, :3] , a__ , atol=1e-4 ) )
@slow
@require_accelerate
@require_torch_gpu
def __A ( self ):
_lowerCAmelCase : Optional[int] = ViTModel.from_pretrained("""facebook/dino-vits8""" , torch_dtype=torch.floataa , device_map="""auto""" )
_lowerCAmelCase : Union[str, Any] = self.default_image_processor
_lowerCAmelCase : Dict = prepare_img()
_lowerCAmelCase : List[str] = image_processor(images=a__ , return_tensors="""pt""" )
_lowerCAmelCase : Any = inputs.pixel_values.to(a__ )
# forward pass to make sure inference works in fp16
with torch.no_grad():
_lowerCAmelCase : Optional[int] = model(a__ )
| 44
|
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
snake_case_ : Optional[Any] = "pt"
elif is_tf_available():
snake_case_ : Union[str, Any] = "tf"
else:
snake_case_ : str = "jax"
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = ByTaTokenizer
UpperCAmelCase__ : int = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().setUp()
UpperCAmelCase_ = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname)
@cached_property
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
return ByTaTokenizer.from_pretrained('''google/byt5-small''')
def lowerCamelCase ( self : List[str] , **_snake_case : Union[str, Any]):
"""simple docstring"""
return self.tokenizer_class.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Dict , _snake_case : int , _snake_case : Tuple=False , _snake_case : Dict=20 , _snake_case : Optional[Any]=5):
"""simple docstring"""
UpperCAmelCase_ = []
for i in range(len(_snake_case)):
try:
UpperCAmelCase_ = tokenizer.decode([i] , clean_up_tokenization_spaces=_snake_case)
except UnicodeDecodeError:
pass
toks.append((i, tok))
UpperCAmelCase_ = list(filter(lambda _snake_case: re.match(r'''^[ a-zA-Z]+$''' , t[1]) , _snake_case))
UpperCAmelCase_ = list(filter(lambda _snake_case: [t[0]] == tokenizer.encode(t[1] , add_special_tokens=_snake_case) , _snake_case))
if max_length is not None and len(_snake_case) > max_length:
UpperCAmelCase_ = toks[:max_length]
if min_length is not None and len(_snake_case) < min_length and len(_snake_case) > 0:
while len(_snake_case) < min_length:
UpperCAmelCase_ = toks + toks
# toks_str = [t[1] for t in toks]
UpperCAmelCase_ = [t[0] for t in toks]
# Ensure consistency
UpperCAmelCase_ = tokenizer.decode(_snake_case , clean_up_tokenization_spaces=_snake_case)
if " " not in output_txt and len(_snake_case) > 1:
UpperCAmelCase_ = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=_snake_case)
+ ''' '''
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=_snake_case)
)
if with_prefix_space:
UpperCAmelCase_ = ''' ''' + output_txt
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
return output_txt, output_ids
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = tokenizer(['''hi</s>''', '''I went to the gym</s>''', '''</s>'''])
UpperCAmelCase_ = tokenizer(['''hi''', '''I went to the gym''', ''''''])
self.assertListEqual(batch_with_eos_added['''input_ids'''] , batch_without_eos_added['''input_ids'''])
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = '''Unicode €.'''
UpperCAmelCase_ = tokenizer(_snake_case)
UpperCAmelCase_ = [88, 113, 108, 102, 114, 103, 104, 35, 229, 133, 175, 49, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''Unicode €.</s>''')
UpperCAmelCase_ = tokenizer('''e è é ê ë''')
UpperCAmelCase_ = [104, 35, 198, 171, 35, 198, 172, 35, 198, 173, 35, 198, 174, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''e è é ê ë</s>''')
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('''e è é ê ë''')) , '''e è é ê ë</s>''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 1, 0]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
if FRAMEWORK != "jax":
UpperCAmelCase_ = list(batch.input_ids.numpy()[0])
else:
UpperCAmelCase_ = list(batch.input_ids.tolist()[0])
self.assertListEqual(_snake_case , _snake_case)
self.assertEqual((2, 37) , batch.input_ids.shape)
self.assertEqual((2, 37) , batch.attention_mask.shape)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
# check if input_ids are returned and no decoder_input_ids
self.assertIn('''input_ids''' , _snake_case)
self.assertIn('''attention_mask''' , _snake_case)
self.assertNotIn('''decoder_input_ids''' , _snake_case)
self.assertNotIn('''decoder_attention_mask''' , _snake_case)
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = [
'''Summary of the text.''',
'''Another summary.''',
]
UpperCAmelCase_ = tokenizer(
text_target=_snake_case , max_length=32 , padding='''max_length''' , truncation=_snake_case , return_tensors=_snake_case)
self.assertEqual(32 , targets['''input_ids'''].shape[1])
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization. </s>''']
UpperCAmelCase_ = ['''Summary of the text. </s>''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 35, 1]
UpperCAmelCase_ = [86, 120, 112, 112, 100, 117, 124, 35, 114, 105, 35, 119, 107, 104, 35, 119, 104, 123, 119, 49, 35, 1]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , text_target=_snake_case)
self.assertEqual(_snake_case , batch['''input_ids'''][0])
self.assertEqual(_snake_case , batch['''labels'''][0])
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
self.assertNotEqual(tokenizer.model_max_length , 42)
# Now let's start the test
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
shutil.rmtree(_snake_case)
UpperCAmelCase_ = self.get_tokenizers(model_max_length=42)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
tokenizer.add_tokens(['''bim''', '''bambam'''])
UpperCAmelCase_ = tokenizer.additional_special_tokens
additional_special_tokens.append('''new_additional_special_token''')
tokenizer.add_special_tokens({'''additional_special_tokens''': additional_special_tokens})
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
self.assertIn('''new_additional_special_token''' , after_tokenizer.additional_special_tokens)
self.assertEqual(after_tokenizer.model_max_length , 42)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case , model_max_length=43)
self.assertEqual(tokenizer.model_max_length , 43)
shutil.rmtree(_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
UpperCAmelCase_ = [F"""<extra_id_{i}>""" for i in range(125)]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , )
self.assertIn(
'''an_additional_special_token''' , tokenizer_without_change_in_init.additional_special_tokens)
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['''an_additional_special_token'''] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['''an_additional_special_token'''])) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
UpperCAmelCase_ = added_tokens_extra_ids + [AddedToken('''a_new_additional_special_token''' , lstrip=_snake_case)]
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , additional_special_tokens=_snake_case , )
self.assertIn('''a_new_additional_special_token''' , tokenizer.additional_special_tokens)
self.assertEqual(
['''a_new_additional_special_token'''] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['''a_new_additional_special_token'''])) , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer_class.from_pretrained(_snake_case)
self.assertTrue(tokenizer.decode([255]) == '''''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers(fast=_snake_case , do_lower_case=_snake_case)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = ['''t''', '''h''', '''i''', '''s''', ''' ''', '''i''', '''s''', ''' ''', '''a''', ''' ''', '''t''', '''e''', '''x''', '''t''', '''</s>''']
UpperCAmelCase_ = tokenizer.convert_tokens_to_string(_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = [
'''bos_token''',
'''eos_token''',
'''unk_token''',
'''sep_token''',
'''pad_token''',
'''cls_token''',
'''mask_token''',
]
UpperCAmelCase_ = 0
UpperCAmelCase_ = tokenizer.convert_ids_to_tokens(
_snake_case , skip_special_tokens=_snake_case)
for attr in attributes_list:
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , '''additional_special_tokens_ids''' , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [])
setattr(_snake_case , '''additional_special_tokens_ids''' , [token_id_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [token_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [token_id_to_test_setters])
| 51
| 0
|
"""simple docstring"""
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
lowercase_ = logging.get_logger(__name__)
lowercase_ = {"vocab_file": "spm_char.model"}
lowercase_ = {
"vocab_file": {
"microsoft/speecht5_asr": "https://huggingface.co/microsoft/speecht5_asr/resolve/main/spm_char.model",
"microsoft/speecht5_tts": "https://huggingface.co/microsoft/speecht5_tts/resolve/main/spm_char.model",
"microsoft/speecht5_vc": "https://huggingface.co/microsoft/speecht5_vc/resolve/main/spm_char.model",
}
}
lowercase_ = {
"microsoft/speecht5_asr": 1_0_2_4,
"microsoft/speecht5_tts": 1_0_2_4,
"microsoft/speecht5_vc": 1_0_2_4,
}
class __lowerCAmelCase ( __SCREAMING_SNAKE_CASE ):
'''simple docstring'''
__UpperCAmelCase : int = VOCAB_FILES_NAMES
__UpperCAmelCase : Union[str, Any] = PRETRAINED_VOCAB_FILES_MAP
__UpperCAmelCase : Optional[int] = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__UpperCAmelCase : Tuple = ['input_ids', 'attention_mask']
def __init__( self , _a , _a="<s>" , _a="</s>" , _a="<unk>" , _a="<pad>" , _a = None , **_a , ):
__a = {} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
bos_token=_a , eos_token=_a , unk_token=_a , pad_token=_a , sp_model_kwargs=self.sp_model_kwargs , **_a , )
__a = vocab_file
__a = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(_a )
@property
def __UpperCAmelCase ( self ):
return self.sp_model.get_piece_size()
def __UpperCAmelCase ( self ):
__a = {self.convert_ids_to_tokens(_a ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def __getstate__( self ):
__a = self.__dict__.copy()
__a = None
return state
def __setstate__( self , _a ):
__a = d
# for backward compatibility
if not hasattr(self , '''sp_model_kwargs''' ):
__a = {}
__a = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(self.vocab_file )
def __UpperCAmelCase ( self , _a ):
return self.sp_model.encode(_a , out_type=_a )
def __UpperCAmelCase ( self , _a ):
return self.sp_model.piece_to_id(_a )
def __UpperCAmelCase ( self , _a ):
__a = self.sp_model.IdToPiece(_a )
return token
def __UpperCAmelCase ( self , _a ):
__a = []
__a = ''''''
for token in tokens:
# make sure that special tokens are not decoded using sentencepiece model
if token in self.all_special_tokens:
out_string += self.sp_model.decode(_a ) + token
__a = []
else:
current_sub_tokens.append(_a )
out_string += self.sp_model.decode(_a )
return out_string.strip()
def __UpperCAmelCase ( self , _a , _a=None ):
if token_ids_a is None:
return token_ids_a + [self.eos_token_id]
# We don't expect to process pairs, but leave the pair logic for API consistency
return token_ids_a + token_ids_a + [self.eos_token_id]
def __UpperCAmelCase ( self , _a , _a = None , _a = False ):
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=_a , token_ids_a=_a , already_has_special_tokens=_a )
__a = [1]
if token_ids_a is None:
return ([0] * len(_a )) + suffix_ones
return ([0] * len(_a )) + ([0] * len(_a )) + suffix_ones
def __UpperCAmelCase ( self , _a , _a = None ):
if not os.path.isdir(_a ):
logger.error(f'''Vocabulary path ({save_directory}) should be a directory''' )
return
__a = os.path.join(
_a , (filename_prefix + '''-''' if filename_prefix else '''''') + VOCAB_FILES_NAMES['''vocab_file'''] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(_a ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , _a )
elif not os.path.isfile(self.vocab_file ):
with open(_a , '''wb''' ) as fi:
__a = self.sp_model.serialized_model_proto()
fi.write(_a )
return (out_vocab_file,)
| 45
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_sentencepiece_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : Dict = {"configuration_mbart": ["MBART_PRETRAINED_CONFIG_ARCHIVE_MAP", "MBartConfig", "MBartOnnxConfig"]}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Tuple = ["MBartTokenizer"]
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = ["MBartTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[Any] = [
"MBART_PRETRAINED_MODEL_ARCHIVE_LIST",
"MBartForCausalLM",
"MBartForConditionalGeneration",
"MBartForQuestionAnswering",
"MBartForSequenceClassification",
"MBartModel",
"MBartPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TFMBartForConditionalGeneration",
"TFMBartModel",
"TFMBartPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"FlaxMBartForConditionalGeneration",
"FlaxMBartForQuestionAnswering",
"FlaxMBartForSequenceClassification",
"FlaxMBartModel",
"FlaxMBartPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_mbart import MBART_PRETRAINED_CONFIG_ARCHIVE_MAP, MBartConfig, MBartOnnxConfig
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart import MBartTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart_fast import MBartTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_mbart import (
MBART_PRETRAINED_MODEL_ARCHIVE_LIST,
MBartForCausalLM,
MBartForConditionalGeneration,
MBartForQuestionAnswering,
MBartForSequenceClassification,
MBartModel,
MBartPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_mbart import TFMBartForConditionalGeneration, TFMBartModel, TFMBartPreTrainedModel
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_mbart import (
FlaxMBartForConditionalGeneration,
FlaxMBartForQuestionAnswering,
FlaxMBartForSequenceClassification,
FlaxMBartModel,
FlaxMBartPreTrainedModel,
)
else:
import sys
snake_case_ : Any = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
"""simple docstring"""
import logging
import os
from logging import (
CRITICAL, # NOQA
DEBUG, # NOQA
ERROR, # NOQA
FATAL, # NOQA
INFO, # NOQA
NOTSET, # NOQA
WARN, # NOQA
WARNING, # NOQA
)
from typing import Optional
from tqdm import auto as tqdm_lib
SCREAMING_SNAKE_CASE__ = {
"debug": logging.DEBUG,
"info": logging.INFO,
"warning": logging.WARNING,
"error": logging.ERROR,
"critical": logging.CRITICAL,
}
SCREAMING_SNAKE_CASE__ = logging.WARNING
def UpperCAmelCase__ ( ):
'''simple docstring'''
lowerCAmelCase = os.getenv("""DATASETS_VERBOSITY""" , SCREAMING_SNAKE_CASE )
if env_level_str:
if env_level_str in log_levels:
return log_levels[env_level_str]
else:
logging.getLogger().warning(
F'Unknown option DATASETS_VERBOSITY={env_level_str}, '
F'has to be one of: { ", ".join(log_levels.keys() ) }' )
return _default_log_level
def UpperCAmelCase__ ( ):
'''simple docstring'''
return __name__.split(""".""" )[0]
def UpperCAmelCase__ ( ):
'''simple docstring'''
return logging.getLogger(_get_library_name() )
def UpperCAmelCase__ ( ):
'''simple docstring'''
lowerCAmelCase = _get_library_root_logger()
library_root_logger.setLevel(_get_default_logging_level() )
def UpperCAmelCase__ ( ):
'''simple docstring'''
lowerCAmelCase = _get_library_root_logger()
library_root_logger.setLevel(logging.NOTSET )
def UpperCAmelCase__ ( SCREAMING_SNAKE_CASE : Optional[str] = None ):
'''simple docstring'''
if name is None:
lowerCAmelCase = _get_library_name()
return logging.getLogger(SCREAMING_SNAKE_CASE )
def UpperCAmelCase__ ( ):
'''simple docstring'''
return _get_library_root_logger().getEffectiveLevel()
def UpperCAmelCase__ ( SCREAMING_SNAKE_CASE : int ):
'''simple docstring'''
_get_library_root_logger().setLevel(SCREAMING_SNAKE_CASE )
def UpperCAmelCase__ ( ):
'''simple docstring'''
return set_verbosity(SCREAMING_SNAKE_CASE )
def UpperCAmelCase__ ( ):
'''simple docstring'''
return set_verbosity(SCREAMING_SNAKE_CASE )
def UpperCAmelCase__ ( ):
'''simple docstring'''
return set_verbosity(SCREAMING_SNAKE_CASE )
def UpperCAmelCase__ ( ):
'''simple docstring'''
return set_verbosity(SCREAMING_SNAKE_CASE )
def UpperCAmelCase__ ( ):
'''simple docstring'''
lowerCAmelCase = False
def UpperCAmelCase__ ( ):
'''simple docstring'''
lowerCAmelCase = True
# Configure the library root logger at the module level (singleton-like)
_configure_library_root_logger()
class lowercase :
def __init__( self , *lowercase , **lowercase ) -> List[Any]: # pylint: disable=unused-argument
lowerCAmelCase = args[0] if args else None
def __iter__( self ) -> Dict:
return iter(self._iterator )
def __getattr__( self , lowercase ) -> int:
def empty_fn(*lowercase , **lowercase ): # pylint: disable=unused-argument
return
return empty_fn
def __enter__( self ) -> Tuple:
return self
def __exit__( self , lowercase , lowercase , lowercase ) -> List[str]:
return
SCREAMING_SNAKE_CASE__ = True
class lowercase :
def __call__( self , *lowercase , lowercase=False , **lowercase ) -> Union[str, Any]:
if _tqdm_active and not disable:
return tqdm_lib.tqdm(*lowercase , **lowercase )
else:
return EmptyTqdm(*lowercase , **lowercase )
def _snake_case ( self , *lowercase , **lowercase ) -> Tuple:
lowerCAmelCase = None
if _tqdm_active:
return tqdm_lib.tqdm.set_lock(*lowercase , **lowercase )
def _snake_case ( self ) -> List[Any]:
if _tqdm_active:
return tqdm_lib.tqdm.get_lock()
SCREAMING_SNAKE_CASE__ = _tqdm_cls()
def UpperCAmelCase__ ( ):
'''simple docstring'''
global _tqdm_active
return bool(_tqdm_active )
def UpperCAmelCase__ ( ):
'''simple docstring'''
global _tqdm_active
lowerCAmelCase = True
def UpperCAmelCase__ ( ):
'''simple docstring'''
global _tqdm_active
lowerCAmelCase = False
| 46
|
import warnings
from typing import List, Optional, Union
from ...image_utils import ImageInput
from ...processing_utils import ProcessorMixin
from ...tokenization_utils_base import BatchEncoding, PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy
from ...utils import TensorType
class __snake_case ( a ):
UpperCAmelCase__ : Dict = ['''image_processor''', '''tokenizer''']
UpperCAmelCase__ : Dict = '''FlavaImageProcessor'''
UpperCAmelCase__ : Dict = ('''BertTokenizer''', '''BertTokenizerFast''')
def __init__( self : Union[str, Any] , _snake_case : List[str]=None , _snake_case : str=None , **_snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = None
if "feature_extractor" in kwargs:
warnings.warn(
'''The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor`'''
''' instead.''' , _snake_case , )
UpperCAmelCase_ = kwargs.pop('''feature_extractor''')
UpperCAmelCase_ = image_processor if image_processor is not None else feature_extractor
if image_processor is None:
raise ValueError('''You need to specify an `image_processor`.''')
if tokenizer is None:
raise ValueError('''You need to specify a `tokenizer`.''')
super().__init__(_snake_case , _snake_case)
UpperCAmelCase_ = self.image_processor
def __call__( self : List[Any] , _snake_case : Optional[ImageInput] = None , _snake_case : Optional[Union[TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]]] = None , _snake_case : bool = True , _snake_case : Union[bool, str, PaddingStrategy] = False , _snake_case : Union[bool, str, TruncationStrategy] = False , _snake_case : Optional[int] = None , _snake_case : int = 0 , _snake_case : Optional[int] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = True , _snake_case : Optional[Union[str, TensorType]] = None , **_snake_case : Any , ):
"""simple docstring"""
if text is None and images is None:
raise ValueError('''You have to specify either text or images. Both cannot be none.''')
if text is not None:
UpperCAmelCase_ = self.tokenizer(
text=_snake_case , add_special_tokens=_snake_case , padding=_snake_case , truncation=_snake_case , max_length=_snake_case , stride=_snake_case , pad_to_multiple_of=_snake_case , return_token_type_ids=_snake_case , return_attention_mask=_snake_case , return_overflowing_tokens=_snake_case , return_special_tokens_mask=_snake_case , return_offsets_mapping=_snake_case , return_length=_snake_case , verbose=_snake_case , return_tensors=_snake_case , **_snake_case , )
if images is not None:
UpperCAmelCase_ = self.image_processor(
_snake_case , return_image_mask=_snake_case , return_codebook_pixels=_snake_case , return_tensors=_snake_case , **_snake_case , )
if text is not None and images is not None:
encoding.update(_snake_case)
return encoding
elif text is not None:
return encoding
else:
return BatchEncoding(data=dict(**_snake_case) , tensor_type=_snake_case)
def lowerCamelCase ( self : Any , *_snake_case : Optional[Any] , **_snake_case : int):
"""simple docstring"""
return self.tokenizer.batch_decode(*_snake_case , **_snake_case)
def lowerCamelCase ( self : Optional[int] , *_snake_case : int , **_snake_case : Dict):
"""simple docstring"""
return self.tokenizer.decode(*_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.tokenizer.model_input_names
UpperCAmelCase_ = self.image_processor.model_input_names
return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names))
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
warnings.warn(
'''`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.''' , _snake_case , )
return self.image_processor_class
@property
def lowerCamelCase ( self : Any):
"""simple docstring"""
warnings.warn(
'''`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.''' , _snake_case , )
return self.image_processor
| 51
| 0
|
'''simple docstring'''
import logging
import os
import random
import sys
from dataclasses import dataclass, field
from typing import Optional
import datasets
import evaluate
import numpy as np
from datasets import load_dataset
import transformers
from transformers import (
AutoConfig,
AutoModelForSequenceClassification,
AutoTokenizer,
DataCollatorWithPadding,
EvalPrediction,
HfArgumentParser,
Trainer,
TrainingArguments,
default_data_collator,
set_seed,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import check_min_version, send_example_telemetry
from transformers.utils.versions import require_version
# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.31.0")
require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/text-classification/requirements.txt")
lowerCamelCase : Tuple = logging.getLogger(__name__)
@dataclass
class A__ :
A__ = field(
default=1_28 , metadata={
'help': (
'The maximum total input sequence length after tokenization. Sequences longer '
'than this will be truncated, sequences shorter will be padded.'
)
} , )
A__ = field(
default=A__ , metadata={'help': 'Overwrite the cached preprocessed datasets or not.'} )
A__ = field(
default=A__ , metadata={
'help': (
'Whether to pad all samples to `max_seq_length`. '
'If False, will pad the samples dynamically when batching to the maximum length in the batch.'
)
} , )
A__ = field(
default=A__ , metadata={
'help': (
'For debugging purposes or quicker training, truncate the number of training examples to this '
'value if set.'
)
} , )
A__ = field(
default=A__ , metadata={
'help': (
'For debugging purposes or quicker training, truncate the number of evaluation examples to this '
'value if set.'
)
} , )
A__ = field(
default=A__ , metadata={
'help': (
'For debugging purposes or quicker training, truncate the number of prediction examples to this '
'value if set.'
)
} , )
@dataclass
class A__ :
A__ = field(
default=A__ , metadata={'help': 'Path to pretrained model or model identifier from huggingface.co/models'} )
A__ = field(
default=A__ , metadata={'help': 'Evaluation language. Also train language if `train_language` is set to None.'} )
A__ = field(
default=A__ , metadata={'help': 'Train language if it is different from the evaluation language.'} )
A__ = field(
default=A__ , metadata={'help': 'Pretrained config name or path if not the same as model_name'} )
A__ = field(
default=A__ , metadata={'help': 'Pretrained tokenizer name or path if not the same as model_name'} )
A__ = field(
default=A__ , metadata={'help': 'Where do you want to store the pretrained models downloaded from huggingface.co'} , )
A__ = field(
default=A__ , metadata={'help': 'arg to indicate if tokenizer should do lower case in AutoTokenizer.from_pretrained()'} , )
A__ = field(
default=A__ , metadata={'help': 'Whether to use one of the fast tokenizer (backed by the tokenizers library) or not.'} , )
A__ = field(
default='main' , metadata={'help': 'The specific model version to use (can be a branch name, tag name or commit id).'} , )
A__ = field(
default=A__ , metadata={
'help': (
'Will use the token generated when running `huggingface-cli login` (necessary to use this script '
'with private models).'
)
} , )
A__ = field(
default=A__ , metadata={'help': 'Will enable to load a pretrained model whose head dimensions are different.'} , )
def _lowerCAmelCase ( ) -> Optional[int]:
"""simple docstring"""
_SCREAMING_SNAKE_CASE =HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) )
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE =parser.parse_args_into_dataclasses()
# Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The
# information sent is the one passed as arguments along with your Python/PyTorch versions.
send_example_telemetry('run_xnli' , _UpperCamelCase )
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s' , datefmt='%m/%d/%Y %H:%M:%S' , handlers=[logging.StreamHandler(sys.stdout )] , )
if training_args.should_log:
# The default of training_args.log_level is passive, so we set log level at info here to have that default.
transformers.utils.logging.set_verbosity_info()
_SCREAMING_SNAKE_CASE =training_args.get_process_log_level()
logger.setLevel(_UpperCamelCase )
datasets.utils.logging.set_verbosity(_UpperCamelCase )
transformers.utils.logging.set_verbosity(_UpperCamelCase )
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
+ f"distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}" )
logger.info(f"Training/evaluation parameters {training_args}" )
# Detecting last checkpoint.
_SCREAMING_SNAKE_CASE =None
if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir:
_SCREAMING_SNAKE_CASE =get_last_checkpoint(training_args.output_dir )
if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0:
raise ValueError(
f"Output directory ({training_args.output_dir}) already exists and is not empty. "
'Use --overwrite_output_dir to overcome.' )
elif last_checkpoint is not None:
logger.info(
f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change "
'the `--output_dir` or add `--overwrite_output_dir` to train from scratch.' )
# Set seed before initializing model.
set_seed(training_args.seed )
# In distributed training, the load_dataset function guarantees that only one local process can concurrently
# download the dataset.
# Downloading and loading xnli dataset from the hub.
if training_args.do_train:
if model_args.train_language is None:
_SCREAMING_SNAKE_CASE =load_dataset(
'xnli' , model_args.language , split='train' , cache_dir=model_args.cache_dir , use_auth_token=True if model_args.use_auth_token else None , )
else:
_SCREAMING_SNAKE_CASE =load_dataset(
'xnli' , model_args.train_language , split='train' , cache_dir=model_args.cache_dir , use_auth_token=True if model_args.use_auth_token else None , )
_SCREAMING_SNAKE_CASE =train_dataset.features['label'].names
if training_args.do_eval:
_SCREAMING_SNAKE_CASE =load_dataset(
'xnli' , model_args.language , split='validation' , cache_dir=model_args.cache_dir , use_auth_token=True if model_args.use_auth_token else None , )
_SCREAMING_SNAKE_CASE =eval_dataset.features['label'].names
if training_args.do_predict:
_SCREAMING_SNAKE_CASE =load_dataset(
'xnli' , model_args.language , split='test' , cache_dir=model_args.cache_dir , use_auth_token=True if model_args.use_auth_token else None , )
_SCREAMING_SNAKE_CASE =predict_dataset.features['label'].names
# Labels
_SCREAMING_SNAKE_CASE =len(_UpperCamelCase )
# Load pretrained model and tokenizer
# In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
_SCREAMING_SNAKE_CASE =AutoConfig.from_pretrained(
model_args.config_name if model_args.config_name else model_args.model_name_or_path , num_labels=_UpperCamelCase , idalabel={str(_UpperCamelCase ): label for i, label in enumerate(_UpperCamelCase )} , labelaid={label: i for i, label in enumerate(_UpperCamelCase )} , finetuning_task='xnli' , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
_SCREAMING_SNAKE_CASE =AutoTokenizer.from_pretrained(
model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path , do_lower_case=model_args.do_lower_case , cache_dir=model_args.cache_dir , use_fast=model_args.use_fast_tokenizer , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
_SCREAMING_SNAKE_CASE =AutoModelForSequenceClassification.from_pretrained(
model_args.model_name_or_path , from_tf=bool('.ckpt' in model_args.model_name_or_path ) , config=_UpperCamelCase , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , ignore_mismatched_sizes=model_args.ignore_mismatched_sizes , )
# Preprocessing the datasets
# Padding strategy
if data_args.pad_to_max_length:
_SCREAMING_SNAKE_CASE ='max_length'
else:
# We will pad later, dynamically at batch creation, to the max sequence length in each batch
_SCREAMING_SNAKE_CASE =False
def preprocess_function(_UpperCamelCase : str ):
# Tokenize the texts
return tokenizer(
examples['premise'] , examples['hypothesis'] , padding=_UpperCamelCase , max_length=data_args.max_seq_length , truncation=_UpperCamelCase , )
if training_args.do_train:
if data_args.max_train_samples is not None:
_SCREAMING_SNAKE_CASE =min(len(_UpperCamelCase ) , data_args.max_train_samples )
_SCREAMING_SNAKE_CASE =train_dataset.select(range(_UpperCamelCase ) )
with training_args.main_process_first(desc='train dataset map pre-processing' ):
_SCREAMING_SNAKE_CASE =train_dataset.map(
_UpperCamelCase , batched=_UpperCamelCase , load_from_cache_file=not data_args.overwrite_cache , desc='Running tokenizer on train dataset' , )
# Log a few random samples from the training set:
for index in random.sample(range(len(_UpperCamelCase ) ) , 3 ):
logger.info(f"Sample {index} of the training set: {train_dataset[index]}." )
if training_args.do_eval:
if data_args.max_eval_samples is not None:
_SCREAMING_SNAKE_CASE =min(len(_UpperCamelCase ) , data_args.max_eval_samples )
_SCREAMING_SNAKE_CASE =eval_dataset.select(range(_UpperCamelCase ) )
with training_args.main_process_first(desc='validation dataset map pre-processing' ):
_SCREAMING_SNAKE_CASE =eval_dataset.map(
_UpperCamelCase , batched=_UpperCamelCase , load_from_cache_file=not data_args.overwrite_cache , desc='Running tokenizer on validation dataset' , )
if training_args.do_predict:
if data_args.max_predict_samples is not None:
_SCREAMING_SNAKE_CASE =min(len(_UpperCamelCase ) , data_args.max_predict_samples )
_SCREAMING_SNAKE_CASE =predict_dataset.select(range(_UpperCamelCase ) )
with training_args.main_process_first(desc='prediction dataset map pre-processing' ):
_SCREAMING_SNAKE_CASE =predict_dataset.map(
_UpperCamelCase , batched=_UpperCamelCase , load_from_cache_file=not data_args.overwrite_cache , desc='Running tokenizer on prediction dataset' , )
# Get the metric function
_SCREAMING_SNAKE_CASE =evaluate.load('xnli' )
# You can define your custom compute_metrics function. It takes an `EvalPrediction` object (a namedtuple with a
# predictions and label_ids field) and has to return a dictionary string to float.
def compute_metrics(_UpperCamelCase : EvalPrediction ):
_SCREAMING_SNAKE_CASE =p.predictions[0] if isinstance(p.predictions , _UpperCamelCase ) else p.predictions
_SCREAMING_SNAKE_CASE =np.argmax(_UpperCamelCase , axis=1 )
return metric.compute(predictions=_UpperCamelCase , references=p.label_ids )
# Data collator will default to DataCollatorWithPadding, so we change it if we already did the padding.
if data_args.pad_to_max_length:
_SCREAMING_SNAKE_CASE =default_data_collator
elif training_args.fpaa:
_SCREAMING_SNAKE_CASE =DataCollatorWithPadding(_UpperCamelCase , pad_to_multiple_of=8 )
else:
_SCREAMING_SNAKE_CASE =None
# Initialize our Trainer
_SCREAMING_SNAKE_CASE =Trainer(
model=_UpperCamelCase , args=_UpperCamelCase , train_dataset=train_dataset if training_args.do_train else None , eval_dataset=eval_dataset if training_args.do_eval else None , compute_metrics=_UpperCamelCase , tokenizer=_UpperCamelCase , data_collator=_UpperCamelCase , )
# Training
if training_args.do_train:
_SCREAMING_SNAKE_CASE =None
if training_args.resume_from_checkpoint is not None:
_SCREAMING_SNAKE_CASE =training_args.resume_from_checkpoint
elif last_checkpoint is not None:
_SCREAMING_SNAKE_CASE =last_checkpoint
_SCREAMING_SNAKE_CASE =trainer.train(resume_from_checkpoint=_UpperCamelCase )
_SCREAMING_SNAKE_CASE =train_result.metrics
_SCREAMING_SNAKE_CASE =(
data_args.max_train_samples if data_args.max_train_samples is not None else len(_UpperCamelCase )
)
_SCREAMING_SNAKE_CASE =min(_UpperCamelCase , len(_UpperCamelCase ) )
trainer.save_model() # Saves the tokenizer too for easy upload
trainer.log_metrics('train' , _UpperCamelCase )
trainer.save_metrics('train' , _UpperCamelCase )
trainer.save_state()
# Evaluation
if training_args.do_eval:
logger.info('*** Evaluate ***' )
_SCREAMING_SNAKE_CASE =trainer.evaluate(eval_dataset=_UpperCamelCase )
_SCREAMING_SNAKE_CASE =data_args.max_eval_samples if data_args.max_eval_samples is not None else len(_UpperCamelCase )
_SCREAMING_SNAKE_CASE =min(_UpperCamelCase , len(_UpperCamelCase ) )
trainer.log_metrics('eval' , _UpperCamelCase )
trainer.save_metrics('eval' , _UpperCamelCase )
# Prediction
if training_args.do_predict:
logger.info('*** Predict ***' )
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE =trainer.predict(_UpperCamelCase , metric_key_prefix='predict' )
_SCREAMING_SNAKE_CASE =(
data_args.max_predict_samples if data_args.max_predict_samples is not None else len(_UpperCamelCase )
)
_SCREAMING_SNAKE_CASE =min(_UpperCamelCase , len(_UpperCamelCase ) )
trainer.log_metrics('predict' , _UpperCamelCase )
trainer.save_metrics('predict' , _UpperCamelCase )
_SCREAMING_SNAKE_CASE =np.argmax(_UpperCamelCase , axis=1 )
_SCREAMING_SNAKE_CASE =os.path.join(training_args.output_dir , 'predictions.txt' )
if trainer.is_world_process_zero():
with open(_UpperCamelCase , 'w' ) as writer:
writer.write('index\tprediction\n' )
for index, item in enumerate(_UpperCamelCase ):
_SCREAMING_SNAKE_CASE =label_list[item]
writer.write(f"{index}\t{item}\n" )
if __name__ == "__main__":
main()
| 47
|
from diffusers.utils.testing_utils import require_onnxruntime
@require_onnxruntime
class __snake_case :
pass
| 51
| 0
|
import warnings
from ...utils import logging
from .image_processing_clip import CLIPImageProcessor
SCREAMING_SNAKE_CASE__ : List[str] = logging.get_logger(__name__)
class UpperCamelCase__ (lowerCAmelCase__ ):
'''simple docstring'''
def __init__( self , *UpperCamelCase__ , **UpperCamelCase__ ) -> None:
warnings.warn(
"The class CLIPFeatureExtractor is deprecated and will be removed in version 5 of Transformers. Please"
" use CLIPImageProcessor instead." , UpperCamelCase__ , )
super().__init__(*UpperCamelCase__ , **UpperCamelCase__ )
| 48
|
import argparse
import os
import pickle
import sys
import torch
from transformers import TransfoXLConfig, TransfoXLLMHeadModel, load_tf_weights_in_transfo_xl
from transformers.models.transfo_xl import tokenization_transfo_xl as data_utils
from transformers.models.transfo_xl.tokenization_transfo_xl import CORPUS_NAME, VOCAB_FILES_NAMES
from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging
logging.set_verbosity_info()
# We do this to be able to load python 2 datasets pickles
# See e.g. https://stackoverflow.com/questions/2121874/python-pickling-after-changing-a-modules-directory/2121918#2121918
snake_case_ : List[Any] = data_utils.TransfoXLTokenizer
snake_case_ : int = data_utils.TransfoXLCorpus
snake_case_ : List[Any] = data_utils
snake_case_ : int = data_utils
def A (__A : Dict , __A : List[Any] , __A : Union[str, Any] , __A : Tuple ) -> Union[str, Any]:
"""simple docstring"""
if transfo_xl_dataset_file:
# Convert a pre-processed corpus (see original TensorFlow repo)
with open(__A , '''rb''' ) as fp:
UpperCAmelCase_ = pickle.load(__A , encoding='''latin1''' )
# Save vocabulary and dataset cache as Dictionaries (should be better than pickles for the long-term)
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + VOCAB_FILES_NAMES['''pretrained_vocab_file''']
print(F"""Save vocabulary to {pytorch_vocab_dump_path}""" )
UpperCAmelCase_ = corpus.vocab.__dict__
torch.save(__A , __A )
UpperCAmelCase_ = corpus.__dict__
corpus_dict_no_vocab.pop('''vocab''' , __A )
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + CORPUS_NAME
print(F"""Save dataset to {pytorch_dataset_dump_path}""" )
torch.save(__A , __A )
if tf_checkpoint_path:
# Convert a pre-trained TensorFlow model
UpperCAmelCase_ = os.path.abspath(__A )
UpperCAmelCase_ = os.path.abspath(__A )
print(F"""Converting Transformer XL checkpoint from {tf_path} with config at {config_path}.""" )
# Initialise PyTorch model
if transfo_xl_config_file == "":
UpperCAmelCase_ = TransfoXLConfig()
else:
UpperCAmelCase_ = TransfoXLConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = TransfoXLLMHeadModel(__A )
UpperCAmelCase_ = load_tf_weights_in_transfo_xl(__A , __A , __A )
# Save pytorch-model
UpperCAmelCase_ = os.path.join(__A , __A )
UpperCAmelCase_ = os.path.join(__A , __A )
print(F"""Save PyTorch model to {os.path.abspath(__A )}""" )
torch.save(model.state_dict() , __A )
print(F"""Save configuration file to {os.path.abspath(__A )}""" )
with open(__A , '''w''' , encoding='''utf-8''' ) as f:
f.write(config.to_json_string() )
if __name__ == "__main__":
snake_case_ : List[str] = argparse.ArgumentParser()
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
type=str,
required=True,
help="Path to the folder to store the PyTorch model or dataset/vocab.",
)
parser.add_argument(
"--tf_checkpoint_path",
default="",
type=str,
help="An optional path to a TensorFlow checkpoint path to be converted.",
)
parser.add_argument(
"--transfo_xl_config_file",
default="",
type=str,
help=(
"An optional config json file corresponding to the pre-trained BERT model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--transfo_xl_dataset_file",
default="",
type=str,
help="An optional dataset file to be converted in a vocabulary.",
)
snake_case_ : int = parser.parse_args()
convert_transfo_xl_checkpoint_to_pytorch(
args.tf_checkpoint_path,
args.transfo_xl_config_file,
args.pytorch_dump_folder_path,
args.transfo_xl_dataset_file,
)
| 51
| 0
|
import logging
import random
import ray
from transformers import RagConfig, RagRetriever, RagTokenizer
from transformers.models.rag.retrieval_rag import CustomHFIndex
__snake_case :List[Any] = logging.getLogger(__name__)
class _A :
def __init__( self : List[str]):
'''simple docstring'''
__a = False
def _lowerCamelCase ( self : Any , __SCREAMING_SNAKE_CASE : Optional[Any] , __SCREAMING_SNAKE_CASE : Optional[Any] , __SCREAMING_SNAKE_CASE : List[str] , __SCREAMING_SNAKE_CASE : Union[str, Any]):
'''simple docstring'''
if not self.initialized:
__a = RagRetriever(
__SCREAMING_SNAKE_CASE , question_encoder_tokenizer=__SCREAMING_SNAKE_CASE , generator_tokenizer=__SCREAMING_SNAKE_CASE , index=__SCREAMING_SNAKE_CASE , init_retrieval=__SCREAMING_SNAKE_CASE , )
__a = True
def _lowerCamelCase ( self : List[str]):
'''simple docstring'''
self.retriever.index.init_index()
def _lowerCamelCase ( self : Tuple , __SCREAMING_SNAKE_CASE : Union[str, Any] , __SCREAMING_SNAKE_CASE : Union[str, Any]):
'''simple docstring'''
__a , __a = self.retriever._main_retrieve(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE)
return doc_ids, retrieved_doc_embeds
class _A ( __UpperCAmelCase ):
def __init__( self : Optional[Any] , __SCREAMING_SNAKE_CASE : List[Any] , __SCREAMING_SNAKE_CASE : List[str] , __SCREAMING_SNAKE_CASE : List[str] , __SCREAMING_SNAKE_CASE : int , __SCREAMING_SNAKE_CASE : Union[str, Any]=None):
'''simple docstring'''
if index is not None and index.is_initialized() and len(__SCREAMING_SNAKE_CASE) > 0:
raise ValueError(
'''When using Ray for distributed fine-tuning, '''
'''you\'ll need to provide the paths instead, '''
'''as the dataset and the index are loaded '''
'''separately. More info in examples/rag/use_own_knowledge_dataset.py ''')
super().__init__(
__SCREAMING_SNAKE_CASE , question_encoder_tokenizer=__SCREAMING_SNAKE_CASE , generator_tokenizer=__SCREAMING_SNAKE_CASE , index=__SCREAMING_SNAKE_CASE , init_retrieval=__SCREAMING_SNAKE_CASE , )
__a = retrieval_workers
if len(self.retrieval_workers) > 0:
ray.get(
[
worker.create_rag_retriever.remote(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE)
for worker in self.retrieval_workers
])
def _lowerCamelCase ( self : List[Any]):
'''simple docstring'''
logger.info('''initializing retrieval''')
if len(self.retrieval_workers) > 0:
ray.get([worker.init_retrieval.remote() for worker in self.retrieval_workers])
else:
# Non-distributed training. Load index into this same process.
self.index.init_index()
def _lowerCamelCase ( self : Dict , __SCREAMING_SNAKE_CASE : Optional[int] , __SCREAMING_SNAKE_CASE : List[Any]):
'''simple docstring'''
if len(self.retrieval_workers) > 0:
# Select a random retrieval actor.
__a = self.retrieval_workers[random.randint(0 , len(self.retrieval_workers) - 1)]
__a , __a = ray.get(random_worker.retrieve.remote(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE))
else:
__a , __a = self._main_retrieve(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE)
return retrieved_doc_embeds, doc_ids, self.index.get_doc_dicts(__SCREAMING_SNAKE_CASE)
@classmethod
def _lowerCamelCase ( cls : Any , __SCREAMING_SNAKE_CASE : Tuple , __SCREAMING_SNAKE_CASE : Tuple=None , **__SCREAMING_SNAKE_CASE : Optional[int]):
'''simple docstring'''
return super(__SCREAMING_SNAKE_CASE , cls).get_tokenizers(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , **__SCREAMING_SNAKE_CASE)
@classmethod
def _lowerCamelCase ( cls : Tuple , __SCREAMING_SNAKE_CASE : Optional[int] , __SCREAMING_SNAKE_CASE : str , __SCREAMING_SNAKE_CASE : str=None , **__SCREAMING_SNAKE_CASE : List[Any]):
'''simple docstring'''
__a = kwargs.pop('''config''' , __SCREAMING_SNAKE_CASE) or RagConfig.from_pretrained(__SCREAMING_SNAKE_CASE , **__SCREAMING_SNAKE_CASE)
__a = RagTokenizer.from_pretrained(__SCREAMING_SNAKE_CASE , config=__SCREAMING_SNAKE_CASE)
__a = rag_tokenizer.question_encoder
__a = rag_tokenizer.generator
if indexed_dataset is not None:
__a = '''custom'''
__a = CustomHFIndex(config.retrieval_vector_size , __SCREAMING_SNAKE_CASE)
else:
__a = cls._build_index(__SCREAMING_SNAKE_CASE)
return cls(
__SCREAMING_SNAKE_CASE , question_encoder_tokenizer=__SCREAMING_SNAKE_CASE , generator_tokenizer=__SCREAMING_SNAKE_CASE , retrieval_workers=__SCREAMING_SNAKE_CASE , index=__SCREAMING_SNAKE_CASE , )
| 49
|
from typing import Optional, Tuple, Union
import torch
from einops import rearrange, reduce
from diffusers import DDIMScheduler, DDPMScheduler, DiffusionPipeline, ImagePipelineOutput, UNetaDConditionModel
from diffusers.schedulers.scheduling_ddim import DDIMSchedulerOutput
from diffusers.schedulers.scheduling_ddpm import DDPMSchedulerOutput
snake_case_ : List[str] = 8
def A (__A : Union[str, Any] , __A : List[Any]=BITS ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x * 255).int().clamp(0 , 255 )
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b c h w -> b c 1 h w''' )
UpperCAmelCase_ = ((x & mask) != 0).float()
UpperCAmelCase_ = rearrange(__A , '''b c d h w -> b (c d) h w''' )
UpperCAmelCase_ = bits * 2 - 1
return bits
def A (__A : Dict , __A : Tuple=BITS ) -> List[str]:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x > 0).int()
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A , dtype=torch.intaa )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b (c d) h w -> b c d h w''' , d=8 )
UpperCAmelCase_ = reduce(x * mask , '''b c d h w -> b c h w''' , '''sum''' )
return (dec / 255).clamp(0.0 , 1.0 )
def A (self : List[Any] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : float = 0.0 , __A : bool = True , __A : Tuple=None , __A : bool = True , ) -> Union[DDIMSchedulerOutput, Tuple]:
"""simple docstring"""
if self.num_inference_steps is None:
raise ValueError(
'''Number of inference steps is \'None\', you need to run \'set_timesteps\' after creating the scheduler''' )
# See formulas (12) and (16) of DDIM paper https://arxiv.org/pdf/2010.02502.pdf
# Ideally, read DDIM paper in-detail understanding
# Notation (<variable name> -> <name in paper>
# - pred_noise_t -> e_theta(x_t, t)
# - pred_original_sample -> f_theta(x_t, t) or x_0
# - std_dev_t -> sigma_t
# - eta -> η
# - pred_sample_direction -> "direction pointing to x_t"
# - pred_prev_sample -> "x_t-1"
# 1. get previous step value (=t-1)
UpperCAmelCase_ = timestep - self.config.num_train_timesteps // self.num_inference_steps
# 2. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[timestep]
UpperCAmelCase_ = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
UpperCAmelCase_ = 1 - alpha_prod_t
# 3. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
# 4. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 5. compute variance: "sigma_t(η)" -> see formula (16)
# σ_t = sqrt((1 − α_t−1)/(1 − α_t)) * sqrt(1 − α_t/α_t−1)
UpperCAmelCase_ = self._get_variance(__A , __A )
UpperCAmelCase_ = eta * variance ** 0.5
if use_clipped_model_output:
# the model_output is always re-derived from the clipped x_0 in Glide
UpperCAmelCase_ = (sample - alpha_prod_t ** 0.5 * pred_original_sample) / beta_prod_t ** 0.5
# 6. compute "direction pointing to x_t" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (1 - alpha_prod_t_prev - std_dev_t**2) ** 0.5 * model_output
# 7. compute x_t without "random noise" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = alpha_prod_t_prev ** 0.5 * pred_original_sample + pred_sample_direction
if eta > 0:
# randn_like does not support generator https://github.com/pytorch/pytorch/issues/27072
UpperCAmelCase_ = model_output.device if torch.is_tensor(__A ) else '''cpu'''
UpperCAmelCase_ = torch.randn(model_output.shape , dtype=model_output.dtype , generator=__A ).to(__A )
UpperCAmelCase_ = self._get_variance(__A , __A ) ** 0.5 * eta * noise
UpperCAmelCase_ = prev_sample + variance
if not return_dict:
return (prev_sample,)
return DDIMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
def A (self : Optional[int] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : int="epsilon" , __A : Optional[Any]=None , __A : bool = True , ) -> Union[DDPMSchedulerOutput, Tuple]:
"""simple docstring"""
UpperCAmelCase_ = timestep
if model_output.shape[1] == sample.shape[1] * 2 and self.variance_type in ["learned", "learned_range"]:
UpperCAmelCase_ , UpperCAmelCase_ = torch.split(__A , sample.shape[1] , dim=1 )
else:
UpperCAmelCase_ = None
# 1. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[t]
UpperCAmelCase_ = self.alphas_cumprod[t - 1] if t > 0 else self.one
UpperCAmelCase_ = 1 - alpha_prod_t
UpperCAmelCase_ = 1 - alpha_prod_t_prev
# 2. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
if prediction_type == "epsilon":
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
elif prediction_type == "sample":
UpperCAmelCase_ = model_output
else:
raise ValueError(F"""Unsupported prediction_type {prediction_type}.""" )
# 3. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = (alpha_prod_t_prev ** 0.5 * self.betas[t]) / beta_prod_t
UpperCAmelCase_ = self.alphas[t] ** 0.5 * beta_prod_t_prev / beta_prod_t
# 5. Compute predicted previous sample µ_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * sample
# 6. Add noise
UpperCAmelCase_ = 0
if t > 0:
UpperCAmelCase_ = torch.randn(
model_output.size() , dtype=model_output.dtype , layout=model_output.layout , generator=__A ).to(model_output.device )
UpperCAmelCase_ = (self._get_variance(__A , predicted_variance=__A ) ** 0.5) * noise
UpperCAmelCase_ = pred_prev_sample + variance
if not return_dict:
return (pred_prev_sample,)
return DDPMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
class __snake_case ( a ):
def __init__( self : Union[str, Any] , _snake_case : UNetaDConditionModel , _snake_case : Union[DDIMScheduler, DDPMScheduler] , _snake_case : Optional[float] = 1.0 , ):
"""simple docstring"""
super().__init__()
UpperCAmelCase_ = bit_scale
UpperCAmelCase_ = (
ddim_bit_scheduler_step if isinstance(_snake_case , _snake_case) else ddpm_bit_scheduler_step
)
self.register_modules(unet=_snake_case , scheduler=_snake_case)
@torch.no_grad()
def __call__( self : Union[str, Any] , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 50 , _snake_case : Optional[torch.Generator] = None , _snake_case : Optional[int] = 1 , _snake_case : Optional[str] = "pil" , _snake_case : bool = True , **_snake_case : Optional[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = torch.randn(
(batch_size, self.unet.config.in_channels, height, width) , generator=_snake_case , )
UpperCAmelCase_ = decimal_to_bits(_snake_case) * self.bit_scale
UpperCAmelCase_ = latents.to(self.device)
self.scheduler.set_timesteps(_snake_case)
for t in self.progress_bar(self.scheduler.timesteps):
# predict the noise residual
UpperCAmelCase_ = self.unet(_snake_case , _snake_case).sample
# compute the previous noisy sample x_t -> x_t-1
UpperCAmelCase_ = self.scheduler.step(_snake_case , _snake_case , _snake_case).prev_sample
UpperCAmelCase_ = bits_to_decimal(_snake_case)
if output_type == "pil":
UpperCAmelCase_ = self.numpy_to_pil(_snake_case)
if not return_dict:
return (image,)
return ImagePipelineOutput(images=_snake_case)
| 51
| 0
|
import argparse
import evaluate
import torch
from datasets import load_dataset
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
_UpperCAmelCase : str = 16
_UpperCAmelCase : str = 32
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase = 16 ) -> str:
lowerCamelCase__ : int = AutoTokenizer.from_pretrained('bert-base-cased' )
lowerCamelCase__ : Union[str, Any] = load_dataset('glue' , 'mrpc' )
def tokenize_function(_UpperCAmelCase ):
# max_length=None => use the model max length (it's actually the default)
lowerCamelCase__ : int = tokenizer(examples['sentence1'] , examples['sentence2'] , truncation=_UpperCAmelCase , max_length=_UpperCAmelCase )
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
lowerCamelCase__ : Any = datasets.map(
_UpperCAmelCase , batched=_UpperCAmelCase , remove_columns=['idx', 'sentence1', 'sentence2'] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
lowerCamelCase__ : Dict = tokenized_datasets.rename_column('label' , 'labels' )
def collate_fn(_UpperCAmelCase ):
# On TPU it's best to pad everything to the same length or training will be very slow.
lowerCamelCase__ : Dict = 128 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
lowerCamelCase__ : List[Any] = 16
elif accelerator.mixed_precision != "no":
lowerCamelCase__ : Dict = 8
else:
lowerCamelCase__ : Any = None
return tokenizer.pad(
_UpperCAmelCase , padding='longest' , max_length=_UpperCAmelCase , pad_to_multiple_of=_UpperCAmelCase , return_tensors='pt' , )
# Instantiate dataloaders.
lowerCamelCase__ : List[str] = DataLoader(
tokenized_datasets['train'] , shuffle=_UpperCAmelCase , collate_fn=_UpperCAmelCase , batch_size=_UpperCAmelCase , drop_last=_UpperCAmelCase )
lowerCamelCase__ : str = DataLoader(
tokenized_datasets['validation'] , shuffle=_UpperCAmelCase , collate_fn=_UpperCAmelCase , batch_size=_UpperCAmelCase , drop_last=(accelerator.mixed_precision == 'fp8') , )
return train_dataloader, eval_dataloader
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
# Initialize accelerator
lowerCamelCase__ : Union[str, Any] = Accelerator(cpu=args.cpu , mixed_precision=args.mixed_precision )
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
lowerCamelCase__ : int = config['lr']
lowerCamelCase__ : Tuple = int(config['num_epochs'] )
lowerCamelCase__ : Union[str, Any] = int(config['seed'] )
lowerCamelCase__ : List[str] = int(config['batch_size'] )
lowerCamelCase__ : List[str] = evaluate.load('glue' , 'mrpc' )
# If the batch size is too big we use gradient accumulation
lowerCamelCase__ : Dict = 1
if batch_size > MAX_GPU_BATCH_SIZE and accelerator.distributed_type != DistributedType.TPU:
lowerCamelCase__ : Optional[Any] = batch_size // MAX_GPU_BATCH_SIZE
lowerCamelCase__ : Any = MAX_GPU_BATCH_SIZE
set_seed(_UpperCAmelCase )
lowerCamelCase__ , lowerCamelCase__ : Dict = get_dataloaders(_UpperCAmelCase , _UpperCAmelCase )
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
lowerCamelCase__ : Any = AutoModelForSequenceClassification.from_pretrained('bert-base-cased' , return_dict=_UpperCAmelCase )
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
lowerCamelCase__ : Dict = model.to(accelerator.device )
# Instantiate optimizer
lowerCamelCase__ : int = AdamW(params=model.parameters() , lr=_UpperCAmelCase )
# Instantiate scheduler
lowerCamelCase__ : Dict = get_linear_schedule_with_warmup(
optimizer=_UpperCAmelCase , num_warmup_steps=100 , num_training_steps=(len(_UpperCAmelCase ) * num_epochs) // gradient_accumulation_steps , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ : int = accelerator.prepare(
_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
# Now we train the model
for epoch in range(_UpperCAmelCase ):
model.train()
for step, batch in enumerate(_UpperCAmelCase ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
lowerCamelCase__ : Optional[int] = model(**_UpperCAmelCase )
lowerCamelCase__ : Any = outputs.loss
lowerCamelCase__ : Optional[Any] = loss / gradient_accumulation_steps
accelerator.backward(_UpperCAmelCase )
if step % gradient_accumulation_steps == 0:
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
for step, batch in enumerate(_UpperCAmelCase ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
with torch.no_grad():
lowerCamelCase__ : List[Any] = model(**_UpperCAmelCase )
lowerCamelCase__ : Any = outputs.logits.argmax(dim=-1 )
lowerCamelCase__ , lowerCamelCase__ : Optional[int] = accelerator.gather_for_metrics((predictions, batch['labels']) )
metric.add_batch(
predictions=_UpperCAmelCase , references=_UpperCAmelCase , )
lowerCamelCase__ : List[Any] = metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(F"""epoch {epoch}:""" , _UpperCAmelCase )
def SCREAMING_SNAKE_CASE ( ) -> Optional[Any]:
lowerCamelCase__ : str = argparse.ArgumentParser(description='Simple example of training script.' )
parser.add_argument(
'--mixed_precision' , type=_UpperCAmelCase , default=_UpperCAmelCase , choices=['no', 'fp16', 'bf16', 'fp8'] , help='Whether to use mixed precision. Choose'
'between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.'
'and an Nvidia Ampere GPU.' , )
parser.add_argument('--cpu' , action='store_true' , help='If passed, will train on the CPU.' )
lowerCamelCase__ : Tuple = parser.parse_args()
lowerCamelCase__ : Dict = {'lr': 2e-5, 'num_epochs': 3, 'seed': 42, 'batch_size': 16}
training_function(_UpperCAmelCase , _UpperCAmelCase )
if __name__ == "__main__":
main()
| 50
|
snake_case_ : Dict = {
"Pillow": "Pillow",
"accelerate": "accelerate>=0.11.0",
"compel": "compel==0.1.8",
"black": "black~=23.1",
"datasets": "datasets",
"filelock": "filelock",
"flax": "flax>=0.4.1",
"hf-doc-builder": "hf-doc-builder>=0.3.0",
"huggingface-hub": "huggingface-hub>=0.13.2",
"requests-mock": "requests-mock==1.10.0",
"importlib_metadata": "importlib_metadata",
"invisible-watermark": "invisible-watermark",
"isort": "isort>=5.5.4",
"jax": "jax>=0.2.8,!=0.3.2",
"jaxlib": "jaxlib>=0.1.65",
"Jinja2": "Jinja2",
"k-diffusion": "k-diffusion>=0.0.12",
"torchsde": "torchsde",
"note_seq": "note_seq",
"librosa": "librosa",
"numpy": "numpy",
"omegaconf": "omegaconf",
"parameterized": "parameterized",
"protobuf": "protobuf>=3.20.3,<4",
"pytest": "pytest",
"pytest-timeout": "pytest-timeout",
"pytest-xdist": "pytest-xdist",
"ruff": "ruff>=0.0.241",
"safetensors": "safetensors",
"sentencepiece": "sentencepiece>=0.1.91,!=0.1.92",
"scipy": "scipy",
"onnx": "onnx",
"regex": "regex!=2019.12.17",
"requests": "requests",
"tensorboard": "tensorboard",
"torch": "torch>=1.4",
"torchvision": "torchvision",
"transformers": "transformers>=4.25.1",
"urllib3": "urllib3<=2.0.0",
}
| 51
| 0
|
import argparse
import torch
# Step 1. clone https://github.com/microsoft/unilm
# Step 2. git checkout to https://github.com/microsoft/unilm/commit/b94ec76c36f02fb2b0bf0dcb0b8554a2185173cd
# Step 3. cd unilm
# Step 4. ln -s $(realpath wavlm/modules.py) ./ # create simlink
# import classes
from unilm.wavlm.WavLM import WavLM as WavLMOrig
from unilm.wavlm.WavLM import WavLMConfig as WavLMConfigOrig
from transformers import WavLMConfig, WavLMModel, logging
logging.set_verbosity_info()
__lowerCamelCase : List[str] = logging.get_logger(__name__)
__lowerCamelCase : List[Any] = {
"""post_extract_proj""": """feature_projection.projection""",
"""encoder.pos_conv.0""": """encoder.pos_conv_embed.conv""",
"""self_attn.k_proj""": """encoder.layers.*.attention.k_proj""",
"""self_attn.v_proj""": """encoder.layers.*.attention.v_proj""",
"""self_attn.q_proj""": """encoder.layers.*.attention.q_proj""",
"""self_attn.out_proj""": """encoder.layers.*.attention.out_proj""",
"""self_attn.grep_linear""": """encoder.layers.*.attention.gru_rel_pos_linear""",
"""self_attn.relative_attention_bias""": """encoder.layers.*.attention.rel_attn_embed""",
"""self_attn.grep_a""": """encoder.layers.*.attention.gru_rel_pos_const""",
"""self_attn_layer_norm""": """encoder.layers.*.layer_norm""",
"""fc1""": """encoder.layers.*.feed_forward.intermediate_dense""",
"""fc2""": """encoder.layers.*.feed_forward.output_dense""",
"""final_layer_norm""": """encoder.layers.*.final_layer_norm""",
"""encoder.layer_norm""": """encoder.layer_norm""",
"""w2v_model.layer_norm""": """feature_projection.layer_norm""",
"""quantizer.weight_proj""": """quantizer.weight_proj""",
"""quantizer.vars""": """quantizer.codevectors""",
"""project_q""": """project_q""",
"""final_proj""": """project_hid""",
"""w2v_encoder.proj""": """ctc_proj""",
"""mask_emb""": """masked_spec_embed""",
}
__lowerCamelCase : str = [
"""ctc_proj""",
"""quantizer.weight_proj""",
"""quantizer.codevectors""",
"""project_q""",
"""project_hid""",
]
def A_ ( _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase ) -> Dict:
for attribute in key.split("." ):
UpperCamelCase : Dict = getattr(_lowerCAmelCase , _lowerCAmelCase )
if weight_type is not None:
UpperCamelCase : Union[str, Any] = getattr(_lowerCAmelCase , _lowerCAmelCase ).shape
else:
UpperCamelCase : Tuple = hf_pointer.shape
assert hf_shape == value.shape, (
F"""Shape of hf {key + "." + weight_type if weight_type is not None else ""} is {hf_shape}, but should be"""
F""" {value.shape} for {full_name}"""
)
if weight_type == "weight":
UpperCamelCase : int = value
elif weight_type == "weight_g":
UpperCamelCase : Optional[int] = value
elif weight_type == "weight_v":
UpperCamelCase : str = value
elif weight_type == "bias":
UpperCamelCase : Union[str, Any] = value
else:
UpperCamelCase : List[str] = value
logger.info(F"""{key + "." + weight_type if weight_type is not None else ""} was initialized from {full_name}.""" )
def A_ ( _lowerCAmelCase , _lowerCAmelCase ) -> str:
UpperCamelCase : Optional[int] = []
UpperCamelCase : Optional[int] = fairseq_model.state_dict()
UpperCamelCase : Optional[Any] = hf_model.feature_extractor
for name, value in fairseq_dict.items():
UpperCamelCase : Dict = False
if "conv_layers" in name:
load_conv_layer(
_lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , hf_model.config.feat_extract_norm == "group" , )
UpperCamelCase : Any = True
else:
for key, mapped_key in MAPPING.items():
if key in name or key.split("w2v_model." )[-1] == name.split("." )[0]:
UpperCamelCase : Dict = True
if "*" in mapped_key:
UpperCamelCase : Any = name.split(_lowerCAmelCase )[0].split("." )[-2]
UpperCamelCase : Optional[int] = mapped_key.replace("*" , _lowerCAmelCase )
if "weight_g" in name:
UpperCamelCase : List[str] = "weight_g"
elif "weight_v" in name:
UpperCamelCase : str = "weight_v"
elif "bias" in name and "relative_attention_bias" not in name:
UpperCamelCase : Optional[Any] = "bias"
elif "weight" in name:
# TODO: don't match quantizer.weight_proj
UpperCamelCase : int = "weight"
else:
UpperCamelCase : str = None
set_recursively(_lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase )
continue
if not is_used:
unused_weights.append(_lowerCAmelCase )
logger.warning(F"""Unused weights: {unused_weights}""" )
def A_ ( _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase ) -> Tuple:
UpperCamelCase : str = full_name.split("conv_layers." )[-1]
UpperCamelCase : List[Any] = name.split("." )
UpperCamelCase : Union[str, Any] = int(items[0] )
UpperCamelCase : Tuple = int(items[1] )
if type_id == 0:
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.bias.data.shape, (
F"""{full_name} has size {value.shape}, but"""
F""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found."""
)
UpperCamelCase : List[Any] = value
logger.info(F"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.weight.data.shape, (
F"""{full_name} has size {value.shape}, but"""
F""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found."""
)
UpperCamelCase : str = value
logger.info(F"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape, (
F"""{full_name} has size {value.shape}, but {feature_extractor[layer_id].layer_norm.bias.data.shape} was"""
" found."
)
UpperCamelCase : Any = value
logger.info(F"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape, (
F"""{full_name} has size {value.shape}, but"""
F""" {feature_extractor[layer_id].layer_norm.weight.data.shape} was found."""
)
UpperCamelCase : Optional[Any] = value
logger.info(F"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(_lowerCAmelCase )
@torch.no_grad()
def A_ ( _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase=None ) -> str:
# load the pre-trained checkpoints
UpperCamelCase : Tuple = torch.load(_lowerCAmelCase )
UpperCamelCase : List[Any] = WavLMConfigOrig(checkpoint["cfg"] )
UpperCamelCase : Any = WavLMOrig(_lowerCAmelCase )
model.load_state_dict(checkpoint["model"] )
model.eval()
if config_path is not None:
UpperCamelCase : Dict = WavLMConfig.from_pretrained(_lowerCAmelCase )
else:
UpperCamelCase : Tuple = WavLMConfig()
UpperCamelCase : Any = WavLMModel(_lowerCAmelCase )
recursively_load_weights(_lowerCAmelCase , _lowerCAmelCase )
hf_wavlm.save_pretrained(_lowerCAmelCase )
if __name__ == "__main__":
__lowerCamelCase : Tuple = argparse.ArgumentParser()
parser.add_argument("""--pytorch_dump_folder_path""", default=None, type=str, help="""Path to the output PyTorch model.""")
parser.add_argument("""--checkpoint_path""", default=None, type=str, help="""Path to fairseq checkpoint""")
parser.add_argument("""--config_path""", default=None, type=str, help="""Path to hf config.json of model to convert""")
__lowerCamelCase : int = parser.parse_args()
convert_wavlm_checkpoint(args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path)
| 52
|
from datetime import datetime
import requests
def A (__A : str ) -> bytes:
"""simple docstring"""
UpperCAmelCase_ = '''https://downloadgram.net/wp-json/wppress/video-downloader/video?url='''
UpperCAmelCase_ = requests.get(base_url + url ).json()[0]['''urls'''][0]['''src''']
return requests.get(__A ).content
if __name__ == "__main__":
snake_case_ : Optional[Any] = input("Enter Video/IGTV url: ").strip()
snake_case_ : Any = f"{datetime.now():%Y-%m-%d_%H:%M:%S}.mp4"
with open(file_name, "wb") as fp:
fp.write(download_video(url))
print(f"Done. Video saved to disk as {file_name}.")
| 51
| 0
|
'''simple docstring'''
import os
from tempfile import TemporaryDirectory
from unittest import TestCase
import pytest
from absl.testing import parameterized
from datasets import config
from datasets.arrow_reader import HF_GCP_BASE_URL
from datasets.builder import DatasetBuilder
from datasets.dataset_dict import IterableDatasetDict
from datasets.iterable_dataset import IterableDataset
from datasets.load import dataset_module_factory, import_main_class
from datasets.utils.file_utils import cached_path
a__ : Optional[int] =[
{'''dataset''': '''wikipedia''', '''config_name''': '''20220301.de'''},
{'''dataset''': '''wikipedia''', '''config_name''': '''20220301.en'''},
{'''dataset''': '''wikipedia''', '''config_name''': '''20220301.fr'''},
{'''dataset''': '''wikipedia''', '''config_name''': '''20220301.frr'''},
{'''dataset''': '''wikipedia''', '''config_name''': '''20220301.it'''},
{'''dataset''': '''wikipedia''', '''config_name''': '''20220301.simple'''},
{'''dataset''': '''snli''', '''config_name''': '''plain_text'''},
{'''dataset''': '''eli5''', '''config_name''': '''LFQA_reddit'''},
{'''dataset''': '''wiki40b''', '''config_name''': '''en'''},
{'''dataset''': '''wiki_dpr''', '''config_name''': '''psgs_w100.nq.compressed'''},
{'''dataset''': '''wiki_dpr''', '''config_name''': '''psgs_w100.nq.no_index'''},
{'''dataset''': '''wiki_dpr''', '''config_name''': '''psgs_w100.multiset.no_index'''},
{'''dataset''': '''natural_questions''', '''config_name''': '''default'''},
]
def lowercase__ ( __lowercase : Optional[int]=True ) -> List[Any]:
"""simple docstring"""
if with_config:
return [
{
"testcase_name": d["dataset"] + "/" + d["config_name"],
"dataset": d["dataset"],
"config_name": d["config_name"],
}
for d in DATASETS_ON_HF_GCP
]
else:
return [
{"testcase_name": dataset, "dataset": dataset} for dataset in {d["dataset"] for d in DATASETS_ON_HF_GCP}
]
@parameterized.named_parameters(list_datasets_on_hf_gcp_parameters(with_config=__lowerCamelCase ) )
class snake_case ( __lowerCamelCase ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : List[Any] =None
SCREAMING_SNAKE_CASE_ : Tuple =None
def _lowerCamelCase ( self : int , __A : Dict , __A : List[Any] ):
with TemporaryDirectory() as tmp_dir:
__UpperCamelCase = dataset_module_factory(__A , cache_dir=__A )
__UpperCamelCase = import_main_class(dataset_module.module_path , dataset=__A )
__UpperCamelCase = builder_cls(
cache_dir=__A , config_name=__A , hash=dataset_module.hash , )
__UpperCamelCase = '/'.join(
[
HF_GCP_BASE_URL,
builder_instance._relative_data_dir(with_hash=__A ).replace(os.sep , '/' ),
config.DATASET_INFO_FILENAME,
] )
__UpperCamelCase = cached_path(__A , cache_dir=__A )
self.assertTrue(os.path.exists(__A ) )
@pytest.mark.integration
def lowercase__ ( __lowercase : Union[str, Any] ) -> Optional[Any]:
"""simple docstring"""
__UpperCamelCase = tmp_path_factory.mktemp('test_hf_gcp' ) / 'test_wikipedia_simple'
__UpperCamelCase = dataset_module_factory('wikipedia' , cache_dir=__lowercase )
__UpperCamelCase = import_main_class(dataset_module.module_path )
__UpperCamelCase = builder_cls(
cache_dir=__lowercase , config_name='20220301.frr' , hash=dataset_module.hash , )
# use the HF cloud storage, not the original download_and_prepare that uses apache-beam
__UpperCamelCase = None
builder_instance.download_and_prepare()
__UpperCamelCase = builder_instance.as_dataset()
assert ds
@pytest.mark.integration
def lowercase__ ( __lowercase : Dict ) -> Any:
"""simple docstring"""
__UpperCamelCase = dataset_module_factory('wikipedia' , cache_dir=__lowercase )
__UpperCamelCase = import_main_class(dataset_module.module_path , dataset=__lowercase )
__UpperCamelCase = builder_cls(
cache_dir=__lowercase , config_name='20220301.frr' , hash=dataset_module.hash , )
__UpperCamelCase = builder_instance.as_streaming_dataset()
assert ds
assert isinstance(__lowercase , __lowercase )
assert "train" in ds
assert isinstance(ds['train'] , __lowercase )
assert next(iter(ds['train'] ) )
| 53
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[Any] = {
"tiiuae/falcon-40b": "https://huggingface.co/tiiuae/falcon-40b/resolve/main/config.json",
"tiiuae/falcon-7b": "https://huggingface.co/tiiuae/falcon-7b/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : Optional[Any] = '''falcon'''
UpperCAmelCase__ : List[Any] = ['''past_key_values''']
def __init__( self : Union[str, Any] , _snake_case : List[str]=65024 , _snake_case : int=4544 , _snake_case : int=32 , _snake_case : Any=71 , _snake_case : int=1e-5 , _snake_case : Dict=0.0_2 , _snake_case : int=True , _snake_case : List[Any]=0.0 , _snake_case : Tuple=0.0 , _snake_case : int=None , _snake_case : Tuple=False , _snake_case : Any=False , _snake_case : str=True , _snake_case : Any=True , _snake_case : List[str]=False , _snake_case : Tuple=11 , _snake_case : Dict=11 , **_snake_case : Optional[int] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
# Backward compatibility with n_embed kwarg
UpperCAmelCase_ = kwargs.pop('''n_embed''' , _snake_case)
UpperCAmelCase_ = hidden_size if n_embed is None else n_embed
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = hidden_dropout
UpperCAmelCase_ = attention_dropout
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
UpperCAmelCase_ = num_attention_heads if num_kv_heads is None else num_kv_heads
UpperCAmelCase_ = alibi
UpperCAmelCase_ = new_decoder_architecture
UpperCAmelCase_ = multi_query # Ignored when new_decoder_architecture is True
UpperCAmelCase_ = parallel_attn
UpperCAmelCase_ = bias
super().__init__(bos_token_id=_snake_case , eos_token_id=_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
return self.hidden_size // self.num_attention_heads
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return not self.alibi
| 51
| 0
|
"""simple docstring"""
def UpperCAmelCase__ (lowerCAmelCase_ ):
'''simple docstring'''
__SCREAMING_SNAKE_CASE = len(lowerCAmelCase_ )
for _ in range(lowerCAmelCase_ ):
for i in range(_ % 2 , arr_size - 1 , 2 ):
if arr[i + 1] < arr[i]:
__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE = arr[i + 1], arr[i]
return arr
if __name__ == "__main__":
a__ : str = list(range(1_0, 0, -1))
print(F"Original: {arr}. Sorted: {odd_even_transposition(arr)}")
| 54
|
from __future__ import annotations
import time
from math import sqrt
# 1 for manhattan, 0 for euclidean
snake_case_ : str = 0
snake_case_ : Union[str, Any] = [
[0, 0, 0, 0, 0, 0, 0],
[0, 1, 0, 0, 0, 0, 0], # 0 are free path whereas 1's are obstacles
[0, 0, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 0],
[1, 0, 1, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 0, 0],
]
snake_case_ : str = [[-1, 0], [0, -1], [1, 0], [0, 1]] # up, left, down, right
snake_case_ : List[Any] = tuple[int, int]
class __snake_case :
def __init__( self : Any , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : Node | None , ):
"""simple docstring"""
UpperCAmelCase_ = pos_x
UpperCAmelCase_ = pos_y
UpperCAmelCase_ = (pos_y, pos_x)
UpperCAmelCase_ = goal_x
UpperCAmelCase_ = goal_y
UpperCAmelCase_ = g_cost
UpperCAmelCase_ = parent
UpperCAmelCase_ = self.calculate_heuristic()
UpperCAmelCase_ = self.g_cost + self.h_cost
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = self.pos_x - self.goal_x
UpperCAmelCase_ = self.pos_y - self.goal_y
if HEURISTIC == 1:
return abs(_snake_case) + abs(_snake_case)
else:
return sqrt(dy**2 + dx**2)
def __lt__( self : Union[str, Any] , _snake_case : Node):
"""simple docstring"""
return self.f_cost < other.f_cost
class __snake_case :
def __init__( self : str , _snake_case : TPosition , _snake_case : TPosition):
"""simple docstring"""
UpperCAmelCase_ = Node(start[1] , start[0] , goal[1] , goal[0] , 0 , _snake_case)
UpperCAmelCase_ = Node(goal[1] , goal[0] , goal[1] , goal[0] , 99999 , _snake_case)
UpperCAmelCase_ = [self.start]
UpperCAmelCase_ = []
UpperCAmelCase_ = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
while self.open_nodes:
# Open Nodes are sorted using __lt__
self.open_nodes.sort()
UpperCAmelCase_ = self.open_nodes.pop(0)
if current_node.pos == self.target.pos:
return self.retrace_path(_snake_case)
self.closed_nodes.append(_snake_case)
UpperCAmelCase_ = self.get_successors(_snake_case)
for child_node in successors:
if child_node in self.closed_nodes:
continue
if child_node not in self.open_nodes:
self.open_nodes.append(_snake_case)
else:
# retrieve the best current path
UpperCAmelCase_ = self.open_nodes.pop(self.open_nodes.index(_snake_case))
if child_node.g_cost < better_node.g_cost:
self.open_nodes.append(_snake_case)
else:
self.open_nodes.append(_snake_case)
return [self.start.pos]
def lowerCamelCase ( self : Tuple , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = []
for action in delta:
UpperCAmelCase_ = parent.pos_x + action[1]
UpperCAmelCase_ = parent.pos_y + action[0]
if not (0 <= pos_x <= len(grid[0]) - 1 and 0 <= pos_y <= len(_snake_case) - 1):
continue
if grid[pos_y][pos_x] != 0:
continue
successors.append(
Node(
_snake_case , _snake_case , self.target.pos_y , self.target.pos_x , parent.g_cost + 1 , _snake_case , ))
return successors
def lowerCamelCase ( self : Any , _snake_case : Node | None):
"""simple docstring"""
UpperCAmelCase_ = node
UpperCAmelCase_ = []
while current_node is not None:
path.append((current_node.pos_y, current_node.pos_x))
UpperCAmelCase_ = current_node.parent
path.reverse()
return path
class __snake_case :
def __init__( self : Any , _snake_case : TPosition , _snake_case : TPosition):
"""simple docstring"""
UpperCAmelCase_ = AStar(_snake_case , _snake_case)
UpperCAmelCase_ = AStar(_snake_case , _snake_case)
UpperCAmelCase_ = False
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
while self.fwd_astar.open_nodes or self.bwd_astar.open_nodes:
self.fwd_astar.open_nodes.sort()
self.bwd_astar.open_nodes.sort()
UpperCAmelCase_ = self.fwd_astar.open_nodes.pop(0)
UpperCAmelCase_ = self.bwd_astar.open_nodes.pop(0)
if current_bwd_node.pos == current_fwd_node.pos:
return self.retrace_bidirectional_path(
_snake_case , _snake_case)
self.fwd_astar.closed_nodes.append(_snake_case)
self.bwd_astar.closed_nodes.append(_snake_case)
UpperCAmelCase_ = current_bwd_node
UpperCAmelCase_ = current_fwd_node
UpperCAmelCase_ = {
self.fwd_astar: self.fwd_astar.get_successors(_snake_case),
self.bwd_astar: self.bwd_astar.get_successors(_snake_case),
}
for astar in [self.fwd_astar, self.bwd_astar]:
for child_node in successors[astar]:
if child_node in astar.closed_nodes:
continue
if child_node not in astar.open_nodes:
astar.open_nodes.append(_snake_case)
else:
# retrieve the best current path
UpperCAmelCase_ = astar.open_nodes.pop(
astar.open_nodes.index(_snake_case))
if child_node.g_cost < better_node.g_cost:
astar.open_nodes.append(_snake_case)
else:
astar.open_nodes.append(_snake_case)
return [self.fwd_astar.start.pos]
def lowerCamelCase ( self : int , _snake_case : Node , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = self.fwd_astar.retrace_path(_snake_case)
UpperCAmelCase_ = self.bwd_astar.retrace_path(_snake_case)
bwd_path.pop()
bwd_path.reverse()
UpperCAmelCase_ = fwd_path + bwd_path
return path
if __name__ == "__main__":
# all coordinates are given in format [y,x]
snake_case_ : Any = (0, 0)
snake_case_ : Union[str, Any] = (len(grid) - 1, len(grid[0]) - 1)
for elem in grid:
print(elem)
snake_case_ : str = time.time()
snake_case_ : List[str] = AStar(init, goal)
snake_case_ : Optional[int] = a_star.search()
snake_case_ : Optional[Any] = time.time() - start_time
print(f"AStar execution time = {end_time:f} seconds")
snake_case_ : int = time.time()
snake_case_ : Dict = BidirectionalAStar(init, goal)
snake_case_ : str = time.time() - bd_start_time
print(f"BidirectionalAStar execution time = {bd_end_time:f} seconds")
| 51
| 0
|
'''simple docstring'''
import unittest
import torch
from torch import nn
from accelerate.test_utils import require_cuda
from accelerate.utils.memory import find_executable_batch_size, release_memory
def __snake_case ( ):
raise RuntimeError("CUDA out of memory." )
class snake_case ( nn.Module ):
"""simple docstring"""
def __init__( self ):
"""simple docstring"""
super().__init__()
lowerCamelCase_ = nn.Linear(3 , 4 )
lowerCamelCase_ = nn.BatchNormad(4 )
lowerCamelCase_ = nn.Linear(4 , 5 )
def snake_case ( self , UpperCamelCase ):
"""simple docstring"""
return self.lineara(self.batchnorm(self.lineara(UpperCamelCase ) ) )
class snake_case ( unittest.TestCase ):
"""simple docstring"""
def snake_case ( self ):
"""simple docstring"""
lowerCamelCase_ = []
@find_executable_batch_size(starting_batch_size=128 )
def mock_training_loop_function(UpperCamelCase ):
nonlocal batch_sizes
batch_sizes.append(UpperCamelCase )
if batch_size != 8:
raise_fake_out_of_memory()
mock_training_loop_function()
self.assertListEqual(UpperCamelCase , [128, 64, 32, 16, 8] )
def snake_case ( self ):
"""simple docstring"""
lowerCamelCase_ = []
@find_executable_batch_size(starting_batch_size=128 )
def mock_training_loop_function(UpperCamelCase , UpperCamelCase ):
nonlocal batch_sizes
batch_sizes.append(UpperCamelCase )
if batch_size != 8:
raise_fake_out_of_memory()
return batch_size, arga
lowerCamelCase_ ,lowerCamelCase_ = mock_training_loop_function("hello" )
self.assertListEqual(UpperCamelCase , [128, 64, 32, 16, 8] )
self.assertListEqual([bs, arga] , [8, "hello"] )
def snake_case ( self ):
"""simple docstring"""
@find_executable_batch_size(starting_batch_size=0 )
def mock_training_loop_function(UpperCamelCase ):
pass
with self.assertRaises(UpperCamelCase ) as cm:
mock_training_loop_function()
self.assertIn("No executable batch size found, reached zero." , cm.exception.args[0] )
def snake_case ( self ):
"""simple docstring"""
@find_executable_batch_size(starting_batch_size=16 )
def mock_training_loop_function(UpperCamelCase ):
if batch_size > 0:
raise_fake_out_of_memory()
pass
with self.assertRaises(UpperCamelCase ) as cm:
mock_training_loop_function()
self.assertIn("No executable batch size found, reached zero." , cm.exception.args[0] )
def snake_case ( self ):
"""simple docstring"""
@find_executable_batch_size(starting_batch_size=128 )
def mock_training_loop_function(UpperCamelCase , UpperCamelCase , UpperCamelCase ):
if batch_size != 8:
raise raise_fake_out_of_memory()
with self.assertRaises(UpperCamelCase ) as cm:
mock_training_loop_function(128 , "hello" , "world" )
self.assertIn("Batch size was passed into `f`" , cm.exception.args[0] )
self.assertIn("`f(arg1='hello', arg2='world')" , cm.exception.args[0] )
def snake_case ( self ):
"""simple docstring"""
@find_executable_batch_size(starting_batch_size=16 )
def mock_training_loop_function(UpperCamelCase ):
raise ValueError("Oops, we had an error!" )
with self.assertRaises(UpperCamelCase ) as cm:
mock_training_loop_function()
self.assertIn("Oops, we had an error!" , cm.exception.args[0] )
@require_cuda
def snake_case ( self ):
"""simple docstring"""
lowerCamelCase_ = torch.cuda.memory_allocated()
lowerCamelCase_ = ModelForTest()
model.cuda()
self.assertGreater(torch.cuda.memory_allocated() , UpperCamelCase )
lowerCamelCase_ = release_memory(UpperCamelCase )
self.assertEqual(torch.cuda.memory_allocated() , UpperCamelCase )
| 55
|
import inspect
import unittest
import numpy as np
from tests.test_modeling_common import floats_tensor
from transformers import DetrConfig, MaskFormerConfig, SwinConfig, is_torch_available, is_vision_available
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import MaskFormerForInstanceSegmentation, MaskFormerModel
if is_vision_available():
from transformers import MaskFormerImageProcessor
if is_vision_available():
from PIL import Image
class __snake_case :
def __init__( self : List[str] , _snake_case : Union[str, Any] , _snake_case : List[str]=2 , _snake_case : Any=True , _snake_case : Any=False , _snake_case : List[str]=10 , _snake_case : Any=3 , _snake_case : Union[str, Any]=32 * 4 , _snake_case : List[Any]=32 * 6 , _snake_case : Tuple=4 , _snake_case : Dict=32 , ):
"""simple docstring"""
UpperCAmelCase_ = parent
UpperCAmelCase_ = batch_size
UpperCAmelCase_ = is_training
UpperCAmelCase_ = use_auxiliary_loss
UpperCAmelCase_ = num_queries
UpperCAmelCase_ = num_channels
UpperCAmelCase_ = min_size
UpperCAmelCase_ = max_size
UpperCAmelCase_ = num_labels
UpperCAmelCase_ = mask_feature_size
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = floats_tensor([self.batch_size, self.num_channels, self.min_size, self.max_size]).to(
_snake_case)
UpperCAmelCase_ = torch.ones([self.batch_size, self.min_size, self.max_size] , device=_snake_case)
UpperCAmelCase_ = (
torch.rand([self.batch_size, self.num_labels, self.min_size, self.max_size] , device=_snake_case) > 0.5
).float()
UpperCAmelCase_ = (torch.rand((self.batch_size, self.num_labels) , device=_snake_case) > 0.5).long()
UpperCAmelCase_ = self.get_config()
return config, pixel_values, pixel_mask, mask_labels, class_labels
def lowerCamelCase ( self : Any):
"""simple docstring"""
return MaskFormerConfig.from_backbone_and_decoder_configs(
backbone_config=SwinConfig(
depths=[1, 1, 1, 1] , ) , decoder_config=DetrConfig(
decoder_ffn_dim=128 , num_queries=self.num_queries , decoder_attention_heads=2 , d_model=self.mask_feature_size , ) , mask_feature_size=self.mask_feature_size , fpn_feature_size=self.mask_feature_size , num_channels=self.num_channels , num_labels=self.num_labels , )
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.prepare_config_and_inputs()
UpperCAmelCase_ = {'''pixel_values''': pixel_values, '''pixel_mask''': pixel_mask}
return config, inputs_dict
def lowerCamelCase ( self : str , _snake_case : List[Any] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = output.encoder_hidden_states
UpperCAmelCase_ = output.pixel_decoder_hidden_states
UpperCAmelCase_ = output.transformer_decoder_hidden_states
self.parent.assertTrue(len(_snake_case) , len(config.backbone_config.depths))
self.parent.assertTrue(len(_snake_case) , len(config.backbone_config.depths))
self.parent.assertTrue(len(_snake_case) , config.decoder_config.decoder_layers)
def lowerCamelCase ( self : Union[str, Any] , _snake_case : List[str] , _snake_case : int , _snake_case : Optional[Any] , _snake_case : str=False):
"""simple docstring"""
with torch.no_grad():
UpperCAmelCase_ = MaskFormerModel(config=_snake_case)
model.to(_snake_case)
model.eval()
UpperCAmelCase_ = model(pixel_values=_snake_case , pixel_mask=_snake_case)
UpperCAmelCase_ = model(_snake_case , output_hidden_states=_snake_case)
# the correct shape of output.transformer_decoder_hidden_states ensure the correcteness of the
# encoder and pixel decoder
self.parent.assertEqual(
output.transformer_decoder_last_hidden_state.shape , (self.batch_size, self.num_queries, self.mask_feature_size) , )
# let's ensure the other two hidden state exists
self.parent.assertTrue(output.pixel_decoder_last_hidden_state is not None)
self.parent.assertTrue(output.encoder_last_hidden_state is not None)
if output_hidden_states:
self.check_output_hidden_state(_snake_case , _snake_case)
def lowerCamelCase ( self : List[Any] , _snake_case : List[Any] , _snake_case : List[Any] , _snake_case : str , _snake_case : Optional[int] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerForInstanceSegmentation(config=_snake_case)
model.to(_snake_case)
model.eval()
def comm_check_on_output(_snake_case : Tuple):
# let's still check that all the required stuff is there
self.parent.assertTrue(result.transformer_decoder_last_hidden_state is not None)
self.parent.assertTrue(result.pixel_decoder_last_hidden_state is not None)
self.parent.assertTrue(result.encoder_last_hidden_state is not None)
# okay, now we need to check the logits shape
# due to the encoder compression, masks have a //4 spatial size
self.parent.assertEqual(
result.masks_queries_logits.shape , (self.batch_size, self.num_queries, self.min_size // 4, self.max_size // 4) , )
# + 1 for null class
self.parent.assertEqual(
result.class_queries_logits.shape , (self.batch_size, self.num_queries, self.num_labels + 1))
with torch.no_grad():
UpperCAmelCase_ = model(pixel_values=_snake_case , pixel_mask=_snake_case)
UpperCAmelCase_ = model(_snake_case)
comm_check_on_output(_snake_case)
UpperCAmelCase_ = model(
pixel_values=_snake_case , pixel_mask=_snake_case , mask_labels=_snake_case , class_labels=_snake_case)
comm_check_on_output(_snake_case)
self.parent.assertTrue(result.loss is not None)
self.parent.assertEqual(result.loss.shape , torch.Size([1]))
@require_torch
class __snake_case ( a , a , unittest.TestCase ):
UpperCAmelCase__ : Union[str, Any] = (MaskFormerModel, MaskFormerForInstanceSegmentation) if is_torch_available() else ()
UpperCAmelCase__ : Optional[Any] = (
{'''feature-extraction''': MaskFormerModel, '''image-segmentation''': MaskFormerForInstanceSegmentation}
if is_torch_available()
else {}
)
UpperCAmelCase__ : Dict = False
UpperCAmelCase__ : List[str] = False
UpperCAmelCase__ : Optional[Any] = False
UpperCAmelCase__ : Union[str, Any] = False
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerModelTester(self)
UpperCAmelCase_ = ConfigTester(self , config_class=_snake_case , has_text_modality=_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
self.config_tester.run_common_tests()
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskformer_model(_snake_case , **_snake_case , output_hidden_states=_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_maskformer_instance_segmentation_head_model(*_snake_case)
@unittest.skip(reason='''MaskFormer does not use inputs_embeds''')
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer does not have a get_input_embeddings method''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer is not a generative model''')
def lowerCamelCase ( self : str):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer does not use token embeddings''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
@require_torch_multi_gpu
@unittest.skip(
reason='''MaskFormer has some layers using `add_module` which doesn\'t work well with `nn.DataParallel`''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
pass
@unittest.skip('''Will be fixed soon by reducing the size of the model used for common tests.''')
def lowerCamelCase ( self : str):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ = model_class(_snake_case)
UpperCAmelCase_ = inspect.signature(model.forward)
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCAmelCase_ = [*signature.parameters.keys()]
UpperCAmelCase_ = ['''pixel_values''']
self.assertListEqual(arg_names[:1] , _snake_case)
@slow
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
for model_name in ["facebook/maskformer-swin-small-coco"]:
UpperCAmelCase_ = MaskFormerModel.from_pretrained(_snake_case)
self.assertIsNotNone(_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = (self.model_tester.min_size,) * 2
UpperCAmelCase_ = {
'''pixel_values''': torch.randn((2, 3, *size) , device=_snake_case),
'''mask_labels''': torch.randn((2, 10, *size) , device=_snake_case),
'''class_labels''': torch.zeros(2 , 10 , device=_snake_case).long(),
}
UpperCAmelCase_ = MaskFormerForInstanceSegmentation(MaskFormerConfig()).to(_snake_case)
UpperCAmelCase_ = model(**_snake_case)
self.assertTrue(outputs.loss is not None)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskformer_model(_snake_case , **_snake_case , output_hidden_states=_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ = model_class(_snake_case).to(_snake_case)
UpperCAmelCase_ = model(**_snake_case , output_attentions=_snake_case)
self.assertTrue(outputs.attentions is not None)
def lowerCamelCase ( self : int):
"""simple docstring"""
if not self.model_tester.is_training:
return
# only MaskFormerForInstanceSegmentation has the loss
UpperCAmelCase_ = self.all_model_classes[1]
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCAmelCase_ = model_class(_snake_case)
model.to(_snake_case)
model.train()
UpperCAmelCase_ = model(_snake_case , mask_labels=_snake_case , class_labels=_snake_case).loss
loss.backward()
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = self.all_model_classes[1]
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCAmelCase_ = True
UpperCAmelCase_ = True
UpperCAmelCase_ = model_class(_snake_case)
model.to(_snake_case)
model.train()
UpperCAmelCase_ = model(_snake_case , mask_labels=_snake_case , class_labels=_snake_case)
UpperCAmelCase_ = outputs.encoder_hidden_states[0]
encoder_hidden_states.retain_grad()
UpperCAmelCase_ = outputs.pixel_decoder_hidden_states[0]
pixel_decoder_hidden_states.retain_grad()
# we requires_grad=True in inputs_embeds (line 2152), the original implementation don't
UpperCAmelCase_ = outputs.transformer_decoder_hidden_states[0]
transformer_decoder_hidden_states.retain_grad()
UpperCAmelCase_ = outputs.attentions[0]
attentions.retain_grad()
outputs.loss.backward(retain_graph=_snake_case)
self.assertIsNotNone(encoder_hidden_states.grad)
self.assertIsNotNone(pixel_decoder_hidden_states.grad)
self.assertIsNotNone(transformer_decoder_hidden_states.grad)
self.assertIsNotNone(attentions.grad)
snake_case_ : Dict = 1e-4
def A () -> Union[str, Any]:
"""simple docstring"""
UpperCAmelCase_ = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
return image
@require_vision
@slow
class __snake_case ( unittest.TestCase ):
@cached_property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return (
MaskFormerImageProcessor.from_pretrained('''facebook/maskformer-swin-small-coco''')
if is_vision_available()
else None
)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerModel.from_pretrained('''facebook/maskformer-swin-small-coco''').to(_snake_case)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
UpperCAmelCase_ = torch.tensor(
[[-0.0_4_8_2, 0.9_2_2_8, 0.4_9_5_1], [-0.2_5_4_7, 0.8_0_1_7, 0.8_5_2_7], [-0.0_0_6_9, 0.3_3_8_5, -0.0_0_8_9]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.encoder_last_hidden_state[0, 0, :3, :3] , _snake_case , atol=_snake_case))
UpperCAmelCase_ = torch.tensor(
[[-0.8_4_2_2, -0.8_4_3_4, -0.9_7_1_8], [-1.0_1_4_4, -0.5_5_6_5, -0.4_1_9_5], [-1.0_0_3_8, -0.4_4_8_4, -0.1_9_6_1]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.pixel_decoder_last_hidden_state[0, 0, :3, :3] , _snake_case , atol=_snake_case))
UpperCAmelCase_ = torch.tensor(
[[0.2_8_5_2, -0.0_1_5_9, 0.9_7_3_5], [0.6_2_5_4, 0.1_8_5_8, 0.8_5_2_9], [-0.0_6_8_0, -0.4_1_1_6, 1.8_4_1_3]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.transformer_decoder_last_hidden_state[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-swin-small-coco''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
# masks_queries_logits
UpperCAmelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.decoder_config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) , )
UpperCAmelCase_ = [
[-1.3_7_3_7_1_2_4, -1.7_7_2_4_9_3_7, -1.9_3_6_4_2_3_3],
[-1.5_9_7_7_2_8_1, -1.9_8_6_7_9_3_9, -2.1_5_2_3_6_9_5],
[-1.5_7_9_5_3_9_8, -1.9_2_6_9_8_3_2, -2.0_9_3_9_4_2],
]
UpperCAmelCase_ = torch.tensor(_snake_case).to(_snake_case)
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _snake_case , atol=_snake_case))
# class_queries_logits
UpperCAmelCase_ = outputs.class_queries_logits
self.assertEqual(
class_queries_logits.shape , (1, model.config.decoder_config.num_queries, model.config.num_labels + 1))
UpperCAmelCase_ = torch.tensor(
[
[1.65_12e00, -5.25_72e00, -3.35_19e00],
[3.61_69e-02, -5.90_25e00, -2.93_13e00],
[1.07_66e-04, -7.76_30e00, -5.12_63e00],
]).to(_snake_case)
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-resnet101-coco-stuff''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
# masks_queries_logits
UpperCAmelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.decoder_config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) , )
UpperCAmelCase_ = [[-0.9_0_4_6, -2.6_3_6_6, -4.6_0_6_2], [-3.4_1_7_9, -5.7_8_9_0, -8.8_0_5_7], [-4.9_1_7_9, -7.6_5_6_0, -1_0.7_7_1_1]]
UpperCAmelCase_ = torch.tensor(_snake_case).to(_snake_case)
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _snake_case , atol=_snake_case))
# class_queries_logits
UpperCAmelCase_ = outputs.class_queries_logits
self.assertEqual(
class_queries_logits.shape , (1, model.config.decoder_config.num_queries, model.config.num_labels + 1))
UpperCAmelCase_ = torch.tensor(
[[4.7_1_8_8, -3.2_5_8_5, -2.8_8_5_7], [6.6_8_7_1, -2.9_1_8_1, -1.2_4_8_7], [7.2_4_4_9, -2.2_7_6_4, -2.1_8_7_4]]).to(_snake_case)
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-swin-small-coco''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = image_processor(
[np.zeros((3, 800, 1333)), np.zeros((3, 800, 1333))] , segmentation_maps=[np.zeros((384, 384)).astype(np.floataa), np.zeros((384, 384)).astype(np.floataa)] , return_tensors='''pt''' , )
UpperCAmelCase_ = inputs['''pixel_values'''].to(_snake_case)
UpperCAmelCase_ = [el.to(_snake_case) for el in inputs['''mask_labels''']]
UpperCAmelCase_ = [el.to(_snake_case) for el in inputs['''class_labels''']]
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
self.assertTrue(outputs.loss is not None)
| 51
| 0
|
'''simple docstring'''
from __future__ import annotations
def __magic_name__ ( __UpperCAmelCase, __UpperCAmelCase, __UpperCAmelCase, __UpperCAmelCase ) -> str: # noqa: E741
'''simple docstring'''
while r - l > 1:
snake_case_ = (l + r) // 2
if v[m] >= key:
snake_case_ = m
else:
snake_case_ = m # noqa: E741
return r
def __magic_name__ ( __UpperCAmelCase ) -> int:
'''simple docstring'''
if len(__UpperCAmelCase ) == 0:
return 0
snake_case_ = [0] * len(__UpperCAmelCase )
snake_case_ = 1
snake_case_ = v[0]
for i in range(1, len(__UpperCAmelCase ) ):
if v[i] < tail[0]:
snake_case_ = v[i]
elif v[i] > tail[length - 1]:
snake_case_ = v[i]
length += 1
else:
snake_case_ = v[i]
return length
if __name__ == "__main__":
import doctest
doctest.testmod()
| 56
|
import argparse
import pickle
import numpy as np
import torch
from torch import nn
from transformers import ReformerConfig, ReformerModelWithLMHead
from transformers.utils import logging
logging.set_verbosity_info()
def A (__A : Optional[int] , __A : int , __A : str=None ) -> List[Any]:
"""simple docstring"""
assert torch_layer.weight.shape == weight.shape, F"""{torch_layer} layer.weight does not match"""
UpperCAmelCase_ = nn.Parameter(__A )
if bias is not None:
assert torch_layer.bias.shape == bias.shape, F"""{torch_layer} layer.bias does not match"""
UpperCAmelCase_ = nn.Parameter(__A )
def A (__A : Tuple , __A : Dict , __A : str ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = np.asarray(weights[0] )
UpperCAmelCase_ = np.asarray(weights[1] )
UpperCAmelCase_ = np.asarray(weights[2] )
set_param(
torch_layer.self_attention.query_key , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.value , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.output.dense , torch.tensor(__A ).view(-1 , __A ).contiguous().transpose(0 , 1 ) , )
def A (__A : Optional[Any] , __A : Any , __A : List[Any] ) -> int:
"""simple docstring"""
UpperCAmelCase_ = np.asarray(weights[0] )
UpperCAmelCase_ = np.asarray(weights[1] )
UpperCAmelCase_ = np.asarray(weights[2] )
UpperCAmelCase_ = np.asarray(weights[3] )
set_param(
torch_layer.self_attention.query , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.key , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.value , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.output.dense , torch.tensor(__A ).view(-1 , __A ).contiguous().transpose(0 , 1 ) , )
def A (__A : int , __A : Union[str, Any] , __A : List[str] ) -> List[Any]:
"""simple docstring"""
UpperCAmelCase_ = weights[0][0][0]
UpperCAmelCase_ = np.asarray(layer_norm_a[0] )
UpperCAmelCase_ = np.asarray(layer_norm_a[1] )
set_param(
torch_block.attention.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# lsh weights + output
UpperCAmelCase_ = weights[0][1]
if len(__A ) < 4:
set_layer_weights_in_torch_lsh(__A , torch_block.attention , __A )
else:
set_layer_weights_in_torch_local(__A , torch_block.attention , __A )
# intermediate weighs
UpperCAmelCase_ = weights[2][0][1][2]
# Chunked Feed Forward
if len(__A ) == 4:
UpperCAmelCase_ = intermediate_weights[2]
# layernorm 2
UpperCAmelCase_ = np.asarray(intermediate_weights[0][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[0][1] )
set_param(
torch_block.feed_forward.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# intermediate dense
UpperCAmelCase_ = np.asarray(intermediate_weights[1][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[1][1] )
set_param(
torch_block.feed_forward.dense.dense , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
# intermediate out
UpperCAmelCase_ = np.asarray(intermediate_weights[4][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[4][1] )
set_param(
torch_block.feed_forward.output.dense , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
def A (__A : Optional[int] , __A : Tuple , __A : Any ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = torch_model.reformer
# word embeds
UpperCAmelCase_ = np.asarray(weights[1] )
set_param(
torch_model_reformer.embeddings.word_embeddings , torch.tensor(__A ) , )
if isinstance(weights[3] , __A ):
UpperCAmelCase_ = torch_model_reformer.embeddings.position_embeddings
for emb_idx in range(len(position_embeddings.weights ) ):
UpperCAmelCase_ = np.asarray(weights[3][emb_idx][0] )
assert (
position_embeddings.weights[emb_idx].shape == emb_weights.shape
), F"""{position_embeddings[emb_idx]} emb does not match"""
UpperCAmelCase_ = nn.Parameter(torch.tensor(__A ) )
UpperCAmelCase_ = weights[5]
assert len(torch_model_reformer.encoder.layers ) * 4 == len(
__A ), "HF and trax model do not have the same number of layers"
for layer_idx, layer in enumerate(torch_model_reformer.encoder.layers ):
UpperCAmelCase_ = trax_layer_weights[4 * layer_idx : 4 * (layer_idx + 1)]
set_block_weights_in_torch(__A , __A , __A )
# output layer norm
UpperCAmelCase_ = np.asarray(weights[7][0] )
UpperCAmelCase_ = np.asarray(weights[7][1] )
set_param(
torch_model_reformer.encoder.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# output embeddings
UpperCAmelCase_ = np.asarray(weights[9][0] )
UpperCAmelCase_ = np.asarray(weights[9][1] )
set_param(
torch_model.lm_head.decoder , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
def A (__A : Tuple , __A : int , __A : str ) -> Union[str, Any]:
"""simple docstring"""
UpperCAmelCase_ = ReformerConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = ReformerModelWithLMHead(__A )
with open(__A , '''rb''' ) as f:
UpperCAmelCase_ = pickle.load(__A )['''weights''']
set_model_weights_in_torch(__A , __A , config.hidden_size )
# Save pytorch-model
print(F"""Save PyTorch model to {pytorch_dump_path}""" )
torch.save(model.state_dict() , __A )
if __name__ == "__main__":
snake_case_ : Optional[Any] = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--trax_model_pkl_path", default=None, type=str, required=True, help="Path to the TensorFlow checkpoint path."
)
parser.add_argument(
"--config_file",
default=None,
type=str,
required=True,
help=(
"The config json file corresponding to the pre-trained Reformer model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--pytorch_dump_path", default=None, type=str, required=True, help="Path to the output PyTorch model."
)
snake_case_ : List[Any] = parser.parse_args()
convert_trax_checkpoint_to_pytorch(args.trax_model_pkl_path, args.config_file, args.pytorch_dump_path)
| 51
| 0
|
"""simple docstring"""
import os
import unittest
from transformers import LayoutLMTokenizer, LayoutLMTokenizerFast
from transformers.models.layoutlm.tokenization_layoutlm import VOCAB_FILES_NAMES
from transformers.testing_utils import require_tokenizers
from ...test_tokenization_common import TokenizerTesterMixin
@require_tokenizers
class _UpperCamelCase ( lowerCAmelCase__ ,unittest.TestCase ):
'''simple docstring'''
__UpperCAmelCase : int =LayoutLMTokenizer
__UpperCAmelCase : Optional[int] =LayoutLMTokenizerFast
__UpperCAmelCase : List[Any] =True
__UpperCAmelCase : Dict =True
def snake_case ( self ):
super().setUp()
__lowerCAmelCase = [
"[UNK]",
"[CLS]",
"[SEP]",
"want",
"##want",
"##ed",
"wa",
"un",
"runn",
"##ing",
",",
"low",
"lowest",
]
__lowerCAmelCase = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["vocab_file"] )
with open(self.vocab_file , "w" , encoding="utf-8" ) as vocab_writer:
vocab_writer.write("".join([x + "\n" for x in vocab_tokens] ) )
def snake_case ( self , **__a ):
return LayoutLMTokenizer.from_pretrained(self.tmpdirname , **__a )
def snake_case ( self , __a ):
__lowerCAmelCase = "UNwant\u00E9d,running"
__lowerCAmelCase = "unwanted, running"
return input_text, output_text
def snake_case ( self ):
__lowerCAmelCase = self.tokenizer_class(self.vocab_file )
__lowerCAmelCase = tokenizer.tokenize("UNwant\u00E9d,running" )
self.assertListEqual(__a , ["un", "##want", "##ed", ",", "runn", "##ing"] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(__a ) , [7, 4, 5, 10, 8, 9] )
def snake_case ( self ):
pass
| 57
|
# This model implementation is heavily inspired by https://github.com/haofanwang/ControlNet-for-Diffusers/
import gc
import random
import tempfile
import unittest
import numpy as np
import torch
from PIL import Image
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
ControlNetModel,
DDIMScheduler,
StableDiffusionControlNetImgaImgPipeline,
UNetaDConditionModel,
)
from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_controlnet import MultiControlNetModel
from diffusers.utils import floats_tensor, load_image, load_numpy, randn_tensor, slow, torch_device
from diffusers.utils.import_utils import is_xformers_available
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import (
IMAGE_TO_IMAGE_IMAGE_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_PARAMS,
)
from ..test_pipelines_common import (
PipelineKarrasSchedulerTesterMixin,
PipelineLatentTesterMixin,
PipelineTesterMixin,
)
enable_full_determinism()
class __snake_case ( a , a , a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = StableDiffusionControlNetImgaImgPipeline
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''}
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
UpperCAmelCase__ : Union[str, Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS.union({'''control_image'''} )
UpperCAmelCase__ : Optional[Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS
def lowerCamelCase ( self : int):
"""simple docstring"""
torch.manual_seed(0)
UpperCAmelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , )
torch.manual_seed(0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
torch.manual_seed(0)
UpperCAmelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='''scaled_linear''' , clip_sample=_snake_case , set_alpha_to_one=_snake_case , )
torch.manual_seed(0)
UpperCAmelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , )
torch.manual_seed(0)
UpperCAmelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCAmelCase_ = CLIPTextModel(_snake_case)
UpperCAmelCase_ = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''')
UpperCAmelCase_ = {
'''unet''': unet,
'''controlnet''': controlnet,
'''scheduler''': scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def lowerCamelCase ( self : Union[str, Any] , _snake_case : Any , _snake_case : Dict=0):
"""simple docstring"""
if str(_snake_case).startswith('''mps'''):
UpperCAmelCase_ = torch.manual_seed(_snake_case)
else:
UpperCAmelCase_ = torch.Generator(device=_snake_case).manual_seed(_snake_case)
UpperCAmelCase_ = 2
UpperCAmelCase_ = randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , )
UpperCAmelCase_ = floats_tensor(control_image.shape , rng=random.Random(_snake_case)).to(_snake_case)
UpperCAmelCase_ = image.cpu().permute(0 , 2 , 3 , 1)[0]
UpperCAmelCase_ = Image.fromarray(np.uinta(_snake_case)).convert('''RGB''').resize((64, 64))
UpperCAmelCase_ = {
'''prompt''': '''A painting of a squirrel eating a burger''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
'''image''': image,
'''control_image''': control_image,
}
return inputs
def lowerCamelCase ( self : Any):
"""simple docstring"""
return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3)
@unittest.skipIf(
torch_device != '''cuda''' or not is_xformers_available() , reason='''XFormers attention is only available with CUDA and `xformers` installed''' , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
self._test_inference_batch_single_identical(expected_max_diff=2e-3)
class __snake_case ( a , a , unittest.TestCase ):
UpperCAmelCase__ : str = StableDiffusionControlNetImgaImgPipeline
UpperCAmelCase__ : List[str] = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''}
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
UpperCAmelCase__ : str = frozenset([] ) # TO_DO: add image_params once refactored VaeImageProcessor.preprocess
def lowerCamelCase ( self : str):
"""simple docstring"""
torch.manual_seed(0)
UpperCAmelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , )
torch.manual_seed(0)
def init_weights(_snake_case : Optional[int]):
if isinstance(_snake_case , torch.nn.Convad):
torch.nn.init.normal(m.weight)
m.bias.data.fill_(1.0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
controlneta.controlnet_down_blocks.apply(_snake_case)
torch.manual_seed(0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
controlneta.controlnet_down_blocks.apply(_snake_case)
torch.manual_seed(0)
UpperCAmelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='''scaled_linear''' , clip_sample=_snake_case , set_alpha_to_one=_snake_case , )
torch.manual_seed(0)
UpperCAmelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , )
torch.manual_seed(0)
UpperCAmelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCAmelCase_ = CLIPTextModel(_snake_case)
UpperCAmelCase_ = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''')
UpperCAmelCase_ = MultiControlNetModel([controlneta, controlneta])
UpperCAmelCase_ = {
'''unet''': unet,
'''controlnet''': controlnet,
'''scheduler''': scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def lowerCamelCase ( self : int , _snake_case : Union[str, Any] , _snake_case : str=0):
"""simple docstring"""
if str(_snake_case).startswith('''mps'''):
UpperCAmelCase_ = torch.manual_seed(_snake_case)
else:
UpperCAmelCase_ = torch.Generator(device=_snake_case).manual_seed(_snake_case)
UpperCAmelCase_ = 2
UpperCAmelCase_ = [
randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , ),
randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , ),
]
UpperCAmelCase_ = floats_tensor(control_image[0].shape , rng=random.Random(_snake_case)).to(_snake_case)
UpperCAmelCase_ = image.cpu().permute(0 , 2 , 3 , 1)[0]
UpperCAmelCase_ = Image.fromarray(np.uinta(_snake_case)).convert('''RGB''').resize((64, 64))
UpperCAmelCase_ = {
'''prompt''': '''A painting of a squirrel eating a burger''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
'''image''': image,
'''control_image''': control_image,
}
return inputs
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_dummy_components()
UpperCAmelCase_ = self.pipeline_class(**_snake_case)
pipe.to(_snake_case)
UpperCAmelCase_ = 1_0.0
UpperCAmelCase_ = 4
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case)[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=0.1 , control_guidance_end=0.2)[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=[0.1, 0.3] , control_guidance_end=[0.2, 0.7])[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=0.4 , control_guidance_end=[0.5, 0.8])[0]
# make sure that all outputs are different
assert np.sum(np.abs(output_a - output_a)) > 1e-3
assert np.sum(np.abs(output_a - output_a)) > 1e-3
assert np.sum(np.abs(output_a - output_a)) > 1e-3
def lowerCamelCase ( self : Dict):
"""simple docstring"""
return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3)
@unittest.skipIf(
torch_device != '''cuda''' or not is_xformers_available() , reason='''XFormers attention is only available with CUDA and `xformers` installed''' , )
def lowerCamelCase ( self : int):
"""simple docstring"""
self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3)
def lowerCamelCase ( self : int):
"""simple docstring"""
self._test_inference_batch_single_identical(expected_max_diff=2e-3)
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = self.get_dummy_components()
UpperCAmelCase_ = self.pipeline_class(**_snake_case)
pipe.to(_snake_case)
pipe.set_progress_bar_config(disable=_snake_case)
with tempfile.TemporaryDirectory() as tmpdir:
try:
# save_pretrained is not implemented for Multi-ControlNet
pipe.save_pretrained(_snake_case)
except NotImplementedError:
pass
@slow
@require_torch_gpu
class __snake_case ( unittest.TestCase ):
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = ControlNetModel.from_pretrained('''lllyasviel/sd-controlnet-canny''')
UpperCAmelCase_ = StableDiffusionControlNetImgaImgPipeline.from_pretrained(
'''runwayml/stable-diffusion-v1-5''' , safety_checker=_snake_case , controlnet=_snake_case)
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=_snake_case)
UpperCAmelCase_ = torch.Generator(device='''cpu''').manual_seed(0)
UpperCAmelCase_ = '''evil space-punk bird'''
UpperCAmelCase_ = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/bird_canny.png''').resize((512, 512))
UpperCAmelCase_ = load_image(
'''https://huggingface.co/lllyasviel/sd-controlnet-canny/resolve/main/images/bird.png''').resize((512, 512))
UpperCAmelCase_ = pipe(
_snake_case , _snake_case , control_image=_snake_case , generator=_snake_case , output_type='''np''' , num_inference_steps=50 , strength=0.6 , )
UpperCAmelCase_ = output.images[0]
assert image.shape == (512, 512, 3)
UpperCAmelCase_ = load_numpy(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/img2img.npy''')
assert np.abs(expected_image - image).max() < 9e-2
| 51
| 0
|
'''simple docstring'''
class a_ :
'''simple docstring'''
def __init__( self , A , A ) -> int:
_SCREAMING_SNAKE_CASE = name
_SCREAMING_SNAKE_CASE = val
def __str__( self ) -> Dict:
return f'{self.__class__.__name__}({self.name}, {self.val})'
def __lt__( self , A ) -> Any:
return self.val < other.val
class a_ :
'''simple docstring'''
def __init__( self , A ) -> Optional[Any]:
_SCREAMING_SNAKE_CASE = {}
_SCREAMING_SNAKE_CASE = {}
_SCREAMING_SNAKE_CASE = self.build_heap(A )
def __getitem__( self , A ) -> Any:
return self.get_value(A )
def snake_case_( self , A ) -> Optional[int]:
return (idx - 1) // 2
def snake_case_( self , A ) -> List[Any]:
return idx * 2 + 1
def snake_case_( self , A ) -> Dict:
return idx * 2 + 2
def snake_case_( self , A ) -> Tuple:
return self.heap_dict[key]
def snake_case_( self , A ) -> Optional[int]:
_SCREAMING_SNAKE_CASE = len(A ) - 1
_SCREAMING_SNAKE_CASE = self.get_parent_idx(A )
for idx, i in enumerate(A ):
_SCREAMING_SNAKE_CASE = idx
_SCREAMING_SNAKE_CASE = i.val
for i in range(A , -1 , -1 ):
self.sift_down(A , A )
return array
def snake_case_( self , A , A ) -> List[str]:
while True:
_SCREAMING_SNAKE_CASE = self.get_left_child_idx(A ) # noqa: E741
_SCREAMING_SNAKE_CASE = self.get_right_child_idx(A )
_SCREAMING_SNAKE_CASE = idx
if l < len(A ) and array[l] < array[idx]:
_SCREAMING_SNAKE_CASE = l
if r < len(A ) and array[r] < array[smallest]:
_SCREAMING_SNAKE_CASE = r
if smallest != idx:
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = array[smallest], array[idx]
(
(
_SCREAMING_SNAKE_CASE
) , (
_SCREAMING_SNAKE_CASE
) ,
) = (
self.idx_of_element[array[smallest]],
self.idx_of_element[array[idx]],
)
_SCREAMING_SNAKE_CASE = smallest
else:
break
def snake_case_( self , A ) -> int:
_SCREAMING_SNAKE_CASE = self.get_parent_idx(A )
while p >= 0 and self.heap[p] > self.heap[idx]:
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = self.heap[idx], self.heap[p]
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = (
self.idx_of_element[self.heap[idx]],
self.idx_of_element[self.heap[p]],
)
_SCREAMING_SNAKE_CASE = p
_SCREAMING_SNAKE_CASE = self.get_parent_idx(A )
def snake_case_( self ) -> Optional[Any]:
return self.heap[0]
def snake_case_( self ) -> Dict:
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = self.heap[-1], self.heap[0]
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = (
self.idx_of_element[self.heap[-1]],
self.idx_of_element[self.heap[0]],
)
_SCREAMING_SNAKE_CASE = self.heap.pop()
del self.idx_of_element[x]
self.sift_down(0 , self.heap )
return x
def snake_case_( self , A ) -> Optional[int]:
self.heap.append(A )
_SCREAMING_SNAKE_CASE = len(self.heap ) - 1
_SCREAMING_SNAKE_CASE = node.val
self.sift_up(len(self.heap ) - 1 )
def snake_case_( self ) -> Optional[int]:
return len(self.heap ) == 0
def snake_case_( self , A , A ) -> Any:
assert (
self.heap[self.idx_of_element[node]].val > new_value
), "newValue must be less that current value"
_SCREAMING_SNAKE_CASE = new_value
_SCREAMING_SNAKE_CASE = new_value
self.sift_up(self.idx_of_element[node] )
lowercase_ = Node("""R""", -1)
lowercase_ = Node("""B""", 6)
lowercase_ = Node("""A""", 3)
lowercase_ = Node("""X""", 1)
lowercase_ = Node("""E""", 4)
# Use one of these two ways to generate Min-Heap
# Generating Min-Heap from array
lowercase_ = MinHeap([r, b, a, x, e])
# Generating Min-Heap by Insert method
# myMinHeap.insert(a)
# myMinHeap.insert(b)
# myMinHeap.insert(x)
# myMinHeap.insert(r)
# myMinHeap.insert(e)
# Before
print("""Min Heap - before decrease key""")
for i in my_min_heap.heap:
print(i)
print("""Min Heap - After decrease key of node [B -> -17]""")
my_min_heap.decrease_key(b, -17)
# After
for i in my_min_heap.heap:
print(i)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 58
|
import random
import timeit
from functools import wraps
from typing import Callable, Optional
from ..configuration_utils import PretrainedConfig
from ..models.auto.modeling_tf_auto import TF_MODEL_MAPPING, TF_MODEL_WITH_LM_HEAD_MAPPING
from ..utils import is_pyanvml_available, is_tf_available, logging
from .benchmark_utils import (
Benchmark,
Memory,
MemorySummary,
measure_peak_memory_cpu,
start_memory_tracing,
stop_memory_tracing,
)
if is_tf_available():
import tensorflow as tf
from tensorflow.python.framework.errors_impl import ResourceExhaustedError
from .benchmark_args_tf import TensorFlowBenchmarkArguments
if is_pyanvml_available():
import pyanvml.pyanvml as nvml
snake_case_ : Tuple = logging.get_logger(__name__)
def A (__A : bool , __A : bool ) -> Optional[Any]:
"""simple docstring"""
def run_func(__A : Optional[Any] ):
@wraps(__A )
def run_in_eager_mode(*__A : Dict , **__A : List[Any] ):
return func(*__A , **__A )
@wraps(__A )
@tf.function(experimental_compile=__A )
def run_in_graph_mode(*__A : Optional[Any] , **__A : Any ):
return func(*__A , **__A )
if do_eager_mode is True:
if use_xla is not False:
raise ValueError(
'''Cannot run model in XLA, if `args.eager_mode` is set to `True`. Please set `args.eager_mode=False`.''' )
return run_in_eager_mode
else:
return run_in_graph_mode
return run_func
def A (__A : int , __A : int , __A : int ) -> ["tf.Tensor"]:
"""simple docstring"""
UpperCAmelCase_ = random.Random()
UpperCAmelCase_ = [rng.randint(0 , vocab_size - 1 ) for i in range(batch_size * sequence_length )]
return tf.constant(__A , shape=(batch_size, sequence_length) , dtype=tf.intaa )
class __snake_case ( a ):
UpperCAmelCase__ : TensorFlowBenchmarkArguments
UpperCAmelCase__ : PretrainedConfig
UpperCAmelCase__ : str = "TensorFlow"
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return tf.__version__
def lowerCamelCase ( self : Dict , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case)
return self._measure_speed(_inference)
def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case)
return self._measure_speed(_train)
def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
if self.args.is_gpu:
tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case)
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case)
return self._measure_memory(_inference)
def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
if self.args.is_gpu:
tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case)
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case)
return self._measure_memory(_train)
def lowerCamelCase ( self : Optional[int] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.config_dict[model_name]
if self.args.fpaa:
raise NotImplementedError('''Mixed precision is currently not supported.''')
UpperCAmelCase_ = (
hasattr(_snake_case , '''architectures''')
and isinstance(config.architectures , _snake_case)
and len(config.architectures) > 0
)
if not self.args.only_pretrain_model and has_model_class_in_config:
try:
UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model
UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class])
UpperCAmelCase_ = getattr(_snake_case , _snake_case)
UpperCAmelCase_ = model_cls(_snake_case)
except ImportError:
raise ImportError(
F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to"""
''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''')
else:
UpperCAmelCase_ = TF_MODEL_MAPPING[config.__class__](_snake_case)
# encoder-decoder has vocab size saved differently
UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size
UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_decoder_forward():
return model(_snake_case , decoder_input_ids=_snake_case , training=_snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_forward():
return model(_snake_case , training=_snake_case)
UpperCAmelCase_ = encoder_decoder_forward if config.is_encoder_decoder else encoder_forward
return _inference
def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.config_dict[model_name]
if self.args.eager_mode is not False:
raise ValueError('''Training cannot be done in eager mode. Please make sure that `args.eager_mode = False`.''')
if self.args.fpaa:
raise NotImplementedError('''Mixed precision is currently not supported.''')
UpperCAmelCase_ = (
hasattr(_snake_case , '''architectures''')
and isinstance(config.architectures , _snake_case)
and len(config.architectures) > 0
)
if not self.args.only_pretrain_model and has_model_class_in_config:
try:
UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model
UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class])
UpperCAmelCase_ = getattr(_snake_case , _snake_case)
UpperCAmelCase_ = model_cls(_snake_case)
except ImportError:
raise ImportError(
F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to"""
''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''')
else:
UpperCAmelCase_ = TF_MODEL_WITH_LM_HEAD_MAPPING[config.__class__](_snake_case)
# encoder-decoder has vocab size saved differently
UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size
UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_decoder_train():
UpperCAmelCase_ = model(_snake_case , decoder_input_ids=_snake_case , labels=_snake_case , training=_snake_case)[0]
UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables)
return gradients
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_train():
UpperCAmelCase_ = model(_snake_case , labels=_snake_case , training=_snake_case)[0]
UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables)
return gradients
UpperCAmelCase_ = encoder_decoder_train if config.is_encoder_decoder else encoder_train
return _train
def lowerCamelCase ( self : Any , _snake_case : Optional[Any]):
"""simple docstring"""
with self.args.strategy.scope():
try:
if self.args.is_tpu or self.args.use_xla:
# run additional 10 times to stabilize compilation for tpu
logger.info('''Do inference on TPU. Running model 5 times to stabilize compilation''')
timeit.repeat(_snake_case , repeat=1 , number=5)
# as written in https://docs.python.org/2/library/timeit.html#timeit.Timer.repeat, min should be taken rather than the average
UpperCAmelCase_ = timeit.repeat(
_snake_case , repeat=self.args.repeat , number=10 , )
return min(_snake_case) / 1_0.0
except ResourceExhaustedError as e:
self.print_fn(F"""Doesn't fit on GPU. {e}""")
def lowerCamelCase ( self : Dict , _snake_case : Callable[[], None]):
"""simple docstring"""
logger.info(
'''Note that TensorFlow allocates more memory than '''
'''it might need to speed up computation. '''
'''The memory reported here corresponds to the memory '''
'''reported by `nvidia-smi`, which can vary depending '''
'''on total available memory on the GPU that is used.''')
with self.args.strategy.scope():
try:
if self.args.trace_memory_line_by_line:
if not self.args.eager_mode:
raise ValueError(
'''`args.eager_mode` is set to `False`. Make sure to run model in eager mode to measure memory'''
''' consumption line by line.''')
UpperCAmelCase_ = start_memory_tracing('''transformers''')
if self.args.is_tpu:
# tpu
raise NotImplementedError(
'''Memory Benchmarking is currently not implemented for TPU. Please disable memory benchmarking'''
''' with `args.memory=False`''')
elif self.args.is_gpu:
# gpu
if not is_pyanvml_available():
logger.warning(
'''py3nvml not installed, we won\'t log GPU memory usage. '''
'''Install py3nvml (pip install py3nvml) to log information about GPU.''')
UpperCAmelCase_ = '''N/A'''
else:
logger.info(
'''Measuring total GPU usage on GPU device. Make sure to not have additional processes'''
''' running on the same GPU.''')
# init nvml
nvml.nvmlInit()
func()
UpperCAmelCase_ = nvml.nvmlDeviceGetHandleByIndex(self.args.device_idx)
UpperCAmelCase_ = nvml.nvmlDeviceGetMemoryInfo(_snake_case)
UpperCAmelCase_ = meminfo.used
UpperCAmelCase_ = Memory(_snake_case)
# shutdown nvml
nvml.nvmlShutdown()
else:
# cpu
if self.args.trace_memory_line_by_line:
logger.info(
'''When enabling line by line tracing, the max peak memory for CPU is inaccurate in'''
''' TensorFlow.''')
UpperCAmelCase_ = None
else:
UpperCAmelCase_ = measure_peak_memory_cpu(_snake_case)
UpperCAmelCase_ = Memory(_snake_case) if isinstance(_snake_case , _snake_case) else memory_bytes
if self.args.trace_memory_line_by_line:
UpperCAmelCase_ = stop_memory_tracing(_snake_case)
if memory is None:
UpperCAmelCase_ = summary.total
else:
UpperCAmelCase_ = None
return memory, summary
except ResourceExhaustedError as e:
self.print_fn(F"""Doesn't fit on GPU. {e}""")
return "N/A", None
| 51
| 0
|
import multiprocessing
from typing import TYPE_CHECKING, Optional, Union
from .. import Dataset, Features, config
from ..formatting import query_table
from ..packaged_modules.sql.sql import Sql
from ..utils import logging
from .abc import AbstractDatasetInputStream
if TYPE_CHECKING:
import sqlitea
import sqlalchemy
class UpperCAmelCase ( A_ ):
def __init__(self : Tuple , snake_case__ : Union[str, "sqlalchemy.sql.Selectable"] , snake_case__ : Union[str, "sqlalchemy.engine.Connection", "sqlalchemy.engine.Engine", "sqlite3.Connection"] , snake_case__ : Optional[Features] = None , snake_case__ : str = None , snake_case__ : bool = False , **snake_case__ : Dict , ) -> Optional[int]:
'''simple docstring'''
super().__init__(features=snake_case__ , cache_dir=snake_case__ , keep_in_memory=snake_case__ , **snake_case__ )
snake_case : Tuple = Sql(
cache_dir=snake_case__ , features=snake_case__ , sql=snake_case__ , con=snake_case__ , **snake_case__ , )
def _SCREAMING_SNAKE_CASE (self : Any ) -> Optional[int]:
'''simple docstring'''
snake_case : Optional[int] = None
snake_case : int = None
snake_case : List[str] = None
snake_case : int = None
self.builder.download_and_prepare(
download_config=snake_case__ , download_mode=snake_case__ , verification_mode=snake_case__ , base_path=snake_case__ , )
# Build dataset for splits
snake_case : Optional[int] = self.builder.as_dataset(
split="train" , verification_mode=snake_case__ , in_memory=self.keep_in_memory )
return dataset
class UpperCAmelCase :
def __init__(self : Union[str, Any] , snake_case__ : Dataset , snake_case__ : str , snake_case__ : Union[str, "sqlalchemy.engine.Connection", "sqlalchemy.engine.Engine", "sqlite3.Connection"] , snake_case__ : Optional[int] = None , snake_case__ : Optional[int] = None , **snake_case__ : Dict , ) -> Optional[Any]:
'''simple docstring'''
if num_proc is not None and num_proc <= 0:
raise ValueError(f"""num_proc {num_proc} must be an integer > 0.""" )
snake_case : Any = dataset
snake_case : Dict = name
snake_case : Optional[Any] = con
snake_case : Any = batch_size if batch_size else config.DEFAULT_MAX_BATCH_SIZE
snake_case : Optional[int] = num_proc
snake_case : str = to_sql_kwargs
def _SCREAMING_SNAKE_CASE (self : List[Any] ) -> int:
'''simple docstring'''
snake_case : int = self.to_sql_kwargs.pop("sql" , snake_case__ )
snake_case : Tuple = self.to_sql_kwargs.pop("con" , snake_case__ )
snake_case : Tuple = self.to_sql_kwargs.pop("index" , snake_case__ )
snake_case : Union[str, Any] = self._write(index=snake_case__ , **self.to_sql_kwargs )
return written
def _SCREAMING_SNAKE_CASE (self : str , snake_case__ : Tuple ) -> Optional[Any]:
'''simple docstring'''
snake_case , snake_case , snake_case : Tuple = args
snake_case : Optional[int] = {**to_sql_kwargs, "if_exists": "append"} if offset > 0 else to_sql_kwargs
snake_case : List[Any] = query_table(
table=self.dataset.data , key=slice(snake_case__ , offset + self.batch_size ) , indices=self.dataset._indices , )
snake_case : List[Any] = batch.to_pandas()
snake_case : Optional[Any] = df.to_sql(self.name , self.con , index=snake_case__ , **snake_case__ )
return num_rows or len(snake_case__ )
def _SCREAMING_SNAKE_CASE (self : List[Any] , snake_case__ : str , **snake_case__ : Union[str, Any] ) -> int:
'''simple docstring'''
snake_case : List[str] = 0
if self.num_proc is None or self.num_proc == 1:
for offset in logging.tqdm(
range(0 , len(self.dataset ) , self.batch_size ) , unit="ba" , disable=not logging.is_progress_bar_enabled() , desc="Creating SQL from Arrow format" , ):
written += self._batch_sql((offset, index, to_sql_kwargs) )
else:
snake_case , snake_case : List[str] = len(self.dataset ), self.batch_size
with multiprocessing.Pool(self.num_proc ) as pool:
for num_rows in logging.tqdm(
pool.imap(
self._batch_sql , [(offset, index, to_sql_kwargs) for offset in range(0 , snake_case__ , snake_case__ )] , ) , total=(num_rows // batch_size) + 1 if num_rows % batch_size else num_rows // batch_size , unit="ba" , disable=not logging.is_progress_bar_enabled() , desc="Creating SQL from Arrow format" , ):
written += num_rows
return written
| 59
|
import hashlib
import unittest
from transformers import MODEL_FOR_DEPTH_ESTIMATION_MAPPING, is_torch_available, is_vision_available
from transformers.pipelines import DepthEstimationPipeline, pipeline
from transformers.testing_utils import (
is_pipeline_test,
nested_simplify,
require_tf,
require_timm,
require_torch,
require_vision,
slow,
)
from .test_pipelines_common import ANY
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
else:
class __snake_case :
@staticmethod
def lowerCamelCase ( *_snake_case : Optional[int] , **_snake_case : int):
"""simple docstring"""
pass
def A (__A : Image ) -> str:
"""simple docstring"""
UpperCAmelCase_ = hashlib.mda(image.tobytes() )
return m.hexdigest()
@is_pipeline_test
@require_vision
@require_timm
@require_torch
class __snake_case ( unittest.TestCase ):
UpperCAmelCase__ : Tuple = MODEL_FOR_DEPTH_ESTIMATION_MAPPING
def lowerCamelCase ( self : Tuple , _snake_case : Optional[Any] , _snake_case : Optional[Any] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = DepthEstimationPipeline(model=_snake_case , image_processor=_snake_case)
return depth_estimator, [
"./tests/fixtures/tests_samples/COCO/000000039769.png",
"./tests/fixtures/tests_samples/COCO/000000039769.png",
]
def lowerCamelCase ( self : str , _snake_case : Optional[int] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = depth_estimator('''./tests/fixtures/tests_samples/COCO/000000039769.png''')
self.assertEqual({'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)} , _snake_case)
import datasets
UpperCAmelCase_ = datasets.load_dataset('''hf-internal-testing/fixtures_image_utils''' , '''image''' , split='''test''')
UpperCAmelCase_ = depth_estimator(
[
Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png'''),
'''http://images.cocodataset.org/val2017/000000039769.jpg''',
# RGBA
dataset[0]['''file'''],
# LA
dataset[1]['''file'''],
# L
dataset[2]['''file'''],
])
self.assertEqual(
[
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
] , _snake_case , )
@require_tf
@unittest.skip('''Depth estimation is not implemented in TF''')
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
pass
@slow
@require_torch
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = '''Intel/dpt-large'''
UpperCAmelCase_ = pipeline('''depth-estimation''' , model=_snake_case)
UpperCAmelCase_ = depth_estimator('''http://images.cocodataset.org/val2017/000000039769.jpg''')
UpperCAmelCase_ = hashimage(outputs['''depth'''])
# This seems flaky.
# self.assertEqual(outputs["depth"], "1a39394e282e9f3b0741a90b9f108977")
self.assertEqual(nested_simplify(outputs['''predicted_depth'''].max().item()) , 2_9.3_0_4)
self.assertEqual(nested_simplify(outputs['''predicted_depth'''].min().item()) , 2.6_6_2)
@require_torch
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
self.skipTest('''There is not hf-internal-testing tiny model for either GLPN nor DPT''')
| 51
| 0
|
"""simple docstring"""
from __future__ import annotations
import copy
import inspect
import unittest
import numpy as np
from transformers import is_tf_available, is_vision_available
from transformers.models.auto import get_values
from transformers.testing_utils import require_tf, slow
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, floats_tensor, ids_tensor, random_attention_mask
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import tensorflow as tf
from transformers import (
TF_LAYOUTLMV3_PRETRAINED_MODEL_ARCHIVE_LIST,
TF_MODEL_FOR_MULTIPLE_CHOICE_MAPPING,
TF_MODEL_FOR_QUESTION_ANSWERING_MAPPING,
TF_MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING,
TF_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING,
LayoutLMvaConfig,
TFLayoutLMvaForQuestionAnswering,
TFLayoutLMvaForSequenceClassification,
TFLayoutLMvaForTokenClassification,
TFLayoutLMvaModel,
)
if is_vision_available():
from PIL import Image
from transformers import LayoutLMvaImageProcessor
class snake_case_:
def __init__( self : Tuple , UpperCamelCase_ : Any , UpperCamelCase_ : List[Any]=2 , UpperCamelCase_ : Dict=3 , UpperCamelCase_ : List[str]=4 , UpperCamelCase_ : Any=2 , UpperCamelCase_ : List[str]=7 , UpperCamelCase_ : Dict=True , UpperCamelCase_ : int=True , UpperCamelCase_ : int=True , UpperCamelCase_ : Union[str, Any]=True , UpperCamelCase_ : List[str]=9_9 , UpperCamelCase_ : Optional[Any]=3_6 , UpperCamelCase_ : Dict=2 , UpperCamelCase_ : Optional[Any]=4 , UpperCamelCase_ : Tuple=3_7 , UpperCamelCase_ : str="gelu" , UpperCamelCase_ : Union[str, Any]=0.1 , UpperCamelCase_ : Optional[int]=0.1 , UpperCamelCase_ : Dict=5_1_2 , UpperCamelCase_ : Dict=1_6 , UpperCamelCase_ : List[Any]=2 , UpperCamelCase_ : str=0.02 , UpperCamelCase_ : Tuple=6 , UpperCamelCase_ : List[Any]=6 , UpperCamelCase_ : Optional[int]=3 , UpperCamelCase_ : Any=4 , UpperCamelCase_ : Union[str, Any]=None , UpperCamelCase_ : str=1_0_0_0 , ):
lowerCAmelCase : Tuple = parent
lowerCAmelCase : Dict = batch_size
lowerCAmelCase : Optional[int] = num_channels
lowerCAmelCase : List[Any] = image_size
lowerCAmelCase : Optional[Any] = patch_size
lowerCAmelCase : Union[str, Any] = is_training
lowerCAmelCase : Optional[int] = use_input_mask
lowerCAmelCase : List[Any] = use_token_type_ids
lowerCAmelCase : Dict = use_labels
lowerCAmelCase : str = vocab_size
lowerCAmelCase : Any = hidden_size
lowerCAmelCase : List[str] = num_hidden_layers
lowerCAmelCase : str = num_attention_heads
lowerCAmelCase : Optional[int] = intermediate_size
lowerCAmelCase : Tuple = hidden_act
lowerCAmelCase : Optional[Any] = hidden_dropout_prob
lowerCAmelCase : Optional[int] = attention_probs_dropout_prob
lowerCAmelCase : Union[str, Any] = max_position_embeddings
lowerCAmelCase : int = type_vocab_size
lowerCAmelCase : List[Any] = type_sequence_label_size
lowerCAmelCase : List[Any] = initializer_range
lowerCAmelCase : List[str] = coordinate_size
lowerCAmelCase : Any = shape_size
lowerCAmelCase : Tuple = num_labels
lowerCAmelCase : Union[str, Any] = num_choices
lowerCAmelCase : List[str] = scope
lowerCAmelCase : Any = range_bbox
# LayoutLMv3's sequence length equals the number of text tokens + number of patches + 1 (we add 1 for the CLS token)
lowerCAmelCase : Dict = text_seq_length
lowerCAmelCase : Tuple = (image_size // patch_size) ** 2 + 1
lowerCAmelCase : Tuple = self.text_seq_length + self.image_seq_length
def lowerCamelCase__ ( self : List[str] ):
lowerCAmelCase : List[str] = ids_tensor([self.batch_size, self.text_seq_length] , self.vocab_size )
lowerCAmelCase : Union[str, Any] = ids_tensor([self.batch_size, self.text_seq_length, 4] , self.range_bbox )
lowerCAmelCase : Optional[int] = bbox.numpy()
# Ensure that bbox is legal
for i in range(bbox.shape[0] ):
for j in range(bbox.shape[1] ):
if bbox[i, j, 3] < bbox[i, j, 1]:
lowerCAmelCase : List[str] = bbox[i, j, 3]
lowerCAmelCase : str = bbox[i, j, 1]
lowerCAmelCase : Dict = tmp_coordinate
if bbox[i, j, 2] < bbox[i, j, 0]:
lowerCAmelCase : int = bbox[i, j, 2]
lowerCAmelCase : Union[str, Any] = bbox[i, j, 0]
lowerCAmelCase : List[str] = tmp_coordinate
lowerCAmelCase : Dict = tf.constant(UpperCamelCase_ )
lowerCAmelCase : str = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] )
lowerCAmelCase : Optional[int] = None
if self.use_input_mask:
lowerCAmelCase : Any = random_attention_mask([self.batch_size, self.text_seq_length] )
lowerCAmelCase : str = None
if self.use_token_type_ids:
lowerCAmelCase : Any = ids_tensor([self.batch_size, self.text_seq_length] , self.type_vocab_size )
lowerCAmelCase : List[str] = None
lowerCAmelCase : Optional[Any] = None
if self.use_labels:
lowerCAmelCase : Tuple = ids_tensor([self.batch_size] , self.type_sequence_label_size )
lowerCAmelCase : Any = ids_tensor([self.batch_size, self.text_seq_length] , self.num_labels )
lowerCAmelCase : List[str] = LayoutLMvaConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , initializer_range=self.initializer_range , coordinate_size=self.coordinate_size , shape_size=self.shape_size , input_size=self.image_size , patch_size=self.patch_size , )
return config, input_ids, bbox, pixel_values, token_type_ids, input_mask, sequence_labels, token_labels
def lowerCamelCase__ ( self : Dict , UpperCamelCase_ : Tuple , UpperCamelCase_ : Optional[int] , UpperCamelCase_ : Any , UpperCamelCase_ : List[Any] , UpperCamelCase_ : List[str] , UpperCamelCase_ : List[Any] ):
lowerCAmelCase : List[Any] = TFLayoutLMvaModel(config=UpperCamelCase_ )
# text + image
lowerCAmelCase : List[Any] = model(UpperCamelCase_ , pixel_values=UpperCamelCase_ , training=UpperCamelCase_ )
lowerCAmelCase : str = model(
UpperCamelCase_ , bbox=UpperCamelCase_ , pixel_values=UpperCamelCase_ , attention_mask=UpperCamelCase_ , token_type_ids=UpperCamelCase_ , training=UpperCamelCase_ , )
lowerCAmelCase : Dict = model(UpperCamelCase_ , bbox=UpperCamelCase_ , pixel_values=UpperCamelCase_ , training=UpperCamelCase_ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
# text only
lowerCAmelCase : Optional[int] = model(UpperCamelCase_ , training=UpperCamelCase_ )
self.parent.assertEqual(
result.last_hidden_state.shape , (self.batch_size, self.text_seq_length, self.hidden_size) )
# image only
lowerCAmelCase : int = model({'''pixel_values''': pixel_values} , training=UpperCamelCase_ )
self.parent.assertEqual(
result.last_hidden_state.shape , (self.batch_size, self.image_seq_length, self.hidden_size) )
def lowerCamelCase__ ( self : Any , UpperCamelCase_ : Dict , UpperCamelCase_ : List[str] , UpperCamelCase_ : int , UpperCamelCase_ : List[Any] , UpperCamelCase_ : Union[str, Any] , UpperCamelCase_ : str , UpperCamelCase_ : Optional[int] ):
lowerCAmelCase : int = self.num_labels
lowerCAmelCase : str = TFLayoutLMvaForSequenceClassification(config=UpperCamelCase_ )
lowerCAmelCase : Dict = model(
UpperCamelCase_ , bbox=UpperCamelCase_ , pixel_values=UpperCamelCase_ , attention_mask=UpperCamelCase_ , token_type_ids=UpperCamelCase_ , labels=UpperCamelCase_ , training=UpperCamelCase_ , )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) )
def lowerCamelCase__ ( self : Any , UpperCamelCase_ : Union[str, Any] , UpperCamelCase_ : List[str] , UpperCamelCase_ : Optional[Any] , UpperCamelCase_ : Dict , UpperCamelCase_ : int , UpperCamelCase_ : int , UpperCamelCase_ : Union[str, Any] ):
lowerCAmelCase : Union[str, Any] = self.num_labels
lowerCAmelCase : Tuple = TFLayoutLMvaForTokenClassification(config=UpperCamelCase_ )
lowerCAmelCase : List[Any] = model(
UpperCamelCase_ , bbox=UpperCamelCase_ , pixel_values=UpperCamelCase_ , attention_mask=UpperCamelCase_ , token_type_ids=UpperCamelCase_ , labels=UpperCamelCase_ , training=UpperCamelCase_ , )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.text_seq_length, self.num_labels) )
def lowerCamelCase__ ( self : Any , UpperCamelCase_ : int , UpperCamelCase_ : str , UpperCamelCase_ : int , UpperCamelCase_ : Union[str, Any] , UpperCamelCase_ : Optional[int] , UpperCamelCase_ : Tuple , UpperCamelCase_ : Optional[Any] ):
lowerCAmelCase : Optional[int] = 2
lowerCAmelCase : int = TFLayoutLMvaForQuestionAnswering(config=UpperCamelCase_ )
lowerCAmelCase : Tuple = model(
UpperCamelCase_ , bbox=UpperCamelCase_ , pixel_values=UpperCamelCase_ , attention_mask=UpperCamelCase_ , token_type_ids=UpperCamelCase_ , start_positions=UpperCamelCase_ , end_positions=UpperCamelCase_ , training=UpperCamelCase_ , )
self.parent.assertEqual(result.start_logits.shape , (self.batch_size, self.seq_length) )
self.parent.assertEqual(result.end_logits.shape , (self.batch_size, self.seq_length) )
def lowerCamelCase__ ( self : Any ):
lowerCAmelCase : Union[str, Any] = self.prepare_config_and_inputs()
((lowerCAmelCase), (lowerCAmelCase), (lowerCAmelCase), (lowerCAmelCase), (lowerCAmelCase), (lowerCAmelCase), (lowerCAmelCase), (lowerCAmelCase)) : Optional[Any] = config_and_inputs
lowerCAmelCase : List[str] = {
'''input_ids''': input_ids,
'''bbox''': bbox,
'''pixel_values''': pixel_values,
'''token_type_ids''': token_type_ids,
'''attention_mask''': input_mask,
}
return config, inputs_dict
@require_tf
class snake_case_( a__ , a__ , unittest.TestCase ):
__UpperCamelCase = (
(
TFLayoutLMvaModel,
TFLayoutLMvaForQuestionAnswering,
TFLayoutLMvaForSequenceClassification,
TFLayoutLMvaForTokenClassification,
)
if is_tf_available()
else ()
)
__UpperCamelCase = (
{'''document-question-answering''': TFLayoutLMvaForQuestionAnswering, '''feature-extraction''': TFLayoutLMvaModel}
if is_tf_available()
else {}
)
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
def lowerCamelCase__ ( self : int , UpperCamelCase_ : int , UpperCamelCase_ : Tuple , UpperCamelCase_ : List[Any] , UpperCamelCase_ : List[Any] , UpperCamelCase_ : List[str] ):
return True
def lowerCamelCase__ ( self : List[Any] , UpperCamelCase_ : Dict , UpperCamelCase_ : int , UpperCamelCase_ : Optional[Any]=False ):
lowerCAmelCase : List[Any] = copy.deepcopy(UpperCamelCase_ )
if model_class in get_values(UpperCamelCase_ ):
lowerCAmelCase : List[str] = {
k: tf.tile(tf.expand_dims(UpperCamelCase_ , 1 ) , (1, self.model_tester.num_choices) + (1,) * (v.ndim - 1) )
if isinstance(UpperCamelCase_ , tf.Tensor ) and v.ndim > 0
else v
for k, v in inputs_dict.items()
}
if return_labels:
if model_class in get_values(UpperCamelCase_ ):
lowerCAmelCase : List[Any] = tf.ones(self.model_tester.batch_size , dtype=tf.intaa )
elif model_class in get_values(UpperCamelCase_ ):
lowerCAmelCase : Optional[int] = tf.zeros(self.model_tester.batch_size , dtype=tf.intaa )
lowerCAmelCase : Union[str, Any] = tf.zeros(self.model_tester.batch_size , dtype=tf.intaa )
elif model_class in get_values(UpperCamelCase_ ):
lowerCAmelCase : List[Any] = tf.zeros(self.model_tester.batch_size , dtype=tf.intaa )
elif model_class in get_values(UpperCamelCase_ ):
lowerCAmelCase : Optional[Any] = tf.zeros(
(self.model_tester.batch_size, self.model_tester.text_seq_length) , dtype=tf.intaa )
return inputs_dict
def lowerCamelCase__ ( self : Optional[int] ):
lowerCAmelCase : Any = TFLayoutLMvaModelTester(self )
lowerCAmelCase : Tuple = ConfigTester(self , config_class=UpperCamelCase_ , hidden_size=3_7 )
def lowerCamelCase__ ( self : Any ):
self.config_tester.run_common_tests()
def lowerCamelCase__ ( self : List[str] ):
lowerCAmelCase, lowerCAmelCase : List[str] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
lowerCAmelCase : int = model_class(UpperCamelCase_ )
if getattr(UpperCamelCase_ , '''hf_compute_loss''' , UpperCamelCase_ ):
# The number of elements in the loss should be the same as the number of elements in the label
lowerCAmelCase : List[Any] = self._prepare_for_class(inputs_dict.copy() , UpperCamelCase_ , return_labels=UpperCamelCase_ )
lowerCAmelCase : Union[str, Any] = prepared_for_class[
sorted(prepared_for_class.keys() - inputs_dict.keys() , reverse=UpperCamelCase_ )[0]
]
lowerCAmelCase : int = added_label.shape.as_list()[:1]
# Test that model correctly compute the loss with kwargs
lowerCAmelCase : Dict = self._prepare_for_class(inputs_dict.copy() , UpperCamelCase_ , return_labels=UpperCamelCase_ )
lowerCAmelCase : List[str] = prepared_for_class.pop('''input_ids''' )
lowerCAmelCase : Optional[Any] = model(UpperCamelCase_ , **UpperCamelCase_ )[0]
self.assertTrue(loss.shape.as_list() == expected_loss_size or loss.shape.as_list() == [1] )
# Test that model correctly compute the loss when we mask some positions
lowerCAmelCase : Optional[int] = self._prepare_for_class(inputs_dict.copy() , UpperCamelCase_ , return_labels=UpperCamelCase_ )
lowerCAmelCase : Union[str, Any] = prepared_for_class.pop('''input_ids''' )
if "labels" in prepared_for_class:
lowerCAmelCase : List[Any] = prepared_for_class['''labels'''].numpy()
if len(labels.shape ) > 1 and labels.shape[1] != 1:
lowerCAmelCase : List[str] = -1_0_0
lowerCAmelCase : Tuple = tf.convert_to_tensor(UpperCamelCase_ )
lowerCAmelCase : Optional[Any] = model(UpperCamelCase_ , **UpperCamelCase_ )[0]
self.assertTrue(loss.shape.as_list() == expected_loss_size or loss.shape.as_list() == [1] )
self.assertTrue(not np.any(np.isnan(loss.numpy() ) ) )
# Test that model correctly compute the loss with a dict
lowerCAmelCase : Optional[int] = self._prepare_for_class(inputs_dict.copy() , UpperCamelCase_ , return_labels=UpperCamelCase_ )
lowerCAmelCase : str = model(UpperCamelCase_ )[0]
self.assertTrue(loss.shape.as_list() == expected_loss_size or loss.shape.as_list() == [1] )
# Test that model correctly compute the loss with a tuple
lowerCAmelCase : List[Any] = self._prepare_for_class(inputs_dict.copy() , UpperCamelCase_ , return_labels=UpperCamelCase_ )
# Get keys that were added with the _prepare_for_class function
lowerCAmelCase : List[str] = prepared_for_class.keys() - inputs_dict.keys()
lowerCAmelCase : Dict = inspect.signature(model.call ).parameters
lowerCAmelCase : Optional[int] = list(signature.keys() )
# Create a dictionary holding the location of the tensors in the tuple
lowerCAmelCase : List[str] = {0: '''input_ids'''}
for label_key in label_keys:
lowerCAmelCase : str = signature_names.index(UpperCamelCase_ )
lowerCAmelCase : Any = label_key
lowerCAmelCase : Union[str, Any] = sorted(tuple_index_mapping.items() )
# Initialize a list with their default values, update the values and convert to a tuple
lowerCAmelCase : List[Any] = []
for name in signature_names:
if name != "kwargs":
list_input.append(signature[name].default )
for index, value in sorted_tuple_index_mapping:
lowerCAmelCase : Any = prepared_for_class[value]
lowerCAmelCase : Optional[Any] = tuple(UpperCamelCase_ )
# Send to model
lowerCAmelCase : Dict = model(tuple_input[:-1] )[0]
self.assertTrue(loss.shape.as_list() == expected_loss_size or loss.shape.as_list() == [1] )
def lowerCamelCase__ ( self : Union[str, Any] ):
(
(
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
),
) : Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ )
def lowerCamelCase__ ( self : List[str] ):
(
(
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
),
) : int = self.model_tester.prepare_config_and_inputs()
for type in ["absolute", "relative_key", "relative_key_query"]:
lowerCAmelCase : List[Any] = type
self.model_tester.create_and_check_model(UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ )
def lowerCamelCase__ ( self : Optional[int] ):
(
(
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
),
) : Any = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_sequence_classification(
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ )
def lowerCamelCase__ ( self : List[str] ):
(
(
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
),
) : int = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_token_classification(
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ )
def lowerCamelCase__ ( self : int ):
(
(
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
), (
lowerCAmelCase
),
) : str = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_question_answering(
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ )
@slow
def lowerCamelCase__ ( self : str ):
for model_name in TF_LAYOUTLMV3_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
lowerCAmelCase : Tuple = TFLayoutLMvaModel.from_pretrained(UpperCamelCase_ )
self.assertIsNotNone(UpperCamelCase_ )
def _snake_case ( ):
lowerCAmelCase : Optional[int] = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
return image
@require_tf
class snake_case_( unittest.TestCase ):
@cached_property
def lowerCamelCase__ ( self : Tuple ):
return LayoutLMvaImageProcessor(apply_ocr=UpperCamelCase_ ) if is_vision_available() else None
@slow
def lowerCamelCase__ ( self : List[Any] ):
lowerCAmelCase : Any = TFLayoutLMvaModel.from_pretrained('''microsoft/layoutlmv3-base''' )
lowerCAmelCase : Optional[Any] = self.default_image_processor
lowerCAmelCase : Any = prepare_img()
lowerCAmelCase : Optional[Any] = image_processor(images=UpperCamelCase_ , return_tensors='''tf''' ).pixel_values
lowerCAmelCase : List[Any] = tf.constant([[1, 2]] )
lowerCAmelCase : Optional[Any] = tf.expand_dims(tf.constant([[1, 2, 3, 4], [5, 6, 7, 8]] ) , axis=0 )
# forward pass
lowerCAmelCase : str = model(input_ids=UpperCamelCase_ , bbox=UpperCamelCase_ , pixel_values=UpperCamelCase_ , training=UpperCamelCase_ )
# verify the logits
lowerCAmelCase : Any = (1, 1_9_9, 7_6_8)
self.assertEqual(outputs.last_hidden_state.shape , UpperCamelCase_ )
lowerCAmelCase : Tuple = tf.constant(
[[-0.0_529, 0.3_618, 0.1_632], [-0.1_587, -0.1_667, -0.0_400], [-0.1_557, -0.1_671, -0.0_505]] )
self.assertTrue(np.allclose(outputs.last_hidden_state[0, :3, :3] , UpperCamelCase_ , atol=1E-4 ) )
| 60
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : int = {
"configuration_deberta": ["DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP", "DebertaConfig", "DebertaOnnxConfig"],
"tokenization_deberta": ["DebertaTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : int = ["DebertaTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"DebertaForMaskedLM",
"DebertaForQuestionAnswering",
"DebertaForSequenceClassification",
"DebertaForTokenClassification",
"DebertaModel",
"DebertaPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFDebertaForMaskedLM",
"TFDebertaForQuestionAnswering",
"TFDebertaForSequenceClassification",
"TFDebertaForTokenClassification",
"TFDebertaModel",
"TFDebertaPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_deberta import DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP, DebertaConfig, DebertaOnnxConfig
from .tokenization_deberta import DebertaTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_deberta_fast import DebertaTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_deberta import (
DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
DebertaForMaskedLM,
DebertaForQuestionAnswering,
DebertaForSequenceClassification,
DebertaForTokenClassification,
DebertaModel,
DebertaPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_deberta import (
TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
TFDebertaForMaskedLM,
TFDebertaForQuestionAnswering,
TFDebertaForSequenceClassification,
TFDebertaForTokenClassification,
TFDebertaModel,
TFDebertaPreTrainedModel,
)
else:
import sys
snake_case_ : Dict = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
"""simple docstring"""
def __a ( __lowerCamelCase, __lowerCamelCase ):
while b:
UpperCAmelCase_ , UpperCAmelCase_ : Dict = b, a % b
return a
def __a ( __lowerCamelCase, __lowerCamelCase ):
return a if b == 0 else euclidean_gcd_recursive(__lowerCamelCase, a % b )
def __a ( ):
print(f"""euclidean_gcd(3, 5) = {euclidean_gcd(3, 5 )}""" )
print(f"""euclidean_gcd(5, 3) = {euclidean_gcd(5, 3 )}""" )
print(f"""euclidean_gcd(1, 3) = {euclidean_gcd(1, 3 )}""" )
print(f"""euclidean_gcd(3, 6) = {euclidean_gcd(3, 6 )}""" )
print(f"""euclidean_gcd(6, 3) = {euclidean_gcd(6, 3 )}""" )
print(f"""euclidean_gcd_recursive(3, 5) = {euclidean_gcd_recursive(3, 5 )}""" )
print(f"""euclidean_gcd_recursive(5, 3) = {euclidean_gcd_recursive(5, 3 )}""" )
print(f"""euclidean_gcd_recursive(1, 3) = {euclidean_gcd_recursive(1, 3 )}""" )
print(f"""euclidean_gcd_recursive(3, 6) = {euclidean_gcd_recursive(3, 6 )}""" )
print(f"""euclidean_gcd_recursive(6, 3) = {euclidean_gcd_recursive(6, 3 )}""" )
if __name__ == "__main__":
main()
| 61
|
from typing import TYPE_CHECKING
from ...file_utils import _LazyModule, is_tokenizers_available, is_torch_available
from ...utils import OptionalDependencyNotAvailable
snake_case_ : Union[str, Any] = {"configuration_gpt_neox": ["GPT_NEOX_PRETRAINED_CONFIG_ARCHIVE_MAP", "GPTNeoXConfig"]}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Dict = ["GPTNeoXTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = [
"GPT_NEOX_PRETRAINED_MODEL_ARCHIVE_LIST",
"GPTNeoXForCausalLM",
"GPTNeoXForQuestionAnswering",
"GPTNeoXForSequenceClassification",
"GPTNeoXForTokenClassification",
"GPTNeoXLayer",
"GPTNeoXModel",
"GPTNeoXPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_gpt_neox import GPT_NEOX_PRETRAINED_CONFIG_ARCHIVE_MAP, GPTNeoXConfig
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_gpt_neox_fast import GPTNeoXTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_gpt_neox import (
GPT_NEOX_PRETRAINED_MODEL_ARCHIVE_LIST,
GPTNeoXForCausalLM,
GPTNeoXForQuestionAnswering,
GPTNeoXForSequenceClassification,
GPTNeoXForTokenClassification,
GPTNeoXLayer,
GPTNeoXModel,
GPTNeoXPreTrainedModel,
)
else:
import sys
snake_case_ : Optional[int] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
from __future__ import annotations
import numpy as np
from numpy import floataa
from numpy.typing import NDArray
def _UpperCAmelCase ( SCREAMING_SNAKE_CASE__ : NDArray[floataa] , SCREAMING_SNAKE_CASE__ : NDArray[floataa] , SCREAMING_SNAKE_CASE__ : list[int] , SCREAMING_SNAKE_CASE__ : int , ):
__UpperCamelCase , __UpperCamelCase =coefficient_matrix.shape
__UpperCamelCase , __UpperCamelCase =constant_matrix.shape
if rowsa != colsa:
__UpperCamelCase =F'Coefficient matrix dimensions must be nxn but received {rowsa}x{colsa}'
raise ValueError(SCREAMING_SNAKE_CASE__ )
if colsa != 1:
__UpperCamelCase =F'Constant matrix must be nx1 but received {rowsa}x{colsa}'
raise ValueError(SCREAMING_SNAKE_CASE__ )
if rowsa != rowsa:
__UpperCamelCase =(
'Coefficient and constant matrices dimensions must be nxn and nx1 but '
F'received {rowsa}x{colsa} and {rowsa}x{colsa}'
)
raise ValueError(SCREAMING_SNAKE_CASE__ )
if len(SCREAMING_SNAKE_CASE__ ) != rowsa:
__UpperCamelCase =(
'Number of initial values must be equal to number of rows in coefficient '
F'matrix but received {len(SCREAMING_SNAKE_CASE__ )} and {rowsa}'
)
raise ValueError(SCREAMING_SNAKE_CASE__ )
if iterations <= 0:
raise ValueError('Iterations must be at least 1' )
__UpperCamelCase =np.concatenate(
(coefficient_matrix, constant_matrix) , axis=1 )
__UpperCamelCase , __UpperCamelCase =table.shape
strictly_diagonally_dominant(SCREAMING_SNAKE_CASE__ )
# Iterates the whole matrix for given number of times
for _ in range(SCREAMING_SNAKE_CASE__ ):
__UpperCamelCase =[]
for row in range(SCREAMING_SNAKE_CASE__ ):
__UpperCamelCase =0
for col in range(SCREAMING_SNAKE_CASE__ ):
if col == row:
__UpperCamelCase =table[row][col]
elif col == cols - 1:
__UpperCamelCase =table[row][col]
else:
temp += (-1) * table[row][col] * init_val[col]
__UpperCamelCase =(temp + val) / denom
new_val.append(SCREAMING_SNAKE_CASE__ )
__UpperCamelCase =new_val
return [float(SCREAMING_SNAKE_CASE__ ) for i in new_val]
def _UpperCAmelCase ( SCREAMING_SNAKE_CASE__ : NDArray[floataa] ):
__UpperCamelCase , __UpperCamelCase =table.shape
__UpperCamelCase =True
for i in range(0 , SCREAMING_SNAKE_CASE__ ):
__UpperCamelCase =0
for j in range(0 , cols - 1 ):
if i == j:
continue
else:
total += table[i][j]
if table[i][i] <= total:
raise ValueError('Coefficient matrix is not strictly diagonally dominant' )
return is_diagonally_dominant
# Test Cases
if __name__ == "__main__":
import doctest
doctest.testmod()
| 62
|
def A (__A : list , __A : int , __A : int = 0 , __A : int = 0 ) -> int:
"""simple docstring"""
UpperCAmelCase_ = right or len(__A ) - 1
if left > right:
return -1
elif list_data[left] == key:
return left
elif list_data[right] == key:
return right
else:
return search(__A , __A , left + 1 , right - 1 )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 51
| 0
|
'''simple docstring'''
import argparse
import json
from pathlib import Path
import torch
import torchaudio
from datasets import load_dataset
from huggingface_hub import hf_hub_download
from transformers import ASTConfig, ASTFeatureExtractor, ASTForAudioClassification
from transformers.utils import logging
logging.set_verbosity_info()
lowerCAmelCase_ : Union[str, Any] = logging.get_logger(__name__)
def _lowerCamelCase ( lowercase : List[Any] ) -> str:
_a = ASTConfig()
if "10-10" in model_name:
pass
elif "speech-commands" in model_name:
_a = 128
elif "12-12" in model_name:
_a = 12
_a = 12
elif "14-14" in model_name:
_a = 14
_a = 14
elif "16-16" in model_name:
_a = 16
_a = 16
else:
raise ValueError("Model not supported" )
_a = "huggingface/label-files"
if "speech-commands" in model_name:
_a = 35
_a = "speech-commands-v2-id2label.json"
else:
_a = 527
_a = "audioset-id2label.json"
_a = json.load(open(hf_hub_download(lowercase , lowercase , repo_type="dataset" ) , "r" ) )
_a = {int(lowercase ): v for k, v in idalabel.items()}
_a = idalabel
_a = {v: k for k, v in idalabel.items()}
return config
def _lowerCamelCase ( lowercase : Dict ) -> Optional[Any]:
if "module.v" in name:
_a = name.replace("module.v" , "audio_spectrogram_transformer" )
if "cls_token" in name:
_a = name.replace("cls_token" , "embeddings.cls_token" )
if "dist_token" in name:
_a = name.replace("dist_token" , "embeddings.distillation_token" )
if "pos_embed" in name:
_a = name.replace("pos_embed" , "embeddings.position_embeddings" )
if "patch_embed.proj" in name:
_a = name.replace("patch_embed.proj" , "embeddings.patch_embeddings.projection" )
# transformer blocks
if "blocks" in name:
_a = name.replace("blocks" , "encoder.layer" )
if "attn.proj" in name:
_a = name.replace("attn.proj" , "attention.output.dense" )
if "attn" in name:
_a = name.replace("attn" , "attention.self" )
if "norm1" in name:
_a = name.replace("norm1" , "layernorm_before" )
if "norm2" in name:
_a = name.replace("norm2" , "layernorm_after" )
if "mlp.fc1" in name:
_a = name.replace("mlp.fc1" , "intermediate.dense" )
if "mlp.fc2" in name:
_a = name.replace("mlp.fc2" , "output.dense" )
# final layernorm
if "audio_spectrogram_transformer.norm" in name:
_a = name.replace("audio_spectrogram_transformer.norm" , "audio_spectrogram_transformer.layernorm" )
# classifier head
if "module.mlp_head.0" in name:
_a = name.replace("module.mlp_head.0" , "classifier.layernorm" )
if "module.mlp_head.1" in name:
_a = name.replace("module.mlp_head.1" , "classifier.dense" )
return name
def _lowerCamelCase ( lowercase : Any , lowercase : Tuple ) -> List[str]:
for key in orig_state_dict.copy().keys():
_a = orig_state_dict.pop(lowercase )
if "qkv" in key:
_a = key.split("." )
_a = int(key_split[3] )
_a = config.hidden_size
if "weight" in key:
_a = val[:dim, :]
_a = val[dim : dim * 2, :]
_a = val[-dim:, :]
else:
_a = val[:dim]
_a = val[dim : dim * 2]
_a = val[-dim:]
else:
_a = val
return orig_state_dict
def _lowerCamelCase ( lowercase : Optional[int] ) -> Optional[int]:
_a = [
"module.v.head.weight",
"module.v.head.bias",
"module.v.head_dist.weight",
"module.v.head_dist.bias",
]
for k in ignore_keys:
state_dict.pop(lowercase , lowercase )
@torch.no_grad()
def _lowerCamelCase ( lowercase : Union[str, Any] , lowercase : List[Any] , lowercase : Dict=False ) -> int:
_a = get_audio_spectrogram_transformer_config(lowercase )
_a = {
"ast-finetuned-audioset-10-10-0.4593": (
"https://www.dropbox.com/s/ca0b1v2nlxzyeb4/audioset_10_10_0.4593.pth?dl=1"
),
"ast-finetuned-audioset-10-10-0.450": (
"https://www.dropbox.com/s/1tv0hovue1bxupk/audioset_10_10_0.4495.pth?dl=1"
),
"ast-finetuned-audioset-10-10-0.448": (
"https://www.dropbox.com/s/6u5sikl4b9wo4u5/audioset_10_10_0.4483.pth?dl=1"
),
"ast-finetuned-audioset-10-10-0.448-v2": (
"https://www.dropbox.com/s/kt6i0v9fvfm1mbq/audioset_10_10_0.4475.pth?dl=1"
),
"ast-finetuned-audioset-12-12-0.447": (
"https://www.dropbox.com/s/snfhx3tizr4nuc8/audioset_12_12_0.4467.pth?dl=1"
),
"ast-finetuned-audioset-14-14-0.443": (
"https://www.dropbox.com/s/z18s6pemtnxm4k7/audioset_14_14_0.4431.pth?dl=1"
),
"ast-finetuned-audioset-16-16-0.442": (
"https://www.dropbox.com/s/mdsa4t1xmcimia6/audioset_16_16_0.4422.pth?dl=1"
),
"ast-finetuned-speech-commands-v2": (
"https://www.dropbox.com/s/q0tbqpwv44pquwy/speechcommands_10_10_0.9812.pth?dl=1"
),
}
# load original state_dict
_a = model_name_to_url[model_name]
_a = torch.hub.load_state_dict_from_url(lowercase , map_location="cpu" )
# remove some keys
remove_keys(lowercase )
# rename some keys
_a = convert_state_dict(lowercase , lowercase )
# load 🤗 model
_a = ASTForAudioClassification(lowercase )
model.eval()
model.load_state_dict(lowercase )
# verify outputs on dummy input
# source: https://github.com/YuanGongND/ast/blob/79e873b8a54d0a3b330dd522584ff2b9926cd581/src/run.py#L62
_a = -4.2_67_73_93 if "speech-commands" not in model_name else -6.84_59_78
_a = 4.5_68_99_74 if "speech-commands" not in model_name else 5.5_65_45_26
_a = 1024 if "speech-commands" not in model_name else 128
_a = ASTFeatureExtractor(mean=lowercase , std=lowercase , max_length=lowercase )
if "speech-commands" in model_name:
_a = load_dataset("speech_commands" , "v0.02" , split="validation" )
_a = dataset[0]["audio"]["array"]
else:
_a = hf_hub_download(
repo_id="nielsr/audio-spectogram-transformer-checkpoint" , filename="sample_audio.flac" , repo_type="dataset" , )
_a , _a = torchaudio.load(lowercase )
_a = waveform.squeeze().numpy()
_a = feature_extractor(lowercase , sampling_rate=1_6000 , return_tensors="pt" )
# forward pass
_a = model(**lowercase )
_a = outputs.logits
if model_name == "ast-finetuned-audioset-10-10-0.4593":
_a = torch.tensor([-0.87_60, -7.00_42, -8.66_02] )
elif model_name == "ast-finetuned-audioset-10-10-0.450":
_a = torch.tensor([-1.19_86, -7.09_03, -8.27_18] )
elif model_name == "ast-finetuned-audioset-10-10-0.448":
_a = torch.tensor([-2.61_28, -8.00_80, -9.43_44] )
elif model_name == "ast-finetuned-audioset-10-10-0.448-v2":
_a = torch.tensor([-1.50_80, -7.45_34, -8.89_17] )
elif model_name == "ast-finetuned-audioset-12-12-0.447":
_a = torch.tensor([-0.50_50, -6.58_33, -8.08_43] )
elif model_name == "ast-finetuned-audioset-14-14-0.443":
_a = torch.tensor([-0.38_26, -7.03_36, -8.24_13] )
elif model_name == "ast-finetuned-audioset-16-16-0.442":
_a = torch.tensor([-1.21_13, -6.91_01, -8.34_70] )
elif model_name == "ast-finetuned-speech-commands-v2":
_a = torch.tensor([6.15_89, -8.05_66, -8.79_84] )
else:
raise ValueError("Unknown model name" )
if not torch.allclose(logits[0, :3] , lowercase , atol=1E-4 ):
raise ValueError("Logits don't match" )
print("Looks ok!" )
if pytorch_dump_folder_path is not None:
Path(lowercase ).mkdir(exist_ok=lowercase )
print(F'Saving model {model_name} to {pytorch_dump_folder_path}' )
model.save_pretrained(lowercase )
print(F'Saving feature extractor to {pytorch_dump_folder_path}' )
feature_extractor.save_pretrained(lowercase )
if push_to_hub:
print("Pushing model and feature extractor to the hub..." )
model.push_to_hub(F'MIT/{model_name}' )
feature_extractor.push_to_hub(F'MIT/{model_name}' )
if __name__ == "__main__":
lowerCAmelCase_ : Tuple = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'--model_name',
default='ast-finetuned-audioset-10-10-0.4593',
type=str,
help='Name of the Audio Spectrogram Transformer model you\'d like to convert.',
)
parser.add_argument(
'--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.'
)
parser.add_argument(
'--push_to_hub', action='store_true', help='Whether or not to push the converted model to the 🤗 hub.'
)
lowerCAmelCase_ : Dict = parser.parse_args()
convert_audio_spectrogram_transformer_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub)
| 63
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : int = logging.get_logger(__name__)
snake_case_ : str = {}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''llama'''
UpperCAmelCase__ : Dict = ['''past_key_values''']
def __init__( self : str , _snake_case : List[str]=32000 , _snake_case : int=4096 , _snake_case : List[str]=11008 , _snake_case : Optional[int]=32 , _snake_case : List[Any]=32 , _snake_case : Tuple=None , _snake_case : int="silu" , _snake_case : List[Any]=2048 , _snake_case : List[str]=0.0_2 , _snake_case : Any=1e-6 , _snake_case : List[str]=True , _snake_case : Optional[Any]=0 , _snake_case : Dict=1 , _snake_case : List[Any]=2 , _snake_case : str=1 , _snake_case : Union[str, Any]=False , _snake_case : str=None , **_snake_case : List[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = max_position_embeddings
UpperCAmelCase_ = hidden_size
UpperCAmelCase_ = intermediate_size
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
# for backward compatibility
if num_key_value_heads is None:
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = num_key_value_heads
UpperCAmelCase_ = hidden_act
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = rms_norm_eps
UpperCAmelCase_ = pretraining_tp
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = rope_scaling
self._rope_scaling_validation()
super().__init__(
pad_token_id=_snake_case , bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case , )
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
if self.rope_scaling is None:
return
if not isinstance(self.rope_scaling , _snake_case) or len(self.rope_scaling) != 2:
raise ValueError(
'''`rope_scaling` must be a dictionary with with two fields, `name` and `factor`, '''
F"""got {self.rope_scaling}""")
UpperCAmelCase_ = self.rope_scaling.get('''type''' , _snake_case)
UpperCAmelCase_ = self.rope_scaling.get('''factor''' , _snake_case)
if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic"]:
raise ValueError(
F"""`rope_scaling`'s name field must be one of ['linear', 'dynamic'], got {rope_scaling_type}""")
if rope_scaling_factor is None or not isinstance(_snake_case , _snake_case) or rope_scaling_factor <= 1.0:
raise ValueError(F"""`rope_scaling`'s factor field must be an float > 1, got {rope_scaling_factor}""")
| 51
| 0
|
"""simple docstring"""
import fire
from transformers import AutoConfig, AutoModelForSeqaSeqLM, AutoTokenizer
def UpperCAmelCase__ (snake_case__ : str , snake_case__ : str , **snake_case__ : Tuple ):
"""simple docstring"""
_snake_case : List[Any] = AutoConfig.from_pretrained(snake_case__ , **snake_case__ )
_snake_case : Dict = AutoModelForSeqaSeqLM.from_config(snake_case__ )
model.save_pretrained(snake_case__ )
AutoTokenizer.from_pretrained(snake_case__ ).save_pretrained(snake_case__ )
return model
if __name__ == "__main__":
fire.Fire(save_randomly_initialized_version)
| 64
|
from collections import OrderedDict
from typing import Any, List, Mapping, Optional
from ... import PreTrainedTokenizer, TensorType, is_torch_available
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfigWithPast, PatchingSpec
from ...utils import logging
snake_case_ : List[str] = logging.get_logger(__name__)
snake_case_ : Tuple = {
"Salesforce/codegen-350M-nl": "https://huggingface.co/Salesforce/codegen-350M-nl/resolve/main/config.json",
"Salesforce/codegen-350M-multi": "https://huggingface.co/Salesforce/codegen-350M-multi/resolve/main/config.json",
"Salesforce/codegen-350M-mono": "https://huggingface.co/Salesforce/codegen-350M-mono/resolve/main/config.json",
"Salesforce/codegen-2B-nl": "https://huggingface.co/Salesforce/codegen-2B-nl/resolve/main/config.json",
"Salesforce/codegen-2B-multi": "https://huggingface.co/Salesforce/codegen-2B-multi/resolve/main/config.json",
"Salesforce/codegen-2B-mono": "https://huggingface.co/Salesforce/codegen-2B-mono/resolve/main/config.json",
"Salesforce/codegen-6B-nl": "https://huggingface.co/Salesforce/codegen-6B-nl/resolve/main/config.json",
"Salesforce/codegen-6B-multi": "https://huggingface.co/Salesforce/codegen-6B-multi/resolve/main/config.json",
"Salesforce/codegen-6B-mono": "https://huggingface.co/Salesforce/codegen-6B-mono/resolve/main/config.json",
"Salesforce/codegen-16B-nl": "https://huggingface.co/Salesforce/codegen-16B-nl/resolve/main/config.json",
"Salesforce/codegen-16B-multi": "https://huggingface.co/Salesforce/codegen-16B-multi/resolve/main/config.json",
"Salesforce/codegen-16B-mono": "https://huggingface.co/Salesforce/codegen-16B-mono/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''codegen'''
UpperCAmelCase__ : int = {
'''max_position_embeddings''': '''n_positions''',
'''hidden_size''': '''n_embd''',
'''num_attention_heads''': '''n_head''',
'''num_hidden_layers''': '''n_layer''',
}
def __init__( self : Union[str, Any] , _snake_case : Union[str, Any]=50400 , _snake_case : Optional[int]=2048 , _snake_case : Union[str, Any]=2048 , _snake_case : List[str]=4096 , _snake_case : Any=28 , _snake_case : List[str]=16 , _snake_case : int=64 , _snake_case : Tuple=None , _snake_case : Dict="gelu_new" , _snake_case : Union[str, Any]=0.0 , _snake_case : Optional[Any]=0.0 , _snake_case : List[Any]=0.0 , _snake_case : List[Any]=1e-5 , _snake_case : List[str]=0.0_2 , _snake_case : Optional[Any]=True , _snake_case : int=50256 , _snake_case : Tuple=50256 , _snake_case : int=False , **_snake_case : Any , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = n_ctx
UpperCAmelCase_ = n_positions
UpperCAmelCase_ = n_embd
UpperCAmelCase_ = n_layer
UpperCAmelCase_ = n_head
UpperCAmelCase_ = n_inner
UpperCAmelCase_ = rotary_dim
UpperCAmelCase_ = activation_function
UpperCAmelCase_ = resid_pdrop
UpperCAmelCase_ = embd_pdrop
UpperCAmelCase_ = attn_pdrop
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
super().__init__(
bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case)
class __snake_case ( a ):
def __init__( self : Tuple , _snake_case : PretrainedConfig , _snake_case : str = "default" , _snake_case : List[PatchingSpec] = None , _snake_case : bool = False , ):
"""simple docstring"""
super().__init__(_snake_case , task=_snake_case , patching_specs=_snake_case , use_past=_snake_case)
if not getattr(self._config , '''pad_token_id''' , _snake_case):
# TODO: how to do that better?
UpperCAmelCase_ = 0
@property
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = OrderedDict({'''input_ids''': {0: '''batch''', 1: '''sequence'''}})
if self.use_past:
self.fill_with_past_key_values_(_snake_case , direction='''inputs''')
UpperCAmelCase_ = {0: '''batch''', 1: '''past_sequence + sequence'''}
else:
UpperCAmelCase_ = {0: '''batch''', 1: '''sequence'''}
return common_inputs
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return self._config.n_layer
@property
def lowerCamelCase ( self : int):
"""simple docstring"""
return self._config.n_head
def lowerCamelCase ( self : Optional[int] , _snake_case : PreTrainedTokenizer , _snake_case : int = -1 , _snake_case : int = -1 , _snake_case : bool = False , _snake_case : Optional[TensorType] = None , ):
"""simple docstring"""
UpperCAmelCase_ = super(_snake_case , self).generate_dummy_inputs(
_snake_case , batch_size=_snake_case , seq_length=_snake_case , is_pair=_snake_case , framework=_snake_case)
# We need to order the input in the way they appears in the forward()
UpperCAmelCase_ = OrderedDict({'''input_ids''': common_inputs['''input_ids''']})
# Need to add the past_keys
if self.use_past:
if not is_torch_available():
raise ValueError('''Cannot generate dummy past_keys inputs without PyTorch installed.''')
else:
import torch
UpperCAmelCase_ , UpperCAmelCase_ = common_inputs['''input_ids'''].shape
# Not using the same length for past_key_values
UpperCAmelCase_ = seqlen + 2
UpperCAmelCase_ = (
batch,
self.num_attention_heads,
past_key_values_length,
self._config.hidden_size // self.num_attention_heads,
)
UpperCAmelCase_ = [
(torch.zeros(_snake_case), torch.zeros(_snake_case)) for _ in range(self.num_layers)
]
UpperCAmelCase_ = common_inputs['''attention_mask''']
if self.use_past:
UpperCAmelCase_ = ordered_inputs['''attention_mask'''].dtype
UpperCAmelCase_ = torch.cat(
[ordered_inputs['''attention_mask'''], torch.ones(_snake_case , _snake_case , dtype=_snake_case)] , dim=1)
return ordered_inputs
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return 13
| 51
| 0
|
from __future__ import annotations
def lowerCAmelCase_ ( __A ) -> list[int]:
'''simple docstring'''
return [ord(__A ) - 96 for elem in plain]
def lowerCAmelCase_ ( __A ) -> str:
'''simple docstring'''
return "".join(chr(elem + 96 ) for elem in encoded )
def lowerCAmelCase_ ( ) -> None:
'''simple docstring'''
UpperCAmelCase__ = encode(input("-> " ).strip().lower() )
print("Encoded: ", __A )
print("Decoded:", decode(__A ) )
if __name__ == "__main__":
main()
| 65
|
import os
import unittest
from transformers.models.phobert.tokenization_phobert import VOCAB_FILES_NAMES, PhobertTokenizer
from ...test_tokenization_common import TokenizerTesterMixin
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : Any = PhobertTokenizer
UpperCAmelCase__ : List[str] = False
def lowerCamelCase ( self : str):
"""simple docstring"""
super().setUp()
# Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt
UpperCAmelCase_ = ['''T@@''', '''i''', '''I''', '''R@@''', '''r''', '''e@@''']
UpperCAmelCase_ = dict(zip(_snake_case , range(len(_snake_case))))
UpperCAmelCase_ = ['''#version: 0.2''', '''l à</w>''']
UpperCAmelCase_ = {'''unk_token''': '''<unk>'''}
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''vocab_file'''])
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''merges_file'''])
with open(self.vocab_file , '''w''' , encoding='''utf-8''') as fp:
for token in vocab_tokens:
fp.write(F"""{token} {vocab_tokens[token]}\n""")
with open(self.merges_file , '''w''' , encoding='''utf-8''') as fp:
fp.write('''\n'''.join(_snake_case))
def lowerCamelCase ( self : int , **_snake_case : Any):
"""simple docstring"""
kwargs.update(self.special_tokens_map)
return PhobertTokenizer.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Optional[Any] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T<unk> i <unk> <unk> <unk> <unk> <unk> <unk> I Re<unk> e<unk> <unk> <unk> <unk>'''
return input_text, output_text
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = PhobertTokenizer(self.vocab_file , self.merges_file , **self.special_tokens_map)
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T@@ ô@@ i l@@ à V@@ i@@ n@@ A@@ I R@@ e@@ s@@ e@@ a@@ r@@ c@@ h'''.split()
UpperCAmelCase_ = tokenizer.tokenize(_snake_case)
print(_snake_case)
self.assertListEqual(_snake_case , _snake_case)
UpperCAmelCase_ = tokens + [tokenizer.unk_token]
UpperCAmelCase_ = [4, 3, 5, 3, 3, 3, 3, 3, 3, 6, 7, 9, 3, 9, 3, 3, 3, 3, 3]
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case) , _snake_case)
| 51
| 0
|
"""simple docstring"""
import json
import os
import unittest
from transformers.models.ctrl.tokenization_ctrl import VOCAB_FILES_NAMES, CTRLTokenizer
from ...test_tokenization_common import TokenizerTesterMixin
class lowerCamelCase ( _lowerCAmelCase , unittest.TestCase ):
'''simple docstring'''
_A : str = CTRLTokenizer
_A : Union[str, Any] = False
_A : Union[str, Any] = False
def lowerCAmelCase_ ( self: Tuple ) -> List[Any]:
super().setUp()
# Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt
snake_case_ :str = ["""adapt""", """re@@""", """a@@""", """apt""", """c@@""", """t""", """<unk>"""]
snake_case_ :str = dict(zip(snake_case , range(len(snake_case ) ) ) )
snake_case_ :Optional[int] = ["""#version: 0.2""", """a p""", """ap t</w>""", """r e""", """a d""", """ad apt</w>""", """"""]
snake_case_ :Any = {"""unk_token""": """<unk>"""}
snake_case_ :List[str] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""vocab_file"""] )
snake_case_ :Dict = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""merges_file"""] )
with open(self.vocab_file , """w""" , encoding="""utf-8""" ) as fp:
fp.write(json.dumps(snake_case ) + """\n""" )
with open(self.merges_file , """w""" , encoding="""utf-8""" ) as fp:
fp.write("""\n""".join(snake_case ) )
def lowerCAmelCase_ ( self: Dict , **snake_case: Optional[int] ) -> str:
kwargs.update(self.special_tokens_map )
return CTRLTokenizer.from_pretrained(self.tmpdirname , **snake_case )
def lowerCAmelCase_ ( self: Tuple , snake_case: Tuple ) -> Optional[Any]:
snake_case_ :Optional[int] = """adapt react readapt apt"""
snake_case_ :Tuple = """adapt react readapt apt"""
return input_text, output_text
def lowerCAmelCase_ ( self: str ) -> int:
snake_case_ :int = CTRLTokenizer(self.vocab_file , self.merges_file , **self.special_tokens_map )
snake_case_ :Optional[Any] = """adapt react readapt apt"""
snake_case_ :List[Any] = """adapt re@@ a@@ c@@ t re@@ adapt apt""".split()
snake_case_ :List[str] = tokenizer.tokenize(snake_case )
self.assertListEqual(snake_case , snake_case )
snake_case_ :Dict = tokens + [tokenizer.unk_token]
snake_case_ :Optional[Any] = [0, 1, 2, 4, 5, 1, 0, 3, 6]
self.assertListEqual(tokenizer.convert_tokens_to_ids(snake_case ) , snake_case )
| 66
|
from typing import List, Optional, TypeVar
from .arrow_dataset import Dataset, _concatenate_map_style_datasets, _interleave_map_style_datasets
from .dataset_dict import DatasetDict, IterableDatasetDict
from .info import DatasetInfo
from .iterable_dataset import IterableDataset, _concatenate_iterable_datasets, _interleave_iterable_datasets
from .splits import NamedSplit
from .utils import logging
from .utils.py_utils import Literal
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[int] = TypeVar("DatasetType", Dataset, IterableDataset)
def A (__A : List[DatasetType] , __A : Optional[List[float]] = None , __A : Optional[int] = None , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : Literal["first_exhausted", "all_exhausted"] = "first_exhausted" , ) -> DatasetType:
"""simple docstring"""
from .arrow_dataset import Dataset
from .iterable_dataset import IterableDataset
if not datasets:
raise ValueError('''Unable to interleave an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if stopping_strategy not in ["first_exhausted", "all_exhausted"]:
raise ValueError(F"""{stopping_strategy} is not supported. Please enter a valid stopping_strategy.""" )
if dataset_type is Dataset:
return _interleave_map_style_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
else:
return _interleave_iterable_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
def A (__A : List[DatasetType] , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : int = 0 , ) -> DatasetType:
"""simple docstring"""
if not dsets:
raise ValueError('''Unable to concatenate an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if dataset_type is Dataset:
return _concatenate_map_style_datasets(__A , info=__A , split=__A , axis=__A )
else:
return _concatenate_iterable_datasets(__A , info=__A , split=__A , axis=__A )
| 51
| 0
|
'''simple docstring'''
import unittest
import numpy as np
from transformers import is_flax_available
from transformers.testing_utils import require_flax
from ..test_modeling_flax_common import ids_tensor
if is_flax_available():
import jax
import jax.numpy as jnp
from transformers.generation import (
FlaxForcedBOSTokenLogitsProcessor,
FlaxForcedEOSTokenLogitsProcessor,
FlaxLogitsProcessorList,
FlaxMinLengthLogitsProcessor,
FlaxTemperatureLogitsWarper,
FlaxTopKLogitsWarper,
FlaxTopPLogitsWarper,
)
@require_flax
class a__ ( unittest.TestCase ):
def SCREAMING_SNAKE_CASE__ ( self : Any , a : int , a : int ):
"""simple docstring"""
__lowerCamelCase = jnp.ones((batch_size, length) ) / length
return scores
def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] ):
"""simple docstring"""
__lowerCamelCase = None
__lowerCamelCase = 20
__lowerCamelCase = self._get_uniform_logits(batch_size=2 , length=a )
# tweak scores to not be uniform anymore
__lowerCamelCase = scores.at[1, 5].set((1 / length) + 0.1 ) # peak, 1st batch
__lowerCamelCase = scores.at[1, 10].set((1 / length) - 0.4 ) # valley, 1st batch
# compute softmax
__lowerCamelCase = jax.nn.softmax(a , axis=-1 )
__lowerCamelCase = FlaxTemperatureLogitsWarper(temperature=0.5 )
__lowerCamelCase = FlaxTemperatureLogitsWarper(temperature=1.3 )
__lowerCamelCase = jax.nn.softmax(temp_dist_warper_sharper(a , scores.copy() , cur_len=a ) , axis=-1 )
__lowerCamelCase = jax.nn.softmax(temp_dist_warper_smoother(a , scores.copy() , cur_len=a ) , axis=-1 )
# uniform distribution stays uniform
self.assertTrue(jnp.allclose(probs[0, :] , warped_prob_sharp[0, :] , atol=1e-3 ) )
self.assertTrue(jnp.allclose(probs[0, :] , warped_prob_smooth[0, :] , atol=1e-3 ) )
# sharp peaks get higher, valleys get lower
self.assertLess(probs[1, :].max() , warped_prob_sharp[1, :].max() )
self.assertGreater(probs[1, :].min() , warped_prob_sharp[1, :].min() )
# smooth peaks get lower, valleys get higher
self.assertGreater(probs[1, :].max() , warped_prob_smooth[1, :].max() )
self.assertLess(probs[1, :].min() , warped_prob_smooth[1, :].min() )
def SCREAMING_SNAKE_CASE__ ( self : Optional[int] ):
"""simple docstring"""
__lowerCamelCase = None
__lowerCamelCase = 10
__lowerCamelCase = 2
# create ramp distribution
__lowerCamelCase = np.broadcast_to(np.arange(a )[None, :] , (batch_size, vocab_size) ).copy()
__lowerCamelCase = ramp_logits[1:, : vocab_size // 2] + vocab_size
__lowerCamelCase = FlaxTopKLogitsWarper(3 )
__lowerCamelCase = top_k_warp(a , a , cur_len=a )
# check that correct tokens are filtered
self.assertListEqual(jnp.isinf(scores[0] ).tolist() , 7 * [True] + 3 * [False] )
self.assertListEqual(jnp.isinf(scores[1] ).tolist() , 2 * [True] + 3 * [False] + 5 * [True] )
# check special case
__lowerCamelCase = 5
__lowerCamelCase = FlaxTopKLogitsWarper(top_k=1 , filter_value=0.0 , min_tokens_to_keep=3 )
__lowerCamelCase = np.broadcast_to(np.arange(a )[None, :] , (batch_size, length) ).copy()
__lowerCamelCase = top_k_warp_safety_check(a , a , cur_len=a )
# min_tokens overwrites k: 3 tokens are kept => 2 tokens are nullified
self.assertListEqual((scores == 0.0).sum(axis=-1 ).tolist() , [2, 2] )
def SCREAMING_SNAKE_CASE__ ( self : int ):
"""simple docstring"""
__lowerCamelCase = None
__lowerCamelCase = 10
__lowerCamelCase = 2
# create distribution and take log (inverse to Softmax as taken in TopPLogitsWarper)
__lowerCamelCase = np.log(np.array([[0.3, 0.1, 0.1, 0.5], [0.15, 0.3, 0.3, 0.25]] ) )
__lowerCamelCase = FlaxTopPLogitsWarper(0.8 )
__lowerCamelCase = np.exp(top_p_warp(a , a , cur_len=a ) )
# dist should be filtered to keep min num values so that sum is >= top_p
# exp (-inf) => 0
__lowerCamelCase = np.array([[0.3, 0.0, 0.0, 0.5], [0.0, 0.3, 0.3, 0.25]] )
self.assertTrue(np.allclose(a , a , atol=1e-3 ) )
# check edge cases with negative and extreme logits
__lowerCamelCase = np.broadcast_to(np.arange(a )[None, :] , (batch_size, vocab_size) ).copy() - (
vocab_size // 2
)
# make ramp_logits more extreme
__lowerCamelCase = ramp_logits[1] * 1_00.0
# make sure at least 2 tokens are kept
__lowerCamelCase = FlaxTopPLogitsWarper(0.9 , min_tokens_to_keep=2 , filter_value=0.0 )
__lowerCamelCase = top_p_warp(a , a , cur_len=a )
# first batch should keep three tokens, second batch would keep only 1, but due to `min_tokens_to_keep=2` keeps 2.
self.assertListEqual((filtered_dist != 0.0).sum(axis=-1 ).tolist() , [3, 2] )
def SCREAMING_SNAKE_CASE__ ( self : Tuple ):
"""simple docstring"""
__lowerCamelCase = 20
__lowerCamelCase = 4
__lowerCamelCase = 0
__lowerCamelCase = FlaxMinLengthLogitsProcessor(min_length=10 , eos_token_id=a )
# check that min length is applied at length 5
__lowerCamelCase = ids_tensor((batch_size, 20) , vocab_size=20 )
__lowerCamelCase = 5
__lowerCamelCase = self._get_uniform_logits(a , a )
__lowerCamelCase = min_dist_processor(a , a , cur_len=a )
self.assertListEqual(scores_before_min_length[:, eos_token_id].tolist() , 4 * [-float('''inf''' )] )
# check that min length is not applied anymore at length 15
__lowerCamelCase = self._get_uniform_logits(a , a )
__lowerCamelCase = 15
__lowerCamelCase = min_dist_processor(a , a , cur_len=a )
self.assertFalse(jnp.isinf(a ).any() )
def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] ):
"""simple docstring"""
__lowerCamelCase = 20
__lowerCamelCase = 4
__lowerCamelCase = 0
__lowerCamelCase = FlaxForcedBOSTokenLogitsProcessor(bos_token_id=a )
# check that all scores are -inf except the bos_token_id score
__lowerCamelCase = ids_tensor((batch_size, 1) , vocab_size=20 )
__lowerCamelCase = 1
__lowerCamelCase = self._get_uniform_logits(a , a )
__lowerCamelCase = logits_processor(a , a , cur_len=a )
self.assertTrue(jnp.isneginf(scores[:, bos_token_id + 1 :] ).all() )
self.assertListEqual(scores[:, bos_token_id].tolist() , 4 * [0] ) # score for bos_token_id shold be zero
# check that bos_token_id is not forced if current length is greater than 1
__lowerCamelCase = 3
__lowerCamelCase = self._get_uniform_logits(a , a )
__lowerCamelCase = logits_processor(a , a , cur_len=a )
self.assertFalse(jnp.isinf(a ).any() )
def SCREAMING_SNAKE_CASE__ ( self : List[str] ):
"""simple docstring"""
__lowerCamelCase = 20
__lowerCamelCase = 4
__lowerCamelCase = 0
__lowerCamelCase = 5
__lowerCamelCase = FlaxForcedEOSTokenLogitsProcessor(max_length=a , eos_token_id=a )
# check that all scores are -inf except the eos_token_id when max_length is reached
__lowerCamelCase = ids_tensor((batch_size, 4) , vocab_size=20 )
__lowerCamelCase = 4
__lowerCamelCase = self._get_uniform_logits(a , a )
__lowerCamelCase = logits_processor(a , a , cur_len=a )
self.assertTrue(jnp.isneginf(scores[:, eos_token_id + 1 :] ).all() )
self.assertListEqual(scores[:, eos_token_id].tolist() , 4 * [0] ) # score for eos_token_id should be zero
# check that eos_token_id is not forced if max_length is not reached
__lowerCamelCase = 3
__lowerCamelCase = self._get_uniform_logits(a , a )
__lowerCamelCase = logits_processor(a , a , cur_len=a )
self.assertFalse(jnp.isinf(a ).any() )
def SCREAMING_SNAKE_CASE__ ( self : Union[str, Any] ):
"""simple docstring"""
__lowerCamelCase = 4
__lowerCamelCase = 10
__lowerCamelCase = 15
__lowerCamelCase = 2
__lowerCamelCase = 1
__lowerCamelCase = 15
# dummy input_ids and scores
__lowerCamelCase = ids_tensor((batch_size, sequence_length) , a )
__lowerCamelCase = input_ids.copy()
__lowerCamelCase = self._get_uniform_logits(a , a )
__lowerCamelCase = scores.copy()
# instantiate all dist processors
__lowerCamelCase = FlaxTemperatureLogitsWarper(temperature=0.5 )
__lowerCamelCase = FlaxTopKLogitsWarper(3 )
__lowerCamelCase = FlaxTopPLogitsWarper(0.8 )
# instantiate all logits processors
__lowerCamelCase = FlaxMinLengthLogitsProcessor(min_length=10 , eos_token_id=a )
__lowerCamelCase = FlaxForcedBOSTokenLogitsProcessor(bos_token_id=a )
__lowerCamelCase = FlaxForcedEOSTokenLogitsProcessor(max_length=a , eos_token_id=a )
__lowerCamelCase = 10
# no processor list
__lowerCamelCase = temp_dist_warp(a , a , cur_len=a )
__lowerCamelCase = top_k_warp(a , a , cur_len=a )
__lowerCamelCase = top_p_warp(a , a , cur_len=a )
__lowerCamelCase = min_dist_proc(a , a , cur_len=a )
__lowerCamelCase = bos_dist_proc(a , a , cur_len=a )
__lowerCamelCase = eos_dist_proc(a , a , cur_len=a )
# with processor list
__lowerCamelCase = FlaxLogitsProcessorList(
[temp_dist_warp, top_k_warp, top_p_warp, min_dist_proc, bos_dist_proc, eos_dist_proc] )
__lowerCamelCase = processor(a , a , cur_len=a )
# scores should be equal
self.assertTrue(jnp.allclose(a , a , atol=1e-3 ) )
# input_ids should never be changed
self.assertListEqual(input_ids.tolist() , input_ids_comp.tolist() )
def SCREAMING_SNAKE_CASE__ ( self : str ):
"""simple docstring"""
__lowerCamelCase = 4
__lowerCamelCase = 10
__lowerCamelCase = 15
__lowerCamelCase = 2
__lowerCamelCase = 1
__lowerCamelCase = 15
# dummy input_ids and scores
__lowerCamelCase = ids_tensor((batch_size, sequence_length) , a )
__lowerCamelCase = input_ids.copy()
__lowerCamelCase = self._get_uniform_logits(a , a )
__lowerCamelCase = scores.copy()
# instantiate all dist processors
__lowerCamelCase = FlaxTemperatureLogitsWarper(temperature=0.5 )
__lowerCamelCase = FlaxTopKLogitsWarper(3 )
__lowerCamelCase = FlaxTopPLogitsWarper(0.8 )
# instantiate all logits processors
__lowerCamelCase = FlaxMinLengthLogitsProcessor(min_length=10 , eos_token_id=a )
__lowerCamelCase = FlaxForcedBOSTokenLogitsProcessor(bos_token_id=a )
__lowerCamelCase = FlaxForcedEOSTokenLogitsProcessor(max_length=a , eos_token_id=a )
__lowerCamelCase = 10
# no processor list
def run_no_processor_list(a : Dict , a : List[str] , a : Dict ):
__lowerCamelCase = temp_dist_warp(a , a , cur_len=a )
__lowerCamelCase = top_k_warp(a , a , cur_len=a )
__lowerCamelCase = top_p_warp(a , a , cur_len=a )
__lowerCamelCase = min_dist_proc(a , a , cur_len=a )
__lowerCamelCase = bos_dist_proc(a , a , cur_len=a )
__lowerCamelCase = eos_dist_proc(a , a , cur_len=a )
return scores
# with processor list
def run_processor_list(a : Optional[int] , a : Tuple , a : str ):
__lowerCamelCase = FlaxLogitsProcessorList(
[temp_dist_warp, top_k_warp, top_p_warp, min_dist_proc, bos_dist_proc, eos_dist_proc] )
__lowerCamelCase = processor(a , a , cur_len=a )
return scores
__lowerCamelCase = jax.jit(a )
__lowerCamelCase = jax.jit(a )
__lowerCamelCase = jitted_run_no_processor_list(a , a , a )
__lowerCamelCase = jitted_run_processor_list(a , a , a )
# scores should be equal
self.assertTrue(jnp.allclose(a , a , atol=1e-3 ) )
# input_ids should never be changed
self.assertListEqual(input_ids.tolist() , input_ids_comp.tolist() )
| 67
|
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
snake_case_ : Optional[Any] = "pt"
elif is_tf_available():
snake_case_ : Union[str, Any] = "tf"
else:
snake_case_ : str = "jax"
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = ByTaTokenizer
UpperCAmelCase__ : int = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().setUp()
UpperCAmelCase_ = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname)
@cached_property
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
return ByTaTokenizer.from_pretrained('''google/byt5-small''')
def lowerCamelCase ( self : List[str] , **_snake_case : Union[str, Any]):
"""simple docstring"""
return self.tokenizer_class.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Dict , _snake_case : int , _snake_case : Tuple=False , _snake_case : Dict=20 , _snake_case : Optional[Any]=5):
"""simple docstring"""
UpperCAmelCase_ = []
for i in range(len(_snake_case)):
try:
UpperCAmelCase_ = tokenizer.decode([i] , clean_up_tokenization_spaces=_snake_case)
except UnicodeDecodeError:
pass
toks.append((i, tok))
UpperCAmelCase_ = list(filter(lambda _snake_case: re.match(r'''^[ a-zA-Z]+$''' , t[1]) , _snake_case))
UpperCAmelCase_ = list(filter(lambda _snake_case: [t[0]] == tokenizer.encode(t[1] , add_special_tokens=_snake_case) , _snake_case))
if max_length is not None and len(_snake_case) > max_length:
UpperCAmelCase_ = toks[:max_length]
if min_length is not None and len(_snake_case) < min_length and len(_snake_case) > 0:
while len(_snake_case) < min_length:
UpperCAmelCase_ = toks + toks
# toks_str = [t[1] for t in toks]
UpperCAmelCase_ = [t[0] for t in toks]
# Ensure consistency
UpperCAmelCase_ = tokenizer.decode(_snake_case , clean_up_tokenization_spaces=_snake_case)
if " " not in output_txt and len(_snake_case) > 1:
UpperCAmelCase_ = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=_snake_case)
+ ''' '''
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=_snake_case)
)
if with_prefix_space:
UpperCAmelCase_ = ''' ''' + output_txt
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
return output_txt, output_ids
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = tokenizer(['''hi</s>''', '''I went to the gym</s>''', '''</s>'''])
UpperCAmelCase_ = tokenizer(['''hi''', '''I went to the gym''', ''''''])
self.assertListEqual(batch_with_eos_added['''input_ids'''] , batch_without_eos_added['''input_ids'''])
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = '''Unicode €.'''
UpperCAmelCase_ = tokenizer(_snake_case)
UpperCAmelCase_ = [88, 113, 108, 102, 114, 103, 104, 35, 229, 133, 175, 49, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''Unicode €.</s>''')
UpperCAmelCase_ = tokenizer('''e è é ê ë''')
UpperCAmelCase_ = [104, 35, 198, 171, 35, 198, 172, 35, 198, 173, 35, 198, 174, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''e è é ê ë</s>''')
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('''e è é ê ë''')) , '''e è é ê ë</s>''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 1, 0]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
if FRAMEWORK != "jax":
UpperCAmelCase_ = list(batch.input_ids.numpy()[0])
else:
UpperCAmelCase_ = list(batch.input_ids.tolist()[0])
self.assertListEqual(_snake_case , _snake_case)
self.assertEqual((2, 37) , batch.input_ids.shape)
self.assertEqual((2, 37) , batch.attention_mask.shape)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
# check if input_ids are returned and no decoder_input_ids
self.assertIn('''input_ids''' , _snake_case)
self.assertIn('''attention_mask''' , _snake_case)
self.assertNotIn('''decoder_input_ids''' , _snake_case)
self.assertNotIn('''decoder_attention_mask''' , _snake_case)
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = [
'''Summary of the text.''',
'''Another summary.''',
]
UpperCAmelCase_ = tokenizer(
text_target=_snake_case , max_length=32 , padding='''max_length''' , truncation=_snake_case , return_tensors=_snake_case)
self.assertEqual(32 , targets['''input_ids'''].shape[1])
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization. </s>''']
UpperCAmelCase_ = ['''Summary of the text. </s>''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 35, 1]
UpperCAmelCase_ = [86, 120, 112, 112, 100, 117, 124, 35, 114, 105, 35, 119, 107, 104, 35, 119, 104, 123, 119, 49, 35, 1]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , text_target=_snake_case)
self.assertEqual(_snake_case , batch['''input_ids'''][0])
self.assertEqual(_snake_case , batch['''labels'''][0])
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
self.assertNotEqual(tokenizer.model_max_length , 42)
# Now let's start the test
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
shutil.rmtree(_snake_case)
UpperCAmelCase_ = self.get_tokenizers(model_max_length=42)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
tokenizer.add_tokens(['''bim''', '''bambam'''])
UpperCAmelCase_ = tokenizer.additional_special_tokens
additional_special_tokens.append('''new_additional_special_token''')
tokenizer.add_special_tokens({'''additional_special_tokens''': additional_special_tokens})
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
self.assertIn('''new_additional_special_token''' , after_tokenizer.additional_special_tokens)
self.assertEqual(after_tokenizer.model_max_length , 42)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case , model_max_length=43)
self.assertEqual(tokenizer.model_max_length , 43)
shutil.rmtree(_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
UpperCAmelCase_ = [F"""<extra_id_{i}>""" for i in range(125)]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , )
self.assertIn(
'''an_additional_special_token''' , tokenizer_without_change_in_init.additional_special_tokens)
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['''an_additional_special_token'''] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['''an_additional_special_token'''])) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
UpperCAmelCase_ = added_tokens_extra_ids + [AddedToken('''a_new_additional_special_token''' , lstrip=_snake_case)]
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , additional_special_tokens=_snake_case , )
self.assertIn('''a_new_additional_special_token''' , tokenizer.additional_special_tokens)
self.assertEqual(
['''a_new_additional_special_token'''] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['''a_new_additional_special_token'''])) , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer_class.from_pretrained(_snake_case)
self.assertTrue(tokenizer.decode([255]) == '''''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers(fast=_snake_case , do_lower_case=_snake_case)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = ['''t''', '''h''', '''i''', '''s''', ''' ''', '''i''', '''s''', ''' ''', '''a''', ''' ''', '''t''', '''e''', '''x''', '''t''', '''</s>''']
UpperCAmelCase_ = tokenizer.convert_tokens_to_string(_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = [
'''bos_token''',
'''eos_token''',
'''unk_token''',
'''sep_token''',
'''pad_token''',
'''cls_token''',
'''mask_token''',
]
UpperCAmelCase_ = 0
UpperCAmelCase_ = tokenizer.convert_ids_to_tokens(
_snake_case , skip_special_tokens=_snake_case)
for attr in attributes_list:
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , '''additional_special_tokens_ids''' , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [])
setattr(_snake_case , '''additional_special_tokens_ids''' , [token_id_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [token_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [token_id_to_test_setters])
| 51
| 0
|
import argparse
import requests
import torch
from PIL import Image
from transformers import SwinConfig, SwinForMaskedImageModeling, ViTImageProcessor
def lowerCAmelCase__ ( SCREAMING_SNAKE_CASE_: Tuple ) -> Union[str, Any]:
'''simple docstring'''
A__ = SwinConfig(image_size=1_9_2 )
if "base" in model_name:
A__ = 6
A__ = 1_2_8
A__ = (2, 2, 1_8, 2)
A__ = (4, 8, 1_6, 3_2)
elif "large" in model_name:
A__ = 1_2
A__ = 1_9_2
A__ = (2, 2, 1_8, 2)
A__ = (6, 1_2, 2_4, 4_8)
else:
raise ValueError("Model not supported, only supports base and large variants" )
A__ = window_size
A__ = embed_dim
A__ = depths
A__ = num_heads
return config
def lowerCAmelCase__ ( SCREAMING_SNAKE_CASE_: List[Any] ) -> Any:
'''simple docstring'''
if "encoder.mask_token" in name:
A__ = name.replace("encoder.mask_token" , "embeddings.mask_token" )
if "encoder.patch_embed.proj" in name:
A__ = name.replace("encoder.patch_embed.proj" , "embeddings.patch_embeddings.projection" )
if "encoder.patch_embed.norm" in name:
A__ = name.replace("encoder.patch_embed.norm" , "embeddings.norm" )
if "attn.proj" in name:
A__ = name.replace("attn.proj" , "attention.output.dense" )
if "attn" in name:
A__ = name.replace("attn" , "attention.self" )
if "norm1" in name:
A__ = name.replace("norm1" , "layernorm_before" )
if "norm2" in name:
A__ = name.replace("norm2" , "layernorm_after" )
if "mlp.fc1" in name:
A__ = name.replace("mlp.fc1" , "intermediate.dense" )
if "mlp.fc2" in name:
A__ = name.replace("mlp.fc2" , "output.dense" )
if name == "encoder.norm.weight":
A__ = "layernorm.weight"
if name == "encoder.norm.bias":
A__ = "layernorm.bias"
if "decoder" in name:
pass
else:
A__ = "swin." + name
return name
def lowerCAmelCase__ ( SCREAMING_SNAKE_CASE_: Dict , SCREAMING_SNAKE_CASE_: Union[str, Any] ) -> List[str]:
'''simple docstring'''
for key in orig_state_dict.copy().keys():
A__ = orig_state_dict.pop(SCREAMING_SNAKE_CASE_ )
if "attn_mask" in key:
pass
elif "qkv" in key:
A__ = key.split("." )
A__ = int(key_split[2] )
A__ = int(key_split[4] )
A__ = model.swin.encoder.layers[layer_num].blocks[block_num].attention.self.all_head_size
if "weight" in key:
A__ = val[:dim, :]
A__ = val[
dim : dim * 2, :
]
A__ = val[-dim:, :]
else:
A__ = val[
:dim
]
A__ = val[
dim : dim * 2
]
A__ = val[
-dim:
]
else:
A__ = val
return orig_state_dict
def lowerCAmelCase__ ( SCREAMING_SNAKE_CASE_: Optional[int] , SCREAMING_SNAKE_CASE_: int , SCREAMING_SNAKE_CASE_: Tuple , SCREAMING_SNAKE_CASE_: Optional[Any] ) -> List[str]:
'''simple docstring'''
A__ = torch.load(SCREAMING_SNAKE_CASE_ , map_location="cpu" )["model"]
A__ = get_swin_config(SCREAMING_SNAKE_CASE_ )
A__ = SwinForMaskedImageModeling(SCREAMING_SNAKE_CASE_ )
model.eval()
A__ = convert_state_dict(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ )
model.load_state_dict(SCREAMING_SNAKE_CASE_ )
A__ = "http://images.cocodataset.org/val2017/000000039769.jpg"
A__ = ViTImageProcessor(size={"height": 1_9_2, "width": 1_9_2} )
A__ = Image.open(requests.get(SCREAMING_SNAKE_CASE_ , stream=SCREAMING_SNAKE_CASE_ ).raw )
A__ = image_processor(images=SCREAMING_SNAKE_CASE_ , return_tensors="pt" )
with torch.no_grad():
A__ = model(**SCREAMING_SNAKE_CASE_ ).logits
print(outputs.keys() )
print("Looks ok!" )
if pytorch_dump_folder_path is not None:
print(F'Saving model {model_name} to {pytorch_dump_folder_path}' )
model.save_pretrained(SCREAMING_SNAKE_CASE_ )
print(F'Saving image processor to {pytorch_dump_folder_path}' )
image_processor.save_pretrained(SCREAMING_SNAKE_CASE_ )
if push_to_hub:
print(F'Pushing model and image processor for {model_name} to hub' )
model.push_to_hub(F'microsoft/{model_name}' )
image_processor.push_to_hub(F'microsoft/{model_name}' )
if __name__ == "__main__":
lowerCAmelCase__ = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--model_name""",
default="""swin-base-simmim-window6-192""",
type=str,
choices=["""swin-base-simmim-window6-192""", """swin-large-simmim-window12-192"""],
help="""Name of the Swin SimMIM model you'd like to convert.""",
)
parser.add_argument(
"""--checkpoint_path""",
default="""/Users/nielsrogge/Documents/SwinSimMIM/simmim_pretrain__swin_base__img192_window6__100ep.pth""",
type=str,
help="""Path to the original PyTorch checkpoint (.pth file).""",
)
parser.add_argument(
"""--pytorch_dump_folder_path""", default=None, type=str, help="""Path to the output PyTorch model directory."""
)
parser.add_argument(
"""--push_to_hub""", action="""store_true""", help="""Whether or not to push the converted model to the 🤗 hub."""
)
lowerCAmelCase__ = parser.parse_args()
convert_swin_checkpoint(args.model_name, args.checkpoint_path, args.pytorch_dump_folder_path, args.push_to_hub)
| 68
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_sentencepiece_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : Dict = {"configuration_mbart": ["MBART_PRETRAINED_CONFIG_ARCHIVE_MAP", "MBartConfig", "MBartOnnxConfig"]}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Tuple = ["MBartTokenizer"]
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = ["MBartTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[Any] = [
"MBART_PRETRAINED_MODEL_ARCHIVE_LIST",
"MBartForCausalLM",
"MBartForConditionalGeneration",
"MBartForQuestionAnswering",
"MBartForSequenceClassification",
"MBartModel",
"MBartPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TFMBartForConditionalGeneration",
"TFMBartModel",
"TFMBartPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"FlaxMBartForConditionalGeneration",
"FlaxMBartForQuestionAnswering",
"FlaxMBartForSequenceClassification",
"FlaxMBartModel",
"FlaxMBartPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_mbart import MBART_PRETRAINED_CONFIG_ARCHIVE_MAP, MBartConfig, MBartOnnxConfig
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart import MBartTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart_fast import MBartTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_mbart import (
MBART_PRETRAINED_MODEL_ARCHIVE_LIST,
MBartForCausalLM,
MBartForConditionalGeneration,
MBartForQuestionAnswering,
MBartForSequenceClassification,
MBartModel,
MBartPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_mbart import TFMBartForConditionalGeneration, TFMBartModel, TFMBartPreTrainedModel
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_mbart import (
FlaxMBartForConditionalGeneration,
FlaxMBartForQuestionAnswering,
FlaxMBartForSequenceClassification,
FlaxMBartModel,
FlaxMBartPreTrainedModel,
)
else:
import sys
snake_case_ : Any = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
"""simple docstring"""
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, PerceiverTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
__UpperCamelCase = '''pt'''
elif is_tf_available():
__UpperCamelCase = '''tf'''
else:
__UpperCamelCase = '''jax'''
class UpperCamelCase ( lowerCAmelCase__ , unittest.TestCase ):
SCREAMING_SNAKE_CASE_ = PerceiverTokenizer
SCREAMING_SNAKE_CASE_ = False
def a_ ( self) -> Union[str, Any]:
super().setUp()
snake_case_ = PerceiverTokenizer()
tokenizer.save_pretrained(self.tmpdirname)
@cached_property
def a_ ( self) -> Union[str, Any]:
return PerceiverTokenizer.from_pretrained('deepmind/language-perceiver')
def a_ ( self, **lowerCAmelCase__) -> PerceiverTokenizer:
return self.tokenizer_class.from_pretrained(self.tmpdirname, **lowerCAmelCase__)
def a_ ( self, lowerCAmelCase__, lowerCAmelCase__=False, lowerCAmelCase__=20, lowerCAmelCase__=5) -> Tuple[str, list]:
# XXX The default common tokenizer tests assume that every ID is decodable on its own.
# This assumption is invalid for Perceiver because single bytes might not be
# valid utf-8 (byte 128 for instance).
# Here we're overriding the smallest possible method to provide
# a clean sequence without making the same assumption.
snake_case_ = []
for i in range(len(lowerCAmelCase__)):
try:
snake_case_ = tokenizer.decode([i], clean_up_tokenization_spaces=lowerCAmelCase__)
except UnicodeDecodeError:
pass
toks.append((i, tok))
snake_case_ = list(filter(lambda lowerCAmelCase__: re.match(R'^[ a-zA-Z]+$', t[1]), lowerCAmelCase__))
snake_case_ = list(filter(lambda lowerCAmelCase__: [t[0]] == tokenizer.encode(t[1], add_special_tokens=lowerCAmelCase__), lowerCAmelCase__))
if max_length is not None and len(lowerCAmelCase__) > max_length:
snake_case_ = toks[:max_length]
if min_length is not None and len(lowerCAmelCase__) < min_length and len(lowerCAmelCase__) > 0:
while len(lowerCAmelCase__) < min_length:
snake_case_ = toks + toks
# toks_str = [t[1] for t in toks]
snake_case_ = [t[0] for t in toks]
# Ensure consistency
snake_case_ = tokenizer.decode(lowerCAmelCase__, clean_up_tokenization_spaces=lowerCAmelCase__)
if " " not in output_txt and len(lowerCAmelCase__) > 1:
snake_case_ = (
tokenizer.decode([toks_ids[0]], clean_up_tokenization_spaces=lowerCAmelCase__)
+ ' '
+ tokenizer.decode(toks_ids[1:], clean_up_tokenization_spaces=lowerCAmelCase__)
)
if with_prefix_space:
snake_case_ = ' ' + output_txt
snake_case_ = tokenizer.encode(lowerCAmelCase__, add_special_tokens=lowerCAmelCase__)
return output_txt, output_ids
def a_ ( self) -> Tuple:
snake_case_ = self.perceiver_tokenizer
snake_case_ = 'Unicode €.'
snake_case_ = tokenizer(lowerCAmelCase__)
snake_case_ = [4, 91, 116, 111, 105, 117, 106, 107, 38, 232, 136, 178, 52, 5]
self.assertEqual(encoded['input_ids'], lowerCAmelCase__)
# decoding
snake_case_ = tokenizer.decode(lowerCAmelCase__)
self.assertEqual(lowerCAmelCase__, '[CLS]Unicode €.[SEP]')
snake_case_ = tokenizer('e è é ê ë')
snake_case_ = [4, 107, 38, 201, 174, 38, 201, 175, 38, 201, 176, 38, 201, 177, 5]
self.assertEqual(encoded['input_ids'], lowerCAmelCase__)
# decoding
snake_case_ = tokenizer.decode(lowerCAmelCase__)
self.assertEqual(lowerCAmelCase__, '[CLS]e è é ê ë[SEP]')
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('e è é ê ë')), '[CLS]e è é ê ë[SEP]')
def a_ ( self) -> List[str]:
snake_case_ = self.perceiver_tokenizer
snake_case_ = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
# fmt: off
snake_case_ = [4, 71, 38, 114, 117, 116, 109, 38, 118, 103, 120, 103, 109, 120, 103, 118, 110, 38, 108, 117, 120, 38, 121, 123, 115, 115, 103, 120, 111, 128, 103, 122, 111, 117, 116, 52, 5, 0]
# fmt: on
snake_case_ = tokenizer(lowerCAmelCase__, padding=lowerCAmelCase__, return_tensors=lowerCAmelCase__)
self.assertIsInstance(lowerCAmelCase__, lowerCAmelCase__)
if FRAMEWORK != "jax":
snake_case_ = list(batch.input_ids.numpy()[0])
else:
snake_case_ = list(batch.input_ids.tolist()[0])
self.assertListEqual(lowerCAmelCase__, lowerCAmelCase__)
self.assertEqual((2, 38), batch.input_ids.shape)
self.assertEqual((2, 38), batch.attention_mask.shape)
def a_ ( self) -> Optional[int]:
snake_case_ = self.perceiver_tokenizer
snake_case_ = ['A long paragraph for summarization.', 'Another paragraph for summarization.']
snake_case_ = tokenizer(lowerCAmelCase__, padding=lowerCAmelCase__, return_tensors=lowerCAmelCase__)
# check if input_ids are returned and no decoder_input_ids
self.assertIn('input_ids', lowerCAmelCase__)
self.assertIn('attention_mask', lowerCAmelCase__)
self.assertNotIn('decoder_input_ids', lowerCAmelCase__)
self.assertNotIn('decoder_attention_mask', lowerCAmelCase__)
def a_ ( self) -> List[str]:
snake_case_ = self.perceiver_tokenizer
snake_case_ = [
'Summary of the text.',
'Another summary.',
]
snake_case_ = tokenizer(
text_target=lowerCAmelCase__, max_length=32, padding='max_length', truncation=lowerCAmelCase__, return_tensors=lowerCAmelCase__)
self.assertEqual(32, targets['input_ids'].shape[1])
def a_ ( self) -> Optional[int]:
# safety check on max_len default value so we are sure the test works
snake_case_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(f'{tokenizer.__class__.__name__}'):
self.assertNotEqual(tokenizer.model_max_length, 42)
# Now let's start the test
snake_case_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(f'{tokenizer.__class__.__name__}'):
# Isolate this from the other tests because we save additional tokens/etc
snake_case_ = tempfile.mkdtemp()
snake_case_ = ' He is very happy, UNwant\u00E9d,running'
snake_case_ = tokenizer.encode(lowerCAmelCase__, add_special_tokens=lowerCAmelCase__)
tokenizer.save_pretrained(lowerCAmelCase__)
snake_case_ = tokenizer.__class__.from_pretrained(lowerCAmelCase__)
snake_case_ = after_tokenizer.encode(lowerCAmelCase__, add_special_tokens=lowerCAmelCase__)
self.assertListEqual(lowerCAmelCase__, lowerCAmelCase__)
shutil.rmtree(lowerCAmelCase__)
snake_case_ = self.get_tokenizers(model_max_length=42)
for tokenizer in tokenizers:
with self.subTest(f'{tokenizer.__class__.__name__}'):
# Isolate this from the other tests because we save additional tokens/etc
snake_case_ = tempfile.mkdtemp()
snake_case_ = ' He is very happy, UNwant\u00E9d,running'
tokenizer.add_tokens(['bim', 'bambam'])
snake_case_ = tokenizer.additional_special_tokens
additional_special_tokens.append('new_additional_special_token')
tokenizer.add_special_tokens({'additional_special_tokens': additional_special_tokens})
snake_case_ = tokenizer.encode(lowerCAmelCase__, add_special_tokens=lowerCAmelCase__)
tokenizer.save_pretrained(lowerCAmelCase__)
snake_case_ = tokenizer.__class__.from_pretrained(lowerCAmelCase__)
snake_case_ = after_tokenizer.encode(lowerCAmelCase__, add_special_tokens=lowerCAmelCase__)
self.assertListEqual(lowerCAmelCase__, lowerCAmelCase__)
self.assertIn('new_additional_special_token', after_tokenizer.additional_special_tokens)
self.assertEqual(after_tokenizer.model_max_length, 42)
snake_case_ = tokenizer.__class__.from_pretrained(lowerCAmelCase__, model_max_length=43)
self.assertEqual(tokenizer.model_max_length, 43)
shutil.rmtree(lowerCAmelCase__)
def a_ ( self) -> Any:
snake_case_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(lowerCAmelCase__)
with open(os.path.join(lowerCAmelCase__, 'special_tokens_map.json'), encoding='utf-8') as json_file:
snake_case_ = json.load(lowerCAmelCase__)
with open(os.path.join(lowerCAmelCase__, 'tokenizer_config.json'), encoding='utf-8') as json_file:
snake_case_ = json.load(lowerCAmelCase__)
snake_case_ = [f'<extra_id_{i}>' for i in range(125)]
snake_case_ = added_tokens_extra_ids + [
'an_additional_special_token'
]
snake_case_ = added_tokens_extra_ids + [
'an_additional_special_token'
]
with open(os.path.join(lowerCAmelCase__, 'special_tokens_map.json'), 'w', encoding='utf-8') as outfile:
json.dump(lowerCAmelCase__, lowerCAmelCase__)
with open(os.path.join(lowerCAmelCase__, 'tokenizer_config.json'), 'w', encoding='utf-8') as outfile:
json.dump(lowerCAmelCase__, lowerCAmelCase__)
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
snake_case_ = tokenizer_class.from_pretrained(
lowerCAmelCase__, )
self.assertIn(
'an_additional_special_token', tokenizer_without_change_in_init.additional_special_tokens)
self.assertEqual(
['an_additional_special_token'], tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['an_additional_special_token'])), )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
snake_case_ = added_tokens_extra_ids + [AddedToken('a_new_additional_special_token', lstrip=lowerCAmelCase__)]
snake_case_ = tokenizer_class.from_pretrained(
lowerCAmelCase__, additional_special_tokens=lowerCAmelCase__, )
self.assertIn('a_new_additional_special_token', tokenizer.additional_special_tokens)
self.assertEqual(
['a_new_additional_special_token'], tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['a_new_additional_special_token'])), )
def a_ ( self) -> Optional[Any]:
snake_case_ = self.perceiver_tokenizer
self.assertEqual(tokenizer.decode([178]), '�')
def a_ ( self) -> Optional[Any]:
pass
def a_ ( self) -> List[str]:
pass
def a_ ( self) -> Union[str, Any]:
pass
def a_ ( self) -> List[Any]:
pass
def a_ ( self) -> Optional[Any]:
# The default common tokenizer tests uses invalid tokens for Perceiver that can only accept one-character
# strings and special added tokens as tokens
snake_case_ = self.get_tokenizers(fast=lowerCAmelCase__, do_lower_case=lowerCAmelCase__)
for tokenizer in tokenizers:
with self.subTest(f'{tokenizer.__class__.__name__}'):
snake_case_ = ['[CLS]', 't', 'h', 'i', 's', ' ', 'i', 's', ' ', 'a', ' ', 't', 'e', 's', 't', '[SEP]']
snake_case_ = tokenizer.convert_tokens_to_string(lowerCAmelCase__)
self.assertIsInstance(lowerCAmelCase__, lowerCAmelCase__)
| 69
|
import warnings
from typing import List, Optional, Union
from ...image_utils import ImageInput
from ...processing_utils import ProcessorMixin
from ...tokenization_utils_base import BatchEncoding, PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy
from ...utils import TensorType
class __snake_case ( a ):
UpperCAmelCase__ : Dict = ['''image_processor''', '''tokenizer''']
UpperCAmelCase__ : Dict = '''FlavaImageProcessor'''
UpperCAmelCase__ : Dict = ('''BertTokenizer''', '''BertTokenizerFast''')
def __init__( self : Union[str, Any] , _snake_case : List[str]=None , _snake_case : str=None , **_snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = None
if "feature_extractor" in kwargs:
warnings.warn(
'''The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor`'''
''' instead.''' , _snake_case , )
UpperCAmelCase_ = kwargs.pop('''feature_extractor''')
UpperCAmelCase_ = image_processor if image_processor is not None else feature_extractor
if image_processor is None:
raise ValueError('''You need to specify an `image_processor`.''')
if tokenizer is None:
raise ValueError('''You need to specify a `tokenizer`.''')
super().__init__(_snake_case , _snake_case)
UpperCAmelCase_ = self.image_processor
def __call__( self : List[Any] , _snake_case : Optional[ImageInput] = None , _snake_case : Optional[Union[TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]]] = None , _snake_case : bool = True , _snake_case : Union[bool, str, PaddingStrategy] = False , _snake_case : Union[bool, str, TruncationStrategy] = False , _snake_case : Optional[int] = None , _snake_case : int = 0 , _snake_case : Optional[int] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = True , _snake_case : Optional[Union[str, TensorType]] = None , **_snake_case : Any , ):
"""simple docstring"""
if text is None and images is None:
raise ValueError('''You have to specify either text or images. Both cannot be none.''')
if text is not None:
UpperCAmelCase_ = self.tokenizer(
text=_snake_case , add_special_tokens=_snake_case , padding=_snake_case , truncation=_snake_case , max_length=_snake_case , stride=_snake_case , pad_to_multiple_of=_snake_case , return_token_type_ids=_snake_case , return_attention_mask=_snake_case , return_overflowing_tokens=_snake_case , return_special_tokens_mask=_snake_case , return_offsets_mapping=_snake_case , return_length=_snake_case , verbose=_snake_case , return_tensors=_snake_case , **_snake_case , )
if images is not None:
UpperCAmelCase_ = self.image_processor(
_snake_case , return_image_mask=_snake_case , return_codebook_pixels=_snake_case , return_tensors=_snake_case , **_snake_case , )
if text is not None and images is not None:
encoding.update(_snake_case)
return encoding
elif text is not None:
return encoding
else:
return BatchEncoding(data=dict(**_snake_case) , tensor_type=_snake_case)
def lowerCamelCase ( self : Any , *_snake_case : Optional[Any] , **_snake_case : int):
"""simple docstring"""
return self.tokenizer.batch_decode(*_snake_case , **_snake_case)
def lowerCamelCase ( self : Optional[int] , *_snake_case : int , **_snake_case : Dict):
"""simple docstring"""
return self.tokenizer.decode(*_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.tokenizer.model_input_names
UpperCAmelCase_ = self.image_processor.model_input_names
return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names))
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
warnings.warn(
'''`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.''' , _snake_case , )
return self.image_processor_class
@property
def lowerCamelCase ( self : Any):
"""simple docstring"""
warnings.warn(
'''`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.''' , _snake_case , )
return self.image_processor
| 51
| 0
|
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
A__ : Tuple ={
'''configuration_lilt''': ['''LILT_PRETRAINED_CONFIG_ARCHIVE_MAP''', '''LiltConfig'''],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
A__ : List[str] =[
'''LILT_PRETRAINED_MODEL_ARCHIVE_LIST''',
'''LiltForQuestionAnswering''',
'''LiltForSequenceClassification''',
'''LiltForTokenClassification''',
'''LiltModel''',
'''LiltPreTrainedModel''',
]
if TYPE_CHECKING:
from .configuration_lilt import LILT_PRETRAINED_CONFIG_ARCHIVE_MAP, LiltConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_lilt import (
LILT_PRETRAINED_MODEL_ARCHIVE_LIST,
LiltForQuestionAnswering,
LiltForSequenceClassification,
LiltForTokenClassification,
LiltModel,
LiltPreTrainedModel,
)
else:
import sys
A__ : Tuple =_LazyModule(__name__, globals()['''__file__'''], _import_structure, module_spec=__spec__)
| 70
|
from diffusers.utils.testing_utils import require_onnxruntime
@require_onnxruntime
class __snake_case :
pass
| 51
| 0
|
import math
import torch
from torch import nn
from ..configuration_utils import ConfigMixin, register_to_config
from .attention_processor import Attention
from .embeddings import get_timestep_embedding
from .modeling_utils import ModelMixin
class __A ( a , a ):
"""simple docstring"""
@register_to_config
def __init__( self , lowerCamelCase__ = 128 , lowerCamelCase__ = 256 , lowerCamelCase__ = 2_000.0 , lowerCamelCase__ = 768 , lowerCamelCase__ = 12 , lowerCamelCase__ = 12 , lowerCamelCase__ = 64 , lowerCamelCase__ = 2048 , lowerCamelCase__ = 0.1 , ):
"""simple docstring"""
super().__init__()
__UpperCamelCase : str =nn.Sequential(
nn.Linear(lowerCamelCase__ , d_model * 4 , bias=lowerCamelCase__ ) , nn.SiLU() , nn.Linear(d_model * 4 , d_model * 4 , bias=lowerCamelCase__ ) , nn.SiLU() , )
__UpperCamelCase : Any =nn.Embedding(lowerCamelCase__ , lowerCamelCase__ )
__UpperCamelCase : Optional[int] =False
__UpperCamelCase : Dict =nn.Linear(lowerCamelCase__ , lowerCamelCase__ , bias=lowerCamelCase__ )
__UpperCamelCase : Optional[Any] =nn.Dropout(p=lowerCamelCase__ )
__UpperCamelCase : int =nn.ModuleList()
for lyr_num in range(lowerCamelCase__ ):
# FiLM conditional T5 decoder
__UpperCamelCase : Tuple =DecoderLayer(d_model=lowerCamelCase__ , d_kv=lowerCamelCase__ , num_heads=lowerCamelCase__ , d_ff=lowerCamelCase__ , dropout_rate=lowerCamelCase__ )
self.decoders.append(lowerCamelCase__ )
__UpperCamelCase : List[str] =TaLayerNorm(lowerCamelCase__ )
__UpperCamelCase : Optional[int] =nn.Dropout(p=lowerCamelCase__ )
__UpperCamelCase : Optional[int] =nn.Linear(lowerCamelCase__ , lowerCamelCase__ , bias=lowerCamelCase__ )
def __lowercase ( self , lowerCamelCase__ , lowerCamelCase__ ):
"""simple docstring"""
__UpperCamelCase : Optional[int] =torch.mul(query_input.unsqueeze(-1 ) , key_input.unsqueeze(-2 ) )
return mask.unsqueeze(-3 )
def __lowercase ( self , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ):
"""simple docstring"""
__UpperCamelCase , __UpperCamelCase , __UpperCamelCase : List[str] =decoder_input_tokens.shape
assert decoder_noise_time.shape == (batch,)
# decoder_noise_time is in [0, 1), so rescale to expected timing range.
__UpperCamelCase : Tuple =get_timestep_embedding(
decoder_noise_time * self.config.max_decoder_noise_time , embedding_dim=self.config.d_model , max_period=self.config.max_decoder_noise_time , ).to(dtype=self.dtype )
__UpperCamelCase : Union[str, Any] =self.conditioning_emb(lowerCamelCase__ ).unsqueeze(1 )
assert conditioning_emb.shape == (batch, 1, self.config.d_model * 4)
__UpperCamelCase : Optional[int] =decoder_input_tokens.shape[1]
# If we want to use relative positions for audio context, we can just offset
# this sequence by the length of encodings_and_masks.
__UpperCamelCase : List[Any] =torch.broadcast_to(
torch.arange(lowerCamelCase__ , device=decoder_input_tokens.device ) , (batch, seq_length) , )
__UpperCamelCase : Any =self.position_encoding(lowerCamelCase__ )
__UpperCamelCase : Any =self.continuous_inputs_projection(lowerCamelCase__ )
inputs += position_encodings
__UpperCamelCase : Optional[Any] =self.dropout(lowerCamelCase__ )
# decoder: No padding present.
__UpperCamelCase : Any =torch.ones(
decoder_input_tokens.shape[:2] , device=decoder_input_tokens.device , dtype=inputs.dtype )
# Translate encoding masks to encoder-decoder masks.
__UpperCamelCase : str =[(x, self.encoder_decoder_mask(lowerCamelCase__ , lowerCamelCase__ )) for x, y in encodings_and_masks]
# cross attend style: concat encodings
__UpperCamelCase : Any =torch.cat([x[0] for x in encodings_and_encdec_masks] , dim=1 )
__UpperCamelCase : List[Any] =torch.cat([x[1] for x in encodings_and_encdec_masks] , dim=-1 )
for lyr in self.decoders:
__UpperCamelCase : Optional[Any] =lyr(
lowerCamelCase__ , conditioning_emb=lowerCamelCase__ , encoder_hidden_states=lowerCamelCase__ , encoder_attention_mask=lowerCamelCase__ , )[0]
__UpperCamelCase : Tuple =self.decoder_norm(lowerCamelCase__ )
__UpperCamelCase : Tuple =self.post_dropout(lowerCamelCase__ )
__UpperCamelCase : Dict =self.spec_out(lowerCamelCase__ )
return spec_out
class __A ( nn.Module ):
"""simple docstring"""
def __init__( self , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__=1E-6 ):
"""simple docstring"""
super().__init__()
__UpperCamelCase : Optional[int] =nn.ModuleList()
# cond self attention: layer 0
self.layer.append(
TaLayerSelfAttentionCond(d_model=lowerCamelCase__ , d_kv=lowerCamelCase__ , num_heads=lowerCamelCase__ , dropout_rate=lowerCamelCase__ ) )
# cross attention: layer 1
self.layer.append(
TaLayerCrossAttention(
d_model=lowerCamelCase__ , d_kv=lowerCamelCase__ , num_heads=lowerCamelCase__ , dropout_rate=lowerCamelCase__ , layer_norm_epsilon=lowerCamelCase__ , ) )
# Film Cond MLP + dropout: last layer
self.layer.append(
TaLayerFFCond(d_model=lowerCamelCase__ , d_ff=lowerCamelCase__ , dropout_rate=lowerCamelCase__ , layer_norm_epsilon=lowerCamelCase__ ) )
def __lowercase ( self , lowerCamelCase__ , lowerCamelCase__=None , lowerCamelCase__=None , lowerCamelCase__=None , lowerCamelCase__=None , lowerCamelCase__=None , ):
"""simple docstring"""
__UpperCamelCase : Optional[Any] =self.layer[0](
lowerCamelCase__ , conditioning_emb=lowerCamelCase__ , attention_mask=lowerCamelCase__ , )
if encoder_hidden_states is not None:
__UpperCamelCase : str =torch.where(encoder_attention_mask > 0 , 0 , -1E10 ).to(
encoder_hidden_states.dtype )
__UpperCamelCase : int =self.layer[1](
lowerCamelCase__ , key_value_states=lowerCamelCase__ , attention_mask=lowerCamelCase__ , )
# Apply Film Conditional Feed Forward layer
__UpperCamelCase : Union[str, Any] =self.layer[-1](lowerCamelCase__ , lowerCamelCase__ )
return (hidden_states,)
class __A ( nn.Module ):
"""simple docstring"""
def __init__( self , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ):
"""simple docstring"""
super().__init__()
__UpperCamelCase : Any =TaLayerNorm(lowerCamelCase__ )
__UpperCamelCase : Tuple =TaFiLMLayer(in_features=d_model * 4 , out_features=lowerCamelCase__ )
__UpperCamelCase : Union[str, Any] =Attention(query_dim=lowerCamelCase__ , heads=lowerCamelCase__ , dim_head=lowerCamelCase__ , out_bias=lowerCamelCase__ , scale_qk=lowerCamelCase__ )
__UpperCamelCase : List[str] =nn.Dropout(lowerCamelCase__ )
def __lowercase ( self , lowerCamelCase__ , lowerCamelCase__=None , lowerCamelCase__=None , ):
"""simple docstring"""
__UpperCamelCase : int =self.layer_norm(lowerCamelCase__ )
if conditioning_emb is not None:
__UpperCamelCase : str =self.FiLMLayer(lowerCamelCase__ , lowerCamelCase__ )
# Self-attention block
__UpperCamelCase : int =self.attention(lowerCamelCase__ )
__UpperCamelCase : Any =hidden_states + self.dropout(lowerCamelCase__ )
return hidden_states
class __A ( nn.Module ):
"""simple docstring"""
def __init__( self , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ):
"""simple docstring"""
super().__init__()
__UpperCamelCase : Tuple =Attention(query_dim=lowerCamelCase__ , heads=lowerCamelCase__ , dim_head=lowerCamelCase__ , out_bias=lowerCamelCase__ , scale_qk=lowerCamelCase__ )
__UpperCamelCase : List[str] =TaLayerNorm(lowerCamelCase__ , eps=lowerCamelCase__ )
__UpperCamelCase : Any =nn.Dropout(lowerCamelCase__ )
def __lowercase ( self , lowerCamelCase__ , lowerCamelCase__=None , lowerCamelCase__=None , ):
"""simple docstring"""
__UpperCamelCase : List[Any] =self.layer_norm(lowerCamelCase__ )
__UpperCamelCase : Tuple =self.attention(
lowerCamelCase__ , encoder_hidden_states=lowerCamelCase__ , attention_mask=attention_mask.squeeze(1 ) , )
__UpperCamelCase : int =hidden_states + self.dropout(lowerCamelCase__ )
return layer_output
class __A ( nn.Module ):
"""simple docstring"""
def __init__( self , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ):
"""simple docstring"""
super().__init__()
__UpperCamelCase : Any =TaDenseGatedActDense(d_model=lowerCamelCase__ , d_ff=lowerCamelCase__ , dropout_rate=lowerCamelCase__ )
__UpperCamelCase : Tuple =TaFiLMLayer(in_features=d_model * 4 , out_features=lowerCamelCase__ )
__UpperCamelCase : Dict =TaLayerNorm(lowerCamelCase__ , eps=lowerCamelCase__ )
__UpperCamelCase : Union[str, Any] =nn.Dropout(lowerCamelCase__ )
def __lowercase ( self , lowerCamelCase__ , lowerCamelCase__=None ):
"""simple docstring"""
__UpperCamelCase : List[str] =self.layer_norm(lowerCamelCase__ )
if conditioning_emb is not None:
__UpperCamelCase : List[str] =self.film(lowerCamelCase__ , lowerCamelCase__ )
__UpperCamelCase : Union[str, Any] =self.DenseReluDense(lowerCamelCase__ )
__UpperCamelCase : Any =hidden_states + self.dropout(lowerCamelCase__ )
return hidden_states
class __A ( nn.Module ):
"""simple docstring"""
def __init__( self , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ):
"""simple docstring"""
super().__init__()
__UpperCamelCase : List[str] =nn.Linear(lowerCamelCase__ , lowerCamelCase__ , bias=lowerCamelCase__ )
__UpperCamelCase : Optional[int] =nn.Linear(lowerCamelCase__ , lowerCamelCase__ , bias=lowerCamelCase__ )
__UpperCamelCase : List[Any] =nn.Linear(lowerCamelCase__ , lowerCamelCase__ , bias=lowerCamelCase__ )
__UpperCamelCase : List[Any] =nn.Dropout(lowerCamelCase__ )
__UpperCamelCase : Any =NewGELUActivation()
def __lowercase ( self , lowerCamelCase__ ):
"""simple docstring"""
__UpperCamelCase : Any =self.act(self.wi_a(lowerCamelCase__ ) )
__UpperCamelCase : Union[str, Any] =self.wi_a(lowerCamelCase__ )
__UpperCamelCase : Any =hidden_gelu * hidden_linear
__UpperCamelCase : List[str] =self.dropout(lowerCamelCase__ )
__UpperCamelCase : Optional[Any] =self.wo(lowerCamelCase__ )
return hidden_states
class __A ( nn.Module ):
"""simple docstring"""
def __init__( self , lowerCamelCase__ , lowerCamelCase__=1E-6 ):
"""simple docstring"""
super().__init__()
__UpperCamelCase : Any =nn.Parameter(torch.ones(lowerCamelCase__ ) )
__UpperCamelCase : str =eps
def __lowercase ( self , lowerCamelCase__ ):
"""simple docstring"""
__UpperCamelCase : Any =hidden_states.to(torch.floataa ).pow(2 ).mean(-1 , keepdim=lowerCamelCase__ )
__UpperCamelCase : Dict =hidden_states * torch.rsqrt(variance + self.variance_epsilon )
# convert into half-precision if necessary
if self.weight.dtype in [torch.floataa, torch.bfloataa]:
__UpperCamelCase : List[str] =hidden_states.to(self.weight.dtype )
return self.weight * hidden_states
class __A ( nn.Module ):
"""simple docstring"""
def __lowercase ( self , lowerCamelCase__ ):
"""simple docstring"""
return 0.5 * input * (1.0 + torch.tanh(math.sqrt(2.0 / math.pi ) * (input + 0.044_715 * torch.pow(lowerCamelCase__ , 3.0 )) ))
class __A ( nn.Module ):
"""simple docstring"""
def __init__( self , lowerCamelCase__ , lowerCamelCase__ ):
"""simple docstring"""
super().__init__()
__UpperCamelCase : Optional[int] =nn.Linear(lowerCamelCase__ , out_features * 2 , bias=lowerCamelCase__ )
def __lowercase ( self , lowerCamelCase__ , lowerCamelCase__ ):
"""simple docstring"""
__UpperCamelCase : Optional[int] =self.scale_bias(lowerCamelCase__ )
__UpperCamelCase , __UpperCamelCase : Optional[int] =torch.chunk(lowerCamelCase__ , 2 , -1 )
__UpperCamelCase : int =x * (1 + scale) + shift
return x
| 71
|
import argparse
import os
import pickle
import sys
import torch
from transformers import TransfoXLConfig, TransfoXLLMHeadModel, load_tf_weights_in_transfo_xl
from transformers.models.transfo_xl import tokenization_transfo_xl as data_utils
from transformers.models.transfo_xl.tokenization_transfo_xl import CORPUS_NAME, VOCAB_FILES_NAMES
from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging
logging.set_verbosity_info()
# We do this to be able to load python 2 datasets pickles
# See e.g. https://stackoverflow.com/questions/2121874/python-pickling-after-changing-a-modules-directory/2121918#2121918
snake_case_ : List[Any] = data_utils.TransfoXLTokenizer
snake_case_ : int = data_utils.TransfoXLCorpus
snake_case_ : List[Any] = data_utils
snake_case_ : int = data_utils
def A (__A : Dict , __A : List[Any] , __A : Union[str, Any] , __A : Tuple ) -> Union[str, Any]:
"""simple docstring"""
if transfo_xl_dataset_file:
# Convert a pre-processed corpus (see original TensorFlow repo)
with open(__A , '''rb''' ) as fp:
UpperCAmelCase_ = pickle.load(__A , encoding='''latin1''' )
# Save vocabulary and dataset cache as Dictionaries (should be better than pickles for the long-term)
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + VOCAB_FILES_NAMES['''pretrained_vocab_file''']
print(F"""Save vocabulary to {pytorch_vocab_dump_path}""" )
UpperCAmelCase_ = corpus.vocab.__dict__
torch.save(__A , __A )
UpperCAmelCase_ = corpus.__dict__
corpus_dict_no_vocab.pop('''vocab''' , __A )
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + CORPUS_NAME
print(F"""Save dataset to {pytorch_dataset_dump_path}""" )
torch.save(__A , __A )
if tf_checkpoint_path:
# Convert a pre-trained TensorFlow model
UpperCAmelCase_ = os.path.abspath(__A )
UpperCAmelCase_ = os.path.abspath(__A )
print(F"""Converting Transformer XL checkpoint from {tf_path} with config at {config_path}.""" )
# Initialise PyTorch model
if transfo_xl_config_file == "":
UpperCAmelCase_ = TransfoXLConfig()
else:
UpperCAmelCase_ = TransfoXLConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = TransfoXLLMHeadModel(__A )
UpperCAmelCase_ = load_tf_weights_in_transfo_xl(__A , __A , __A )
# Save pytorch-model
UpperCAmelCase_ = os.path.join(__A , __A )
UpperCAmelCase_ = os.path.join(__A , __A )
print(F"""Save PyTorch model to {os.path.abspath(__A )}""" )
torch.save(model.state_dict() , __A )
print(F"""Save configuration file to {os.path.abspath(__A )}""" )
with open(__A , '''w''' , encoding='''utf-8''' ) as f:
f.write(config.to_json_string() )
if __name__ == "__main__":
snake_case_ : List[str] = argparse.ArgumentParser()
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
type=str,
required=True,
help="Path to the folder to store the PyTorch model or dataset/vocab.",
)
parser.add_argument(
"--tf_checkpoint_path",
default="",
type=str,
help="An optional path to a TensorFlow checkpoint path to be converted.",
)
parser.add_argument(
"--transfo_xl_config_file",
default="",
type=str,
help=(
"An optional config json file corresponding to the pre-trained BERT model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--transfo_xl_dataset_file",
default="",
type=str,
help="An optional dataset file to be converted in a vocabulary.",
)
snake_case_ : int = parser.parse_args()
convert_transfo_xl_checkpoint_to_pytorch(
args.tf_checkpoint_path,
args.transfo_xl_config_file,
args.pytorch_dump_folder_path,
args.transfo_xl_dataset_file,
)
| 51
| 0
|
"""simple docstring"""
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import AddedToken, PreTrainedTokenizer
from ...utils import logging
lowerCAmelCase__ = logging.get_logger(__name__)
lowerCAmelCase__ = {'''vocab_file''': '''sentencepiece.bpe.model'''}
lowerCAmelCase__ = {
'''vocab_file''': {
'''moussaKam/mbarthez''': '''https://huggingface.co/moussaKam/mbarthez/resolve/main/sentencepiece.bpe.model''',
'''moussaKam/barthez''': '''https://huggingface.co/moussaKam/barthez/resolve/main/sentencepiece.bpe.model''',
'''moussaKam/barthez-orangesum-title''': (
'''https://huggingface.co/moussaKam/barthez-orangesum-title/resolve/main/sentencepiece.bpe.model'''
),
},
}
lowerCAmelCase__ = {
'''moussaKam/mbarthez''': 1024,
'''moussaKam/barthez''': 1024,
'''moussaKam/barthez-orangesum-title''': 1024,
}
lowerCAmelCase__ = '''▁'''
class __snake_case ( _lowercase):
snake_case__ : int = VOCAB_FILES_NAMES
snake_case__ : Any = PRETRAINED_VOCAB_FILES_MAP
snake_case__ : Optional[Any] = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
snake_case__ : Union[str, Any] = ["input_ids", "attention_mask"]
def __init__( self : Union[str, Any] , __lowerCAmelCase : Optional[int] , __lowerCAmelCase : Optional[Any]="<s>" , __lowerCAmelCase : Dict="</s>" , __lowerCAmelCase : Optional[int]="</s>" , __lowerCAmelCase : Optional[int]="<s>" , __lowerCAmelCase : Optional[int]="<unk>" , __lowerCAmelCase : str="<pad>" , __lowerCAmelCase : Optional[Any]="<mask>" , __lowerCAmelCase : Optional[Dict[str, Any]] = None , **__lowerCAmelCase : Optional[int] , ):
"""simple docstring"""
_lowerCamelCase : Dict = AddedToken(__lowerCAmelCase , lstrip=__lowerCAmelCase , rstrip=__lowerCAmelCase ) if isinstance(__lowerCAmelCase , __lowerCAmelCase ) else mask_token
_lowerCamelCase : List[Any] = {} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
bos_token=__lowerCAmelCase , eos_token=__lowerCAmelCase , unk_token=__lowerCAmelCase , sep_token=__lowerCAmelCase , cls_token=__lowerCAmelCase , pad_token=__lowerCAmelCase , mask_token=__lowerCAmelCase , sp_model_kwargs=self.sp_model_kwargs , **__lowerCAmelCase , )
_lowerCamelCase : Dict = vocab_file
_lowerCamelCase : int = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(str(__lowerCAmelCase ) )
_lowerCamelCase : Union[str, Any] = {'''<s>''': 0, '''<pad>''': 1, '''</s>''': 2, '''<unk>''': 3}
_lowerCamelCase : int = len(self.sp_model ) - 1
_lowerCamelCase : List[str] = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
def SCREAMING_SNAKE_CASE ( self : Dict , __lowerCAmelCase : List[int] , __lowerCAmelCase : Optional[List[int]] = None ):
"""simple docstring"""
if token_ids_a is None:
return [self.cls_token_id] + token_ids_a + [self.sep_token_id]
_lowerCamelCase : List[str] = [self.cls_token_id]
_lowerCamelCase : int = [self.sep_token_id]
return cls + token_ids_a + sep + sep + token_ids_a + sep
def SCREAMING_SNAKE_CASE ( self : Dict , __lowerCAmelCase : List[int] , __lowerCAmelCase : Optional[List[int]] = None , __lowerCAmelCase : bool = False ):
"""simple docstring"""
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=__lowerCAmelCase , token_ids_a=__lowerCAmelCase , already_has_special_tokens=__lowerCAmelCase )
if token_ids_a is None:
return [1] + ([0] * len(__lowerCAmelCase )) + [1]
return [1] + ([0] * len(__lowerCAmelCase )) + [1, 1] + ([0] * len(__lowerCAmelCase )) + [1]
def SCREAMING_SNAKE_CASE ( self : List[Any] , __lowerCAmelCase : List[int] , __lowerCAmelCase : Optional[List[int]] = None ):
"""simple docstring"""
_lowerCamelCase : List[Any] = [self.sep_token_id]
_lowerCamelCase : Optional[int] = [self.cls_token_id]
if token_ids_a is None:
return len(cls + token_ids_a + sep ) * [0]
return len(cls + token_ids_a + sep + sep + token_ids_a + sep ) * [0]
@property
def SCREAMING_SNAKE_CASE ( self : Optional[int] ):
"""simple docstring"""
return len(self.sp_model )
def SCREAMING_SNAKE_CASE ( self : Dict ):
"""simple docstring"""
_lowerCamelCase : int = {self.convert_ids_to_tokens(__lowerCAmelCase ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def SCREAMING_SNAKE_CASE ( self : Union[str, Any] , __lowerCAmelCase : str ):
"""simple docstring"""
return self.sp_model.encode(__lowerCAmelCase , out_type=__lowerCAmelCase )
def SCREAMING_SNAKE_CASE ( self : Optional[Any] , __lowerCAmelCase : List[str] ):
"""simple docstring"""
if token in self.fairseq_tokens_to_ids:
return self.fairseq_tokens_to_ids[token]
_lowerCamelCase : List[Any] = self.sp_model.PieceToId(__lowerCAmelCase )
return spm_id if spm_id else self.unk_token_id
def SCREAMING_SNAKE_CASE ( self : List[Any] , __lowerCAmelCase : str ):
"""simple docstring"""
if index in self.fairseq_ids_to_tokens:
return self.fairseq_ids_to_tokens[index]
return self.sp_model.IdToPiece(__lowerCAmelCase )
def SCREAMING_SNAKE_CASE ( self : str , __lowerCAmelCase : Dict ):
"""simple docstring"""
_lowerCamelCase : Tuple = []
_lowerCamelCase : Tuple = ''''''
_lowerCamelCase : Dict = False
for token in tokens:
# make sure that special tokens are not decoded using sentencepiece model
if token in self.all_special_tokens:
if not prev_is_special:
out_string += " "
out_string += self.sp_model.decode(__lowerCAmelCase ) + token
_lowerCamelCase : Union[str, Any] = True
_lowerCamelCase : List[Any] = []
else:
current_sub_tokens.append(__lowerCAmelCase )
_lowerCamelCase : List[str] = False
out_string += self.sp_model.decode(__lowerCAmelCase )
return out_string.strip()
def __getstate__( self : List[str] ):
"""simple docstring"""
_lowerCamelCase : Dict = self.__dict__.copy()
_lowerCamelCase : int = None
return state
def __setstate__( self : int , __lowerCAmelCase : Any ):
"""simple docstring"""
_lowerCamelCase : Union[str, Any] = d
# for backward compatibility
if not hasattr(self , '''sp_model_kwargs''' ):
_lowerCamelCase : Dict = {}
_lowerCamelCase : Optional[Any] = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(self.vocab_file )
def SCREAMING_SNAKE_CASE ( self : Tuple , __lowerCAmelCase : str , __lowerCAmelCase : Optional[str] = None ):
"""simple docstring"""
if not os.path.isdir(__lowerCAmelCase ):
logger.error(f'''Vocabulary path ({save_directory}) should be a directory''' )
return
_lowerCamelCase : Any = os.path.join(
__lowerCAmelCase , (filename_prefix + '''-''' if filename_prefix else '''''') + VOCAB_FILES_NAMES['''vocab_file'''] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(__lowerCAmelCase ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , __lowerCAmelCase )
elif not os.path.isfile(self.vocab_file ):
with open(__lowerCAmelCase , '''wb''' ) as fi:
_lowerCamelCase : Tuple = self.sp_model.serialized_model_proto()
fi.write(__lowerCAmelCase )
return (out_vocab_file,)
| 72
|
from typing import Optional, Tuple, Union
import torch
from einops import rearrange, reduce
from diffusers import DDIMScheduler, DDPMScheduler, DiffusionPipeline, ImagePipelineOutput, UNetaDConditionModel
from diffusers.schedulers.scheduling_ddim import DDIMSchedulerOutput
from diffusers.schedulers.scheduling_ddpm import DDPMSchedulerOutput
snake_case_ : List[str] = 8
def A (__A : Union[str, Any] , __A : List[Any]=BITS ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x * 255).int().clamp(0 , 255 )
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b c h w -> b c 1 h w''' )
UpperCAmelCase_ = ((x & mask) != 0).float()
UpperCAmelCase_ = rearrange(__A , '''b c d h w -> b (c d) h w''' )
UpperCAmelCase_ = bits * 2 - 1
return bits
def A (__A : Dict , __A : Tuple=BITS ) -> List[str]:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x > 0).int()
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A , dtype=torch.intaa )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b (c d) h w -> b c d h w''' , d=8 )
UpperCAmelCase_ = reduce(x * mask , '''b c d h w -> b c h w''' , '''sum''' )
return (dec / 255).clamp(0.0 , 1.0 )
def A (self : List[Any] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : float = 0.0 , __A : bool = True , __A : Tuple=None , __A : bool = True , ) -> Union[DDIMSchedulerOutput, Tuple]:
"""simple docstring"""
if self.num_inference_steps is None:
raise ValueError(
'''Number of inference steps is \'None\', you need to run \'set_timesteps\' after creating the scheduler''' )
# See formulas (12) and (16) of DDIM paper https://arxiv.org/pdf/2010.02502.pdf
# Ideally, read DDIM paper in-detail understanding
# Notation (<variable name> -> <name in paper>
# - pred_noise_t -> e_theta(x_t, t)
# - pred_original_sample -> f_theta(x_t, t) or x_0
# - std_dev_t -> sigma_t
# - eta -> η
# - pred_sample_direction -> "direction pointing to x_t"
# - pred_prev_sample -> "x_t-1"
# 1. get previous step value (=t-1)
UpperCAmelCase_ = timestep - self.config.num_train_timesteps // self.num_inference_steps
# 2. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[timestep]
UpperCAmelCase_ = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
UpperCAmelCase_ = 1 - alpha_prod_t
# 3. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
# 4. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 5. compute variance: "sigma_t(η)" -> see formula (16)
# σ_t = sqrt((1 − α_t−1)/(1 − α_t)) * sqrt(1 − α_t/α_t−1)
UpperCAmelCase_ = self._get_variance(__A , __A )
UpperCAmelCase_ = eta * variance ** 0.5
if use_clipped_model_output:
# the model_output is always re-derived from the clipped x_0 in Glide
UpperCAmelCase_ = (sample - alpha_prod_t ** 0.5 * pred_original_sample) / beta_prod_t ** 0.5
# 6. compute "direction pointing to x_t" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (1 - alpha_prod_t_prev - std_dev_t**2) ** 0.5 * model_output
# 7. compute x_t without "random noise" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = alpha_prod_t_prev ** 0.5 * pred_original_sample + pred_sample_direction
if eta > 0:
# randn_like does not support generator https://github.com/pytorch/pytorch/issues/27072
UpperCAmelCase_ = model_output.device if torch.is_tensor(__A ) else '''cpu'''
UpperCAmelCase_ = torch.randn(model_output.shape , dtype=model_output.dtype , generator=__A ).to(__A )
UpperCAmelCase_ = self._get_variance(__A , __A ) ** 0.5 * eta * noise
UpperCAmelCase_ = prev_sample + variance
if not return_dict:
return (prev_sample,)
return DDIMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
def A (self : Optional[int] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : int="epsilon" , __A : Optional[Any]=None , __A : bool = True , ) -> Union[DDPMSchedulerOutput, Tuple]:
"""simple docstring"""
UpperCAmelCase_ = timestep
if model_output.shape[1] == sample.shape[1] * 2 and self.variance_type in ["learned", "learned_range"]:
UpperCAmelCase_ , UpperCAmelCase_ = torch.split(__A , sample.shape[1] , dim=1 )
else:
UpperCAmelCase_ = None
# 1. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[t]
UpperCAmelCase_ = self.alphas_cumprod[t - 1] if t > 0 else self.one
UpperCAmelCase_ = 1 - alpha_prod_t
UpperCAmelCase_ = 1 - alpha_prod_t_prev
# 2. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
if prediction_type == "epsilon":
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
elif prediction_type == "sample":
UpperCAmelCase_ = model_output
else:
raise ValueError(F"""Unsupported prediction_type {prediction_type}.""" )
# 3. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = (alpha_prod_t_prev ** 0.5 * self.betas[t]) / beta_prod_t
UpperCAmelCase_ = self.alphas[t] ** 0.5 * beta_prod_t_prev / beta_prod_t
# 5. Compute predicted previous sample µ_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * sample
# 6. Add noise
UpperCAmelCase_ = 0
if t > 0:
UpperCAmelCase_ = torch.randn(
model_output.size() , dtype=model_output.dtype , layout=model_output.layout , generator=__A ).to(model_output.device )
UpperCAmelCase_ = (self._get_variance(__A , predicted_variance=__A ) ** 0.5) * noise
UpperCAmelCase_ = pred_prev_sample + variance
if not return_dict:
return (pred_prev_sample,)
return DDPMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
class __snake_case ( a ):
def __init__( self : Union[str, Any] , _snake_case : UNetaDConditionModel , _snake_case : Union[DDIMScheduler, DDPMScheduler] , _snake_case : Optional[float] = 1.0 , ):
"""simple docstring"""
super().__init__()
UpperCAmelCase_ = bit_scale
UpperCAmelCase_ = (
ddim_bit_scheduler_step if isinstance(_snake_case , _snake_case) else ddpm_bit_scheduler_step
)
self.register_modules(unet=_snake_case , scheduler=_snake_case)
@torch.no_grad()
def __call__( self : Union[str, Any] , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 50 , _snake_case : Optional[torch.Generator] = None , _snake_case : Optional[int] = 1 , _snake_case : Optional[str] = "pil" , _snake_case : bool = True , **_snake_case : Optional[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = torch.randn(
(batch_size, self.unet.config.in_channels, height, width) , generator=_snake_case , )
UpperCAmelCase_ = decimal_to_bits(_snake_case) * self.bit_scale
UpperCAmelCase_ = latents.to(self.device)
self.scheduler.set_timesteps(_snake_case)
for t in self.progress_bar(self.scheduler.timesteps):
# predict the noise residual
UpperCAmelCase_ = self.unet(_snake_case , _snake_case).sample
# compute the previous noisy sample x_t -> x_t-1
UpperCAmelCase_ = self.scheduler.step(_snake_case , _snake_case , _snake_case).prev_sample
UpperCAmelCase_ = bits_to_decimal(_snake_case)
if output_type == "pil":
UpperCAmelCase_ = self.numpy_to_pil(_snake_case)
if not return_dict:
return (image,)
return ImagePipelineOutput(images=_snake_case)
| 51
| 0
|
import csv
import tweepy
# Twitter API credentials
a =""""""
a =""""""
a =""""""
a =""""""
def SCREAMING_SNAKE_CASE__ ( lowerCamelCase__ ) -> None:
# authorize twitter, initialize tweepy
__lowerCamelCase : Tuple = tweepy.OAuthHandler(lowerCamelCase__ , lowerCamelCase__ )
auth.set_access_token(lowerCamelCase__ , lowerCamelCase__ )
__lowerCamelCase : Optional[int] = tweepy.API(lowerCamelCase__ )
# initialize a list to hold all the tweepy Tweets
__lowerCamelCase : str = []
# make initial request for most recent tweets (200 is the maximum allowed count)
__lowerCamelCase : Union[str, Any] = api.user_timeline(screen_name=lowerCamelCase__ , count=2_0_0 )
# save most recent tweets
alltweets.extend(lowerCamelCase__ )
# save the id of the oldest tweet less one
__lowerCamelCase : Any = alltweets[-1].id - 1
# keep grabbing tweets until there are no tweets left to grab
while len(lowerCamelCase__ ) > 0:
print(F"getting tweets before {oldest}" )
# all subsequent requests use the max_id param to prevent duplicates
__lowerCamelCase : str = api.user_timeline(
screen_name=lowerCamelCase__ , count=2_0_0 , max_id=lowerCamelCase__ )
# save most recent tweets
alltweets.extend(lowerCamelCase__ )
# update the id of the oldest tweet less one
__lowerCamelCase : Optional[int] = alltweets[-1].id - 1
print(F"...{len(lowerCamelCase__ )} tweets downloaded so far" )
# transform the tweepy tweets into a 2D array that will populate the csv
__lowerCamelCase : str = [[tweet.id_str, tweet.created_at, tweet.text] for tweet in alltweets]
# write the csv
with open(F"new_{screen_name}_tweets.csv" , 'w' ) as f:
__lowerCamelCase : Any = csv.writer(lowerCamelCase__ )
writer.writerow(['id', 'created_at', 'text'] )
writer.writerows(lowerCamelCase__ )
if __name__ == "__main__":
# pass in the username of the account you want to download
get_all_tweets("""FirePing32""")
| 73
|
snake_case_ : Dict = {
"Pillow": "Pillow",
"accelerate": "accelerate>=0.11.0",
"compel": "compel==0.1.8",
"black": "black~=23.1",
"datasets": "datasets",
"filelock": "filelock",
"flax": "flax>=0.4.1",
"hf-doc-builder": "hf-doc-builder>=0.3.0",
"huggingface-hub": "huggingface-hub>=0.13.2",
"requests-mock": "requests-mock==1.10.0",
"importlib_metadata": "importlib_metadata",
"invisible-watermark": "invisible-watermark",
"isort": "isort>=5.5.4",
"jax": "jax>=0.2.8,!=0.3.2",
"jaxlib": "jaxlib>=0.1.65",
"Jinja2": "Jinja2",
"k-diffusion": "k-diffusion>=0.0.12",
"torchsde": "torchsde",
"note_seq": "note_seq",
"librosa": "librosa",
"numpy": "numpy",
"omegaconf": "omegaconf",
"parameterized": "parameterized",
"protobuf": "protobuf>=3.20.3,<4",
"pytest": "pytest",
"pytest-timeout": "pytest-timeout",
"pytest-xdist": "pytest-xdist",
"ruff": "ruff>=0.0.241",
"safetensors": "safetensors",
"sentencepiece": "sentencepiece>=0.1.91,!=0.1.92",
"scipy": "scipy",
"onnx": "onnx",
"regex": "regex!=2019.12.17",
"requests": "requests",
"tensorboard": "tensorboard",
"torch": "torch>=1.4",
"torchvision": "torchvision",
"transformers": "transformers>=4.25.1",
"urllib3": "urllib3<=2.0.0",
}
| 51
| 0
|
"""simple docstring"""
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
is_vision_available,
)
_lowercase = {
'''configuration_owlvit''': [
'''OWLVIT_PRETRAINED_CONFIG_ARCHIVE_MAP''',
'''OwlViTConfig''',
'''OwlViTOnnxConfig''',
'''OwlViTTextConfig''',
'''OwlViTVisionConfig''',
],
'''processing_owlvit''': ['''OwlViTProcessor'''],
}
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_lowercase = ['''OwlViTFeatureExtractor''']
_lowercase = ['''OwlViTImageProcessor''']
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_lowercase = [
'''OWLVIT_PRETRAINED_MODEL_ARCHIVE_LIST''',
'''OwlViTModel''',
'''OwlViTPreTrainedModel''',
'''OwlViTTextModel''',
'''OwlViTVisionModel''',
'''OwlViTForObjectDetection''',
]
if TYPE_CHECKING:
from .configuration_owlvit import (
OWLVIT_PRETRAINED_CONFIG_ARCHIVE_MAP,
OwlViTConfig,
OwlViTOnnxConfig,
OwlViTTextConfig,
OwlViTVisionConfig,
)
from .processing_owlvit import OwlViTProcessor
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .feature_extraction_owlvit import OwlViTFeatureExtractor
from .image_processing_owlvit import OwlViTImageProcessor
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_owlvit import (
OWLVIT_PRETRAINED_MODEL_ARCHIVE_LIST,
OwlViTForObjectDetection,
OwlViTModel,
OwlViTPreTrainedModel,
OwlViTTextModel,
OwlViTVisionModel,
)
else:
import sys
_lowercase = _LazyModule(__name__, globals()['''__file__'''], _import_structure, module_spec=__spec__)
| 74
|
from datetime import datetime
import requests
def A (__A : str ) -> bytes:
"""simple docstring"""
UpperCAmelCase_ = '''https://downloadgram.net/wp-json/wppress/video-downloader/video?url='''
UpperCAmelCase_ = requests.get(base_url + url ).json()[0]['''urls'''][0]['''src''']
return requests.get(__A ).content
if __name__ == "__main__":
snake_case_ : Optional[Any] = input("Enter Video/IGTV url: ").strip()
snake_case_ : Any = f"{datetime.now():%Y-%m-%d_%H:%M:%S}.mp4"
with open(file_name, "wb") as fp:
fp.write(download_video(url))
print(f"Done. Video saved to disk as {file_name}.")
| 51
| 0
|
'''simple docstring'''
import json
import pathlib
import unittest
import numpy as np
from transformers.testing_utils import require_torch, require_vision, slow
from transformers.utils import is_torch_available, is_vision_available
from ...test_image_processing_common import ImageProcessingSavingTestMixin, prepare_image_inputs
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
from transformers import DetaImageProcessor
class __UpperCamelCase ( unittest.TestCase ):
def __init__( self, lowerCAmelCase, lowerCAmelCase=7, lowerCAmelCase=3, lowerCAmelCase=30, lowerCAmelCase=400, lowerCAmelCase=True, lowerCAmelCase=None, lowerCAmelCase=True, lowerCAmelCase=[0.5, 0.5, 0.5], lowerCAmelCase=[0.5, 0.5, 0.5], lowerCAmelCase=True, lowerCAmelCase=1 / 255, lowerCAmelCase=True, ):
"""simple docstring"""
lowerCamelCase_ =size if size is not None else {'''shortest_edge''': 18, '''longest_edge''': 1_333}
lowerCamelCase_ =parent
lowerCamelCase_ =batch_size
lowerCamelCase_ =num_channels
lowerCamelCase_ =min_resolution
lowerCamelCase_ =max_resolution
lowerCamelCase_ =do_resize
lowerCamelCase_ =size
lowerCamelCase_ =do_normalize
lowerCamelCase_ =image_mean
lowerCamelCase_ =image_std
lowerCamelCase_ =do_rescale
lowerCamelCase_ =rescale_factor
lowerCamelCase_ =do_pad
def lowercase__ ( self ):
"""simple docstring"""
return {
"do_resize": self.do_resize,
"size": self.size,
"do_normalize": self.do_normalize,
"image_mean": self.image_mean,
"image_std": self.image_std,
"do_rescale": self.do_rescale,
"rescale_factor": self.rescale_factor,
"do_pad": self.do_pad,
}
def lowercase__ ( self, lowerCAmelCase, lowerCAmelCase=False ):
"""simple docstring"""
if not batched:
lowerCamelCase_ =image_inputs[0]
if isinstance(lowerCAmelCase, Image.Image ):
lowerCamelCase_, lowerCamelCase_ =image.size
else:
lowerCamelCase_, lowerCamelCase_ =image.shape[1], image.shape[2]
if w < h:
lowerCamelCase_ =int(self.size['''shortest_edge'''] * h / w )
lowerCamelCase_ =self.size['''shortest_edge''']
elif w > h:
lowerCamelCase_ =self.size['''shortest_edge''']
lowerCamelCase_ =int(self.size['''shortest_edge'''] * w / h )
else:
lowerCamelCase_ =self.size['''shortest_edge''']
lowerCamelCase_ =self.size['''shortest_edge''']
else:
lowerCamelCase_ =[]
for image in image_inputs:
lowerCamelCase_, lowerCamelCase_ =self.get_expected_values([image] )
expected_values.append((expected_height, expected_width) )
lowerCamelCase_ =max(lowerCAmelCase, key=lambda lowerCAmelCase : item[0] )[0]
lowerCamelCase_ =max(lowerCAmelCase, key=lambda lowerCAmelCase : item[1] )[1]
return expected_height, expected_width
@require_torch
@require_vision
class __UpperCamelCase ( lowerCamelCase__ , unittest.TestCase ):
lowercase : List[str] =DetaImageProcessor if is_vision_available() else None
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =DetaImageProcessingTester(self )
@property
def lowercase__ ( self ):
"""simple docstring"""
return self.image_processor_tester.prepare_image_processor_dict()
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =self.image_processing_class(**self.image_processor_dict )
self.assertTrue(hasattr(lowerCAmelCase, '''image_mean''' ) )
self.assertTrue(hasattr(lowerCAmelCase, '''image_std''' ) )
self.assertTrue(hasattr(lowerCAmelCase, '''do_normalize''' ) )
self.assertTrue(hasattr(lowerCAmelCase, '''do_resize''' ) )
self.assertTrue(hasattr(lowerCAmelCase, '''do_rescale''' ) )
self.assertTrue(hasattr(lowerCAmelCase, '''do_pad''' ) )
self.assertTrue(hasattr(lowerCAmelCase, '''size''' ) )
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =self.image_processing_class.from_dict(self.image_processor_dict )
self.assertEqual(image_processor.size, {'''shortest_edge''': 18, '''longest_edge''': 1_333} )
self.assertEqual(image_processor.do_pad, lowerCAmelCase )
def lowercase__ ( self ):
"""simple docstring"""
pass
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =self.image_processing_class(**self.image_processor_dict )
# create random PIL images
lowerCamelCase_ =prepare_image_inputs(self.image_processor_tester, equal_resolution=lowerCAmelCase )
for image in image_inputs:
self.assertIsInstance(lowerCAmelCase, Image.Image )
# Test not batched input
lowerCamelCase_ =image_processing(image_inputs[0], return_tensors='''pt''' ).pixel_values
lowerCamelCase_, lowerCamelCase_ =self.image_processor_tester.get_expected_values(lowerCAmelCase )
self.assertEqual(
encoded_images.shape, (1, self.image_processor_tester.num_channels, expected_height, expected_width), )
# Test batched
lowerCamelCase_, lowerCamelCase_ =self.image_processor_tester.get_expected_values(lowerCAmelCase, batched=lowerCAmelCase )
lowerCamelCase_ =image_processing(lowerCAmelCase, return_tensors='''pt''' ).pixel_values
self.assertEqual(
encoded_images.shape, (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
expected_height,
expected_width,
), )
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =self.image_processing_class(**self.image_processor_dict )
# create random numpy tensors
lowerCamelCase_ =prepare_image_inputs(self.image_processor_tester, equal_resolution=lowerCAmelCase, numpify=lowerCAmelCase )
for image in image_inputs:
self.assertIsInstance(lowerCAmelCase, np.ndarray )
# Test not batched input
lowerCamelCase_ =image_processing(image_inputs[0], return_tensors='''pt''' ).pixel_values
lowerCamelCase_, lowerCamelCase_ =self.image_processor_tester.get_expected_values(lowerCAmelCase )
self.assertEqual(
encoded_images.shape, (1, self.image_processor_tester.num_channels, expected_height, expected_width), )
# Test batched
lowerCamelCase_ =image_processing(lowerCAmelCase, return_tensors='''pt''' ).pixel_values
lowerCamelCase_, lowerCamelCase_ =self.image_processor_tester.get_expected_values(lowerCAmelCase, batched=lowerCAmelCase )
self.assertEqual(
encoded_images.shape, (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
expected_height,
expected_width,
), )
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =self.image_processing_class(**self.image_processor_dict )
# create random PyTorch tensors
lowerCamelCase_ =prepare_image_inputs(self.image_processor_tester, equal_resolution=lowerCAmelCase, torchify=lowerCAmelCase )
for image in image_inputs:
self.assertIsInstance(lowerCAmelCase, torch.Tensor )
# Test not batched input
lowerCamelCase_ =image_processing(image_inputs[0], return_tensors='''pt''' ).pixel_values
lowerCamelCase_, lowerCamelCase_ =self.image_processor_tester.get_expected_values(lowerCAmelCase )
self.assertEqual(
encoded_images.shape, (1, self.image_processor_tester.num_channels, expected_height, expected_width), )
# Test batched
lowerCamelCase_ =image_processing(lowerCAmelCase, return_tensors='''pt''' ).pixel_values
lowerCamelCase_, lowerCamelCase_ =self.image_processor_tester.get_expected_values(lowerCAmelCase, batched=lowerCAmelCase )
self.assertEqual(
encoded_images.shape, (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
expected_height,
expected_width,
), )
@slow
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
with open('''./tests/fixtures/tests_samples/COCO/coco_annotations.txt''', '''r''' ) as f:
lowerCamelCase_ =json.loads(f.read() )
lowerCamelCase_ ={'''image_id''': 39_769, '''annotations''': target}
# encode them
lowerCamelCase_ =DetaImageProcessor()
lowerCamelCase_ =image_processing(images=lowerCAmelCase, annotations=lowerCAmelCase, return_tensors='''pt''' )
# verify pixel values
lowerCamelCase_ =torch.Size([1, 3, 800, 1_066] )
self.assertEqual(encoding['''pixel_values'''].shape, lowerCAmelCase )
lowerCamelCase_ =torch.tensor([0.2_7_9_6, 0.3_1_3_8, 0.3_4_8_1] )
self.assertTrue(torch.allclose(encoding['''pixel_values'''][0, 0, 0, :3], lowerCAmelCase, atol=1e-4 ) )
# verify area
lowerCamelCase_ =torch.tensor([5_8_8_7.9_6_0_0, 1_1_2_5_0.2_0_6_1, 4_8_9_3_5_3.8_4_3_8, 8_3_7_1_2_2.7_5_0_0, 1_4_7_9_6_7.5_1_5_6, 1_6_5_7_3_2.3_4_3_8] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''area'''], lowerCAmelCase ) )
# verify boxes
lowerCamelCase_ =torch.Size([6, 4] )
self.assertEqual(encoding['''labels'''][0]['''boxes'''].shape, lowerCAmelCase )
lowerCamelCase_ =torch.tensor([0.5_5_0_3, 0.2_7_6_5, 0.0_6_0_4, 0.2_2_1_5] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''boxes'''][0], lowerCAmelCase, atol=1e-3 ) )
# verify image_id
lowerCamelCase_ =torch.tensor([39_769] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''image_id'''], lowerCAmelCase ) )
# verify is_crowd
lowerCamelCase_ =torch.tensor([0, 0, 0, 0, 0, 0] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''iscrowd'''], lowerCAmelCase ) )
# verify class_labels
lowerCamelCase_ =torch.tensor([75, 75, 63, 65, 17, 17] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''class_labels'''], lowerCAmelCase ) )
# verify orig_size
lowerCamelCase_ =torch.tensor([480, 640] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''orig_size'''], lowerCAmelCase ) )
# verify size
lowerCamelCase_ =torch.tensor([800, 1_066] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''size'''], lowerCAmelCase ) )
@slow
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
with open('''./tests/fixtures/tests_samples/COCO/coco_panoptic_annotations.txt''', '''r''' ) as f:
lowerCamelCase_ =json.loads(f.read() )
lowerCamelCase_ ={'''file_name''': '''000000039769.png''', '''image_id''': 39_769, '''segments_info''': target}
lowerCamelCase_ =pathlib.Path('''./tests/fixtures/tests_samples/COCO/coco_panoptic''' )
# encode them
lowerCamelCase_ =DetaImageProcessor(format='''coco_panoptic''' )
lowerCamelCase_ =image_processing(images=lowerCAmelCase, annotations=lowerCAmelCase, masks_path=lowerCAmelCase, return_tensors='''pt''' )
# verify pixel values
lowerCamelCase_ =torch.Size([1, 3, 800, 1_066] )
self.assertEqual(encoding['''pixel_values'''].shape, lowerCAmelCase )
lowerCamelCase_ =torch.tensor([0.2_7_9_6, 0.3_1_3_8, 0.3_4_8_1] )
self.assertTrue(torch.allclose(encoding['''pixel_values'''][0, 0, 0, :3], lowerCAmelCase, atol=1e-4 ) )
# verify area
lowerCamelCase_ =torch.tensor([1_4_7_9_7_9.6_8_7_5, 1_6_5_5_2_7.0_4_6_9, 4_8_4_6_3_8.5_9_3_8, 1_1_2_9_2.9_3_7_5, 5_8_7_9.6_5_6_2, 7_6_3_4.1_1_4_7] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''area'''], lowerCAmelCase ) )
# verify boxes
lowerCamelCase_ =torch.Size([6, 4] )
self.assertEqual(encoding['''labels'''][0]['''boxes'''].shape, lowerCAmelCase )
lowerCamelCase_ =torch.tensor([0.2_6_2_5, 0.5_4_3_7, 0.4_6_8_8, 0.8_6_2_5] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''boxes'''][0], lowerCAmelCase, atol=1e-3 ) )
# verify image_id
lowerCamelCase_ =torch.tensor([39_769] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''image_id'''], lowerCAmelCase ) )
# verify is_crowd
lowerCamelCase_ =torch.tensor([0, 0, 0, 0, 0, 0] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''iscrowd'''], lowerCAmelCase ) )
# verify class_labels
lowerCamelCase_ =torch.tensor([17, 17, 63, 75, 75, 93] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''class_labels'''], lowerCAmelCase ) )
# verify masks
lowerCamelCase_ =822_873
self.assertEqual(encoding['''labels'''][0]['''masks'''].sum().item(), lowerCAmelCase )
# verify orig_size
lowerCamelCase_ =torch.tensor([480, 640] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''orig_size'''], lowerCAmelCase ) )
# verify size
lowerCamelCase_ =torch.tensor([800, 1_066] )
self.assertTrue(torch.allclose(encoding['''labels'''][0]['''size'''], lowerCAmelCase ) )
| 75
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[Any] = {
"tiiuae/falcon-40b": "https://huggingface.co/tiiuae/falcon-40b/resolve/main/config.json",
"tiiuae/falcon-7b": "https://huggingface.co/tiiuae/falcon-7b/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : Optional[Any] = '''falcon'''
UpperCAmelCase__ : List[Any] = ['''past_key_values''']
def __init__( self : Union[str, Any] , _snake_case : List[str]=65024 , _snake_case : int=4544 , _snake_case : int=32 , _snake_case : Any=71 , _snake_case : int=1e-5 , _snake_case : Dict=0.0_2 , _snake_case : int=True , _snake_case : List[Any]=0.0 , _snake_case : Tuple=0.0 , _snake_case : int=None , _snake_case : Tuple=False , _snake_case : Any=False , _snake_case : str=True , _snake_case : Any=True , _snake_case : List[str]=False , _snake_case : Tuple=11 , _snake_case : Dict=11 , **_snake_case : Optional[int] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
# Backward compatibility with n_embed kwarg
UpperCAmelCase_ = kwargs.pop('''n_embed''' , _snake_case)
UpperCAmelCase_ = hidden_size if n_embed is None else n_embed
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = hidden_dropout
UpperCAmelCase_ = attention_dropout
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
UpperCAmelCase_ = num_attention_heads if num_kv_heads is None else num_kv_heads
UpperCAmelCase_ = alibi
UpperCAmelCase_ = new_decoder_architecture
UpperCAmelCase_ = multi_query # Ignored when new_decoder_architecture is True
UpperCAmelCase_ = parallel_attn
UpperCAmelCase_ = bias
super().__init__(bos_token_id=_snake_case , eos_token_id=_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
return self.hidden_size // self.num_attention_heads
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return not self.alibi
| 51
| 0
|
def lowerCamelCase__ ( _a , _a):
return int(input_a == input_a == 0)
def lowerCamelCase__ ( ):
print("Truth Table of NOR Gate:")
print("| Input 1 | Input 2 | Output |")
print(f"| 0 | 0 | {nor_gate(0 , 0)} |")
print(f"| 0 | 1 | {nor_gate(0 , 1)} |")
print(f"| 1 | 0 | {nor_gate(1 , 0)} |")
print(f"| 1 | 1 | {nor_gate(1 , 1)} |")
if __name__ == "__main__":
import doctest
doctest.testmod()
main()
| 76
|
from __future__ import annotations
import time
from math import sqrt
# 1 for manhattan, 0 for euclidean
snake_case_ : str = 0
snake_case_ : Union[str, Any] = [
[0, 0, 0, 0, 0, 0, 0],
[0, 1, 0, 0, 0, 0, 0], # 0 are free path whereas 1's are obstacles
[0, 0, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 0],
[1, 0, 1, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 0, 0],
]
snake_case_ : str = [[-1, 0], [0, -1], [1, 0], [0, 1]] # up, left, down, right
snake_case_ : List[Any] = tuple[int, int]
class __snake_case :
def __init__( self : Any , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : Node | None , ):
"""simple docstring"""
UpperCAmelCase_ = pos_x
UpperCAmelCase_ = pos_y
UpperCAmelCase_ = (pos_y, pos_x)
UpperCAmelCase_ = goal_x
UpperCAmelCase_ = goal_y
UpperCAmelCase_ = g_cost
UpperCAmelCase_ = parent
UpperCAmelCase_ = self.calculate_heuristic()
UpperCAmelCase_ = self.g_cost + self.h_cost
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = self.pos_x - self.goal_x
UpperCAmelCase_ = self.pos_y - self.goal_y
if HEURISTIC == 1:
return abs(_snake_case) + abs(_snake_case)
else:
return sqrt(dy**2 + dx**2)
def __lt__( self : Union[str, Any] , _snake_case : Node):
"""simple docstring"""
return self.f_cost < other.f_cost
class __snake_case :
def __init__( self : str , _snake_case : TPosition , _snake_case : TPosition):
"""simple docstring"""
UpperCAmelCase_ = Node(start[1] , start[0] , goal[1] , goal[0] , 0 , _snake_case)
UpperCAmelCase_ = Node(goal[1] , goal[0] , goal[1] , goal[0] , 99999 , _snake_case)
UpperCAmelCase_ = [self.start]
UpperCAmelCase_ = []
UpperCAmelCase_ = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
while self.open_nodes:
# Open Nodes are sorted using __lt__
self.open_nodes.sort()
UpperCAmelCase_ = self.open_nodes.pop(0)
if current_node.pos == self.target.pos:
return self.retrace_path(_snake_case)
self.closed_nodes.append(_snake_case)
UpperCAmelCase_ = self.get_successors(_snake_case)
for child_node in successors:
if child_node in self.closed_nodes:
continue
if child_node not in self.open_nodes:
self.open_nodes.append(_snake_case)
else:
# retrieve the best current path
UpperCAmelCase_ = self.open_nodes.pop(self.open_nodes.index(_snake_case))
if child_node.g_cost < better_node.g_cost:
self.open_nodes.append(_snake_case)
else:
self.open_nodes.append(_snake_case)
return [self.start.pos]
def lowerCamelCase ( self : Tuple , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = []
for action in delta:
UpperCAmelCase_ = parent.pos_x + action[1]
UpperCAmelCase_ = parent.pos_y + action[0]
if not (0 <= pos_x <= len(grid[0]) - 1 and 0 <= pos_y <= len(_snake_case) - 1):
continue
if grid[pos_y][pos_x] != 0:
continue
successors.append(
Node(
_snake_case , _snake_case , self.target.pos_y , self.target.pos_x , parent.g_cost + 1 , _snake_case , ))
return successors
def lowerCamelCase ( self : Any , _snake_case : Node | None):
"""simple docstring"""
UpperCAmelCase_ = node
UpperCAmelCase_ = []
while current_node is not None:
path.append((current_node.pos_y, current_node.pos_x))
UpperCAmelCase_ = current_node.parent
path.reverse()
return path
class __snake_case :
def __init__( self : Any , _snake_case : TPosition , _snake_case : TPosition):
"""simple docstring"""
UpperCAmelCase_ = AStar(_snake_case , _snake_case)
UpperCAmelCase_ = AStar(_snake_case , _snake_case)
UpperCAmelCase_ = False
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
while self.fwd_astar.open_nodes or self.bwd_astar.open_nodes:
self.fwd_astar.open_nodes.sort()
self.bwd_astar.open_nodes.sort()
UpperCAmelCase_ = self.fwd_astar.open_nodes.pop(0)
UpperCAmelCase_ = self.bwd_astar.open_nodes.pop(0)
if current_bwd_node.pos == current_fwd_node.pos:
return self.retrace_bidirectional_path(
_snake_case , _snake_case)
self.fwd_astar.closed_nodes.append(_snake_case)
self.bwd_astar.closed_nodes.append(_snake_case)
UpperCAmelCase_ = current_bwd_node
UpperCAmelCase_ = current_fwd_node
UpperCAmelCase_ = {
self.fwd_astar: self.fwd_astar.get_successors(_snake_case),
self.bwd_astar: self.bwd_astar.get_successors(_snake_case),
}
for astar in [self.fwd_astar, self.bwd_astar]:
for child_node in successors[astar]:
if child_node in astar.closed_nodes:
continue
if child_node not in astar.open_nodes:
astar.open_nodes.append(_snake_case)
else:
# retrieve the best current path
UpperCAmelCase_ = astar.open_nodes.pop(
astar.open_nodes.index(_snake_case))
if child_node.g_cost < better_node.g_cost:
astar.open_nodes.append(_snake_case)
else:
astar.open_nodes.append(_snake_case)
return [self.fwd_astar.start.pos]
def lowerCamelCase ( self : int , _snake_case : Node , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = self.fwd_astar.retrace_path(_snake_case)
UpperCAmelCase_ = self.bwd_astar.retrace_path(_snake_case)
bwd_path.pop()
bwd_path.reverse()
UpperCAmelCase_ = fwd_path + bwd_path
return path
if __name__ == "__main__":
# all coordinates are given in format [y,x]
snake_case_ : Any = (0, 0)
snake_case_ : Union[str, Any] = (len(grid) - 1, len(grid[0]) - 1)
for elem in grid:
print(elem)
snake_case_ : str = time.time()
snake_case_ : List[str] = AStar(init, goal)
snake_case_ : Optional[int] = a_star.search()
snake_case_ : Optional[Any] = time.time() - start_time
print(f"AStar execution time = {end_time:f} seconds")
snake_case_ : int = time.time()
snake_case_ : Dict = BidirectionalAStar(init, goal)
snake_case_ : str = time.time() - bd_start_time
print(f"BidirectionalAStar execution time = {bd_end_time:f} seconds")
| 51
| 0
|
"""simple docstring"""
from ...configuration_utils import PretrainedConfig
from ...utils import logging
_UpperCamelCase : Optional[Any] = logging.get_logger(__name__)
_UpperCamelCase : Optional[int] = {
"abeja/gpt-neox-japanese-2.7b": "https://huggingface.co/abeja/gpt-neox-japanese-2.7b/resolve/main/config.json",
}
class UpperCAmelCase_ ( _a):
lowerCamelCase__ : Dict = "gpt_neox_japanese"
def __init__( self , a=3_2_0_0_0 , a=2_5_6_0 , a=3_2 , a=3_2 , a=4 , a="gelu" , a=1.00 , a=1_0_0_0_0 , a=2_0_4_8 , a=0.02 , a=1e-5 , a=True , a=3_1_9_9_6 , a=3_1_9_9_9 , a=0.1 , a=0.0 , **a , ) -> Optional[Any]:
super().__init__(bos_token_id=a , eos_token_id=a , **a )
lowercase__ : Optional[int] = vocab_size
lowercase__ : int = max_position_embeddings
lowercase__ : Optional[Any] = hidden_size
lowercase__ : Optional[Any] = num_hidden_layers
lowercase__ : Optional[int] = num_attention_heads
lowercase__ : Optional[int] = intermediate_multiple_size
lowercase__ : Dict = hidden_act
lowercase__ : Union[str, Any] = rotary_pct
lowercase__ : Tuple = rotary_emb_base
lowercase__ : Dict = initializer_range
lowercase__ : List[str] = layer_norm_eps
lowercase__ : List[Any] = use_cache
lowercase__ : Dict = attention_dropout
lowercase__ : Tuple = hidden_dropout
| 77
|
import inspect
import unittest
import numpy as np
from tests.test_modeling_common import floats_tensor
from transformers import DetrConfig, MaskFormerConfig, SwinConfig, is_torch_available, is_vision_available
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import MaskFormerForInstanceSegmentation, MaskFormerModel
if is_vision_available():
from transformers import MaskFormerImageProcessor
if is_vision_available():
from PIL import Image
class __snake_case :
def __init__( self : List[str] , _snake_case : Union[str, Any] , _snake_case : List[str]=2 , _snake_case : Any=True , _snake_case : Any=False , _snake_case : List[str]=10 , _snake_case : Any=3 , _snake_case : Union[str, Any]=32 * 4 , _snake_case : List[Any]=32 * 6 , _snake_case : Tuple=4 , _snake_case : Dict=32 , ):
"""simple docstring"""
UpperCAmelCase_ = parent
UpperCAmelCase_ = batch_size
UpperCAmelCase_ = is_training
UpperCAmelCase_ = use_auxiliary_loss
UpperCAmelCase_ = num_queries
UpperCAmelCase_ = num_channels
UpperCAmelCase_ = min_size
UpperCAmelCase_ = max_size
UpperCAmelCase_ = num_labels
UpperCAmelCase_ = mask_feature_size
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = floats_tensor([self.batch_size, self.num_channels, self.min_size, self.max_size]).to(
_snake_case)
UpperCAmelCase_ = torch.ones([self.batch_size, self.min_size, self.max_size] , device=_snake_case)
UpperCAmelCase_ = (
torch.rand([self.batch_size, self.num_labels, self.min_size, self.max_size] , device=_snake_case) > 0.5
).float()
UpperCAmelCase_ = (torch.rand((self.batch_size, self.num_labels) , device=_snake_case) > 0.5).long()
UpperCAmelCase_ = self.get_config()
return config, pixel_values, pixel_mask, mask_labels, class_labels
def lowerCamelCase ( self : Any):
"""simple docstring"""
return MaskFormerConfig.from_backbone_and_decoder_configs(
backbone_config=SwinConfig(
depths=[1, 1, 1, 1] , ) , decoder_config=DetrConfig(
decoder_ffn_dim=128 , num_queries=self.num_queries , decoder_attention_heads=2 , d_model=self.mask_feature_size , ) , mask_feature_size=self.mask_feature_size , fpn_feature_size=self.mask_feature_size , num_channels=self.num_channels , num_labels=self.num_labels , )
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.prepare_config_and_inputs()
UpperCAmelCase_ = {'''pixel_values''': pixel_values, '''pixel_mask''': pixel_mask}
return config, inputs_dict
def lowerCamelCase ( self : str , _snake_case : List[Any] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = output.encoder_hidden_states
UpperCAmelCase_ = output.pixel_decoder_hidden_states
UpperCAmelCase_ = output.transformer_decoder_hidden_states
self.parent.assertTrue(len(_snake_case) , len(config.backbone_config.depths))
self.parent.assertTrue(len(_snake_case) , len(config.backbone_config.depths))
self.parent.assertTrue(len(_snake_case) , config.decoder_config.decoder_layers)
def lowerCamelCase ( self : Union[str, Any] , _snake_case : List[str] , _snake_case : int , _snake_case : Optional[Any] , _snake_case : str=False):
"""simple docstring"""
with torch.no_grad():
UpperCAmelCase_ = MaskFormerModel(config=_snake_case)
model.to(_snake_case)
model.eval()
UpperCAmelCase_ = model(pixel_values=_snake_case , pixel_mask=_snake_case)
UpperCAmelCase_ = model(_snake_case , output_hidden_states=_snake_case)
# the correct shape of output.transformer_decoder_hidden_states ensure the correcteness of the
# encoder and pixel decoder
self.parent.assertEqual(
output.transformer_decoder_last_hidden_state.shape , (self.batch_size, self.num_queries, self.mask_feature_size) , )
# let's ensure the other two hidden state exists
self.parent.assertTrue(output.pixel_decoder_last_hidden_state is not None)
self.parent.assertTrue(output.encoder_last_hidden_state is not None)
if output_hidden_states:
self.check_output_hidden_state(_snake_case , _snake_case)
def lowerCamelCase ( self : List[Any] , _snake_case : List[Any] , _snake_case : List[Any] , _snake_case : str , _snake_case : Optional[int] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerForInstanceSegmentation(config=_snake_case)
model.to(_snake_case)
model.eval()
def comm_check_on_output(_snake_case : Tuple):
# let's still check that all the required stuff is there
self.parent.assertTrue(result.transformer_decoder_last_hidden_state is not None)
self.parent.assertTrue(result.pixel_decoder_last_hidden_state is not None)
self.parent.assertTrue(result.encoder_last_hidden_state is not None)
# okay, now we need to check the logits shape
# due to the encoder compression, masks have a //4 spatial size
self.parent.assertEqual(
result.masks_queries_logits.shape , (self.batch_size, self.num_queries, self.min_size // 4, self.max_size // 4) , )
# + 1 for null class
self.parent.assertEqual(
result.class_queries_logits.shape , (self.batch_size, self.num_queries, self.num_labels + 1))
with torch.no_grad():
UpperCAmelCase_ = model(pixel_values=_snake_case , pixel_mask=_snake_case)
UpperCAmelCase_ = model(_snake_case)
comm_check_on_output(_snake_case)
UpperCAmelCase_ = model(
pixel_values=_snake_case , pixel_mask=_snake_case , mask_labels=_snake_case , class_labels=_snake_case)
comm_check_on_output(_snake_case)
self.parent.assertTrue(result.loss is not None)
self.parent.assertEqual(result.loss.shape , torch.Size([1]))
@require_torch
class __snake_case ( a , a , unittest.TestCase ):
UpperCAmelCase__ : Union[str, Any] = (MaskFormerModel, MaskFormerForInstanceSegmentation) if is_torch_available() else ()
UpperCAmelCase__ : Optional[Any] = (
{'''feature-extraction''': MaskFormerModel, '''image-segmentation''': MaskFormerForInstanceSegmentation}
if is_torch_available()
else {}
)
UpperCAmelCase__ : Dict = False
UpperCAmelCase__ : List[str] = False
UpperCAmelCase__ : Optional[Any] = False
UpperCAmelCase__ : Union[str, Any] = False
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerModelTester(self)
UpperCAmelCase_ = ConfigTester(self , config_class=_snake_case , has_text_modality=_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
self.config_tester.run_common_tests()
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskformer_model(_snake_case , **_snake_case , output_hidden_states=_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_maskformer_instance_segmentation_head_model(*_snake_case)
@unittest.skip(reason='''MaskFormer does not use inputs_embeds''')
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer does not have a get_input_embeddings method''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer is not a generative model''')
def lowerCamelCase ( self : str):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer does not use token embeddings''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
@require_torch_multi_gpu
@unittest.skip(
reason='''MaskFormer has some layers using `add_module` which doesn\'t work well with `nn.DataParallel`''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
pass
@unittest.skip('''Will be fixed soon by reducing the size of the model used for common tests.''')
def lowerCamelCase ( self : str):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ = model_class(_snake_case)
UpperCAmelCase_ = inspect.signature(model.forward)
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCAmelCase_ = [*signature.parameters.keys()]
UpperCAmelCase_ = ['''pixel_values''']
self.assertListEqual(arg_names[:1] , _snake_case)
@slow
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
for model_name in ["facebook/maskformer-swin-small-coco"]:
UpperCAmelCase_ = MaskFormerModel.from_pretrained(_snake_case)
self.assertIsNotNone(_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = (self.model_tester.min_size,) * 2
UpperCAmelCase_ = {
'''pixel_values''': torch.randn((2, 3, *size) , device=_snake_case),
'''mask_labels''': torch.randn((2, 10, *size) , device=_snake_case),
'''class_labels''': torch.zeros(2 , 10 , device=_snake_case).long(),
}
UpperCAmelCase_ = MaskFormerForInstanceSegmentation(MaskFormerConfig()).to(_snake_case)
UpperCAmelCase_ = model(**_snake_case)
self.assertTrue(outputs.loss is not None)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskformer_model(_snake_case , **_snake_case , output_hidden_states=_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ = model_class(_snake_case).to(_snake_case)
UpperCAmelCase_ = model(**_snake_case , output_attentions=_snake_case)
self.assertTrue(outputs.attentions is not None)
def lowerCamelCase ( self : int):
"""simple docstring"""
if not self.model_tester.is_training:
return
# only MaskFormerForInstanceSegmentation has the loss
UpperCAmelCase_ = self.all_model_classes[1]
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCAmelCase_ = model_class(_snake_case)
model.to(_snake_case)
model.train()
UpperCAmelCase_ = model(_snake_case , mask_labels=_snake_case , class_labels=_snake_case).loss
loss.backward()
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = self.all_model_classes[1]
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCAmelCase_ = True
UpperCAmelCase_ = True
UpperCAmelCase_ = model_class(_snake_case)
model.to(_snake_case)
model.train()
UpperCAmelCase_ = model(_snake_case , mask_labels=_snake_case , class_labels=_snake_case)
UpperCAmelCase_ = outputs.encoder_hidden_states[0]
encoder_hidden_states.retain_grad()
UpperCAmelCase_ = outputs.pixel_decoder_hidden_states[0]
pixel_decoder_hidden_states.retain_grad()
# we requires_grad=True in inputs_embeds (line 2152), the original implementation don't
UpperCAmelCase_ = outputs.transformer_decoder_hidden_states[0]
transformer_decoder_hidden_states.retain_grad()
UpperCAmelCase_ = outputs.attentions[0]
attentions.retain_grad()
outputs.loss.backward(retain_graph=_snake_case)
self.assertIsNotNone(encoder_hidden_states.grad)
self.assertIsNotNone(pixel_decoder_hidden_states.grad)
self.assertIsNotNone(transformer_decoder_hidden_states.grad)
self.assertIsNotNone(attentions.grad)
snake_case_ : Dict = 1e-4
def A () -> Union[str, Any]:
"""simple docstring"""
UpperCAmelCase_ = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
return image
@require_vision
@slow
class __snake_case ( unittest.TestCase ):
@cached_property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return (
MaskFormerImageProcessor.from_pretrained('''facebook/maskformer-swin-small-coco''')
if is_vision_available()
else None
)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerModel.from_pretrained('''facebook/maskformer-swin-small-coco''').to(_snake_case)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
UpperCAmelCase_ = torch.tensor(
[[-0.0_4_8_2, 0.9_2_2_8, 0.4_9_5_1], [-0.2_5_4_7, 0.8_0_1_7, 0.8_5_2_7], [-0.0_0_6_9, 0.3_3_8_5, -0.0_0_8_9]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.encoder_last_hidden_state[0, 0, :3, :3] , _snake_case , atol=_snake_case))
UpperCAmelCase_ = torch.tensor(
[[-0.8_4_2_2, -0.8_4_3_4, -0.9_7_1_8], [-1.0_1_4_4, -0.5_5_6_5, -0.4_1_9_5], [-1.0_0_3_8, -0.4_4_8_4, -0.1_9_6_1]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.pixel_decoder_last_hidden_state[0, 0, :3, :3] , _snake_case , atol=_snake_case))
UpperCAmelCase_ = torch.tensor(
[[0.2_8_5_2, -0.0_1_5_9, 0.9_7_3_5], [0.6_2_5_4, 0.1_8_5_8, 0.8_5_2_9], [-0.0_6_8_0, -0.4_1_1_6, 1.8_4_1_3]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.transformer_decoder_last_hidden_state[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-swin-small-coco''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
# masks_queries_logits
UpperCAmelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.decoder_config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) , )
UpperCAmelCase_ = [
[-1.3_7_3_7_1_2_4, -1.7_7_2_4_9_3_7, -1.9_3_6_4_2_3_3],
[-1.5_9_7_7_2_8_1, -1.9_8_6_7_9_3_9, -2.1_5_2_3_6_9_5],
[-1.5_7_9_5_3_9_8, -1.9_2_6_9_8_3_2, -2.0_9_3_9_4_2],
]
UpperCAmelCase_ = torch.tensor(_snake_case).to(_snake_case)
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _snake_case , atol=_snake_case))
# class_queries_logits
UpperCAmelCase_ = outputs.class_queries_logits
self.assertEqual(
class_queries_logits.shape , (1, model.config.decoder_config.num_queries, model.config.num_labels + 1))
UpperCAmelCase_ = torch.tensor(
[
[1.65_12e00, -5.25_72e00, -3.35_19e00],
[3.61_69e-02, -5.90_25e00, -2.93_13e00],
[1.07_66e-04, -7.76_30e00, -5.12_63e00],
]).to(_snake_case)
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-resnet101-coco-stuff''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
# masks_queries_logits
UpperCAmelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.decoder_config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) , )
UpperCAmelCase_ = [[-0.9_0_4_6, -2.6_3_6_6, -4.6_0_6_2], [-3.4_1_7_9, -5.7_8_9_0, -8.8_0_5_7], [-4.9_1_7_9, -7.6_5_6_0, -1_0.7_7_1_1]]
UpperCAmelCase_ = torch.tensor(_snake_case).to(_snake_case)
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _snake_case , atol=_snake_case))
# class_queries_logits
UpperCAmelCase_ = outputs.class_queries_logits
self.assertEqual(
class_queries_logits.shape , (1, model.config.decoder_config.num_queries, model.config.num_labels + 1))
UpperCAmelCase_ = torch.tensor(
[[4.7_1_8_8, -3.2_5_8_5, -2.8_8_5_7], [6.6_8_7_1, -2.9_1_8_1, -1.2_4_8_7], [7.2_4_4_9, -2.2_7_6_4, -2.1_8_7_4]]).to(_snake_case)
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-swin-small-coco''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = image_processor(
[np.zeros((3, 800, 1333)), np.zeros((3, 800, 1333))] , segmentation_maps=[np.zeros((384, 384)).astype(np.floataa), np.zeros((384, 384)).astype(np.floataa)] , return_tensors='''pt''' , )
UpperCAmelCase_ = inputs['''pixel_values'''].to(_snake_case)
UpperCAmelCase_ = [el.to(_snake_case) for el in inputs['''mask_labels''']]
UpperCAmelCase_ = [el.to(_snake_case) for el in inputs['''class_labels''']]
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
self.assertTrue(outputs.loss is not None)
| 51
| 0
|
"""simple docstring"""
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tf_available, is_torch_available
snake_case_ = {
"""configuration_xlm""": ["""XLM_PRETRAINED_CONFIG_ARCHIVE_MAP""", """XLMConfig""", """XLMOnnxConfig"""],
"""tokenization_xlm""": ["""XLMTokenizer"""],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ = [
"""XLM_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""XLMForMultipleChoice""",
"""XLMForQuestionAnswering""",
"""XLMForQuestionAnsweringSimple""",
"""XLMForSequenceClassification""",
"""XLMForTokenClassification""",
"""XLMModel""",
"""XLMPreTrainedModel""",
"""XLMWithLMHeadModel""",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ = [
"""TF_XLM_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""TFXLMForMultipleChoice""",
"""TFXLMForQuestionAnsweringSimple""",
"""TFXLMForSequenceClassification""",
"""TFXLMForTokenClassification""",
"""TFXLMMainLayer""",
"""TFXLMModel""",
"""TFXLMPreTrainedModel""",
"""TFXLMWithLMHeadModel""",
]
if TYPE_CHECKING:
from .configuration_xlm import XLM_PRETRAINED_CONFIG_ARCHIVE_MAP, XLMConfig, XLMOnnxConfig
from .tokenization_xlm import XLMTokenizer
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_xlm import (
XLM_PRETRAINED_MODEL_ARCHIVE_LIST,
XLMForMultipleChoice,
XLMForQuestionAnswering,
XLMForQuestionAnsweringSimple,
XLMForSequenceClassification,
XLMForTokenClassification,
XLMModel,
XLMPreTrainedModel,
XLMWithLMHeadModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_xlm import (
TF_XLM_PRETRAINED_MODEL_ARCHIVE_LIST,
TFXLMForMultipleChoice,
TFXLMForQuestionAnsweringSimple,
TFXLMForSequenceClassification,
TFXLMForTokenClassification,
TFXLMMainLayer,
TFXLMModel,
TFXLMPreTrainedModel,
TFXLMWithLMHeadModel,
)
else:
import sys
snake_case_ = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 78
|
import argparse
import pickle
import numpy as np
import torch
from torch import nn
from transformers import ReformerConfig, ReformerModelWithLMHead
from transformers.utils import logging
logging.set_verbosity_info()
def A (__A : Optional[int] , __A : int , __A : str=None ) -> List[Any]:
"""simple docstring"""
assert torch_layer.weight.shape == weight.shape, F"""{torch_layer} layer.weight does not match"""
UpperCAmelCase_ = nn.Parameter(__A )
if bias is not None:
assert torch_layer.bias.shape == bias.shape, F"""{torch_layer} layer.bias does not match"""
UpperCAmelCase_ = nn.Parameter(__A )
def A (__A : Tuple , __A : Dict , __A : str ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = np.asarray(weights[0] )
UpperCAmelCase_ = np.asarray(weights[1] )
UpperCAmelCase_ = np.asarray(weights[2] )
set_param(
torch_layer.self_attention.query_key , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.value , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.output.dense , torch.tensor(__A ).view(-1 , __A ).contiguous().transpose(0 , 1 ) , )
def A (__A : Optional[Any] , __A : Any , __A : List[Any] ) -> int:
"""simple docstring"""
UpperCAmelCase_ = np.asarray(weights[0] )
UpperCAmelCase_ = np.asarray(weights[1] )
UpperCAmelCase_ = np.asarray(weights[2] )
UpperCAmelCase_ = np.asarray(weights[3] )
set_param(
torch_layer.self_attention.query , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.key , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.value , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.output.dense , torch.tensor(__A ).view(-1 , __A ).contiguous().transpose(0 , 1 ) , )
def A (__A : int , __A : Union[str, Any] , __A : List[str] ) -> List[Any]:
"""simple docstring"""
UpperCAmelCase_ = weights[0][0][0]
UpperCAmelCase_ = np.asarray(layer_norm_a[0] )
UpperCAmelCase_ = np.asarray(layer_norm_a[1] )
set_param(
torch_block.attention.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# lsh weights + output
UpperCAmelCase_ = weights[0][1]
if len(__A ) < 4:
set_layer_weights_in_torch_lsh(__A , torch_block.attention , __A )
else:
set_layer_weights_in_torch_local(__A , torch_block.attention , __A )
# intermediate weighs
UpperCAmelCase_ = weights[2][0][1][2]
# Chunked Feed Forward
if len(__A ) == 4:
UpperCAmelCase_ = intermediate_weights[2]
# layernorm 2
UpperCAmelCase_ = np.asarray(intermediate_weights[0][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[0][1] )
set_param(
torch_block.feed_forward.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# intermediate dense
UpperCAmelCase_ = np.asarray(intermediate_weights[1][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[1][1] )
set_param(
torch_block.feed_forward.dense.dense , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
# intermediate out
UpperCAmelCase_ = np.asarray(intermediate_weights[4][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[4][1] )
set_param(
torch_block.feed_forward.output.dense , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
def A (__A : Optional[int] , __A : Tuple , __A : Any ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = torch_model.reformer
# word embeds
UpperCAmelCase_ = np.asarray(weights[1] )
set_param(
torch_model_reformer.embeddings.word_embeddings , torch.tensor(__A ) , )
if isinstance(weights[3] , __A ):
UpperCAmelCase_ = torch_model_reformer.embeddings.position_embeddings
for emb_idx in range(len(position_embeddings.weights ) ):
UpperCAmelCase_ = np.asarray(weights[3][emb_idx][0] )
assert (
position_embeddings.weights[emb_idx].shape == emb_weights.shape
), F"""{position_embeddings[emb_idx]} emb does not match"""
UpperCAmelCase_ = nn.Parameter(torch.tensor(__A ) )
UpperCAmelCase_ = weights[5]
assert len(torch_model_reformer.encoder.layers ) * 4 == len(
__A ), "HF and trax model do not have the same number of layers"
for layer_idx, layer in enumerate(torch_model_reformer.encoder.layers ):
UpperCAmelCase_ = trax_layer_weights[4 * layer_idx : 4 * (layer_idx + 1)]
set_block_weights_in_torch(__A , __A , __A )
# output layer norm
UpperCAmelCase_ = np.asarray(weights[7][0] )
UpperCAmelCase_ = np.asarray(weights[7][1] )
set_param(
torch_model_reformer.encoder.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# output embeddings
UpperCAmelCase_ = np.asarray(weights[9][0] )
UpperCAmelCase_ = np.asarray(weights[9][1] )
set_param(
torch_model.lm_head.decoder , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
def A (__A : Tuple , __A : int , __A : str ) -> Union[str, Any]:
"""simple docstring"""
UpperCAmelCase_ = ReformerConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = ReformerModelWithLMHead(__A )
with open(__A , '''rb''' ) as f:
UpperCAmelCase_ = pickle.load(__A )['''weights''']
set_model_weights_in_torch(__A , __A , config.hidden_size )
# Save pytorch-model
print(F"""Save PyTorch model to {pytorch_dump_path}""" )
torch.save(model.state_dict() , __A )
if __name__ == "__main__":
snake_case_ : Optional[Any] = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--trax_model_pkl_path", default=None, type=str, required=True, help="Path to the TensorFlow checkpoint path."
)
parser.add_argument(
"--config_file",
default=None,
type=str,
required=True,
help=(
"The config json file corresponding to the pre-trained Reformer model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--pytorch_dump_path", default=None, type=str, required=True, help="Path to the output PyTorch model."
)
snake_case_ : List[Any] = parser.parse_args()
convert_trax_checkpoint_to_pytorch(args.trax_model_pkl_path, args.config_file, args.pytorch_dump_path)
| 51
| 0
|
'''simple docstring'''
from typing import List
import numpy as np
def __lowercase ( __lowercase ) -> int:
'''simple docstring'''
_A = {key: len(__lowercase ) for key, value in gen_kwargs.items() if isinstance(__lowercase , __lowercase )}
if len(set(lists_lengths.values() ) ) > 1:
raise RuntimeError(
(
"Sharding is ambiguous for this dataset: "
+ "we found several data sources lists of different lengths, and we don't know over which list we should parallelize:\n"
+ "\n".join(F'''\t- key {key} has length {length}''' for key, length in lists_lengths.items() )
+ "\nTo fix this, check the 'gen_kwargs' and make sure to use lists only for data sources, "
+ "and use tuples otherwise. In the end there should only be one single list, or several lists with the same length."
) )
_A = max(lists_lengths.values() , default=0 )
return max(1 , __lowercase )
def __lowercase ( __lowercase , __lowercase ) -> List[range]:
'''simple docstring'''
_A = []
for group_idx in range(__lowercase ):
_A = num_shards // max_num_jobs + (group_idx < (num_shards % max_num_jobs))
if num_shards_to_add == 0:
break
_A = shards_indices_per_group[-1].stop if shards_indices_per_group else 0
_A = range(__lowercase , start + num_shards_to_add )
shards_indices_per_group.append(__lowercase )
return shards_indices_per_group
def __lowercase ( __lowercase , __lowercase ) -> List[dict]:
'''simple docstring'''
_A = _number_of_shards_in_gen_kwargs(__lowercase )
if num_shards == 1:
return [dict(__lowercase )]
else:
_A = _distribute_shards(num_shards=__lowercase , max_num_jobs=__lowercase )
return [
{
key: [value[shard_idx] for shard_idx in shard_indices_per_group[group_idx]]
if isinstance(__lowercase , __lowercase )
else value
for key, value in gen_kwargs.items()
}
for group_idx in range(len(__lowercase ) )
]
def __lowercase ( __lowercase ) -> dict:
'''simple docstring'''
return {
key: [value for gen_kwargs in gen_kwargs_list for value in gen_kwargs[key]]
if isinstance(gen_kwargs_list[0][key] , __lowercase )
else gen_kwargs_list[0][key]
for key in gen_kwargs_list[0]
}
def __lowercase ( __lowercase , __lowercase ) -> dict:
'''simple docstring'''
_A = {len(__lowercase ) for value in gen_kwargs.values() if isinstance(__lowercase , __lowercase )}
_A = {}
for size in list_sizes:
_A = list(range(__lowercase ) )
rng.shuffle(indices_per_size[size] )
# Now let's copy the gen_kwargs and shuffle the lists based on their sizes
_A = dict(__lowercase )
for key, value in shuffled_kwargs.items():
if isinstance(__lowercase , __lowercase ):
_A = [value[i] for i in indices_per_size[len(__lowercase )]]
return shuffled_kwargs
| 79
|
# This model implementation is heavily inspired by https://github.com/haofanwang/ControlNet-for-Diffusers/
import gc
import random
import tempfile
import unittest
import numpy as np
import torch
from PIL import Image
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
ControlNetModel,
DDIMScheduler,
StableDiffusionControlNetImgaImgPipeline,
UNetaDConditionModel,
)
from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_controlnet import MultiControlNetModel
from diffusers.utils import floats_tensor, load_image, load_numpy, randn_tensor, slow, torch_device
from diffusers.utils.import_utils import is_xformers_available
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import (
IMAGE_TO_IMAGE_IMAGE_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_PARAMS,
)
from ..test_pipelines_common import (
PipelineKarrasSchedulerTesterMixin,
PipelineLatentTesterMixin,
PipelineTesterMixin,
)
enable_full_determinism()
class __snake_case ( a , a , a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = StableDiffusionControlNetImgaImgPipeline
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''}
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
UpperCAmelCase__ : Union[str, Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS.union({'''control_image'''} )
UpperCAmelCase__ : Optional[Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS
def lowerCamelCase ( self : int):
"""simple docstring"""
torch.manual_seed(0)
UpperCAmelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , )
torch.manual_seed(0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
torch.manual_seed(0)
UpperCAmelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='''scaled_linear''' , clip_sample=_snake_case , set_alpha_to_one=_snake_case , )
torch.manual_seed(0)
UpperCAmelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , )
torch.manual_seed(0)
UpperCAmelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCAmelCase_ = CLIPTextModel(_snake_case)
UpperCAmelCase_ = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''')
UpperCAmelCase_ = {
'''unet''': unet,
'''controlnet''': controlnet,
'''scheduler''': scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def lowerCamelCase ( self : Union[str, Any] , _snake_case : Any , _snake_case : Dict=0):
"""simple docstring"""
if str(_snake_case).startswith('''mps'''):
UpperCAmelCase_ = torch.manual_seed(_snake_case)
else:
UpperCAmelCase_ = torch.Generator(device=_snake_case).manual_seed(_snake_case)
UpperCAmelCase_ = 2
UpperCAmelCase_ = randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , )
UpperCAmelCase_ = floats_tensor(control_image.shape , rng=random.Random(_snake_case)).to(_snake_case)
UpperCAmelCase_ = image.cpu().permute(0 , 2 , 3 , 1)[0]
UpperCAmelCase_ = Image.fromarray(np.uinta(_snake_case)).convert('''RGB''').resize((64, 64))
UpperCAmelCase_ = {
'''prompt''': '''A painting of a squirrel eating a burger''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
'''image''': image,
'''control_image''': control_image,
}
return inputs
def lowerCamelCase ( self : Any):
"""simple docstring"""
return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3)
@unittest.skipIf(
torch_device != '''cuda''' or not is_xformers_available() , reason='''XFormers attention is only available with CUDA and `xformers` installed''' , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
self._test_inference_batch_single_identical(expected_max_diff=2e-3)
class __snake_case ( a , a , unittest.TestCase ):
UpperCAmelCase__ : str = StableDiffusionControlNetImgaImgPipeline
UpperCAmelCase__ : List[str] = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''}
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
UpperCAmelCase__ : str = frozenset([] ) # TO_DO: add image_params once refactored VaeImageProcessor.preprocess
def lowerCamelCase ( self : str):
"""simple docstring"""
torch.manual_seed(0)
UpperCAmelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , )
torch.manual_seed(0)
def init_weights(_snake_case : Optional[int]):
if isinstance(_snake_case , torch.nn.Convad):
torch.nn.init.normal(m.weight)
m.bias.data.fill_(1.0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
controlneta.controlnet_down_blocks.apply(_snake_case)
torch.manual_seed(0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
controlneta.controlnet_down_blocks.apply(_snake_case)
torch.manual_seed(0)
UpperCAmelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='''scaled_linear''' , clip_sample=_snake_case , set_alpha_to_one=_snake_case , )
torch.manual_seed(0)
UpperCAmelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , )
torch.manual_seed(0)
UpperCAmelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCAmelCase_ = CLIPTextModel(_snake_case)
UpperCAmelCase_ = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''')
UpperCAmelCase_ = MultiControlNetModel([controlneta, controlneta])
UpperCAmelCase_ = {
'''unet''': unet,
'''controlnet''': controlnet,
'''scheduler''': scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def lowerCamelCase ( self : int , _snake_case : Union[str, Any] , _snake_case : str=0):
"""simple docstring"""
if str(_snake_case).startswith('''mps'''):
UpperCAmelCase_ = torch.manual_seed(_snake_case)
else:
UpperCAmelCase_ = torch.Generator(device=_snake_case).manual_seed(_snake_case)
UpperCAmelCase_ = 2
UpperCAmelCase_ = [
randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , ),
randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , ),
]
UpperCAmelCase_ = floats_tensor(control_image[0].shape , rng=random.Random(_snake_case)).to(_snake_case)
UpperCAmelCase_ = image.cpu().permute(0 , 2 , 3 , 1)[0]
UpperCAmelCase_ = Image.fromarray(np.uinta(_snake_case)).convert('''RGB''').resize((64, 64))
UpperCAmelCase_ = {
'''prompt''': '''A painting of a squirrel eating a burger''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
'''image''': image,
'''control_image''': control_image,
}
return inputs
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_dummy_components()
UpperCAmelCase_ = self.pipeline_class(**_snake_case)
pipe.to(_snake_case)
UpperCAmelCase_ = 1_0.0
UpperCAmelCase_ = 4
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case)[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=0.1 , control_guidance_end=0.2)[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=[0.1, 0.3] , control_guidance_end=[0.2, 0.7])[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=0.4 , control_guidance_end=[0.5, 0.8])[0]
# make sure that all outputs are different
assert np.sum(np.abs(output_a - output_a)) > 1e-3
assert np.sum(np.abs(output_a - output_a)) > 1e-3
assert np.sum(np.abs(output_a - output_a)) > 1e-3
def lowerCamelCase ( self : Dict):
"""simple docstring"""
return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3)
@unittest.skipIf(
torch_device != '''cuda''' or not is_xformers_available() , reason='''XFormers attention is only available with CUDA and `xformers` installed''' , )
def lowerCamelCase ( self : int):
"""simple docstring"""
self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3)
def lowerCamelCase ( self : int):
"""simple docstring"""
self._test_inference_batch_single_identical(expected_max_diff=2e-3)
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = self.get_dummy_components()
UpperCAmelCase_ = self.pipeline_class(**_snake_case)
pipe.to(_snake_case)
pipe.set_progress_bar_config(disable=_snake_case)
with tempfile.TemporaryDirectory() as tmpdir:
try:
# save_pretrained is not implemented for Multi-ControlNet
pipe.save_pretrained(_snake_case)
except NotImplementedError:
pass
@slow
@require_torch_gpu
class __snake_case ( unittest.TestCase ):
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = ControlNetModel.from_pretrained('''lllyasviel/sd-controlnet-canny''')
UpperCAmelCase_ = StableDiffusionControlNetImgaImgPipeline.from_pretrained(
'''runwayml/stable-diffusion-v1-5''' , safety_checker=_snake_case , controlnet=_snake_case)
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=_snake_case)
UpperCAmelCase_ = torch.Generator(device='''cpu''').manual_seed(0)
UpperCAmelCase_ = '''evil space-punk bird'''
UpperCAmelCase_ = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/bird_canny.png''').resize((512, 512))
UpperCAmelCase_ = load_image(
'''https://huggingface.co/lllyasviel/sd-controlnet-canny/resolve/main/images/bird.png''').resize((512, 512))
UpperCAmelCase_ = pipe(
_snake_case , _snake_case , control_image=_snake_case , generator=_snake_case , output_type='''np''' , num_inference_steps=50 , strength=0.6 , )
UpperCAmelCase_ = output.images[0]
assert image.shape == (512, 512, 3)
UpperCAmelCase_ = load_numpy(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/img2img.npy''')
assert np.abs(expected_image - image).max() < 9e-2
| 51
| 0
|
'''simple docstring'''
# Lint as: python3
import itertools
import os
import re
a__ : int = re.compile(R'([A-Z]+)([A-Z][a-z])')
a__ : str = re.compile(R'([a-z\d])([A-Z])')
a__ : Tuple = re.compile(R'(?<!_)_(?!_)')
a__ : Union[str, Any] = re.compile(R'(_{2,})')
a__ : Dict = R'^\w+(\.\w+)*$'
a__ : Optional[Any] = R'<>:/\|?*'
def _UpperCamelCase ( __A ) -> Union[str, Any]:
'''simple docstring'''
UpperCamelCase__ = _uppercase_uppercase_re.sub(R"\1_\2" , __A )
UpperCamelCase__ = _lowercase_uppercase_re.sub(R"\1_\2" , __A )
return name.lower()
def _UpperCamelCase ( __A ) -> Optional[int]:
'''simple docstring'''
UpperCamelCase__ = _single_underscore_re.split(__A )
UpperCamelCase__ = [_multiple_underscores_re.split(__A ) for n in name]
return "".join(n.capitalize() for n in itertools.chain.from_iterable(__A ) if n != "" )
def _UpperCamelCase ( __A ) -> int:
'''simple docstring'''
if os.path.basename(__A ) != name:
raise ValueError(F'''Should be a dataset name, not a path: {name}''' )
return camelcase_to_snakecase(__A )
def _UpperCamelCase ( __A , __A ) -> str:
'''simple docstring'''
if os.path.basename(__A ) != name:
raise ValueError(F'''Should be a dataset name, not a path: {name}''' )
if not re.match(_split_re , __A ):
raise ValueError(F'''Split name should match \'{_split_re}\'\' but got \'{split}\'.''' )
return F'''{filename_prefix_for_name(__A )}-{split}'''
def _UpperCamelCase ( __A , __A , __A , __A=None ) -> List[Any]:
'''simple docstring'''
UpperCamelCase__ = filename_prefix_for_split(__A , __A )
if filetype_suffix:
prefix += F'''.{filetype_suffix}'''
UpperCamelCase__ = os.path.join(__A , __A )
return F'''{filepath}*'''
def _UpperCamelCase ( __A , __A , __A , __A=None , __A=None ) -> Any:
'''simple docstring'''
UpperCamelCase__ = filename_prefix_for_split(__A , __A )
UpperCamelCase__ = os.path.join(__A , __A )
if shard_lengths:
UpperCamelCase__ = len(__A )
UpperCamelCase__ = [F'''{prefix}-{shard_id:05d}-of-{num_shards:05d}''' for shard_id in range(__A )]
if filetype_suffix:
UpperCamelCase__ = [filename + F'''.{filetype_suffix}''' for filename in filenames]
return filenames
else:
UpperCamelCase__ = prefix
if filetype_suffix:
filename += F'''.{filetype_suffix}'''
return [filename]
| 80
|
import random
import timeit
from functools import wraps
from typing import Callable, Optional
from ..configuration_utils import PretrainedConfig
from ..models.auto.modeling_tf_auto import TF_MODEL_MAPPING, TF_MODEL_WITH_LM_HEAD_MAPPING
from ..utils import is_pyanvml_available, is_tf_available, logging
from .benchmark_utils import (
Benchmark,
Memory,
MemorySummary,
measure_peak_memory_cpu,
start_memory_tracing,
stop_memory_tracing,
)
if is_tf_available():
import tensorflow as tf
from tensorflow.python.framework.errors_impl import ResourceExhaustedError
from .benchmark_args_tf import TensorFlowBenchmarkArguments
if is_pyanvml_available():
import pyanvml.pyanvml as nvml
snake_case_ : Tuple = logging.get_logger(__name__)
def A (__A : bool , __A : bool ) -> Optional[Any]:
"""simple docstring"""
def run_func(__A : Optional[Any] ):
@wraps(__A )
def run_in_eager_mode(*__A : Dict , **__A : List[Any] ):
return func(*__A , **__A )
@wraps(__A )
@tf.function(experimental_compile=__A )
def run_in_graph_mode(*__A : Optional[Any] , **__A : Any ):
return func(*__A , **__A )
if do_eager_mode is True:
if use_xla is not False:
raise ValueError(
'''Cannot run model in XLA, if `args.eager_mode` is set to `True`. Please set `args.eager_mode=False`.''' )
return run_in_eager_mode
else:
return run_in_graph_mode
return run_func
def A (__A : int , __A : int , __A : int ) -> ["tf.Tensor"]:
"""simple docstring"""
UpperCAmelCase_ = random.Random()
UpperCAmelCase_ = [rng.randint(0 , vocab_size - 1 ) for i in range(batch_size * sequence_length )]
return tf.constant(__A , shape=(batch_size, sequence_length) , dtype=tf.intaa )
class __snake_case ( a ):
UpperCAmelCase__ : TensorFlowBenchmarkArguments
UpperCAmelCase__ : PretrainedConfig
UpperCAmelCase__ : str = "TensorFlow"
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return tf.__version__
def lowerCamelCase ( self : Dict , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case)
return self._measure_speed(_inference)
def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case)
return self._measure_speed(_train)
def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
if self.args.is_gpu:
tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case)
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case)
return self._measure_memory(_inference)
def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
if self.args.is_gpu:
tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case)
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case)
return self._measure_memory(_train)
def lowerCamelCase ( self : Optional[int] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.config_dict[model_name]
if self.args.fpaa:
raise NotImplementedError('''Mixed precision is currently not supported.''')
UpperCAmelCase_ = (
hasattr(_snake_case , '''architectures''')
and isinstance(config.architectures , _snake_case)
and len(config.architectures) > 0
)
if not self.args.only_pretrain_model and has_model_class_in_config:
try:
UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model
UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class])
UpperCAmelCase_ = getattr(_snake_case , _snake_case)
UpperCAmelCase_ = model_cls(_snake_case)
except ImportError:
raise ImportError(
F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to"""
''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''')
else:
UpperCAmelCase_ = TF_MODEL_MAPPING[config.__class__](_snake_case)
# encoder-decoder has vocab size saved differently
UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size
UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_decoder_forward():
return model(_snake_case , decoder_input_ids=_snake_case , training=_snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_forward():
return model(_snake_case , training=_snake_case)
UpperCAmelCase_ = encoder_decoder_forward if config.is_encoder_decoder else encoder_forward
return _inference
def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.config_dict[model_name]
if self.args.eager_mode is not False:
raise ValueError('''Training cannot be done in eager mode. Please make sure that `args.eager_mode = False`.''')
if self.args.fpaa:
raise NotImplementedError('''Mixed precision is currently not supported.''')
UpperCAmelCase_ = (
hasattr(_snake_case , '''architectures''')
and isinstance(config.architectures , _snake_case)
and len(config.architectures) > 0
)
if not self.args.only_pretrain_model and has_model_class_in_config:
try:
UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model
UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class])
UpperCAmelCase_ = getattr(_snake_case , _snake_case)
UpperCAmelCase_ = model_cls(_snake_case)
except ImportError:
raise ImportError(
F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to"""
''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''')
else:
UpperCAmelCase_ = TF_MODEL_WITH_LM_HEAD_MAPPING[config.__class__](_snake_case)
# encoder-decoder has vocab size saved differently
UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size
UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_decoder_train():
UpperCAmelCase_ = model(_snake_case , decoder_input_ids=_snake_case , labels=_snake_case , training=_snake_case)[0]
UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables)
return gradients
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_train():
UpperCAmelCase_ = model(_snake_case , labels=_snake_case , training=_snake_case)[0]
UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables)
return gradients
UpperCAmelCase_ = encoder_decoder_train if config.is_encoder_decoder else encoder_train
return _train
def lowerCamelCase ( self : Any , _snake_case : Optional[Any]):
"""simple docstring"""
with self.args.strategy.scope():
try:
if self.args.is_tpu or self.args.use_xla:
# run additional 10 times to stabilize compilation for tpu
logger.info('''Do inference on TPU. Running model 5 times to stabilize compilation''')
timeit.repeat(_snake_case , repeat=1 , number=5)
# as written in https://docs.python.org/2/library/timeit.html#timeit.Timer.repeat, min should be taken rather than the average
UpperCAmelCase_ = timeit.repeat(
_snake_case , repeat=self.args.repeat , number=10 , )
return min(_snake_case) / 1_0.0
except ResourceExhaustedError as e:
self.print_fn(F"""Doesn't fit on GPU. {e}""")
def lowerCamelCase ( self : Dict , _snake_case : Callable[[], None]):
"""simple docstring"""
logger.info(
'''Note that TensorFlow allocates more memory than '''
'''it might need to speed up computation. '''
'''The memory reported here corresponds to the memory '''
'''reported by `nvidia-smi`, which can vary depending '''
'''on total available memory on the GPU that is used.''')
with self.args.strategy.scope():
try:
if self.args.trace_memory_line_by_line:
if not self.args.eager_mode:
raise ValueError(
'''`args.eager_mode` is set to `False`. Make sure to run model in eager mode to measure memory'''
''' consumption line by line.''')
UpperCAmelCase_ = start_memory_tracing('''transformers''')
if self.args.is_tpu:
# tpu
raise NotImplementedError(
'''Memory Benchmarking is currently not implemented for TPU. Please disable memory benchmarking'''
''' with `args.memory=False`''')
elif self.args.is_gpu:
# gpu
if not is_pyanvml_available():
logger.warning(
'''py3nvml not installed, we won\'t log GPU memory usage. '''
'''Install py3nvml (pip install py3nvml) to log information about GPU.''')
UpperCAmelCase_ = '''N/A'''
else:
logger.info(
'''Measuring total GPU usage on GPU device. Make sure to not have additional processes'''
''' running on the same GPU.''')
# init nvml
nvml.nvmlInit()
func()
UpperCAmelCase_ = nvml.nvmlDeviceGetHandleByIndex(self.args.device_idx)
UpperCAmelCase_ = nvml.nvmlDeviceGetMemoryInfo(_snake_case)
UpperCAmelCase_ = meminfo.used
UpperCAmelCase_ = Memory(_snake_case)
# shutdown nvml
nvml.nvmlShutdown()
else:
# cpu
if self.args.trace_memory_line_by_line:
logger.info(
'''When enabling line by line tracing, the max peak memory for CPU is inaccurate in'''
''' TensorFlow.''')
UpperCAmelCase_ = None
else:
UpperCAmelCase_ = measure_peak_memory_cpu(_snake_case)
UpperCAmelCase_ = Memory(_snake_case) if isinstance(_snake_case , _snake_case) else memory_bytes
if self.args.trace_memory_line_by_line:
UpperCAmelCase_ = stop_memory_tracing(_snake_case)
if memory is None:
UpperCAmelCase_ = summary.total
else:
UpperCAmelCase_ = None
return memory, summary
except ResourceExhaustedError as e:
self.print_fn(F"""Doesn't fit on GPU. {e}""")
return "N/A", None
| 51
| 0
|
"""simple docstring"""
import os
from typing import BinaryIO, Optional, Union
import numpy as np
import pyarrow.parquet as pq
from .. import Audio, Dataset, Features, Image, NamedSplit, Value, config
from ..features.features import FeatureType, _visit
from ..formatting import query_table
from ..packaged_modules import _PACKAGED_DATASETS_MODULES
from ..packaged_modules.parquet.parquet import Parquet
from ..utils import logging
from ..utils.typing import NestedDataStructureLike, PathLike
from .abc import AbstractDatasetReader
def _A ( lowercase ):
"""simple docstring"""
a =np.inf
def set_batch_size(lowercase ) -> None:
nonlocal batch_size
if isinstance(lowercase , lowercase ):
a =min(lowercase , config.PARQUET_ROW_GROUP_SIZE_FOR_IMAGE_DATASETS )
elif isinstance(lowercase , lowercase ):
a =min(lowercase , config.PARQUET_ROW_GROUP_SIZE_FOR_AUDIO_DATASETS )
elif isinstance(lowercase , lowercase ) and feature.dtype == "binary":
a =min(lowercase , config.PARQUET_ROW_GROUP_SIZE_FOR_BINARY_DATASETS )
_visit(lowercase , lowercase )
return None if batch_size is np.inf else batch_size
class __A ( _SCREAMING_SNAKE_CASE ):
"""simple docstring"""
def __init__( self , __A , __A = None , __A = None , __A = None , __A = False , __A = False , __A = None , **__A , ) -> Optional[Any]:
super().__init__(
__A , split=__A , features=__A , cache_dir=__A , keep_in_memory=__A , streaming=__A , num_proc=__A , **__A , )
a =path_or_paths if isinstance(__A , __A ) else {self.split: path_or_paths}
a =_PACKAGED_DATASETS_MODULES['''parquet'''][1]
a =Parquet(
cache_dir=__A , data_files=__A , features=__A , hash=__A , **__A , )
def SCREAMING_SNAKE_CASE ( self ) -> int:
# Build iterable dataset
if self.streaming:
a =self.builder.as_streaming_dataset(split=self.split )
# Build regular (map-style) dataset
else:
a =None
a =None
a =None
a =None
self.builder.download_and_prepare(
download_config=__A , download_mode=__A , verification_mode=__A , base_path=__A , num_proc=self.num_proc , )
a =self.builder.as_dataset(
split=self.split , verification_mode=__A , in_memory=self.keep_in_memory )
return dataset
class __A :
"""simple docstring"""
def __init__( self , __A , __A , __A = None , **__A , ) -> Optional[int]:
a =dataset
a =path_or_buf
a =batch_size or get_writer_batch_size(dataset.features )
a =parquet_writer_kwargs
def SCREAMING_SNAKE_CASE ( self ) -> int:
a =self.batch_size if self.batch_size else config.DEFAULT_MAX_BATCH_SIZE
if isinstance(self.path_or_buf , (str, bytes, os.PathLike) ):
with open(self.path_or_buf , '''wb+''' ) as buffer:
a =self._write(file_obj=__A , batch_size=__A , **self.parquet_writer_kwargs )
else:
a =self._write(file_obj=self.path_or_buf , batch_size=__A , **self.parquet_writer_kwargs )
return written
def SCREAMING_SNAKE_CASE ( self , __A , __A , **__A ) -> int:
a =0
a =parquet_writer_kwargs.pop('''path_or_buf''' , __A )
a =self.dataset.features.arrow_schema
a =pq.ParquetWriter(__A , schema=__A , **__A )
for offset in logging.tqdm(
range(0 , len(self.dataset ) , __A ) , unit='''ba''' , disable=not logging.is_progress_bar_enabled() , desc='''Creating parquet from Arrow format''' , ):
a =query_table(
table=self.dataset._data , key=slice(__A , offset + batch_size ) , indices=self.dataset._indices if self.dataset._indices is not None else None , )
writer.write_table(__A )
written += batch.nbytes
writer.close()
return written
| 81
|
import hashlib
import unittest
from transformers import MODEL_FOR_DEPTH_ESTIMATION_MAPPING, is_torch_available, is_vision_available
from transformers.pipelines import DepthEstimationPipeline, pipeline
from transformers.testing_utils import (
is_pipeline_test,
nested_simplify,
require_tf,
require_timm,
require_torch,
require_vision,
slow,
)
from .test_pipelines_common import ANY
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
else:
class __snake_case :
@staticmethod
def lowerCamelCase ( *_snake_case : Optional[int] , **_snake_case : int):
"""simple docstring"""
pass
def A (__A : Image ) -> str:
"""simple docstring"""
UpperCAmelCase_ = hashlib.mda(image.tobytes() )
return m.hexdigest()
@is_pipeline_test
@require_vision
@require_timm
@require_torch
class __snake_case ( unittest.TestCase ):
UpperCAmelCase__ : Tuple = MODEL_FOR_DEPTH_ESTIMATION_MAPPING
def lowerCamelCase ( self : Tuple , _snake_case : Optional[Any] , _snake_case : Optional[Any] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = DepthEstimationPipeline(model=_snake_case , image_processor=_snake_case)
return depth_estimator, [
"./tests/fixtures/tests_samples/COCO/000000039769.png",
"./tests/fixtures/tests_samples/COCO/000000039769.png",
]
def lowerCamelCase ( self : str , _snake_case : Optional[int] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = depth_estimator('''./tests/fixtures/tests_samples/COCO/000000039769.png''')
self.assertEqual({'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)} , _snake_case)
import datasets
UpperCAmelCase_ = datasets.load_dataset('''hf-internal-testing/fixtures_image_utils''' , '''image''' , split='''test''')
UpperCAmelCase_ = depth_estimator(
[
Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png'''),
'''http://images.cocodataset.org/val2017/000000039769.jpg''',
# RGBA
dataset[0]['''file'''],
# LA
dataset[1]['''file'''],
# L
dataset[2]['''file'''],
])
self.assertEqual(
[
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
] , _snake_case , )
@require_tf
@unittest.skip('''Depth estimation is not implemented in TF''')
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
pass
@slow
@require_torch
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = '''Intel/dpt-large'''
UpperCAmelCase_ = pipeline('''depth-estimation''' , model=_snake_case)
UpperCAmelCase_ = depth_estimator('''http://images.cocodataset.org/val2017/000000039769.jpg''')
UpperCAmelCase_ = hashimage(outputs['''depth'''])
# This seems flaky.
# self.assertEqual(outputs["depth"], "1a39394e282e9f3b0741a90b9f108977")
self.assertEqual(nested_simplify(outputs['''predicted_depth'''].max().item()) , 2_9.3_0_4)
self.assertEqual(nested_simplify(outputs['''predicted_depth'''].min().item()) , 2.6_6_2)
@require_torch
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
self.skipTest('''There is not hf-internal-testing tiny model for either GLPN nor DPT''')
| 51
| 0
|
from __future__ import annotations
from random import random
from typing import Generic, TypeVar
A__ = TypeVar("""KT""")
A__ = TypeVar("""VT""")
class __lowerCAmelCase ( Generic[KT, VT] ):
def __init__( self , _snake_case = "root" , _snake_case = None ):
"""simple docstring"""
_lowerCAmelCase = key
_lowerCAmelCase = value
_lowerCAmelCase = []
def __repr__( self ):
"""simple docstring"""
return F'Node({self.key}: {self.value})'
@property
def snake_case ( self ):
"""simple docstring"""
return len(self.forward )
class __lowerCAmelCase ( Generic[KT, VT] ):
def __init__( self , _snake_case = 0.5 , _snake_case = 16 ):
"""simple docstring"""
_lowerCAmelCase = Node[KT, VT]()
_lowerCAmelCase = 0
_lowerCAmelCase = p
_lowerCAmelCase = max_level
def __str__( self ):
"""simple docstring"""
_lowerCAmelCase = list(self )
if len(_snake_case ) == 0:
return F'SkipList(level={self.level})'
_lowerCAmelCase = max((len(str(_snake_case ) ) for item in items) , default=4 )
_lowerCAmelCase = max(_snake_case , 4 ) + 4
_lowerCAmelCase = self.head
_lowerCAmelCase = []
_lowerCAmelCase = node.forward.copy()
lines.append(F'[{node.key}]'.ljust(_snake_case , """-""" ) + """* """ * len(_snake_case ) )
lines.append(""" """ * label_size + """| """ * len(_snake_case ) )
while len(node.forward ) != 0:
_lowerCAmelCase = node.forward[0]
lines.append(
F'[{node.key}]'.ljust(_snake_case , """-""" )
+ """ """.join(str(n.key ) if n.key == node.key else """|""" for n in forwards ) )
lines.append(""" """ * label_size + """| """ * len(_snake_case ) )
_lowerCAmelCase = node.forward
lines.append("""None""".ljust(_snake_case ) + """* """ * len(_snake_case ) )
return F'SkipList(level={self.level})\n' + "\n".join(_snake_case )
def __iter__( self ):
"""simple docstring"""
_lowerCAmelCase = self.head
while len(node.forward ) != 0:
yield node.forward[0].key
_lowerCAmelCase = node.forward[0]
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = 1
while random() < self.p and level < self.max_level:
level += 1
return level
def snake_case ( self , _snake_case ):
"""simple docstring"""
_lowerCAmelCase = []
_lowerCAmelCase = self.head
for i in reversed(range(self.level ) ):
# i < node.level - When node level is lesser than `i` decrement `i`.
# node.forward[i].key < key - Jumping to node with key value higher
# or equal to searched key would result
# in skipping searched key.
while i < node.level and node.forward[i].key < key:
_lowerCAmelCase = node.forward[i]
# Each leftmost node (relative to searched node) will potentially have to
# be updated.
update_vector.append(_snake_case )
update_vector.reverse() # Note that we were inserting values in reverse order.
# len(node.forward) != 0 - If current node doesn't contain any further
# references then searched key is not present.
# node.forward[0].key == key - Next node key should be equal to search key
# if key is present.
if len(node.forward ) != 0 and node.forward[0].key == key:
return node.forward[0], update_vector
else:
return None, update_vector
def snake_case ( self , _snake_case ):
"""simple docstring"""
_lowerCAmelCase , _lowerCAmelCase = self._locate_node(_snake_case )
if node is not None:
for i, update_node in enumerate(_snake_case ):
# Remove or replace all references to removed node.
if update_node.level > i and update_node.forward[i].key == key:
if node.level > i:
_lowerCAmelCase = node.forward[i]
else:
_lowerCAmelCase = update_node.forward[:i]
def snake_case ( self , _snake_case , _snake_case ):
"""simple docstring"""
_lowerCAmelCase , _lowerCAmelCase = self._locate_node(_snake_case )
if node is not None:
_lowerCAmelCase = value
else:
_lowerCAmelCase = self.random_level()
if level > self.level:
# After level increase we have to add additional nodes to head.
for _ in range(self.level - 1 , _snake_case ):
update_vector.append(self.head )
_lowerCAmelCase = level
_lowerCAmelCase = Node(_snake_case , _snake_case )
for i, update_node in enumerate(update_vector[:level] ):
# Change references to pass through new node.
if update_node.level > i:
new_node.forward.append(update_node.forward[i] )
if update_node.level < i + 1:
update_node.forward.append(_snake_case )
else:
_lowerCAmelCase = new_node
def snake_case ( self , _snake_case ):
"""simple docstring"""
_lowerCAmelCase , _lowerCAmelCase = self._locate_node(_snake_case )
if node is not None:
return node.value
return None
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
skip_list.insert("""Key1""" , 3 )
skip_list.insert("""Key2""" , 12 )
skip_list.insert("""Key3""" , 41 )
skip_list.insert("""Key4""" , -19 )
_lowerCAmelCase = skip_list.head
_lowerCAmelCase = {}
while node.level != 0:
_lowerCAmelCase = node.forward[0]
_lowerCAmelCase = node.value
assert len(snake_case ) == 4
assert all_values["Key1"] == 3
assert all_values["Key2"] == 12
assert all_values["Key3"] == 41
assert all_values["Key4"] == -19
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
skip_list.insert("""Key1""" , 10 )
skip_list.insert("""Key1""" , 12 )
skip_list.insert("""Key5""" , 7 )
skip_list.insert("""Key7""" , 10 )
skip_list.insert("""Key10""" , 5 )
skip_list.insert("""Key7""" , 7 )
skip_list.insert("""Key5""" , 5 )
skip_list.insert("""Key10""" , 10 )
_lowerCAmelCase = skip_list.head
_lowerCAmelCase = {}
while node.level != 0:
_lowerCAmelCase = node.forward[0]
_lowerCAmelCase = node.value
if len(snake_case ) != 4:
print()
assert len(snake_case ) == 4
assert all_values["Key1"] == 12
assert all_values["Key7"] == 7
assert all_values["Key5"] == 5
assert all_values["Key10"] == 10
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
assert skip_list.find("""Some key""" ) is None
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
skip_list.insert("""Key2""" , 20 )
assert skip_list.find("""Key2""" ) == 20
skip_list.insert("""Some Key""" , 10 )
skip_list.insert("""Key2""" , 8 )
skip_list.insert("""V""" , 13 )
assert skip_list.find("""Y""" ) is None
assert skip_list.find("""Key2""" ) == 8
assert skip_list.find("""Some Key""" ) == 10
assert skip_list.find("""V""" ) == 13
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
skip_list.delete("""Some key""" )
assert len(skip_list.head.forward ) == 0
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
skip_list.insert("""Key1""" , 12 )
skip_list.insert("""V""" , 13 )
skip_list.insert("""X""" , 14 )
skip_list.insert("""Key2""" , 15 )
skip_list.delete("""V""" )
skip_list.delete("""Key2""" )
assert skip_list.find("""V""" ) is None
assert skip_list.find("""Key2""" ) is None
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
skip_list.insert("""Key1""" , 12 )
skip_list.insert("""V""" , 13 )
skip_list.insert("""X""" , 14 )
skip_list.insert("""Key2""" , 15 )
skip_list.delete("""V""" )
assert skip_list.find("""V""" ) is None
assert skip_list.find("""X""" ) == 14
assert skip_list.find("""Key1""" ) == 12
assert skip_list.find("""Key2""" ) == 15
skip_list.delete("""X""" )
assert skip_list.find("""V""" ) is None
assert skip_list.find("""X""" ) is None
assert skip_list.find("""Key1""" ) == 12
assert skip_list.find("""Key2""" ) == 15
skip_list.delete("""Key1""" )
assert skip_list.find("""V""" ) is None
assert skip_list.find("""X""" ) is None
assert skip_list.find("""Key1""" ) is None
assert skip_list.find("""Key2""" ) == 15
skip_list.delete("""Key2""" )
assert skip_list.find("""V""" ) is None
assert skip_list.find("""X""" ) is None
assert skip_list.find("""Key1""" ) is None
assert skip_list.find("""Key2""" ) is None
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
skip_list.insert("""Key1""" , 12 )
skip_list.insert("""V""" , 13 )
skip_list.insert("""X""" , 1_42 )
skip_list.insert("""Key2""" , 15 )
skip_list.delete("""X""" )
def traverse_keys(snake_case ):
yield node.key
for forward_node in node.forward:
yield from traverse_keys(snake_case )
assert len(set(traverse_keys(skip_list.head ) ) ) == 4
def _UpperCAmelCase ( ):
"""simple docstring"""
def is_sorted(snake_case ):
return all(next_item >= item for item, next_item in zip(snake_case , lst[1:] ) )
_lowerCAmelCase = SkipList()
for i in range(10 ):
skip_list.insert(snake_case , snake_case )
assert is_sorted(list(snake_case ) )
skip_list.delete(5 )
skip_list.delete(8 )
skip_list.delete(2 )
assert is_sorted(list(snake_case ) )
skip_list.insert(-12 , -12 )
skip_list.insert(77 , 77 )
assert is_sorted(list(snake_case ) )
def _UpperCAmelCase ( ):
"""simple docstring"""
for _ in range(1_00 ):
# Repeat test 100 times due to the probabilistic nature of skip list
# random values == random bugs
test_insert()
test_insert_overrides_existing_value()
test_searching_empty_list_returns_none()
test_search()
test_deleting_item_from_empty_list_do_nothing()
test_deleted_items_are_not_founded_by_find_method()
test_delete_removes_only_given_key()
test_delete_doesnt_leave_dead_nodes()
test_iter_always_yields_sorted_values()
def _UpperCAmelCase ( ):
"""simple docstring"""
_lowerCAmelCase = SkipList()
skip_list.insert(2 , """2""" )
skip_list.insert(4 , """4""" )
skip_list.insert(6 , """4""" )
skip_list.insert(4 , """5""" )
skip_list.insert(8 , """4""" )
skip_list.insert(9 , """4""" )
skip_list.delete(4 )
print(snake_case )
if __name__ == "__main__":
import doctest
doctest.testmod()
main()
| 82
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : int = {
"configuration_deberta": ["DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP", "DebertaConfig", "DebertaOnnxConfig"],
"tokenization_deberta": ["DebertaTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : int = ["DebertaTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"DebertaForMaskedLM",
"DebertaForQuestionAnswering",
"DebertaForSequenceClassification",
"DebertaForTokenClassification",
"DebertaModel",
"DebertaPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFDebertaForMaskedLM",
"TFDebertaForQuestionAnswering",
"TFDebertaForSequenceClassification",
"TFDebertaForTokenClassification",
"TFDebertaModel",
"TFDebertaPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_deberta import DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP, DebertaConfig, DebertaOnnxConfig
from .tokenization_deberta import DebertaTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_deberta_fast import DebertaTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_deberta import (
DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
DebertaForMaskedLM,
DebertaForQuestionAnswering,
DebertaForSequenceClassification,
DebertaForTokenClassification,
DebertaModel,
DebertaPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_deberta import (
TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
TFDebertaForMaskedLM,
TFDebertaForQuestionAnswering,
TFDebertaForSequenceClassification,
TFDebertaForTokenClassification,
TFDebertaModel,
TFDebertaPreTrainedModel,
)
else:
import sys
snake_case_ : Dict = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
'''simple docstring'''
import os
def A__ ( UpperCAmelCase_ = "matrix.txt" ):
with open(os.path.join(os.path.dirname(UpperCAmelCase_ ) , UpperCAmelCase_ ) ) as in_file:
_UpperCamelCase : str = in_file.read()
_UpperCamelCase : List[str] = [[int(UpperCAmelCase_ ) for cell in row.split(',' )] for row in data.strip().splitlines()]
_UpperCamelCase : List[str] = [[0 for cell in row] for row in grid]
_UpperCamelCase : str = len(grid[0] )
_UpperCamelCase : str = [[0 for i in range(UpperCAmelCase_ )] for j in range(UpperCAmelCase_ )]
_UpperCamelCase : Union[str, Any] = grid[0][0]
for i in range(1 , UpperCAmelCase_ ):
_UpperCamelCase : Any = grid[0][i] + dp[0][i - 1]
for i in range(1 , UpperCAmelCase_ ):
_UpperCamelCase : Union[str, Any] = grid[i][0] + dp[i - 1][0]
for i in range(1 , UpperCAmelCase_ ):
for j in range(1 , UpperCAmelCase_ ):
_UpperCamelCase : Tuple = grid[i][j] + min(dp[i - 1][j] , dp[i][j - 1] )
return dp[-1][-1]
if __name__ == "__main__":
print(F"""{solution() = }""")
| 83
|
from typing import TYPE_CHECKING
from ...file_utils import _LazyModule, is_tokenizers_available, is_torch_available
from ...utils import OptionalDependencyNotAvailable
snake_case_ : Union[str, Any] = {"configuration_gpt_neox": ["GPT_NEOX_PRETRAINED_CONFIG_ARCHIVE_MAP", "GPTNeoXConfig"]}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Dict = ["GPTNeoXTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = [
"GPT_NEOX_PRETRAINED_MODEL_ARCHIVE_LIST",
"GPTNeoXForCausalLM",
"GPTNeoXForQuestionAnswering",
"GPTNeoXForSequenceClassification",
"GPTNeoXForTokenClassification",
"GPTNeoXLayer",
"GPTNeoXModel",
"GPTNeoXPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_gpt_neox import GPT_NEOX_PRETRAINED_CONFIG_ARCHIVE_MAP, GPTNeoXConfig
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_gpt_neox_fast import GPTNeoXTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_gpt_neox import (
GPT_NEOX_PRETRAINED_MODEL_ARCHIVE_LIST,
GPTNeoXForCausalLM,
GPTNeoXForQuestionAnswering,
GPTNeoXForSequenceClassification,
GPTNeoXForTokenClassification,
GPTNeoXLayer,
GPTNeoXModel,
GPTNeoXPreTrainedModel,
)
else:
import sys
snake_case_ : Optional[int] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
"""simple docstring"""
import collections
import importlib.util
import os
import re
from pathlib import Path
__UpperCAmelCase = 'src/transformers'
# Matches is_xxx_available()
__UpperCAmelCase = re.compile(R'is\_([a-z_]*)_available()')
# Catches a one-line _import_struct = {xxx}
__UpperCAmelCase = re.compile(R'^_import_structure\s+=\s+\{([^\}]+)\}')
# Catches a line with a key-values pattern: "bla": ["foo", "bar"]
__UpperCAmelCase = re.compile(R'\s+"\S*":\s+\[([^\]]*)\]')
# Catches a line if not is_foo_available
__UpperCAmelCase = re.compile(R'^\s*if\s+not\s+is\_[a-z_]*\_available\(\)')
# Catches a line _import_struct["bla"].append("foo")
__UpperCAmelCase = re.compile(R'^\s*_import_structure\["\S*"\]\.append\("(\S*)"\)')
# Catches a line _import_struct["bla"].extend(["foo", "bar"]) or _import_struct["bla"] = ["foo", "bar"]
__UpperCAmelCase = re.compile(R'^\s*_import_structure\[\S*\](?:\.extend\(|\s*=\s+)\[([^\]]*)\]')
# Catches a line with an object between quotes and a comma: "MyModel",
__UpperCAmelCase = re.compile('^\s+"([^"]+)",')
# Catches a line with objects between brackets only: ["foo", "bar"],
__UpperCAmelCase = re.compile('^\s+\[([^\]]+)\]')
# Catches a line with from foo import bar, bla, boo
__UpperCAmelCase = re.compile(R'\s+from\s+\S*\s+import\s+([^\(\s].*)\n')
# Catches a line with try:
__UpperCAmelCase = re.compile(R'^\s*try:')
# Catches a line with else:
__UpperCAmelCase = re.compile(R'^\s*else:')
def _snake_case ( lowercase__ : Optional[Any] ) -> List[str]:
'''simple docstring'''
if _re_test_backend.search(lowercase__ ) is None:
return None
lowerCAmelCase_ :List[str] = [b[0] for b in _re_backend.findall(lowercase__ )]
backends.sort()
return "_and_".join(lowercase__ )
def _snake_case ( lowercase__ : Union[str, Any] ) -> Union[str, Any]:
'''simple docstring'''
with open(lowercase__ , """r""" , encoding="""utf-8""" , newline="""\n""" ) as f:
lowerCAmelCase_ :int = f.readlines()
lowerCAmelCase_ :Any = 0
while line_index < len(lowercase__ ) and not lines[line_index].startswith("""_import_structure = {""" ):
line_index += 1
# If this is a traditional init, just return.
if line_index >= len(lowercase__ ):
return None
# First grab the objects without a specific backend in _import_structure
lowerCAmelCase_ :Optional[int] = []
while not lines[line_index].startswith("""if TYPE_CHECKING""" ) and find_backend(lines[line_index] ) is None:
lowerCAmelCase_ :Dict = lines[line_index]
# If we have everything on a single line, let's deal with it.
if _re_one_line_import_struct.search(lowercase__ ):
lowerCAmelCase_ :List[str] = _re_one_line_import_struct.search(lowercase__ ).groups()[0]
lowerCAmelCase_ :Optional[Any] = re.findall("""\[([^\]]+)\]""" , lowercase__ )
for imp in imports:
objects.extend([obj[1:-1] for obj in imp.split(""", """ )] )
line_index += 1
continue
lowerCAmelCase_ :str = _re_import_struct_key_value.search(lowercase__ )
if single_line_import_search is not None:
lowerCAmelCase_ :int = [obj[1:-1] for obj in single_line_import_search.groups()[0].split(""", """ ) if len(lowercase__ ) > 0]
objects.extend(lowercase__ )
elif line.startswith(""" """ * 8 + """\"""" ):
objects.append(line[9:-3] )
line_index += 1
lowerCAmelCase_ :Dict = {"""none""": objects}
# Let's continue with backend-specific objects in _import_structure
while not lines[line_index].startswith("""if TYPE_CHECKING""" ):
# If the line is an if not is_backend_available, we grab all objects associated.
lowerCAmelCase_ :Tuple = find_backend(lines[line_index] )
# Check if the backend declaration is inside a try block:
if _re_try.search(lines[line_index - 1] ) is None:
lowerCAmelCase_ :List[Any] = None
if backend is not None:
line_index += 1
# Scroll until we hit the else block of try-except-else
while _re_else.search(lines[line_index] ) is None:
line_index += 1
line_index += 1
lowerCAmelCase_ :List[str] = []
# Until we unindent, add backend objects to the list
while len(lines[line_index] ) <= 1 or lines[line_index].startswith(""" """ * 4 ):
lowerCAmelCase_ :List[Any] = lines[line_index]
if _re_import_struct_add_one.search(lowercase__ ) is not None:
objects.append(_re_import_struct_add_one.search(lowercase__ ).groups()[0] )
elif _re_import_struct_add_many.search(lowercase__ ) is not None:
lowerCAmelCase_ :Union[str, Any] = _re_import_struct_add_many.search(lowercase__ ).groups()[0].split(""", """ )
lowerCAmelCase_ :Any = [obj[1:-1] for obj in imports if len(lowercase__ ) > 0]
objects.extend(lowercase__ )
elif _re_between_brackets.search(lowercase__ ) is not None:
lowerCAmelCase_ :Optional[int] = _re_between_brackets.search(lowercase__ ).groups()[0].split(""", """ )
lowerCAmelCase_ :Optional[int] = [obj[1:-1] for obj in imports if len(lowercase__ ) > 0]
objects.extend(lowercase__ )
elif _re_quote_object.search(lowercase__ ) is not None:
objects.append(_re_quote_object.search(lowercase__ ).groups()[0] )
elif line.startswith(""" """ * 8 + """\"""" ):
objects.append(line[9:-3] )
elif line.startswith(""" """ * 1_2 + """\"""" ):
objects.append(line[1_3:-3] )
line_index += 1
lowerCAmelCase_ :List[Any] = objects
else:
line_index += 1
# At this stage we are in the TYPE_CHECKING part, first grab the objects without a specific backend
lowerCAmelCase_ :Union[str, Any] = []
while (
line_index < len(lowercase__ )
and find_backend(lines[line_index] ) is None
and not lines[line_index].startswith("""else""" )
):
lowerCAmelCase_ :str = lines[line_index]
lowerCAmelCase_ :Union[str, Any] = _re_import.search(lowercase__ )
if single_line_import_search is not None:
objects.extend(single_line_import_search.groups()[0].split(""", """ ) )
elif line.startswith(""" """ * 8 ):
objects.append(line[8:-2] )
line_index += 1
lowerCAmelCase_ :Any = {"""none""": objects}
# Let's continue with backend-specific objects
while line_index < len(lowercase__ ):
# If the line is an if is_backend_available, we grab all objects associated.
lowerCAmelCase_ :Union[str, Any] = find_backend(lines[line_index] )
# Check if the backend declaration is inside a try block:
if _re_try.search(lines[line_index - 1] ) is None:
lowerCAmelCase_ :int = None
if backend is not None:
line_index += 1
# Scroll until we hit the else block of try-except-else
while _re_else.search(lines[line_index] ) is None:
line_index += 1
line_index += 1
lowerCAmelCase_ :List[Any] = []
# Until we unindent, add backend objects to the list
while len(lines[line_index] ) <= 1 or lines[line_index].startswith(""" """ * 8 ):
lowerCAmelCase_ :Union[str, Any] = lines[line_index]
lowerCAmelCase_ :Optional[int] = _re_import.search(lowercase__ )
if single_line_import_search is not None:
objects.extend(single_line_import_search.groups()[0].split(""", """ ) )
elif line.startswith(""" """ * 1_2 ):
objects.append(line[1_2:-2] )
line_index += 1
lowerCAmelCase_ :Union[str, Any] = objects
else:
line_index += 1
return import_dict_objects, type_hint_objects
def _snake_case ( lowercase__ : int , lowercase__ : List[Any] ) -> Any:
'''simple docstring'''
def find_duplicates(lowercase__ : Union[str, Any] ):
return [k for k, v in collections.Counter(lowercase__ ).items() if v > 1]
if list(import_dict_objects.keys() ) != list(type_hint_objects.keys() ):
return ["Both sides of the init do not have the same backends!"]
lowerCAmelCase_ :Any = []
for key in import_dict_objects.keys():
lowerCAmelCase_ :Any = find_duplicates(import_dict_objects[key] )
if duplicate_imports:
errors.append(f"""Duplicate _import_structure definitions for: {duplicate_imports}""" )
lowerCAmelCase_ :List[str] = find_duplicates(type_hint_objects[key] )
if duplicate_type_hints:
errors.append(f"""Duplicate TYPE_CHECKING objects for: {duplicate_type_hints}""" )
if sorted(set(import_dict_objects[key] ) ) != sorted(set(type_hint_objects[key] ) ):
lowerCAmelCase_ :List[Any] = """base imports""" if key == """none""" else f"""{key} backend"""
errors.append(f"""Differences for {name}:""" )
for a in type_hint_objects[key]:
if a not in import_dict_objects[key]:
errors.append(f""" {a} in TYPE_HINT but not in _import_structure.""" )
for a in import_dict_objects[key]:
if a not in type_hint_objects[key]:
errors.append(f""" {a} in _import_structure but not in TYPE_HINT.""" )
return errors
def _snake_case ( ) -> Tuple:
'''simple docstring'''
lowerCAmelCase_ :int = []
for root, _, files in os.walk(lowercase__ ):
if "__init__.py" in files:
lowerCAmelCase_ :Any = os.path.join(lowercase__ , """__init__.py""" )
lowerCAmelCase_ :List[Any] = parse_init(lowercase__ )
if objects is not None:
lowerCAmelCase_ :str = analyze_results(*lowercase__ )
if len(lowercase__ ) > 0:
lowerCAmelCase_ :Any = f"""Problem in {fname}, both halves do not define the same objects.\n{errors[0]}"""
failures.append("""\n""".join(lowercase__ ) )
if len(lowercase__ ) > 0:
raise ValueError("""\n\n""".join(lowercase__ ) )
def _snake_case ( ) -> List[str]:
'''simple docstring'''
lowerCAmelCase_ :Dict = []
for path, directories, files in os.walk(lowercase__ ):
for folder in directories:
# Ignore private modules
if folder.startswith("""_""" ):
directories.remove(lowercase__ )
continue
# Ignore leftovers from branches (empty folders apart from pycache)
if len(list((Path(lowercase__ ) / folder).glob("""*.py""" ) ) ) == 0:
continue
lowerCAmelCase_ :Tuple = str((Path(lowercase__ ) / folder).relative_to(lowercase__ ) )
lowerCAmelCase_ :Any = short_path.replace(os.path.sep , """.""" )
submodules.append(lowercase__ )
for fname in files:
if fname == "__init__.py":
continue
lowerCAmelCase_ :str = str((Path(lowercase__ ) / fname).relative_to(lowercase__ ) )
lowerCAmelCase_ :int = short_path.replace(""".py""" , """""" ).replace(os.path.sep , """.""" )
if len(submodule.split(""".""" ) ) == 1:
submodules.append(lowercase__ )
return submodules
__UpperCAmelCase = [
'convert_pytorch_checkpoint_to_tf2',
'modeling_flax_pytorch_utils',
]
def _snake_case ( ) -> Union[str, Any]:
'''simple docstring'''
lowerCAmelCase_ :int = importlib.util.spec_from_file_location(
"""transformers""" , os.path.join(lowercase__ , """__init__.py""" ) , submodule_search_locations=[PATH_TO_TRANSFORMERS] , )
lowerCAmelCase_ :List[Any] = spec.loader.load_module()
lowerCAmelCase_ :List[Any] = [
module
for module in get_transformers_submodules()
if module not in IGNORE_SUBMODULES and module not in transformers._import_structure.keys()
]
if len(lowercase__ ) > 0:
lowerCAmelCase_ :Optional[int] = """\n""".join(f"""- {module}""" for module in module_not_registered )
raise ValueError(
"""The following submodules are not properly registered in the main init of Transformers:\n"""
f"""{list_of_modules}\n"""
"""Make sure they appear somewhere in the keys of `_import_structure` with an empty list as value.""" )
if __name__ == "__main__":
check_all_inits()
check_submodules()
| 84
|
def A (__A : list , __A : int , __A : int = 0 , __A : int = 0 ) -> int:
"""simple docstring"""
UpperCAmelCase_ = right or len(__A ) - 1
if left > right:
return -1
elif list_data[left] == key:
return left
elif list_data[right] == key:
return right
else:
return search(__A , __A , left + 1 , right - 1 )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 51
| 0
|
'''simple docstring'''
from maths.prime_factors import prime_factors
def UpperCamelCase_( snake_case : int ):
'''simple docstring'''
if not isinstance(snake_case , snake_case ):
snake_case_ = f'Input value of [number={number}] must be an integer'
raise TypeError(snake_case )
if number < 1:
raise ValueError("Input must be a positive integer" )
return -1 if len(prime_factors(snake_case ) ) % 2 else 1
if __name__ == "__main__":
import doctest
doctest.testmod()
| 85
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : int = logging.get_logger(__name__)
snake_case_ : str = {}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''llama'''
UpperCAmelCase__ : Dict = ['''past_key_values''']
def __init__( self : str , _snake_case : List[str]=32000 , _snake_case : int=4096 , _snake_case : List[str]=11008 , _snake_case : Optional[int]=32 , _snake_case : List[Any]=32 , _snake_case : Tuple=None , _snake_case : int="silu" , _snake_case : List[Any]=2048 , _snake_case : List[str]=0.0_2 , _snake_case : Any=1e-6 , _snake_case : List[str]=True , _snake_case : Optional[Any]=0 , _snake_case : Dict=1 , _snake_case : List[Any]=2 , _snake_case : str=1 , _snake_case : Union[str, Any]=False , _snake_case : str=None , **_snake_case : List[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = max_position_embeddings
UpperCAmelCase_ = hidden_size
UpperCAmelCase_ = intermediate_size
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
# for backward compatibility
if num_key_value_heads is None:
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = num_key_value_heads
UpperCAmelCase_ = hidden_act
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = rms_norm_eps
UpperCAmelCase_ = pretraining_tp
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = rope_scaling
self._rope_scaling_validation()
super().__init__(
pad_token_id=_snake_case , bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case , )
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
if self.rope_scaling is None:
return
if not isinstance(self.rope_scaling , _snake_case) or len(self.rope_scaling) != 2:
raise ValueError(
'''`rope_scaling` must be a dictionary with with two fields, `name` and `factor`, '''
F"""got {self.rope_scaling}""")
UpperCAmelCase_ = self.rope_scaling.get('''type''' , _snake_case)
UpperCAmelCase_ = self.rope_scaling.get('''factor''' , _snake_case)
if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic"]:
raise ValueError(
F"""`rope_scaling`'s name field must be one of ['linear', 'dynamic'], got {rope_scaling_type}""")
if rope_scaling_factor is None or not isinstance(_snake_case , _snake_case) or rope_scaling_factor <= 1.0:
raise ValueError(F"""`rope_scaling`'s factor field must be an float > 1, got {rope_scaling_factor}""")
| 51
| 0
|
"""simple docstring"""
from binascii import hexlify
from hashlib import shaaaa
from os import urandom
# RFC 3526 - More Modular Exponential (MODP) Diffie-Hellman groups for
# Internet Key Exchange (IKE) https://tools.ietf.org/html/rfc3526
lowerCamelCase__ = {
# 1536-bit
5: {
"""prime""": int(
"""FFFFFFFFFFFFFFFFC90FDAA22168C234C4C6628B80DC1CD1"""
+ """29024E088A67CC74020BBEA63B139B22514A08798E3404DD"""
+ """EF9519B3CD3A431B302B0A6DF25F14374FE1356D6D51C245"""
+ """E485B576625E7EC6F44C42E9A637ED6B0BFF5CB6F406B7ED"""
+ """EE386BFB5A899FA5AE9F24117C4B1FE649286651ECE45B3D"""
+ """C2007CB8A163BF0598DA48361C55D39A69163FA8FD24CF5F"""
+ """83655D23DCA3AD961C62F356208552BB9ED529077096966D"""
+ """670C354E4ABC9804F1746C08CA237327FFFFFFFFFFFFFFFF""",
base=16,
),
"""generator""": 2,
},
# 2048-bit
14: {
"""prime""": int(
"""FFFFFFFFFFFFFFFFC90FDAA22168C234C4C6628B80DC1CD1"""
+ """29024E088A67CC74020BBEA63B139B22514A08798E3404DD"""
+ """EF9519B3CD3A431B302B0A6DF25F14374FE1356D6D51C245"""
+ """E485B576625E7EC6F44C42E9A637ED6B0BFF5CB6F406B7ED"""
+ """EE386BFB5A899FA5AE9F24117C4B1FE649286651ECE45B3D"""
+ """C2007CB8A163BF0598DA48361C55D39A69163FA8FD24CF5F"""
+ """83655D23DCA3AD961C62F356208552BB9ED529077096966D"""
+ """670C354E4ABC9804F1746C08CA18217C32905E462E36CE3B"""
+ """E39E772C180E86039B2783A2EC07A28FB5C55DF06F4C52C9"""
+ """DE2BCBF6955817183995497CEA956AE515D2261898FA0510"""
+ """15728E5A8AACAA68FFFFFFFFFFFFFFFF""",
base=16,
),
"""generator""": 2,
},
# 3072-bit
15: {
"""prime""": int(
"""FFFFFFFFFFFFFFFFC90FDAA22168C234C4C6628B80DC1CD1"""
+ """29024E088A67CC74020BBEA63B139B22514A08798E3404DD"""
+ """EF9519B3CD3A431B302B0A6DF25F14374FE1356D6D51C245"""
+ """E485B576625E7EC6F44C42E9A637ED6B0BFF5CB6F406B7ED"""
+ """EE386BFB5A899FA5AE9F24117C4B1FE649286651ECE45B3D"""
+ """C2007CB8A163BF0598DA48361C55D39A69163FA8FD24CF5F"""
+ """83655D23DCA3AD961C62F356208552BB9ED529077096966D"""
+ """670C354E4ABC9804F1746C08CA18217C32905E462E36CE3B"""
+ """E39E772C180E86039B2783A2EC07A28FB5C55DF06F4C52C9"""
+ """DE2BCBF6955817183995497CEA956AE515D2261898FA0510"""
+ """15728E5A8AAAC42DAD33170D04507A33A85521ABDF1CBA64"""
+ """ECFB850458DBEF0A8AEA71575D060C7DB3970F85A6E1E4C7"""
+ """ABF5AE8CDB0933D71E8C94E04A25619DCEE3D2261AD2EE6B"""
+ """F12FFA06D98A0864D87602733EC86A64521F2B18177B200C"""
+ """BBE117577A615D6C770988C0BAD946E208E24FA074E5AB31"""
+ """43DB5BFCE0FD108E4B82D120A93AD2CAFFFFFFFFFFFFFFFF""",
base=16,
),
"""generator""": 2,
},
# 4096-bit
16: {
"""prime""": int(
"""FFFFFFFFFFFFFFFFC90FDAA22168C234C4C6628B80DC1CD1"""
+ """29024E088A67CC74020BBEA63B139B22514A08798E3404DD"""
+ """EF9519B3CD3A431B302B0A6DF25F14374FE1356D6D51C245"""
+ """E485B576625E7EC6F44C42E9A637ED6B0BFF5CB6F406B7ED"""
+ """EE386BFB5A899FA5AE9F24117C4B1FE649286651ECE45B3D"""
+ """C2007CB8A163BF0598DA48361C55D39A69163FA8FD24CF5F"""
+ """83655D23DCA3AD961C62F356208552BB9ED529077096966D"""
+ """670C354E4ABC9804F1746C08CA18217C32905E462E36CE3B"""
+ """E39E772C180E86039B2783A2EC07A28FB5C55DF06F4C52C9"""
+ """DE2BCBF6955817183995497CEA956AE515D2261898FA0510"""
+ """15728E5A8AAAC42DAD33170D04507A33A85521ABDF1CBA64"""
+ """ECFB850458DBEF0A8AEA71575D060C7DB3970F85A6E1E4C7"""
+ """ABF5AE8CDB0933D71E8C94E04A25619DCEE3D2261AD2EE6B"""
+ """F12FFA06D98A0864D87602733EC86A64521F2B18177B200C"""
+ """BBE117577A615D6C770988C0BAD946E208E24FA074E5AB31"""
+ """43DB5BFCE0FD108E4B82D120A92108011A723C12A787E6D7"""
+ """88719A10BDBA5B2699C327186AF4E23C1A946834B6150BDA"""
+ """2583E9CA2AD44CE8DBBBC2DB04DE8EF92E8EFC141FBECAA6"""
+ """287C59474E6BC05D99B2964FA090C3A2233BA186515BE7ED"""
+ """1F612970CEE2D7AFB81BDD762170481CD0069127D5B05AA9"""
+ """93B4EA988D8FDDC186FFB7DC90A6C08F4DF435C934063199"""
+ """FFFFFFFFFFFFFFFF""",
base=16,
),
"""generator""": 2,
},
# 6144-bit
17: {
"""prime""": int(
"""FFFFFFFFFFFFFFFFC90FDAA22168C234C4C6628B80DC1CD129024E08"""
+ """8A67CC74020BBEA63B139B22514A08798E3404DDEF9519B3CD3A431B"""
+ """302B0A6DF25F14374FE1356D6D51C245E485B576625E7EC6F44C42E9"""
+ """A637ED6B0BFF5CB6F406B7EDEE386BFB5A899FA5AE9F24117C4B1FE6"""
+ """49286651ECE45B3DC2007CB8A163BF0598DA48361C55D39A69163FA8"""
+ """FD24CF5F83655D23DCA3AD961C62F356208552BB9ED529077096966D"""
+ """670C354E4ABC9804F1746C08CA18217C32905E462E36CE3BE39E772C"""
+ """180E86039B2783A2EC07A28FB5C55DF06F4C52C9DE2BCBF695581718"""
+ """3995497CEA956AE515D2261898FA051015728E5A8AAAC42DAD33170D"""
+ """04507A33A85521ABDF1CBA64ECFB850458DBEF0A8AEA71575D060C7D"""
+ """B3970F85A6E1E4C7ABF5AE8CDB0933D71E8C94E04A25619DCEE3D226"""
+ """1AD2EE6BF12FFA06D98A0864D87602733EC86A64521F2B18177B200C"""
+ """BBE117577A615D6C770988C0BAD946E208E24FA074E5AB3143DB5BFC"""
+ """E0FD108E4B82D120A92108011A723C12A787E6D788719A10BDBA5B26"""
+ """99C327186AF4E23C1A946834B6150BDA2583E9CA2AD44CE8DBBBC2DB"""
+ """04DE8EF92E8EFC141FBECAA6287C59474E6BC05D99B2964FA090C3A2"""
+ """233BA186515BE7ED1F612970CEE2D7AFB81BDD762170481CD0069127"""
+ """D5B05AA993B4EA988D8FDDC186FFB7DC90A6C08F4DF435C934028492"""
+ """36C3FAB4D27C7026C1D4DCB2602646DEC9751E763DBA37BDF8FF9406"""
+ """AD9E530EE5DB382F413001AEB06A53ED9027D831179727B0865A8918"""
+ """DA3EDBEBCF9B14ED44CE6CBACED4BB1BDB7F1447E6CC254B33205151"""
+ """2BD7AF426FB8F401378CD2BF5983CA01C64B92ECF032EA15D1721D03"""
+ """F482D7CE6E74FEF6D55E702F46980C82B5A84031900B1C9E59E7C97F"""
+ """BEC7E8F323A97A7E36CC88BE0F1D45B7FF585AC54BD407B22B4154AA"""
+ """CC8F6D7EBF48E1D814CC5ED20F8037E0A79715EEF29BE32806A1D58B"""
+ """B7C5DA76F550AA3D8A1FBFF0EB19CCB1A313D55CDA56C9EC2EF29632"""
+ """387FE8D76E3C0468043E8F663F4860EE12BF2D5B0B7474D6E694F91E"""
+ """6DCC4024FFFFFFFFFFFFFFFF""",
base=16,
),
"""generator""": 2,
},
# 8192-bit
18: {
"""prime""": int(
"""FFFFFFFFFFFFFFFFC90FDAA22168C234C4C6628B80DC1CD1"""
+ """29024E088A67CC74020BBEA63B139B22514A08798E3404DD"""
+ """EF9519B3CD3A431B302B0A6DF25F14374FE1356D6D51C245"""
+ """E485B576625E7EC6F44C42E9A637ED6B0BFF5CB6F406B7ED"""
+ """EE386BFB5A899FA5AE9F24117C4B1FE649286651ECE45B3D"""
+ """C2007CB8A163BF0598DA48361C55D39A69163FA8FD24CF5F"""
+ """83655D23DCA3AD961C62F356208552BB9ED529077096966D"""
+ """670C354E4ABC9804F1746C08CA18217C32905E462E36CE3B"""
+ """E39E772C180E86039B2783A2EC07A28FB5C55DF06F4C52C9"""
+ """DE2BCBF6955817183995497CEA956AE515D2261898FA0510"""
+ """15728E5A8AAAC42DAD33170D04507A33A85521ABDF1CBA64"""
+ """ECFB850458DBEF0A8AEA71575D060C7DB3970F85A6E1E4C7"""
+ """ABF5AE8CDB0933D71E8C94E04A25619DCEE3D2261AD2EE6B"""
+ """F12FFA06D98A0864D87602733EC86A64521F2B18177B200C"""
+ """BBE117577A615D6C770988C0BAD946E208E24FA074E5AB31"""
+ """43DB5BFCE0FD108E4B82D120A92108011A723C12A787E6D7"""
+ """88719A10BDBA5B2699C327186AF4E23C1A946834B6150BDA"""
+ """2583E9CA2AD44CE8DBBBC2DB04DE8EF92E8EFC141FBECAA6"""
+ """287C59474E6BC05D99B2964FA090C3A2233BA186515BE7ED"""
+ """1F612970CEE2D7AFB81BDD762170481CD0069127D5B05AA9"""
+ """93B4EA988D8FDDC186FFB7DC90A6C08F4DF435C934028492"""
+ """36C3FAB4D27C7026C1D4DCB2602646DEC9751E763DBA37BD"""
+ """F8FF9406AD9E530EE5DB382F413001AEB06A53ED9027D831"""
+ """179727B0865A8918DA3EDBEBCF9B14ED44CE6CBACED4BB1B"""
+ """DB7F1447E6CC254B332051512BD7AF426FB8F401378CD2BF"""
+ """5983CA01C64B92ECF032EA15D1721D03F482D7CE6E74FEF6"""
+ """D55E702F46980C82B5A84031900B1C9E59E7C97FBEC7E8F3"""
+ """23A97A7E36CC88BE0F1D45B7FF585AC54BD407B22B4154AA"""
+ """CC8F6D7EBF48E1D814CC5ED20F8037E0A79715EEF29BE328"""
+ """06A1D58BB7C5DA76F550AA3D8A1FBFF0EB19CCB1A313D55C"""
+ """DA56C9EC2EF29632387FE8D76E3C0468043E8F663F4860EE"""
+ """12BF2D5B0B7474D6E694F91E6DBE115974A3926F12FEE5E4"""
+ """38777CB6A932DF8CD8BEC4D073B931BA3BC832B68D9DD300"""
+ """741FA7BF8AFC47ED2576F6936BA424663AAB639C5AE4F568"""
+ """3423B4742BF1C978238F16CBE39D652DE3FDB8BEFC848AD9"""
+ """22222E04A4037C0713EB57A81A23F0C73473FC646CEA306B"""
+ """4BCBC8862F8385DDFA9D4B7FA2C087E879683303ED5BDD3A"""
+ """062B3CF5B3A278A66D2A13F83F44F82DDF310EE074AB6A36"""
+ """4597E899A0255DC164F31CC50846851DF9AB48195DED7EA1"""
+ """B1D510BD7EE74D73FAF36BC31ECFA268359046F4EB879F92"""
+ """4009438B481C6CD7889A002ED5EE382BC9190DA6FC026E47"""
+ """9558E4475677E9AA9E3050E2765694DFC81F56E880B96E71"""
+ """60C980DD98EDD3DFFFFFFFFFFFFFFFFF""",
base=16,
),
"""generator""": 2,
},
}
class A__ :
def __init__( self , _SCREAMING_SNAKE_CASE = 14 ):
if group not in primes:
raise ValueError('Unsupported Group' )
__lowerCAmelCase : Dict = primes[group]['prime']
__lowerCAmelCase : Any = primes[group]['generator']
__lowerCAmelCase : Dict = int(hexlify(urandom(32 ) ) , base=16 )
def __lowerCamelCase ( self ):
return hex(self.__private_key )[2:]
def __lowerCamelCase ( self ):
__lowerCAmelCase : str = pow(self.generator , self.__private_key , self.prime )
return hex(_SCREAMING_SNAKE_CASE )[2:]
def __lowerCamelCase ( self , _SCREAMING_SNAKE_CASE ):
# check if the other public key is valid based on NIST SP800-56
return (
2 <= key <= self.prime - 2
and pow(_SCREAMING_SNAKE_CASE , (self.prime - 1) // 2 , self.prime ) == 1
)
def __lowerCamelCase ( self , _SCREAMING_SNAKE_CASE ):
__lowerCAmelCase : Any = int(_SCREAMING_SNAKE_CASE , base=16 )
if not self.is_valid_public_key(_SCREAMING_SNAKE_CASE ):
raise ValueError('Invalid public key' )
__lowerCAmelCase : Dict = pow(_SCREAMING_SNAKE_CASE , self.__private_key , self.prime )
return shaaaa(str(_SCREAMING_SNAKE_CASE ).encode() ).hexdigest()
@staticmethod
def __lowerCamelCase ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
# check if the other public key is valid based on NIST SP800-56
return (
2 <= remote_public_key_str <= prime - 2
and pow(_SCREAMING_SNAKE_CASE , (prime - 1) // 2 , _SCREAMING_SNAKE_CASE ) == 1
)
@staticmethod
def __lowerCamelCase ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = 14 ):
__lowerCAmelCase : Optional[int] = int(_SCREAMING_SNAKE_CASE , base=16 )
__lowerCAmelCase : List[Any] = int(_SCREAMING_SNAKE_CASE , base=16 )
__lowerCAmelCase : Tuple = primes[group]['prime']
if not DiffieHellman.is_valid_public_key_static(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
raise ValueError('Invalid public key' )
__lowerCAmelCase : Tuple = pow(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
return shaaaa(str(_SCREAMING_SNAKE_CASE ).encode() ).hexdigest()
if __name__ == "__main__":
import doctest
doctest.testmod()
| 86
|
from collections import OrderedDict
from typing import Any, List, Mapping, Optional
from ... import PreTrainedTokenizer, TensorType, is_torch_available
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfigWithPast, PatchingSpec
from ...utils import logging
snake_case_ : List[str] = logging.get_logger(__name__)
snake_case_ : Tuple = {
"Salesforce/codegen-350M-nl": "https://huggingface.co/Salesforce/codegen-350M-nl/resolve/main/config.json",
"Salesforce/codegen-350M-multi": "https://huggingface.co/Salesforce/codegen-350M-multi/resolve/main/config.json",
"Salesforce/codegen-350M-mono": "https://huggingface.co/Salesforce/codegen-350M-mono/resolve/main/config.json",
"Salesforce/codegen-2B-nl": "https://huggingface.co/Salesforce/codegen-2B-nl/resolve/main/config.json",
"Salesforce/codegen-2B-multi": "https://huggingface.co/Salesforce/codegen-2B-multi/resolve/main/config.json",
"Salesforce/codegen-2B-mono": "https://huggingface.co/Salesforce/codegen-2B-mono/resolve/main/config.json",
"Salesforce/codegen-6B-nl": "https://huggingface.co/Salesforce/codegen-6B-nl/resolve/main/config.json",
"Salesforce/codegen-6B-multi": "https://huggingface.co/Salesforce/codegen-6B-multi/resolve/main/config.json",
"Salesforce/codegen-6B-mono": "https://huggingface.co/Salesforce/codegen-6B-mono/resolve/main/config.json",
"Salesforce/codegen-16B-nl": "https://huggingface.co/Salesforce/codegen-16B-nl/resolve/main/config.json",
"Salesforce/codegen-16B-multi": "https://huggingface.co/Salesforce/codegen-16B-multi/resolve/main/config.json",
"Salesforce/codegen-16B-mono": "https://huggingface.co/Salesforce/codegen-16B-mono/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''codegen'''
UpperCAmelCase__ : int = {
'''max_position_embeddings''': '''n_positions''',
'''hidden_size''': '''n_embd''',
'''num_attention_heads''': '''n_head''',
'''num_hidden_layers''': '''n_layer''',
}
def __init__( self : Union[str, Any] , _snake_case : Union[str, Any]=50400 , _snake_case : Optional[int]=2048 , _snake_case : Union[str, Any]=2048 , _snake_case : List[str]=4096 , _snake_case : Any=28 , _snake_case : List[str]=16 , _snake_case : int=64 , _snake_case : Tuple=None , _snake_case : Dict="gelu_new" , _snake_case : Union[str, Any]=0.0 , _snake_case : Optional[Any]=0.0 , _snake_case : List[Any]=0.0 , _snake_case : List[Any]=1e-5 , _snake_case : List[str]=0.0_2 , _snake_case : Optional[Any]=True , _snake_case : int=50256 , _snake_case : Tuple=50256 , _snake_case : int=False , **_snake_case : Any , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = n_ctx
UpperCAmelCase_ = n_positions
UpperCAmelCase_ = n_embd
UpperCAmelCase_ = n_layer
UpperCAmelCase_ = n_head
UpperCAmelCase_ = n_inner
UpperCAmelCase_ = rotary_dim
UpperCAmelCase_ = activation_function
UpperCAmelCase_ = resid_pdrop
UpperCAmelCase_ = embd_pdrop
UpperCAmelCase_ = attn_pdrop
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
super().__init__(
bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case)
class __snake_case ( a ):
def __init__( self : Tuple , _snake_case : PretrainedConfig , _snake_case : str = "default" , _snake_case : List[PatchingSpec] = None , _snake_case : bool = False , ):
"""simple docstring"""
super().__init__(_snake_case , task=_snake_case , patching_specs=_snake_case , use_past=_snake_case)
if not getattr(self._config , '''pad_token_id''' , _snake_case):
# TODO: how to do that better?
UpperCAmelCase_ = 0
@property
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = OrderedDict({'''input_ids''': {0: '''batch''', 1: '''sequence'''}})
if self.use_past:
self.fill_with_past_key_values_(_snake_case , direction='''inputs''')
UpperCAmelCase_ = {0: '''batch''', 1: '''past_sequence + sequence'''}
else:
UpperCAmelCase_ = {0: '''batch''', 1: '''sequence'''}
return common_inputs
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return self._config.n_layer
@property
def lowerCamelCase ( self : int):
"""simple docstring"""
return self._config.n_head
def lowerCamelCase ( self : Optional[int] , _snake_case : PreTrainedTokenizer , _snake_case : int = -1 , _snake_case : int = -1 , _snake_case : bool = False , _snake_case : Optional[TensorType] = None , ):
"""simple docstring"""
UpperCAmelCase_ = super(_snake_case , self).generate_dummy_inputs(
_snake_case , batch_size=_snake_case , seq_length=_snake_case , is_pair=_snake_case , framework=_snake_case)
# We need to order the input in the way they appears in the forward()
UpperCAmelCase_ = OrderedDict({'''input_ids''': common_inputs['''input_ids''']})
# Need to add the past_keys
if self.use_past:
if not is_torch_available():
raise ValueError('''Cannot generate dummy past_keys inputs without PyTorch installed.''')
else:
import torch
UpperCAmelCase_ , UpperCAmelCase_ = common_inputs['''input_ids'''].shape
# Not using the same length for past_key_values
UpperCAmelCase_ = seqlen + 2
UpperCAmelCase_ = (
batch,
self.num_attention_heads,
past_key_values_length,
self._config.hidden_size // self.num_attention_heads,
)
UpperCAmelCase_ = [
(torch.zeros(_snake_case), torch.zeros(_snake_case)) for _ in range(self.num_layers)
]
UpperCAmelCase_ = common_inputs['''attention_mask''']
if self.use_past:
UpperCAmelCase_ = ordered_inputs['''attention_mask'''].dtype
UpperCAmelCase_ = torch.cat(
[ordered_inputs['''attention_mask'''], torch.ones(_snake_case , _snake_case , dtype=_snake_case)] , dim=1)
return ordered_inputs
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return 13
| 51
| 0
|
import logging
from transformers.configuration_utils import PretrainedConfig
UpperCamelCase = logging.getLogger(__name__)
class snake_case_ ( __A ):
__A : List[Any] = "masked_bert"
def __init__( self : Optional[int] , lowercase_ : List[str]=3_05_22 , lowercase_ : Optional[Any]=7_68 , lowercase_ : Tuple=12 , lowercase_ : str=12 , lowercase_ : List[Any]=30_72 , lowercase_ : Union[str, Any]="gelu" , lowercase_ : Dict=0.1 , lowercase_ : Union[str, Any]=0.1 , lowercase_ : List[Any]=5_12 , lowercase_ : List[str]=2 , lowercase_ : List[str]=0.02 , lowercase_ : int=1E-12 , lowercase_ : Any=0 , lowercase_ : List[Any]="topK" , lowercase_ : List[str]="constant" , lowercase_ : Dict=0.0 , **lowercase_ : Union[str, Any] , ) -> Optional[Any]:
super().__init__(pad_token_id=lowercase_ , **lowercase_ )
lowercase__ : int = vocab_size
lowercase__ : Optional[int] = hidden_size
lowercase__ : Any = num_hidden_layers
lowercase__ : Dict = num_attention_heads
lowercase__ : Optional[int] = hidden_act
lowercase__ : Any = intermediate_size
lowercase__ : Optional[Any] = hidden_dropout_prob
lowercase__ : int = attention_probs_dropout_prob
lowercase__ : str = max_position_embeddings
lowercase__ : str = type_vocab_size
lowercase__ : Optional[int] = initializer_range
lowercase__ : Optional[int] = layer_norm_eps
lowercase__ : Tuple = pruning_method
lowercase__ : Union[str, Any] = mask_init
lowercase__ : List[str] = mask_scale
| 87
|
import os
import unittest
from transformers.models.phobert.tokenization_phobert import VOCAB_FILES_NAMES, PhobertTokenizer
from ...test_tokenization_common import TokenizerTesterMixin
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : Any = PhobertTokenizer
UpperCAmelCase__ : List[str] = False
def lowerCamelCase ( self : str):
"""simple docstring"""
super().setUp()
# Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt
UpperCAmelCase_ = ['''T@@''', '''i''', '''I''', '''R@@''', '''r''', '''e@@''']
UpperCAmelCase_ = dict(zip(_snake_case , range(len(_snake_case))))
UpperCAmelCase_ = ['''#version: 0.2''', '''l à</w>''']
UpperCAmelCase_ = {'''unk_token''': '''<unk>'''}
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''vocab_file'''])
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''merges_file'''])
with open(self.vocab_file , '''w''' , encoding='''utf-8''') as fp:
for token in vocab_tokens:
fp.write(F"""{token} {vocab_tokens[token]}\n""")
with open(self.merges_file , '''w''' , encoding='''utf-8''') as fp:
fp.write('''\n'''.join(_snake_case))
def lowerCamelCase ( self : int , **_snake_case : Any):
"""simple docstring"""
kwargs.update(self.special_tokens_map)
return PhobertTokenizer.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Optional[Any] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T<unk> i <unk> <unk> <unk> <unk> <unk> <unk> I Re<unk> e<unk> <unk> <unk> <unk>'''
return input_text, output_text
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = PhobertTokenizer(self.vocab_file , self.merges_file , **self.special_tokens_map)
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T@@ ô@@ i l@@ à V@@ i@@ n@@ A@@ I R@@ e@@ s@@ e@@ a@@ r@@ c@@ h'''.split()
UpperCAmelCase_ = tokenizer.tokenize(_snake_case)
print(_snake_case)
self.assertListEqual(_snake_case , _snake_case)
UpperCAmelCase_ = tokens + [tokenizer.unk_token]
UpperCAmelCase_ = [4, 3, 5, 3, 3, 3, 3, 3, 3, 6, 7, 9, 3, 9, 3, 3, 3, 3, 3]
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case) , _snake_case)
| 51
| 0
|
import copy
from dataclasses import dataclass, field
from typing import ClassVar, Dict
from ..features import ClassLabel, Features, Value
from .base import TaskTemplate
@dataclass(frozen=_A )
class UpperCAmelCase_ ( _A ):
'''simple docstring'''
a__ = field(default="""text-classification""" , metadata={"""include_in_asdict_even_if_is_default""": True} )
a__ = Features({"""text""": Value("""string""" )} )
a__ = Features({"""labels""": ClassLabel} )
a__ = "text"
a__ = "labels"
def _lowercase ( self : Tuple , UpperCamelCase__ : Dict ) -> Dict:
"""simple docstring"""
if self.label_column not in features:
raise ValueError(F'''Column {self.label_column} is not present in features.''' )
if not isinstance(features[self.label_column] , UpperCamelCase__ ):
raise ValueError(F'''Column {self.label_column} is not a ClassLabel.''' )
__magic_name__ = copy.deepcopy(self )
__magic_name__ = self.label_schema.copy()
__magic_name__ = features[self.label_column]
__magic_name__ = label_schema
return task_template
@property
def _lowercase ( self : Optional[int] ) -> Dict[str, str]:
"""simple docstring"""
return {
self.text_column: "text",
self.label_column: "labels",
}
| 88
|
from typing import List, Optional, TypeVar
from .arrow_dataset import Dataset, _concatenate_map_style_datasets, _interleave_map_style_datasets
from .dataset_dict import DatasetDict, IterableDatasetDict
from .info import DatasetInfo
from .iterable_dataset import IterableDataset, _concatenate_iterable_datasets, _interleave_iterable_datasets
from .splits import NamedSplit
from .utils import logging
from .utils.py_utils import Literal
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[int] = TypeVar("DatasetType", Dataset, IterableDataset)
def A (__A : List[DatasetType] , __A : Optional[List[float]] = None , __A : Optional[int] = None , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : Literal["first_exhausted", "all_exhausted"] = "first_exhausted" , ) -> DatasetType:
"""simple docstring"""
from .arrow_dataset import Dataset
from .iterable_dataset import IterableDataset
if not datasets:
raise ValueError('''Unable to interleave an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if stopping_strategy not in ["first_exhausted", "all_exhausted"]:
raise ValueError(F"""{stopping_strategy} is not supported. Please enter a valid stopping_strategy.""" )
if dataset_type is Dataset:
return _interleave_map_style_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
else:
return _interleave_iterable_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
def A (__A : List[DatasetType] , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : int = 0 , ) -> DatasetType:
"""simple docstring"""
if not dsets:
raise ValueError('''Unable to concatenate an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if dataset_type is Dataset:
return _concatenate_map_style_datasets(__A , info=__A , split=__A , axis=__A )
else:
return _concatenate_iterable_datasets(__A , info=__A , split=__A , axis=__A )
| 51
| 0
|
'''simple docstring'''
import copy
import os
from typing import Union
from ...configuration_utils import PretrainedConfig
from ...models.auto.modeling_auto import MODEL_FOR_CAUSAL_LM_MAPPING_NAMES
from ...utils import logging
from ..auto import CONFIG_MAPPING
__lowerCAmelCase = logging.get_logger(__name__)
__lowerCAmelCase = {
'''salesforce/blip2-opt-2.7b''': '''https://huggingface.co/salesforce/blip2-opt-2.7b/resolve/main/config.json''',
}
class __magic_name__ ( _UpperCamelCase ):
lowerCAmelCase : Tuple = 'blip_2_vision_model'
def __init__( self : List[Any] ,_UpperCAmelCase : List[Any]=1408 ,_UpperCAmelCase : Optional[int]=6144 ,_UpperCAmelCase : List[str]=39 ,_UpperCAmelCase : Dict=16 ,_UpperCAmelCase : Optional[Any]=224 ,_UpperCAmelCase : Any=14 ,_UpperCAmelCase : Optional[int]="gelu" ,_UpperCAmelCase : Optional[int]=0.0_00_01 ,_UpperCAmelCase : Tuple=0.0 ,_UpperCAmelCase : Union[str, Any]=1E-10 ,_UpperCAmelCase : Any=True ,**_UpperCAmelCase : Dict ,):
super().__init__(**_UpperCAmelCase )
_a : int = hidden_size
_a : List[str] = intermediate_size
_a : List[Any] = num_hidden_layers
_a : Optional[int] = num_attention_heads
_a : Tuple = patch_size
_a : List[str] = image_size
_a : int = initializer_range
_a : Union[str, Any] = attention_dropout
_a : Union[str, Any] = layer_norm_eps
_a : List[str] = hidden_act
_a : Any = qkv_bias
@classmethod
def __lowercase ( cls : Tuple ,_UpperCAmelCase : Union[str, os.PathLike] ,**_UpperCAmelCase : Any ):
cls._set_token_in_kwargs(_UpperCAmelCase )
_a , _a : List[Any] = cls.get_config_dict(_UpperCAmelCase ,**_UpperCAmelCase )
# get the vision config dict if we are loading from Blip2Config
if config_dict.get('model_type' ) == "blip-2":
_a : Union[str, Any] = config_dict['vision_config']
if "model_type" in config_dict and hasattr(cls ,'model_type' ) and config_dict["model_type"] != cls.model_type:
logger.warning(
F"""You are using a model of type {config_dict['model_type']} to instantiate a model of type """
F"""{cls.model_type}. This is not supported for all configurations of models and can yield errors.""" )
return cls.from_dict(_UpperCAmelCase ,**_UpperCAmelCase )
class __magic_name__ ( _UpperCamelCase ):
lowerCAmelCase : Tuple = 'blip_2_qformer'
def __init__( self : int ,_UpperCAmelCase : List[str]=30522 ,_UpperCAmelCase : Dict=768 ,_UpperCAmelCase : Tuple=12 ,_UpperCAmelCase : Any=12 ,_UpperCAmelCase : int=3072 ,_UpperCAmelCase : str="gelu" ,_UpperCAmelCase : Dict=0.1 ,_UpperCAmelCase : str=0.1 ,_UpperCAmelCase : int=512 ,_UpperCAmelCase : Union[str, Any]=0.02 ,_UpperCAmelCase : List[str]=1E-12 ,_UpperCAmelCase : int=0 ,_UpperCAmelCase : Optional[int]="absolute" ,_UpperCAmelCase : List[str]=2 ,_UpperCAmelCase : Any=1408 ,**_UpperCAmelCase : Optional[int] ,):
super().__init__(pad_token_id=_UpperCAmelCase ,**_UpperCAmelCase )
_a : Any = vocab_size
_a : Optional[int] = hidden_size
_a : Dict = num_hidden_layers
_a : Dict = num_attention_heads
_a : str = hidden_act
_a : Any = intermediate_size
_a : Dict = hidden_dropout_prob
_a : str = attention_probs_dropout_prob
_a : Union[str, Any] = max_position_embeddings
_a : Tuple = initializer_range
_a : Optional[int] = layer_norm_eps
_a : int = position_embedding_type
_a : Optional[Any] = cross_attention_frequency
_a : List[str] = encoder_hidden_size
@classmethod
def __lowercase ( cls : Dict ,_UpperCAmelCase : Union[str, os.PathLike] ,**_UpperCAmelCase : Union[str, Any] ):
cls._set_token_in_kwargs(_UpperCAmelCase )
_a , _a : str = cls.get_config_dict(_UpperCAmelCase ,**_UpperCAmelCase )
# get the qformer config dict if we are loading from Blip2Config
if config_dict.get('model_type' ) == "blip-2":
_a : Any = config_dict['qformer_config']
if "model_type" in config_dict and hasattr(cls ,'model_type' ) and config_dict["model_type"] != cls.model_type:
logger.warning(
F"""You are using a model of type {config_dict['model_type']} to instantiate a model of type """
F"""{cls.model_type}. This is not supported for all configurations of models and can yield errors.""" )
return cls.from_dict(_UpperCAmelCase ,**_UpperCAmelCase )
class __magic_name__ ( _UpperCamelCase ):
lowerCAmelCase : Tuple = 'blip-2'
lowerCAmelCase : Any = True
def __init__( self : Any ,_UpperCAmelCase : Union[str, Any]=None ,_UpperCAmelCase : Tuple=None ,_UpperCAmelCase : Dict=None ,_UpperCAmelCase : int=32 ,**_UpperCAmelCase : str ):
super().__init__(**_UpperCAmelCase )
if vision_config is None:
_a : str = {}
logger.info('vision_config is None. initializing the Blip2VisionConfig with default values.' )
if qformer_config is None:
_a : List[str] = {}
logger.info('qformer_config is None. Initializing the Blip2QFormerConfig with default values.' )
if text_config is None:
_a : List[Any] = {}
logger.info('text_config is None. Initializing the text config with default values (`OPTConfig`).' )
_a : Optional[Any] = BlipaVisionConfig(**_UpperCAmelCase )
_a : Tuple = BlipaQFormerConfig(**_UpperCAmelCase )
_a : Optional[int] = text_config['model_type'] if 'model_type' in text_config else 'opt'
_a : Any = CONFIG_MAPPING[text_model_type](**_UpperCAmelCase )
_a : Dict = self.text_config.tie_word_embeddings
_a : Union[str, Any] = self.text_config.is_encoder_decoder
_a : int = num_query_tokens
_a : str = self.vision_config.hidden_size
_a : Optional[Any] = self.text_config.model_type in MODEL_FOR_CAUSAL_LM_MAPPING_NAMES
_a : Optional[int] = 1.0
_a : str = 0.02
@classmethod
def __lowercase ( cls : str ,_UpperCAmelCase : BlipaVisionConfig ,_UpperCAmelCase : BlipaQFormerConfig ,_UpperCAmelCase : PretrainedConfig ,**_UpperCAmelCase : List[Any] ,):
return cls(
vision_config=vision_config.to_dict() ,qformer_config=qformer_config.to_dict() ,text_config=text_config.to_dict() ,**_UpperCAmelCase ,)
def __lowercase ( self : int ):
_a : Dict = copy.deepcopy(self.__dict__ )
_a : Any = self.vision_config.to_dict()
_a : Optional[int] = self.qformer_config.to_dict()
_a : Dict = self.text_config.to_dict()
_a : Optional[Any] = self.__class__.model_type
return output
| 89
|
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
snake_case_ : Optional[Any] = "pt"
elif is_tf_available():
snake_case_ : Union[str, Any] = "tf"
else:
snake_case_ : str = "jax"
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = ByTaTokenizer
UpperCAmelCase__ : int = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().setUp()
UpperCAmelCase_ = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname)
@cached_property
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
return ByTaTokenizer.from_pretrained('''google/byt5-small''')
def lowerCamelCase ( self : List[str] , **_snake_case : Union[str, Any]):
"""simple docstring"""
return self.tokenizer_class.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Dict , _snake_case : int , _snake_case : Tuple=False , _snake_case : Dict=20 , _snake_case : Optional[Any]=5):
"""simple docstring"""
UpperCAmelCase_ = []
for i in range(len(_snake_case)):
try:
UpperCAmelCase_ = tokenizer.decode([i] , clean_up_tokenization_spaces=_snake_case)
except UnicodeDecodeError:
pass
toks.append((i, tok))
UpperCAmelCase_ = list(filter(lambda _snake_case: re.match(r'''^[ a-zA-Z]+$''' , t[1]) , _snake_case))
UpperCAmelCase_ = list(filter(lambda _snake_case: [t[0]] == tokenizer.encode(t[1] , add_special_tokens=_snake_case) , _snake_case))
if max_length is not None and len(_snake_case) > max_length:
UpperCAmelCase_ = toks[:max_length]
if min_length is not None and len(_snake_case) < min_length and len(_snake_case) > 0:
while len(_snake_case) < min_length:
UpperCAmelCase_ = toks + toks
# toks_str = [t[1] for t in toks]
UpperCAmelCase_ = [t[0] for t in toks]
# Ensure consistency
UpperCAmelCase_ = tokenizer.decode(_snake_case , clean_up_tokenization_spaces=_snake_case)
if " " not in output_txt and len(_snake_case) > 1:
UpperCAmelCase_ = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=_snake_case)
+ ''' '''
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=_snake_case)
)
if with_prefix_space:
UpperCAmelCase_ = ''' ''' + output_txt
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
return output_txt, output_ids
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = tokenizer(['''hi</s>''', '''I went to the gym</s>''', '''</s>'''])
UpperCAmelCase_ = tokenizer(['''hi''', '''I went to the gym''', ''''''])
self.assertListEqual(batch_with_eos_added['''input_ids'''] , batch_without_eos_added['''input_ids'''])
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = '''Unicode €.'''
UpperCAmelCase_ = tokenizer(_snake_case)
UpperCAmelCase_ = [88, 113, 108, 102, 114, 103, 104, 35, 229, 133, 175, 49, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''Unicode €.</s>''')
UpperCAmelCase_ = tokenizer('''e è é ê ë''')
UpperCAmelCase_ = [104, 35, 198, 171, 35, 198, 172, 35, 198, 173, 35, 198, 174, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''e è é ê ë</s>''')
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('''e è é ê ë''')) , '''e è é ê ë</s>''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 1, 0]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
if FRAMEWORK != "jax":
UpperCAmelCase_ = list(batch.input_ids.numpy()[0])
else:
UpperCAmelCase_ = list(batch.input_ids.tolist()[0])
self.assertListEqual(_snake_case , _snake_case)
self.assertEqual((2, 37) , batch.input_ids.shape)
self.assertEqual((2, 37) , batch.attention_mask.shape)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
# check if input_ids are returned and no decoder_input_ids
self.assertIn('''input_ids''' , _snake_case)
self.assertIn('''attention_mask''' , _snake_case)
self.assertNotIn('''decoder_input_ids''' , _snake_case)
self.assertNotIn('''decoder_attention_mask''' , _snake_case)
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = [
'''Summary of the text.''',
'''Another summary.''',
]
UpperCAmelCase_ = tokenizer(
text_target=_snake_case , max_length=32 , padding='''max_length''' , truncation=_snake_case , return_tensors=_snake_case)
self.assertEqual(32 , targets['''input_ids'''].shape[1])
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization. </s>''']
UpperCAmelCase_ = ['''Summary of the text. </s>''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 35, 1]
UpperCAmelCase_ = [86, 120, 112, 112, 100, 117, 124, 35, 114, 105, 35, 119, 107, 104, 35, 119, 104, 123, 119, 49, 35, 1]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , text_target=_snake_case)
self.assertEqual(_snake_case , batch['''input_ids'''][0])
self.assertEqual(_snake_case , batch['''labels'''][0])
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
self.assertNotEqual(tokenizer.model_max_length , 42)
# Now let's start the test
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
shutil.rmtree(_snake_case)
UpperCAmelCase_ = self.get_tokenizers(model_max_length=42)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
tokenizer.add_tokens(['''bim''', '''bambam'''])
UpperCAmelCase_ = tokenizer.additional_special_tokens
additional_special_tokens.append('''new_additional_special_token''')
tokenizer.add_special_tokens({'''additional_special_tokens''': additional_special_tokens})
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
self.assertIn('''new_additional_special_token''' , after_tokenizer.additional_special_tokens)
self.assertEqual(after_tokenizer.model_max_length , 42)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case , model_max_length=43)
self.assertEqual(tokenizer.model_max_length , 43)
shutil.rmtree(_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
UpperCAmelCase_ = [F"""<extra_id_{i}>""" for i in range(125)]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , )
self.assertIn(
'''an_additional_special_token''' , tokenizer_without_change_in_init.additional_special_tokens)
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['''an_additional_special_token'''] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['''an_additional_special_token'''])) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
UpperCAmelCase_ = added_tokens_extra_ids + [AddedToken('''a_new_additional_special_token''' , lstrip=_snake_case)]
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , additional_special_tokens=_snake_case , )
self.assertIn('''a_new_additional_special_token''' , tokenizer.additional_special_tokens)
self.assertEqual(
['''a_new_additional_special_token'''] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['''a_new_additional_special_token'''])) , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer_class.from_pretrained(_snake_case)
self.assertTrue(tokenizer.decode([255]) == '''''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers(fast=_snake_case , do_lower_case=_snake_case)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = ['''t''', '''h''', '''i''', '''s''', ''' ''', '''i''', '''s''', ''' ''', '''a''', ''' ''', '''t''', '''e''', '''x''', '''t''', '''</s>''']
UpperCAmelCase_ = tokenizer.convert_tokens_to_string(_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = [
'''bos_token''',
'''eos_token''',
'''unk_token''',
'''sep_token''',
'''pad_token''',
'''cls_token''',
'''mask_token''',
]
UpperCAmelCase_ = 0
UpperCAmelCase_ = tokenizer.convert_ids_to_tokens(
_snake_case , skip_special_tokens=_snake_case)
for attr in attributes_list:
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , '''additional_special_tokens_ids''' , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [])
setattr(_snake_case , '''additional_special_tokens_ids''' , [token_id_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [token_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [token_id_to_test_setters])
| 51
| 0
|
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
__A = logging.get_logger(__name__)
__A = {
"facebook/data2vec-vision-base-ft": (
"https://huggingface.co/facebook/data2vec-vision-base-ft/resolve/main/config.json"
),
}
class __lowerCAmelCase ( __magic_name__ ):
"""simple docstring"""
snake_case_ = '''data2vec-vision'''
def __init__( self , lowerCamelCase__=768 , lowerCamelCase__=12 , lowerCamelCase__=12 , lowerCamelCase__=3_072 , lowerCamelCase__="gelu" , lowerCamelCase__=0.0 , lowerCamelCase__=0.0 , lowerCamelCase__=0.02 , lowerCamelCase__=1e-12 , lowerCamelCase__=224 , lowerCamelCase__=16 , lowerCamelCase__=3 , lowerCamelCase__=False , lowerCamelCase__=False , lowerCamelCase__=False , lowerCamelCase__=False , lowerCamelCase__=0.1 , lowerCamelCase__=0.1 , lowerCamelCase__=True , lowerCamelCase__=[3, 5, 7, 11] , lowerCamelCase__=[1, 2, 3, 6] , lowerCamelCase__=True , lowerCamelCase__=0.4 , lowerCamelCase__=256 , lowerCamelCase__=1 , lowerCamelCase__=False , lowerCamelCase__=255 , **lowerCamelCase__ , ) -> Optional[Any]:
'''simple docstring'''
super().__init__(**lowerCamelCase__ )
__lowerCamelCase = hidden_size
__lowerCamelCase = num_hidden_layers
__lowerCamelCase = num_attention_heads
__lowerCamelCase = intermediate_size
__lowerCamelCase = hidden_act
__lowerCamelCase = hidden_dropout_prob
__lowerCamelCase = attention_probs_dropout_prob
__lowerCamelCase = initializer_range
__lowerCamelCase = layer_norm_eps
__lowerCamelCase = image_size
__lowerCamelCase = patch_size
__lowerCamelCase = num_channels
__lowerCamelCase = use_mask_token
__lowerCamelCase = use_absolute_position_embeddings
__lowerCamelCase = use_relative_position_bias
__lowerCamelCase = use_shared_relative_position_bias
__lowerCamelCase = layer_scale_init_value
__lowerCamelCase = drop_path_rate
__lowerCamelCase = use_mean_pooling
# decode head attributes (semantic segmentation)
__lowerCamelCase = out_indices
__lowerCamelCase = pool_scales
# auxiliary head attributes (semantic segmentation)
__lowerCamelCase = use_auxiliary_head
__lowerCamelCase = auxiliary_loss_weight
__lowerCamelCase = auxiliary_channels
__lowerCamelCase = auxiliary_num_convs
__lowerCamelCase = auxiliary_concat_input
__lowerCamelCase = semantic_loss_ignore_index
class __lowerCAmelCase ( __magic_name__ ):
"""simple docstring"""
snake_case_ = version.parse('''1.11''' )
@property
def lowercase_ ( self ) -> Mapping[str, Mapping[int, str]]:
'''simple docstring'''
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def lowercase_ ( self ) -> float:
'''simple docstring'''
return 1e-4
| 90
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_sentencepiece_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : Dict = {"configuration_mbart": ["MBART_PRETRAINED_CONFIG_ARCHIVE_MAP", "MBartConfig", "MBartOnnxConfig"]}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Tuple = ["MBartTokenizer"]
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = ["MBartTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[Any] = [
"MBART_PRETRAINED_MODEL_ARCHIVE_LIST",
"MBartForCausalLM",
"MBartForConditionalGeneration",
"MBartForQuestionAnswering",
"MBartForSequenceClassification",
"MBartModel",
"MBartPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TFMBartForConditionalGeneration",
"TFMBartModel",
"TFMBartPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"FlaxMBartForConditionalGeneration",
"FlaxMBartForQuestionAnswering",
"FlaxMBartForSequenceClassification",
"FlaxMBartModel",
"FlaxMBartPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_mbart import MBART_PRETRAINED_CONFIG_ARCHIVE_MAP, MBartConfig, MBartOnnxConfig
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart import MBartTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart_fast import MBartTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_mbart import (
MBART_PRETRAINED_MODEL_ARCHIVE_LIST,
MBartForCausalLM,
MBartForConditionalGeneration,
MBartForQuestionAnswering,
MBartForSequenceClassification,
MBartModel,
MBartPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_mbart import TFMBartForConditionalGeneration, TFMBartModel, TFMBartPreTrainedModel
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_mbart import (
FlaxMBartForConditionalGeneration,
FlaxMBartForQuestionAnswering,
FlaxMBartForSequenceClassification,
FlaxMBartModel,
FlaxMBartPreTrainedModel,
)
else:
import sys
snake_case_ : Any = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
"""simple docstring"""
import os
import re
import shutil
import sys
import tempfile
import unittest
import black
UpperCAmelCase_ : Optional[Any] = os.path.abspath(os.path.dirname(os.path.dirname(os.path.dirname(__file__))))
sys.path.append(os.path.join(git_repo_path, """utils"""))
import check_copies # noqa: E402
# This is the reference code that will be used in the tests.
# If BertLMPredictionHead is changed in modeling_bert.py, this code needs to be manually updated.
UpperCAmelCase_ : List[Any] = """ def __init__(self, config):
super().__init__()
self.transform = BertPredictionHeadTransform(config)
# The output weights are the same as the input embeddings, but there is
# an output-only bias for each token.
self.decoder = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
self.bias = nn.Parameter(torch.zeros(config.vocab_size))
# Need a link between the two variables so that the bias is correctly resized with `resize_token_embeddings`
self.decoder.bias = self.bias
def forward(self, hidden_states):
hidden_states = self.transform(hidden_states)
hidden_states = self.decoder(hidden_states)
return hidden_states
"""
class lowerCAmelCase__ ( unittest.TestCase ):
'''simple docstring'''
def _SCREAMING_SNAKE_CASE ( self : int):
'''simple docstring'''
SCREAMING_SNAKE_CASE_ : List[str] = tempfile.mkdtemp()
os.makedirs(os.path.join(self.transformer_dir , '''models/bert/'''))
SCREAMING_SNAKE_CASE_ : Union[str, Any] = self.transformer_dir
shutil.copy(
os.path.join(lowercase_ , '''src/transformers/models/bert/modeling_bert.py''') , os.path.join(self.transformer_dir , '''models/bert/modeling_bert.py''') , )
def _SCREAMING_SNAKE_CASE ( self : Dict):
'''simple docstring'''
SCREAMING_SNAKE_CASE_ : List[Any] = '''src/transformers'''
shutil.rmtree(self.transformer_dir)
def _SCREAMING_SNAKE_CASE ( self : Dict , lowercase_ : Dict , lowercase_ : str , lowercase_ : int , lowercase_ : Tuple=None):
'''simple docstring'''
SCREAMING_SNAKE_CASE_ : Union[str, Any] = comment + F'\nclass {class_name}(nn.Module):\n' + class_code
if overwrite_result is not None:
SCREAMING_SNAKE_CASE_ : List[Any] = comment + F'\nclass {class_name}(nn.Module):\n' + overwrite_result
SCREAMING_SNAKE_CASE_ : List[Any] = black.Mode(target_versions={black.TargetVersion.PYaa} , line_length=119)
SCREAMING_SNAKE_CASE_ : Optional[int] = black.format_str(lowercase_ , mode=lowercase_)
SCREAMING_SNAKE_CASE_ : Any = os.path.join(self.transformer_dir , '''new_code.py''')
with open(lowercase_ , '''w''' , newline='''\n''') as f:
f.write(lowercase_)
if overwrite_result is None:
self.assertTrue(len(check_copies.is_copy_consistent(lowercase_)) == 0)
else:
check_copies.is_copy_consistent(f.name , overwrite=lowercase_)
with open(lowercase_ , '''r''') as f:
self.assertTrue(f.read() , lowercase_)
def _SCREAMING_SNAKE_CASE ( self : Optional[Any]):
'''simple docstring'''
SCREAMING_SNAKE_CASE_ : str = check_copies.find_code_in_transformers('''models.bert.modeling_bert.BertLMPredictionHead''')
self.assertEqual(lowercase_ , lowercase_)
def _SCREAMING_SNAKE_CASE ( self : Tuple):
'''simple docstring'''
self.check_copy_consistency(
'''# Copied from transformers.models.bert.modeling_bert.BertLMPredictionHead''' , '''BertLMPredictionHead''' , REFERENCE_CODE + '''\n''' , )
# With no empty line at the end
self.check_copy_consistency(
'''# Copied from transformers.models.bert.modeling_bert.BertLMPredictionHead''' , '''BertLMPredictionHead''' , lowercase_ , )
# Copy consistency with rename
self.check_copy_consistency(
'''# Copied from transformers.models.bert.modeling_bert.BertLMPredictionHead with Bert->TestModel''' , '''TestModelLMPredictionHead''' , re.sub('''Bert''' , '''TestModel''' , lowercase_) , )
# Copy consistency with a really long name
SCREAMING_SNAKE_CASE_ : Optional[Any] = '''TestModelWithAReallyLongNameBecauseSomePeopleLikeThatForSomeReason'''
self.check_copy_consistency(
F'# Copied from transformers.models.bert.modeling_bert.BertLMPredictionHead with Bert->{long_class_name}' , F'{long_class_name}LMPredictionHead' , re.sub('''Bert''' , lowercase_ , lowercase_) , )
# Copy consistency with overwrite
self.check_copy_consistency(
'''# Copied from transformers.models.bert.modeling_bert.BertLMPredictionHead with Bert->TestModel''' , '''TestModelLMPredictionHead''' , lowercase_ , overwrite_result=re.sub('''Bert''' , '''TestModel''' , lowercase_) , )
def _SCREAMING_SNAKE_CASE ( self : List[Any]):
'''simple docstring'''
SCREAMING_SNAKE_CASE_ : Any = check_copies.LOCALIZED_READMES['''README_zh-hans.md''']
SCREAMING_SNAKE_CASE_ : Any = (
'''1. **[ALBERT](https://huggingface.co/transformers/model_doc/albert.html)** (from Google Research and the'''
''' Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for'''
''' Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong'''
''' Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.\n1.'''
''' **[DistilBERT](https://huggingface.co/transformers/model_doc/distilbert.html)** (from HuggingFace),'''
''' released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and'''
''' lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same'''
''' method has been applied to compress GPT2 into'''
''' [DistilGPT2](https://github.com/huggingface/transformers/tree/main/examples/distillation), RoBERTa into'''
''' [DistilRoBERTa](https://github.com/huggingface/transformers/tree/main/examples/distillation),'''
''' Multilingual BERT into'''
''' [DistilmBERT](https://github.com/huggingface/transformers/tree/main/examples/distillation) and a German'''
''' version of DistilBERT.\n1. **[ELECTRA](https://huggingface.co/transformers/model_doc/electra.html)**'''
''' (from Google Research/Stanford University) released with the paper [ELECTRA: Pre-training text encoders'''
''' as discriminators rather than generators](https://arxiv.org/abs/2003.10555) by Kevin Clark, Minh-Thang'''
''' Luong, Quoc V. Le, Christopher D. Manning.'''
)
SCREAMING_SNAKE_CASE_ : Tuple = (
'''1. **[ALBERT](https://huggingface.co/transformers/model_doc/albert.html)** (来自 Google Research and the'''
''' Toyota Technological Institute at Chicago) 伴随论文 [ALBERT: A Lite BERT for Self-supervised Learning of'''
''' Language Representations](https://arxiv.org/abs/1909.11942), 由 Zhenzhong Lan, Mingda Chen, Sebastian'''
''' Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut 发布。\n'''
)
SCREAMING_SNAKE_CASE_ : Optional[int] = (
'''1. **[ALBERT](https://huggingface.co/transformers/model_doc/albert.html)** (来自 Google Research and the'''
''' Toyota Technological Institute at Chicago) 伴随论文 [ALBERT: A Lite BERT for Self-supervised Learning of'''
''' Language Representations](https://arxiv.org/abs/1909.11942), 由 Zhenzhong Lan, Mingda Chen, Sebastian'''
''' Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut 发布。\n1.'''
''' **[DistilBERT](https://huggingface.co/transformers/model_doc/distilbert.html)** (来自 HuggingFace) 伴随论文'''
''' [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and'''
''' lighter](https://arxiv.org/abs/1910.01108) 由 Victor Sanh, Lysandre Debut and Thomas Wolf 发布。 The same'''
''' method has been applied to compress GPT2 into'''
''' [DistilGPT2](https://github.com/huggingface/transformers/tree/main/examples/distillation), RoBERTa into'''
''' [DistilRoBERTa](https://github.com/huggingface/transformers/tree/main/examples/distillation),'''
''' Multilingual BERT into'''
''' [DistilmBERT](https://github.com/huggingface/transformers/tree/main/examples/distillation) and a German'''
''' version of DistilBERT.\n1. **[ELECTRA](https://huggingface.co/transformers/model_doc/electra.html)** (来自'''
''' Google Research/Stanford University) 伴随论文 [ELECTRA: Pre-training text encoders as discriminators rather'''
''' than generators](https://arxiv.org/abs/2003.10555) 由 Kevin Clark, Minh-Thang Luong, Quoc V. Le,'''
''' Christopher D. Manning 发布。\n'''
)
SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ : List[str] = check_copies.convert_to_localized_md(
lowercase_ , lowercase_ , localized_readme['''format_model_list'''])
self.assertFalse(lowercase_)
self.assertEqual(lowercase_ , lowercase_)
SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ : Optional[int] = check_copies.convert_to_localized_md(
lowercase_ , lowercase_ , localized_readme['''format_model_list'''])
# Check whether the number of models is equal to README.md after conversion.
self.assertTrue(lowercase_)
SCREAMING_SNAKE_CASE_ : Dict = (
'''1. **[ALBERT](https://huggingface.co/transformers/model_doc/albert.html)** (from Google Research and the'''
''' Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for'''
''' Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong'''
''' Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.'''
)
SCREAMING_SNAKE_CASE_ : Union[str, Any] = (
'''1. **[ALBERT](https://huggingface.co/transformers/main/model_doc/albert.html)** (来自 Google Research and'''
''' the Toyota Technological Institute at Chicago) 伴随论文 [ALBERT: A Lite BERT for Self-supervised Learning of'''
''' Language Representations](https://arxiv.org/abs/1909.11942), 由 Zhenzhong Lan, Mingda Chen, Sebastian'''
''' Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut 发布。\n'''
)
SCREAMING_SNAKE_CASE_ : Optional[Any] = (
'''1. **[ALBERT](https://huggingface.co/transformers/model_doc/albert.html)** (来自 Google Research and the'''
''' Toyota Technological Institute at Chicago) 伴随论文 [ALBERT: A Lite BERT for Self-supervised Learning of'''
''' Language Representations](https://arxiv.org/abs/1909.11942), 由 Zhenzhong Lan, Mingda Chen, Sebastian'''
''' Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut 发布。\n'''
)
SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ : List[str] = check_copies.convert_to_localized_md(
lowercase_ , lowercase_ , localized_readme['''format_model_list'''])
# Check if the model link is synchronized.
self.assertEqual(lowercase_ , lowercase_)
| 91
|
import warnings
from typing import List, Optional, Union
from ...image_utils import ImageInput
from ...processing_utils import ProcessorMixin
from ...tokenization_utils_base import BatchEncoding, PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy
from ...utils import TensorType
class __snake_case ( a ):
UpperCAmelCase__ : Dict = ['''image_processor''', '''tokenizer''']
UpperCAmelCase__ : Dict = '''FlavaImageProcessor'''
UpperCAmelCase__ : Dict = ('''BertTokenizer''', '''BertTokenizerFast''')
def __init__( self : Union[str, Any] , _snake_case : List[str]=None , _snake_case : str=None , **_snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = None
if "feature_extractor" in kwargs:
warnings.warn(
'''The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor`'''
''' instead.''' , _snake_case , )
UpperCAmelCase_ = kwargs.pop('''feature_extractor''')
UpperCAmelCase_ = image_processor if image_processor is not None else feature_extractor
if image_processor is None:
raise ValueError('''You need to specify an `image_processor`.''')
if tokenizer is None:
raise ValueError('''You need to specify a `tokenizer`.''')
super().__init__(_snake_case , _snake_case)
UpperCAmelCase_ = self.image_processor
def __call__( self : List[Any] , _snake_case : Optional[ImageInput] = None , _snake_case : Optional[Union[TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]]] = None , _snake_case : bool = True , _snake_case : Union[bool, str, PaddingStrategy] = False , _snake_case : Union[bool, str, TruncationStrategy] = False , _snake_case : Optional[int] = None , _snake_case : int = 0 , _snake_case : Optional[int] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = True , _snake_case : Optional[Union[str, TensorType]] = None , **_snake_case : Any , ):
"""simple docstring"""
if text is None and images is None:
raise ValueError('''You have to specify either text or images. Both cannot be none.''')
if text is not None:
UpperCAmelCase_ = self.tokenizer(
text=_snake_case , add_special_tokens=_snake_case , padding=_snake_case , truncation=_snake_case , max_length=_snake_case , stride=_snake_case , pad_to_multiple_of=_snake_case , return_token_type_ids=_snake_case , return_attention_mask=_snake_case , return_overflowing_tokens=_snake_case , return_special_tokens_mask=_snake_case , return_offsets_mapping=_snake_case , return_length=_snake_case , verbose=_snake_case , return_tensors=_snake_case , **_snake_case , )
if images is not None:
UpperCAmelCase_ = self.image_processor(
_snake_case , return_image_mask=_snake_case , return_codebook_pixels=_snake_case , return_tensors=_snake_case , **_snake_case , )
if text is not None and images is not None:
encoding.update(_snake_case)
return encoding
elif text is not None:
return encoding
else:
return BatchEncoding(data=dict(**_snake_case) , tensor_type=_snake_case)
def lowerCamelCase ( self : Any , *_snake_case : Optional[Any] , **_snake_case : int):
"""simple docstring"""
return self.tokenizer.batch_decode(*_snake_case , **_snake_case)
def lowerCamelCase ( self : Optional[int] , *_snake_case : int , **_snake_case : Dict):
"""simple docstring"""
return self.tokenizer.decode(*_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.tokenizer.model_input_names
UpperCAmelCase_ = self.image_processor.model_input_names
return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names))
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
warnings.warn(
'''`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.''' , _snake_case , )
return self.image_processor_class
@property
def lowerCamelCase ( self : Any):
"""simple docstring"""
warnings.warn(
'''`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.''' , _snake_case , )
return self.image_processor
| 51
| 0
|
import importlib
import sys
from argparse import REMAINDER, ArgumentParser
from pathlib import Path
import torch_xla.distributed.xla_multiprocessing as xmp
def _a ( ):
__lowerCAmelCase = ArgumentParser(
description=(
"PyTorch TPU distributed training launch helper utility that will spawn up multiple distributed processes"
) )
# Optional arguments for the launch helper
parser.add_argument("--num_cores" , type=SCREAMING_SNAKE_CASE_ , default=1 , help="Number of TPU cores to use (1 or 8)." )
# positional
parser.add_argument(
"training_script" , type=SCREAMING_SNAKE_CASE_ , help=(
"The full path to the single TPU training "
"program/script to be launched in parallel, "
"followed by all the arguments for the "
"training script"
) , )
# rest from the training program
parser.add_argument("training_script_args" , nargs=SCREAMING_SNAKE_CASE_ )
return parser.parse_args()
def _a ( ):
__lowerCAmelCase = parse_args()
# Import training_script as a module.
__lowerCAmelCase = Path(args.training_script )
sys.path.append(str(script_fpath.parent.resolve() ) )
__lowerCAmelCase = script_fpath.stem
__lowerCAmelCase = importlib.import_module(SCREAMING_SNAKE_CASE_ )
# Patch sys.argv
__lowerCAmelCase = [args.training_script] + args.training_script_args + ["--tpu_num_cores", str(args.num_cores )]
xmp.spawn(mod._mp_fn , args=() , nprocs=args.num_cores )
if __name__ == "__main__":
main()
| 92
|
from diffusers.utils.testing_utils import require_onnxruntime
@require_onnxruntime
class __snake_case :
pass
| 51
| 0
|
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tokenizers_available, is_torch_available
_lowercase : List[Any] = {
"configuration_roc_bert": ["ROC_BERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "RoCBertConfig"],
"tokenization_roc_bert": ["RoCBertTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
pass
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_lowercase : Union[str, Any] = [
"ROC_BERT_PRETRAINED_MODEL_ARCHIVE_LIST",
"RoCBertForCausalLM",
"RoCBertForMaskedLM",
"RoCBertForMultipleChoice",
"RoCBertForPreTraining",
"RoCBertForQuestionAnswering",
"RoCBertForSequenceClassification",
"RoCBertForTokenClassification",
"RoCBertLayer",
"RoCBertModel",
"RoCBertPreTrainedModel",
"load_tf_weights_in_roc_bert",
]
if TYPE_CHECKING:
from .configuration_roc_bert import ROC_BERT_PRETRAINED_CONFIG_ARCHIVE_MAP, RoCBertConfig
from .tokenization_roc_bert import RoCBertTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
raise OptionalDependencyNotAvailable()
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_roc_bert import (
ROC_BERT_PRETRAINED_MODEL_ARCHIVE_LIST,
RoCBertForCausalLM,
RoCBertForMaskedLM,
RoCBertForMultipleChoice,
RoCBertForPreTraining,
RoCBertForQuestionAnswering,
RoCBertForSequenceClassification,
RoCBertForTokenClassification,
RoCBertLayer,
RoCBertModel,
RoCBertPreTrainedModel,
load_tf_weights_in_roc_bert,
)
else:
import sys
_lowercase : str = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 93
|
import argparse
import os
import pickle
import sys
import torch
from transformers import TransfoXLConfig, TransfoXLLMHeadModel, load_tf_weights_in_transfo_xl
from transformers.models.transfo_xl import tokenization_transfo_xl as data_utils
from transformers.models.transfo_xl.tokenization_transfo_xl import CORPUS_NAME, VOCAB_FILES_NAMES
from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging
logging.set_verbosity_info()
# We do this to be able to load python 2 datasets pickles
# See e.g. https://stackoverflow.com/questions/2121874/python-pickling-after-changing-a-modules-directory/2121918#2121918
snake_case_ : List[Any] = data_utils.TransfoXLTokenizer
snake_case_ : int = data_utils.TransfoXLCorpus
snake_case_ : List[Any] = data_utils
snake_case_ : int = data_utils
def A (__A : Dict , __A : List[Any] , __A : Union[str, Any] , __A : Tuple ) -> Union[str, Any]:
"""simple docstring"""
if transfo_xl_dataset_file:
# Convert a pre-processed corpus (see original TensorFlow repo)
with open(__A , '''rb''' ) as fp:
UpperCAmelCase_ = pickle.load(__A , encoding='''latin1''' )
# Save vocabulary and dataset cache as Dictionaries (should be better than pickles for the long-term)
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + VOCAB_FILES_NAMES['''pretrained_vocab_file''']
print(F"""Save vocabulary to {pytorch_vocab_dump_path}""" )
UpperCAmelCase_ = corpus.vocab.__dict__
torch.save(__A , __A )
UpperCAmelCase_ = corpus.__dict__
corpus_dict_no_vocab.pop('''vocab''' , __A )
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + CORPUS_NAME
print(F"""Save dataset to {pytorch_dataset_dump_path}""" )
torch.save(__A , __A )
if tf_checkpoint_path:
# Convert a pre-trained TensorFlow model
UpperCAmelCase_ = os.path.abspath(__A )
UpperCAmelCase_ = os.path.abspath(__A )
print(F"""Converting Transformer XL checkpoint from {tf_path} with config at {config_path}.""" )
# Initialise PyTorch model
if transfo_xl_config_file == "":
UpperCAmelCase_ = TransfoXLConfig()
else:
UpperCAmelCase_ = TransfoXLConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = TransfoXLLMHeadModel(__A )
UpperCAmelCase_ = load_tf_weights_in_transfo_xl(__A , __A , __A )
# Save pytorch-model
UpperCAmelCase_ = os.path.join(__A , __A )
UpperCAmelCase_ = os.path.join(__A , __A )
print(F"""Save PyTorch model to {os.path.abspath(__A )}""" )
torch.save(model.state_dict() , __A )
print(F"""Save configuration file to {os.path.abspath(__A )}""" )
with open(__A , '''w''' , encoding='''utf-8''' ) as f:
f.write(config.to_json_string() )
if __name__ == "__main__":
snake_case_ : List[str] = argparse.ArgumentParser()
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
type=str,
required=True,
help="Path to the folder to store the PyTorch model or dataset/vocab.",
)
parser.add_argument(
"--tf_checkpoint_path",
default="",
type=str,
help="An optional path to a TensorFlow checkpoint path to be converted.",
)
parser.add_argument(
"--transfo_xl_config_file",
default="",
type=str,
help=(
"An optional config json file corresponding to the pre-trained BERT model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--transfo_xl_dataset_file",
default="",
type=str,
help="An optional dataset file to be converted in a vocabulary.",
)
snake_case_ : int = parser.parse_args()
convert_transfo_xl_checkpoint_to_pytorch(
args.tf_checkpoint_path,
args.transfo_xl_config_file,
args.pytorch_dump_folder_path,
args.transfo_xl_dataset_file,
)
| 51
| 0
|
import gc
import random
import tempfile
import unittest
import numpy as np
import torch
from PIL import Image
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
DDIMInverseScheduler,
DDIMScheduler,
DPMSolverMultistepInverseScheduler,
DPMSolverMultistepScheduler,
StableDiffusionDiffEditPipeline,
UNetaDConditionModel,
)
from diffusers.utils import load_image, slow
from diffusers.utils.testing_utils import enable_full_determinism, floats_tensor, require_torch_gpu, torch_device
from ..pipeline_params import TEXT_GUIDED_IMAGE_INPAINTING_BATCH_PARAMS, TEXT_GUIDED_IMAGE_INPAINTING_PARAMS
from ..test_pipelines_common import PipelineLatentTesterMixin, PipelineTesterMixin
enable_full_determinism()
class _snake_case ( _snake_case , _snake_case , unittest.TestCase ):
SCREAMING_SNAKE_CASE__ = StableDiffusionDiffEditPipeline
SCREAMING_SNAKE_CASE__ = TEXT_GUIDED_IMAGE_INPAINTING_PARAMS - {'height', 'width', 'image'} | {'image_latents'}
SCREAMING_SNAKE_CASE__ = TEXT_GUIDED_IMAGE_INPAINTING_BATCH_PARAMS - {'image'} | {'image_latents'}
SCREAMING_SNAKE_CASE__ = frozenset(
[] ) # TO-DO: update image_params once pipeline is refactored with VaeImageProcessor.preprocess
SCREAMING_SNAKE_CASE__ = frozenset([] )
def SCREAMING_SNAKE_CASE__ ( self ):
torch.manual_seed(0 )
a :Dict = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , attention_head_dim=(2, 4) , use_linear_projection=_lowerCamelCase , )
a :int = DDIMScheduler(
beta_start=0.0_0085 , beta_end=0.012 , beta_schedule='''scaled_linear''' , clip_sample=_lowerCamelCase , set_alpha_to_one=_lowerCamelCase , )
a :Tuple = DDIMInverseScheduler(
beta_start=0.0_0085 , beta_end=0.012 , beta_schedule='''scaled_linear''' , clip_sample=_lowerCamelCase , set_alpha_to_zero=_lowerCamelCase , )
torch.manual_seed(0 )
a :Dict = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , sample_size=128 , )
torch.manual_seed(0 )
a :int = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , hidden_act='''gelu''' , projection_dim=512 , )
a :Optional[int] = CLIPTextModel(_lowerCamelCase )
a :str = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''' )
a :Optional[Any] = {
'''unet''': unet,
'''scheduler''': scheduler,
'''inverse_scheduler''': inverse_scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def SCREAMING_SNAKE_CASE__ ( self , _lowerCamelCase , _lowerCamelCase=0 ):
a :Optional[Any] = floats_tensor((1, 16, 16) , rng=random.Random(_lowerCamelCase ) ).to(_lowerCamelCase )
a :Tuple = floats_tensor((1, 2, 4, 16, 16) , rng=random.Random(_lowerCamelCase ) ).to(_lowerCamelCase )
if str(_lowerCamelCase ).startswith('''mps''' ):
a :int = torch.manual_seed(_lowerCamelCase )
else:
a :int = torch.Generator(device=_lowerCamelCase ).manual_seed(_lowerCamelCase )
a :Dict = {
'''prompt''': '''a dog and a newt''',
'''mask_image''': mask,
'''image_latents''': latents,
'''generator''': generator,
'''num_inference_steps''': 2,
'''inpaint_strength''': 1.0,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
}
return inputs
def SCREAMING_SNAKE_CASE__ ( self , _lowerCamelCase , _lowerCamelCase=0 ):
a :str = floats_tensor((1, 3, 32, 32) , rng=random.Random(_lowerCamelCase ) ).to(_lowerCamelCase )
a :str = image.cpu().permute(0 , 2 , 3 , 1 )[0]
a :Optional[Any] = Image.fromarray(np.uinta(_lowerCamelCase ) ).convert('''RGB''' )
if str(_lowerCamelCase ).startswith('''mps''' ):
a :Dict = torch.manual_seed(_lowerCamelCase )
else:
a :Any = torch.Generator(device=_lowerCamelCase ).manual_seed(_lowerCamelCase )
a :Union[str, Any] = {
'''image''': image,
'''source_prompt''': '''a cat and a frog''',
'''target_prompt''': '''a dog and a newt''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''num_maps_per_mask''': 2,
'''mask_encode_strength''': 1.0,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
}
return inputs
def SCREAMING_SNAKE_CASE__ ( self , _lowerCamelCase , _lowerCamelCase=0 ):
a :List[str] = floats_tensor((1, 3, 32, 32) , rng=random.Random(_lowerCamelCase ) ).to(_lowerCamelCase )
a :int = image.cpu().permute(0 , 2 , 3 , 1 )[0]
a :Optional[Any] = Image.fromarray(np.uinta(_lowerCamelCase ) ).convert('''RGB''' )
if str(_lowerCamelCase ).startswith('''mps''' ):
a :Dict = torch.manual_seed(_lowerCamelCase )
else:
a :Any = torch.Generator(device=_lowerCamelCase ).manual_seed(_lowerCamelCase )
a :List[Any] = {
'''image''': image,
'''prompt''': '''a cat and a frog''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''inpaint_strength''': 1.0,
'''guidance_scale''': 6.0,
'''decode_latents''': True,
'''output_type''': '''numpy''',
}
return inputs
def SCREAMING_SNAKE_CASE__ ( self ):
if not hasattr(self.pipeline_class , '''_optional_components''' ):
return
a :Any = self.get_dummy_components()
a :str = self.pipeline_class(**_lowerCamelCase )
pipe.to(_lowerCamelCase )
pipe.set_progress_bar_config(disable=_lowerCamelCase )
# set all optional components to None and update pipeline config accordingly
for optional_component in pipe._optional_components:
setattr(_lowerCamelCase , _lowerCamelCase , _lowerCamelCase )
pipe.register_modules(**{optional_component: None for optional_component in pipe._optional_components} )
a :Union[str, Any] = self.get_dummy_inputs(_lowerCamelCase )
a :List[Any] = pipe(**_lowerCamelCase )[0]
with tempfile.TemporaryDirectory() as tmpdir:
pipe.save_pretrained(_lowerCamelCase )
a :int = self.pipeline_class.from_pretrained(_lowerCamelCase )
pipe_loaded.to(_lowerCamelCase )
pipe_loaded.set_progress_bar_config(disable=_lowerCamelCase )
for optional_component in pipe._optional_components:
self.assertTrue(
getattr(_lowerCamelCase , _lowerCamelCase ) is None , F'''`{optional_component}` did not stay set to None after loading.''' , )
a :Dict = self.get_dummy_inputs(_lowerCamelCase )
a :Tuple = pipe_loaded(**_lowerCamelCase )[0]
a :List[str] = np.abs(output - output_loaded ).max()
self.assertLess(_lowerCamelCase , 1e-4 )
def SCREAMING_SNAKE_CASE__ ( self ):
a :Any = '''cpu'''
a :Optional[int] = self.get_dummy_components()
a :List[str] = self.pipeline_class(**_lowerCamelCase )
pipe.to(_lowerCamelCase )
pipe.set_progress_bar_config(disable=_lowerCamelCase )
a :Any = self.get_dummy_mask_inputs(_lowerCamelCase )
a :str = pipe.generate_mask(**_lowerCamelCase )
a :List[str] = mask[0, -3:, -3:]
self.assertEqual(mask.shape , (1, 16, 16) )
a :List[Any] = np.array([0] * 9 )
a :str = np.abs(mask_slice.flatten() - expected_slice ).max()
self.assertLessEqual(_lowerCamelCase , 1e-3 )
self.assertEqual(mask[0, -3, -4] , 0 )
def SCREAMING_SNAKE_CASE__ ( self ):
a :Any = '''cpu'''
a :List[str] = self.get_dummy_components()
a :List[Any] = self.pipeline_class(**_lowerCamelCase )
pipe.to(_lowerCamelCase )
pipe.set_progress_bar_config(disable=_lowerCamelCase )
a :str = self.get_dummy_inversion_inputs(_lowerCamelCase )
a :Any = pipe.invert(**_lowerCamelCase ).images
a :Union[str, Any] = image[0, -1, -3:, -3:]
self.assertEqual(image.shape , (2, 32, 32, 3) )
a :str = np.array(
[0.5150, 0.5134, 0.5043, 0.5376, 0.4694, 0.5_1050, 0.5015, 0.4407, 0.4799] , )
a :str = np.abs(image_slice.flatten() - expected_slice ).max()
self.assertLessEqual(_lowerCamelCase , 1e-3 )
def SCREAMING_SNAKE_CASE__ ( self ):
super().test_inference_batch_single_identical(expected_max_diff=5e-3 )
def SCREAMING_SNAKE_CASE__ ( self ):
a :str = '''cpu'''
a :str = self.get_dummy_components()
a :Union[str, Any] = {'''beta_start''': 0.0_0085, '''beta_end''': 0.012, '''beta_schedule''': '''scaled_linear'''}
a :Dict = DPMSolverMultistepScheduler(**_lowerCamelCase )
a :List[str] = DPMSolverMultistepInverseScheduler(**_lowerCamelCase )
a :int = self.pipeline_class(**_lowerCamelCase )
pipe.to(_lowerCamelCase )
pipe.set_progress_bar_config(disable=_lowerCamelCase )
a :str = self.get_dummy_inversion_inputs(_lowerCamelCase )
a :Tuple = pipe.invert(**_lowerCamelCase ).images
a :Any = image[0, -1, -3:, -3:]
self.assertEqual(image.shape , (2, 32, 32, 3) )
a :int = np.array(
[0.5150, 0.5134, 0.5043, 0.5376, 0.4694, 0.5_1050, 0.5015, 0.4407, 0.4799] , )
a :Tuple = np.abs(image_slice.flatten() - expected_slice ).max()
self.assertLessEqual(_lowerCamelCase , 1e-3 )
@require_torch_gpu
@slow
class _snake_case ( unittest.TestCase ):
def SCREAMING_SNAKE_CASE__ ( self ):
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
@classmethod
def SCREAMING_SNAKE_CASE__ ( cls ):
a :Tuple = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/diffedit/fruit.png''' )
a :Union[str, Any] = raw_image.convert('''RGB''' ).resize((768, 768) )
a :List[Any] = raw_image
def SCREAMING_SNAKE_CASE__ ( self ):
a :Tuple = torch.manual_seed(0 )
a :int = StableDiffusionDiffEditPipeline.from_pretrained(
'''stabilityai/stable-diffusion-2-1''' , safety_checker=_lowerCamelCase , torch_dtype=torch.floataa )
a :Optional[int] = DDIMScheduler.from_config(pipe.scheduler.config )
a :Optional[int] = DDIMInverseScheduler.from_config(pipe.scheduler.config )
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=_lowerCamelCase )
a :Optional[Any] = '''a bowl of fruit'''
a :Any = '''a bowl of pears'''
a :Any = pipe.generate_mask(
image=self.raw_image , source_prompt=_lowerCamelCase , target_prompt=_lowerCamelCase , generator=_lowerCamelCase , )
a :Dict = pipe.invert(
prompt=_lowerCamelCase , image=self.raw_image , inpaint_strength=0.7 , generator=_lowerCamelCase ).latents
a :List[str] = pipe(
prompt=_lowerCamelCase , mask_image=_lowerCamelCase , image_latents=_lowerCamelCase , generator=_lowerCamelCase , negative_prompt=_lowerCamelCase , inpaint_strength=0.7 , output_type='''numpy''' , ).images[0]
a :List[str] = (
np.array(
load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main'''
'''/diffedit/pears.png''' ).resize((768, 768) ) )
/ 255
)
assert np.abs((expected_image - image).max() ) < 5e-1
def SCREAMING_SNAKE_CASE__ ( self ):
a :Optional[int] = torch.manual_seed(0 )
a :List[Any] = StableDiffusionDiffEditPipeline.from_pretrained(
'''stabilityai/stable-diffusion-2-1''' , safety_checker=_lowerCamelCase , torch_dtype=torch.floataa )
a :Any = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config )
a :Union[str, Any] = DPMSolverMultistepInverseScheduler.from_config(pipe.scheduler.config )
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=_lowerCamelCase )
a :Dict = '''a bowl of fruit'''
a :Optional[Any] = '''a bowl of pears'''
a :Tuple = pipe.generate_mask(
image=self.raw_image , source_prompt=_lowerCamelCase , target_prompt=_lowerCamelCase , generator=_lowerCamelCase , )
a :Dict = pipe.invert(
prompt=_lowerCamelCase , image=self.raw_image , inpaint_strength=0.7 , generator=_lowerCamelCase , num_inference_steps=25 , ).latents
a :str = pipe(
prompt=_lowerCamelCase , mask_image=_lowerCamelCase , image_latents=_lowerCamelCase , generator=_lowerCamelCase , negative_prompt=_lowerCamelCase , inpaint_strength=0.7 , num_inference_steps=25 , output_type='''numpy''' , ).images[0]
a :List[Any] = (
np.array(
load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main'''
'''/diffedit/pears.png''' ).resize((768, 768) ) )
/ 255
)
assert np.abs((expected_image - image).max() ) < 5e-1
| 94
|
from typing import Optional, Tuple, Union
import torch
from einops import rearrange, reduce
from diffusers import DDIMScheduler, DDPMScheduler, DiffusionPipeline, ImagePipelineOutput, UNetaDConditionModel
from diffusers.schedulers.scheduling_ddim import DDIMSchedulerOutput
from diffusers.schedulers.scheduling_ddpm import DDPMSchedulerOutput
snake_case_ : List[str] = 8
def A (__A : Union[str, Any] , __A : List[Any]=BITS ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x * 255).int().clamp(0 , 255 )
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b c h w -> b c 1 h w''' )
UpperCAmelCase_ = ((x & mask) != 0).float()
UpperCAmelCase_ = rearrange(__A , '''b c d h w -> b (c d) h w''' )
UpperCAmelCase_ = bits * 2 - 1
return bits
def A (__A : Dict , __A : Tuple=BITS ) -> List[str]:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x > 0).int()
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A , dtype=torch.intaa )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b (c d) h w -> b c d h w''' , d=8 )
UpperCAmelCase_ = reduce(x * mask , '''b c d h w -> b c h w''' , '''sum''' )
return (dec / 255).clamp(0.0 , 1.0 )
def A (self : List[Any] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : float = 0.0 , __A : bool = True , __A : Tuple=None , __A : bool = True , ) -> Union[DDIMSchedulerOutput, Tuple]:
"""simple docstring"""
if self.num_inference_steps is None:
raise ValueError(
'''Number of inference steps is \'None\', you need to run \'set_timesteps\' after creating the scheduler''' )
# See formulas (12) and (16) of DDIM paper https://arxiv.org/pdf/2010.02502.pdf
# Ideally, read DDIM paper in-detail understanding
# Notation (<variable name> -> <name in paper>
# - pred_noise_t -> e_theta(x_t, t)
# - pred_original_sample -> f_theta(x_t, t) or x_0
# - std_dev_t -> sigma_t
# - eta -> η
# - pred_sample_direction -> "direction pointing to x_t"
# - pred_prev_sample -> "x_t-1"
# 1. get previous step value (=t-1)
UpperCAmelCase_ = timestep - self.config.num_train_timesteps // self.num_inference_steps
# 2. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[timestep]
UpperCAmelCase_ = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
UpperCAmelCase_ = 1 - alpha_prod_t
# 3. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
# 4. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 5. compute variance: "sigma_t(η)" -> see formula (16)
# σ_t = sqrt((1 − α_t−1)/(1 − α_t)) * sqrt(1 − α_t/α_t−1)
UpperCAmelCase_ = self._get_variance(__A , __A )
UpperCAmelCase_ = eta * variance ** 0.5
if use_clipped_model_output:
# the model_output is always re-derived from the clipped x_0 in Glide
UpperCAmelCase_ = (sample - alpha_prod_t ** 0.5 * pred_original_sample) / beta_prod_t ** 0.5
# 6. compute "direction pointing to x_t" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (1 - alpha_prod_t_prev - std_dev_t**2) ** 0.5 * model_output
# 7. compute x_t without "random noise" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = alpha_prod_t_prev ** 0.5 * pred_original_sample + pred_sample_direction
if eta > 0:
# randn_like does not support generator https://github.com/pytorch/pytorch/issues/27072
UpperCAmelCase_ = model_output.device if torch.is_tensor(__A ) else '''cpu'''
UpperCAmelCase_ = torch.randn(model_output.shape , dtype=model_output.dtype , generator=__A ).to(__A )
UpperCAmelCase_ = self._get_variance(__A , __A ) ** 0.5 * eta * noise
UpperCAmelCase_ = prev_sample + variance
if not return_dict:
return (prev_sample,)
return DDIMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
def A (self : Optional[int] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : int="epsilon" , __A : Optional[Any]=None , __A : bool = True , ) -> Union[DDPMSchedulerOutput, Tuple]:
"""simple docstring"""
UpperCAmelCase_ = timestep
if model_output.shape[1] == sample.shape[1] * 2 and self.variance_type in ["learned", "learned_range"]:
UpperCAmelCase_ , UpperCAmelCase_ = torch.split(__A , sample.shape[1] , dim=1 )
else:
UpperCAmelCase_ = None
# 1. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[t]
UpperCAmelCase_ = self.alphas_cumprod[t - 1] if t > 0 else self.one
UpperCAmelCase_ = 1 - alpha_prod_t
UpperCAmelCase_ = 1 - alpha_prod_t_prev
# 2. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
if prediction_type == "epsilon":
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
elif prediction_type == "sample":
UpperCAmelCase_ = model_output
else:
raise ValueError(F"""Unsupported prediction_type {prediction_type}.""" )
# 3. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = (alpha_prod_t_prev ** 0.5 * self.betas[t]) / beta_prod_t
UpperCAmelCase_ = self.alphas[t] ** 0.5 * beta_prod_t_prev / beta_prod_t
# 5. Compute predicted previous sample µ_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * sample
# 6. Add noise
UpperCAmelCase_ = 0
if t > 0:
UpperCAmelCase_ = torch.randn(
model_output.size() , dtype=model_output.dtype , layout=model_output.layout , generator=__A ).to(model_output.device )
UpperCAmelCase_ = (self._get_variance(__A , predicted_variance=__A ) ** 0.5) * noise
UpperCAmelCase_ = pred_prev_sample + variance
if not return_dict:
return (pred_prev_sample,)
return DDPMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
class __snake_case ( a ):
def __init__( self : Union[str, Any] , _snake_case : UNetaDConditionModel , _snake_case : Union[DDIMScheduler, DDPMScheduler] , _snake_case : Optional[float] = 1.0 , ):
"""simple docstring"""
super().__init__()
UpperCAmelCase_ = bit_scale
UpperCAmelCase_ = (
ddim_bit_scheduler_step if isinstance(_snake_case , _snake_case) else ddpm_bit_scheduler_step
)
self.register_modules(unet=_snake_case , scheduler=_snake_case)
@torch.no_grad()
def __call__( self : Union[str, Any] , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 50 , _snake_case : Optional[torch.Generator] = None , _snake_case : Optional[int] = 1 , _snake_case : Optional[str] = "pil" , _snake_case : bool = True , **_snake_case : Optional[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = torch.randn(
(batch_size, self.unet.config.in_channels, height, width) , generator=_snake_case , )
UpperCAmelCase_ = decimal_to_bits(_snake_case) * self.bit_scale
UpperCAmelCase_ = latents.to(self.device)
self.scheduler.set_timesteps(_snake_case)
for t in self.progress_bar(self.scheduler.timesteps):
# predict the noise residual
UpperCAmelCase_ = self.unet(_snake_case , _snake_case).sample
# compute the previous noisy sample x_t -> x_t-1
UpperCAmelCase_ = self.scheduler.step(_snake_case , _snake_case , _snake_case).prev_sample
UpperCAmelCase_ = bits_to_decimal(_snake_case)
if output_type == "pil":
UpperCAmelCase_ = self.numpy_to_pil(_snake_case)
if not return_dict:
return (image,)
return ImagePipelineOutput(images=_snake_case)
| 51
| 0
|
import argparse
import os
import evaluate
import torch
from datasets import load_dataset
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate,
# specifically showcasing how to properly calculate the metrics on the
# validation dataset when in a distributed system, and builds off the
# `nlp_example.py` script.
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To help focus on the differences in the code, building `DataLoaders`
# was refactored into its own function.
# New additions from the base script can be found quickly by
# looking for the # New Code # tags
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
UpperCAmelCase : Any = 16
UpperCAmelCase : str = 32
def _A ( SCREAMING_SNAKE_CASE : Accelerator , SCREAMING_SNAKE_CASE : int = 16 ):
"""simple docstring"""
a__ : int =AutoTokenizer.from_pretrained("bert-base-cased" )
a__ : List[str] =load_dataset("glue" , "mrpc" )
def tokenize_function(SCREAMING_SNAKE_CASE : List[Any] ):
# max_length=None => use the model max length (it's actually the default)
a__ : int =tokenizer(examples["sentence1"] , examples["sentence2"] , truncation=SCREAMING_SNAKE_CASE , max_length=SCREAMING_SNAKE_CASE )
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
a__ : Dict =datasets.map(
SCREAMING_SNAKE_CASE , batched=SCREAMING_SNAKE_CASE , remove_columns=["idx", "sentence1", "sentence2"] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
a__ : Dict =tokenized_datasets.rename_column("label" , "labels" )
def collate_fn(SCREAMING_SNAKE_CASE : str ):
# On TPU it's best to pad everything to the same length or training will be very slow.
a__ : Optional[Any] =128 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
a__ : str =16
elif accelerator.mixed_precision != "no":
a__ : Union[str, Any] =8
else:
a__ : List[str] =None
return tokenizer.pad(
SCREAMING_SNAKE_CASE , padding="longest" , max_length=SCREAMING_SNAKE_CASE , pad_to_multiple_of=SCREAMING_SNAKE_CASE , return_tensors="pt" , )
# Instantiate dataloaders.
a__ : Any =DataLoader(
tokenized_datasets["train"] , shuffle=SCREAMING_SNAKE_CASE , collate_fn=SCREAMING_SNAKE_CASE , batch_size=SCREAMING_SNAKE_CASE )
a__ : int =DataLoader(
tokenized_datasets["validation"] , shuffle=SCREAMING_SNAKE_CASE , collate_fn=SCREAMING_SNAKE_CASE , batch_size=SCREAMING_SNAKE_CASE )
return train_dataloader, eval_dataloader
# For testing only
if os.environ.get("""TESTING_MOCKED_DATALOADERS""", None) == "1":
from accelerate.test_utils.training import mocked_dataloaders
UpperCAmelCase : str = mocked_dataloaders # noqa: F811
def _A ( SCREAMING_SNAKE_CASE : List[Any] , SCREAMING_SNAKE_CASE : str ):
"""simple docstring"""
if os.environ.get("TESTING_MOCKED_DATALOADERS" , SCREAMING_SNAKE_CASE ) == "1":
a__ : Tuple =2
# Initialize accelerator
a__ : int =Accelerator(cpu=args.cpu , mixed_precision=args.mixed_precision )
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
a__ : Optional[int] =config["lr"]
a__ : Union[str, Any] =int(config["num_epochs"] )
a__ : Any =int(config["seed"] )
a__ : Dict =int(config["batch_size"] )
a__ : int =evaluate.load("glue" , "mrpc" )
# If the batch size is too big we use gradient accumulation
a__ : int =1
if batch_size > MAX_GPU_BATCH_SIZE and accelerator.distributed_type != DistributedType.TPU:
a__ : Dict =batch_size // MAX_GPU_BATCH_SIZE
a__ : Tuple =MAX_GPU_BATCH_SIZE
set_seed(SCREAMING_SNAKE_CASE )
a__ , a__ : Optional[int] =get_dataloaders(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE )
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
a__ : List[str] =AutoModelForSequenceClassification.from_pretrained("bert-base-cased" , return_dict=SCREAMING_SNAKE_CASE )
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
a__ : List[str] =model.to(accelerator.device )
# Instantiate optimizer
a__ : List[Any] =AdamW(params=model.parameters() , lr=SCREAMING_SNAKE_CASE )
# Instantiate scheduler
a__ : Optional[int] =get_linear_schedule_with_warmup(
optimizer=SCREAMING_SNAKE_CASE , num_warmup_steps=100 , num_training_steps=(len(SCREAMING_SNAKE_CASE ) * num_epochs) // gradient_accumulation_steps , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
a__ , a__ , a__ , a__ , a__ : Optional[int] =accelerator.prepare(
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE )
# Now we train the model
for epoch in range(SCREAMING_SNAKE_CASE ):
model.train()
for step, batch in enumerate(SCREAMING_SNAKE_CASE ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
a__ : Dict =model(**SCREAMING_SNAKE_CASE )
a__ : List[Any] =outputs.loss
a__ : List[str] =loss / gradient_accumulation_steps
accelerator.backward(SCREAMING_SNAKE_CASE )
if step % gradient_accumulation_steps == 0:
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
a__ : Optional[Any] =0
for step, batch in enumerate(SCREAMING_SNAKE_CASE ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
with torch.no_grad():
a__ : Any =model(**SCREAMING_SNAKE_CASE )
a__ : str =outputs.logits.argmax(dim=-1 )
a__ , a__ : List[str] =accelerator.gather((predictions, batch["labels"]) )
# New Code #
# First we check if it's a distributed system
if accelerator.use_distributed:
# Then see if we're on the last batch of our eval dataloader
if step == len(SCREAMING_SNAKE_CASE ) - 1:
# Last batch needs to be truncated on distributed systems as it contains additional samples
a__ : Optional[Any] =predictions[: len(eval_dataloader.dataset ) - samples_seen]
a__ : Dict =references[: len(eval_dataloader.dataset ) - samples_seen]
else:
# Otherwise we add the number of samples seen
samples_seen += references.shape[0]
# All of this can be avoided if you use `Accelerator.gather_for_metrics` instead of `Accelerator.gather`:
# accelerator.gather_for_metrics((predictions, batch["labels"]))
metric.add_batch(
predictions=SCREAMING_SNAKE_CASE , references=SCREAMING_SNAKE_CASE , )
a__ : Tuple =metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(f'''epoch {epoch}:''' , SCREAMING_SNAKE_CASE )
def _A ( ):
"""simple docstring"""
a__ : List[str] =argparse.ArgumentParser(description="Simple example of training script." )
parser.add_argument(
"--mixed_precision" , type=SCREAMING_SNAKE_CASE , default=SCREAMING_SNAKE_CASE , choices=["no", "fp16", "bf16", "fp8"] , help="Whether to use mixed precision. Choose"
"between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10."
"and an Nvidia Ampere GPU." , )
parser.add_argument("--cpu" , action="store_true" , help="If passed, will train on the CPU." )
a__ : str =parser.parse_args()
a__ : Optional[int] ={"lr": 2e-5, "num_epochs": 3, "seed": 42, "batch_size": 16}
training_function(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
main()
| 95
|
snake_case_ : Dict = {
"Pillow": "Pillow",
"accelerate": "accelerate>=0.11.0",
"compel": "compel==0.1.8",
"black": "black~=23.1",
"datasets": "datasets",
"filelock": "filelock",
"flax": "flax>=0.4.1",
"hf-doc-builder": "hf-doc-builder>=0.3.0",
"huggingface-hub": "huggingface-hub>=0.13.2",
"requests-mock": "requests-mock==1.10.0",
"importlib_metadata": "importlib_metadata",
"invisible-watermark": "invisible-watermark",
"isort": "isort>=5.5.4",
"jax": "jax>=0.2.8,!=0.3.2",
"jaxlib": "jaxlib>=0.1.65",
"Jinja2": "Jinja2",
"k-diffusion": "k-diffusion>=0.0.12",
"torchsde": "torchsde",
"note_seq": "note_seq",
"librosa": "librosa",
"numpy": "numpy",
"omegaconf": "omegaconf",
"parameterized": "parameterized",
"protobuf": "protobuf>=3.20.3,<4",
"pytest": "pytest",
"pytest-timeout": "pytest-timeout",
"pytest-xdist": "pytest-xdist",
"ruff": "ruff>=0.0.241",
"safetensors": "safetensors",
"sentencepiece": "sentencepiece>=0.1.91,!=0.1.92",
"scipy": "scipy",
"onnx": "onnx",
"regex": "regex!=2019.12.17",
"requests": "requests",
"tensorboard": "tensorboard",
"torch": "torch>=1.4",
"torchvision": "torchvision",
"transformers": "transformers>=4.25.1",
"urllib3": "urllib3<=2.0.0",
}
| 51
| 0
|
"""simple docstring"""
import numpy as np
import torch
import torch.nn as nn
from transformers import CLIPConfig, CLIPVisionModelWithProjection, PreTrainedModel
from ...utils import logging
lowercase__ = logging.get_logger(__name__)
class lowerCAmelCase__ ( lowercase ):
'''simple docstring'''
lowerCamelCase__ = CLIPConfig
lowerCamelCase__ = ["""CLIPEncoderLayer"""]
def __init__( self , lowercase ):
super().__init__(lowercase )
_lowerCamelCase : List[str] = CLIPVisionModelWithProjection(config.vision_config )
_lowerCamelCase : List[Any] = nn.Linear(config.vision_config.projection_dim , 1 )
_lowerCamelCase : Dict = nn.Linear(config.vision_config.projection_dim , 1 )
@torch.no_grad()
def A_ ( self , lowercase , lowercase , lowercase=0.5 , lowercase=0.5 ):
_lowerCamelCase : Dict = self.vision_model(lowercase )[0]
_lowerCamelCase : Any = self.p_head(lowercase )
_lowerCamelCase : str = nsfw_detected.flatten()
_lowerCamelCase : int = nsfw_detected > p_threshold
_lowerCamelCase : Union[str, Any] = nsfw_detected.tolist()
if any(lowercase ):
logger.warning(
'Potential NSFW content was detected in one or more images. A black image will be returned instead.'
' Try again with a different prompt and/or seed.' )
for idx, nsfw_detected_ in enumerate(lowercase ):
if nsfw_detected_:
_lowerCamelCase : Dict = np.zeros(images[idx].shape )
_lowerCamelCase : Dict = self.w_head(lowercase )
_lowerCamelCase : Tuple = watermark_detected.flatten()
_lowerCamelCase : Dict = watermark_detected > w_threshold
_lowerCamelCase : Dict = watermark_detected.tolist()
if any(lowercase ):
logger.warning(
'Potential watermarked content was detected in one or more images. A black image will be returned instead.'
' Try again with a different prompt and/or seed.' )
for idx, watermark_detected_ in enumerate(lowercase ):
if watermark_detected_:
_lowerCamelCase : List[str] = np.zeros(images[idx].shape )
return images, nsfw_detected, watermark_detected
| 96
|
from datetime import datetime
import requests
def A (__A : str ) -> bytes:
"""simple docstring"""
UpperCAmelCase_ = '''https://downloadgram.net/wp-json/wppress/video-downloader/video?url='''
UpperCAmelCase_ = requests.get(base_url + url ).json()[0]['''urls'''][0]['''src''']
return requests.get(__A ).content
if __name__ == "__main__":
snake_case_ : Optional[Any] = input("Enter Video/IGTV url: ").strip()
snake_case_ : Any = f"{datetime.now():%Y-%m-%d_%H:%M:%S}.mp4"
with open(file_name, "wb") as fp:
fp.write(download_video(url))
print(f"Done. Video saved to disk as {file_name}.")
| 51
| 0
|
'''simple docstring'''
from typing import Any
import numpy as np
def a ( __a ) -> bool:
'''simple docstring'''
return np.array_equal(__a , matrix.conjugate().T )
def a ( __a , __a ) -> Any:
'''simple docstring'''
UpperCamelCase__ :Dict = v.conjugate().T
UpperCamelCase__ :Optional[Any] = v_star.dot(__a )
assert isinstance(__a , np.ndarray )
return (v_star_dot.dot(__a )) / (v_star.dot(__a ))
def a ( ) -> None:
'''simple docstring'''
UpperCamelCase__ :Any = np.array([[2, 2 + 1J, 4], [2 - 1J, 3, 1J], [4, -1J, 1]] )
UpperCamelCase__ :Tuple = np.array([[1], [2], [3]] )
assert is_hermitian(__a ), f'''{a} is not hermitian.'''
print(rayleigh_quotient(__a , __a ) )
UpperCamelCase__ :Any = np.array([[1, 2, 4], [2, 3, -1], [4, -1, 1]] )
assert is_hermitian(__a ), f'''{a} is not hermitian.'''
assert rayleigh_quotient(__a , __a ) == float(3 )
if __name__ == "__main__":
import doctest
doctest.testmod()
tests()
| 97
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[Any] = {
"tiiuae/falcon-40b": "https://huggingface.co/tiiuae/falcon-40b/resolve/main/config.json",
"tiiuae/falcon-7b": "https://huggingface.co/tiiuae/falcon-7b/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : Optional[Any] = '''falcon'''
UpperCAmelCase__ : List[Any] = ['''past_key_values''']
def __init__( self : Union[str, Any] , _snake_case : List[str]=65024 , _snake_case : int=4544 , _snake_case : int=32 , _snake_case : Any=71 , _snake_case : int=1e-5 , _snake_case : Dict=0.0_2 , _snake_case : int=True , _snake_case : List[Any]=0.0 , _snake_case : Tuple=0.0 , _snake_case : int=None , _snake_case : Tuple=False , _snake_case : Any=False , _snake_case : str=True , _snake_case : Any=True , _snake_case : List[str]=False , _snake_case : Tuple=11 , _snake_case : Dict=11 , **_snake_case : Optional[int] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
# Backward compatibility with n_embed kwarg
UpperCAmelCase_ = kwargs.pop('''n_embed''' , _snake_case)
UpperCAmelCase_ = hidden_size if n_embed is None else n_embed
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = hidden_dropout
UpperCAmelCase_ = attention_dropout
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
UpperCAmelCase_ = num_attention_heads if num_kv_heads is None else num_kv_heads
UpperCAmelCase_ = alibi
UpperCAmelCase_ = new_decoder_architecture
UpperCAmelCase_ = multi_query # Ignored when new_decoder_architecture is True
UpperCAmelCase_ = parallel_attn
UpperCAmelCase_ = bias
super().__init__(bos_token_id=_snake_case , eos_token_id=_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
return self.hidden_size // self.num_attention_heads
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return not self.alibi
| 51
| 0
|
"""simple docstring"""
def a_ ( lowerCamelCase , lowerCamelCase ):
UpperCAmelCase__ = len(lowerCamelCase ) + 1
UpperCAmelCase__ = len(lowerCamelCase ) + 1
# dp is a 2d matrix where dp[i][j] denotes whether prefix string of
# length i of input_string matches with prefix string of length j of
# given pattern.
# "dp" stands for dynamic programming.
UpperCAmelCase__ = [[0 for i in range(lowerCamelCase )] for j in range(lowerCamelCase )]
# since string of zero length match pattern of zero length
UpperCAmelCase__ = 1
# since pattern of zero length will never match with string of non-zero length
for i in range(1 , lowerCamelCase ):
UpperCAmelCase__ = 0
# since string of zero length will match with pattern where there
# is at least one * alternatively
for j in range(1 , lowerCamelCase ):
UpperCAmelCase__ = dp[0][j - 2] if pattern[j - 1] == '*' else 0
# now using bottom-up approach to find for all remaining lengths
for i in range(1 , lowerCamelCase ):
for j in range(1 , lowerCamelCase ):
if input_string[i - 1] == pattern[j - 1] or pattern[j - 1] == ".":
UpperCAmelCase__ = dp[i - 1][j - 1]
elif pattern[j - 1] == "*":
if dp[i][j - 2] == 1:
UpperCAmelCase__ = 1
elif pattern[j - 2] in (input_string[i - 1], "."):
UpperCAmelCase__ = dp[i - 1][j]
else:
UpperCAmelCase__ = 0
else:
UpperCAmelCase__ = 0
return bool(dp[-1][-1] )
if __name__ == "__main__":
import doctest
doctest.testmod()
# inputing the strings
# input_string = input("input a string :")
# pattern = input("input a pattern :")
lowerCAmelCase__ : List[Any] = 'aab'
lowerCAmelCase__ : List[Any] = 'c*a*b'
# using function to check whether given string matches the given pattern
if match_pattern(input_string, pattern):
print(F"""{input_string} matches the given pattern {pattern}""")
else:
print(F"""{input_string} does not match with the given pattern {pattern}""")
| 98
|
from __future__ import annotations
import time
from math import sqrt
# 1 for manhattan, 0 for euclidean
snake_case_ : str = 0
snake_case_ : Union[str, Any] = [
[0, 0, 0, 0, 0, 0, 0],
[0, 1, 0, 0, 0, 0, 0], # 0 are free path whereas 1's are obstacles
[0, 0, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 0],
[1, 0, 1, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 0, 0],
]
snake_case_ : str = [[-1, 0], [0, -1], [1, 0], [0, 1]] # up, left, down, right
snake_case_ : List[Any] = tuple[int, int]
class __snake_case :
def __init__( self : Any , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : Node | None , ):
"""simple docstring"""
UpperCAmelCase_ = pos_x
UpperCAmelCase_ = pos_y
UpperCAmelCase_ = (pos_y, pos_x)
UpperCAmelCase_ = goal_x
UpperCAmelCase_ = goal_y
UpperCAmelCase_ = g_cost
UpperCAmelCase_ = parent
UpperCAmelCase_ = self.calculate_heuristic()
UpperCAmelCase_ = self.g_cost + self.h_cost
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = self.pos_x - self.goal_x
UpperCAmelCase_ = self.pos_y - self.goal_y
if HEURISTIC == 1:
return abs(_snake_case) + abs(_snake_case)
else:
return sqrt(dy**2 + dx**2)
def __lt__( self : Union[str, Any] , _snake_case : Node):
"""simple docstring"""
return self.f_cost < other.f_cost
class __snake_case :
def __init__( self : str , _snake_case : TPosition , _snake_case : TPosition):
"""simple docstring"""
UpperCAmelCase_ = Node(start[1] , start[0] , goal[1] , goal[0] , 0 , _snake_case)
UpperCAmelCase_ = Node(goal[1] , goal[0] , goal[1] , goal[0] , 99999 , _snake_case)
UpperCAmelCase_ = [self.start]
UpperCAmelCase_ = []
UpperCAmelCase_ = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
while self.open_nodes:
# Open Nodes are sorted using __lt__
self.open_nodes.sort()
UpperCAmelCase_ = self.open_nodes.pop(0)
if current_node.pos == self.target.pos:
return self.retrace_path(_snake_case)
self.closed_nodes.append(_snake_case)
UpperCAmelCase_ = self.get_successors(_snake_case)
for child_node in successors:
if child_node in self.closed_nodes:
continue
if child_node not in self.open_nodes:
self.open_nodes.append(_snake_case)
else:
# retrieve the best current path
UpperCAmelCase_ = self.open_nodes.pop(self.open_nodes.index(_snake_case))
if child_node.g_cost < better_node.g_cost:
self.open_nodes.append(_snake_case)
else:
self.open_nodes.append(_snake_case)
return [self.start.pos]
def lowerCamelCase ( self : Tuple , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = []
for action in delta:
UpperCAmelCase_ = parent.pos_x + action[1]
UpperCAmelCase_ = parent.pos_y + action[0]
if not (0 <= pos_x <= len(grid[0]) - 1 and 0 <= pos_y <= len(_snake_case) - 1):
continue
if grid[pos_y][pos_x] != 0:
continue
successors.append(
Node(
_snake_case , _snake_case , self.target.pos_y , self.target.pos_x , parent.g_cost + 1 , _snake_case , ))
return successors
def lowerCamelCase ( self : Any , _snake_case : Node | None):
"""simple docstring"""
UpperCAmelCase_ = node
UpperCAmelCase_ = []
while current_node is not None:
path.append((current_node.pos_y, current_node.pos_x))
UpperCAmelCase_ = current_node.parent
path.reverse()
return path
class __snake_case :
def __init__( self : Any , _snake_case : TPosition , _snake_case : TPosition):
"""simple docstring"""
UpperCAmelCase_ = AStar(_snake_case , _snake_case)
UpperCAmelCase_ = AStar(_snake_case , _snake_case)
UpperCAmelCase_ = False
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
while self.fwd_astar.open_nodes or self.bwd_astar.open_nodes:
self.fwd_astar.open_nodes.sort()
self.bwd_astar.open_nodes.sort()
UpperCAmelCase_ = self.fwd_astar.open_nodes.pop(0)
UpperCAmelCase_ = self.bwd_astar.open_nodes.pop(0)
if current_bwd_node.pos == current_fwd_node.pos:
return self.retrace_bidirectional_path(
_snake_case , _snake_case)
self.fwd_astar.closed_nodes.append(_snake_case)
self.bwd_astar.closed_nodes.append(_snake_case)
UpperCAmelCase_ = current_bwd_node
UpperCAmelCase_ = current_fwd_node
UpperCAmelCase_ = {
self.fwd_astar: self.fwd_astar.get_successors(_snake_case),
self.bwd_astar: self.bwd_astar.get_successors(_snake_case),
}
for astar in [self.fwd_astar, self.bwd_astar]:
for child_node in successors[astar]:
if child_node in astar.closed_nodes:
continue
if child_node not in astar.open_nodes:
astar.open_nodes.append(_snake_case)
else:
# retrieve the best current path
UpperCAmelCase_ = astar.open_nodes.pop(
astar.open_nodes.index(_snake_case))
if child_node.g_cost < better_node.g_cost:
astar.open_nodes.append(_snake_case)
else:
astar.open_nodes.append(_snake_case)
return [self.fwd_astar.start.pos]
def lowerCamelCase ( self : int , _snake_case : Node , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = self.fwd_astar.retrace_path(_snake_case)
UpperCAmelCase_ = self.bwd_astar.retrace_path(_snake_case)
bwd_path.pop()
bwd_path.reverse()
UpperCAmelCase_ = fwd_path + bwd_path
return path
if __name__ == "__main__":
# all coordinates are given in format [y,x]
snake_case_ : Any = (0, 0)
snake_case_ : Union[str, Any] = (len(grid) - 1, len(grid[0]) - 1)
for elem in grid:
print(elem)
snake_case_ : str = time.time()
snake_case_ : List[str] = AStar(init, goal)
snake_case_ : Optional[int] = a_star.search()
snake_case_ : Optional[Any] = time.time() - start_time
print(f"AStar execution time = {end_time:f} seconds")
snake_case_ : int = time.time()
snake_case_ : Dict = BidirectionalAStar(init, goal)
snake_case_ : str = time.time() - bd_start_time
print(f"BidirectionalAStar execution time = {bd_end_time:f} seconds")
| 51
| 0
|
import numpy as np
import torch
from imwatermark import WatermarkEncoder
# Copied from https://github.com/Stability-AI/generative-models/blob/613af104c6b85184091d42d374fef420eddb356d/scripts/demo/streamlit_helpers.py#L66
lowercase : List[Any] = 0B10_11_00_11_11_10_11_00_10_01_00_00_01_11_10_11_10_11_00_01_10_01_11_10
# bin(x)[2:] gives bits of x as str, use int to convert them to 0/1
lowercase : str = [int(bit) for bit in bin(WATERMARK_MESSAGE)[2:]]
class A__ :
"""simple docstring"""
def __init__( self) -> Union[str, Any]:
'''simple docstring'''
a__ : Any = WATERMARK_BITS
a__ : Any = WatermarkEncoder()
self.encoder.set_watermark('bits' , self.watermark)
def __lowercase ( self , lowercase) -> Union[str, Any]:
'''simple docstring'''
if images.shape[-1] < 256:
return images
a__ : List[Any] = (255 * (images / 2 + 0.5)).cpu().permute(0 , 2 , 3 , 1).float().numpy()
a__ : Dict = [self.encoder.encode(lowercase , 'dwtDct') for image in images]
a__ : str = torch.from_numpy(np.array(lowercase)).permute(0 , 3 , 1 , 2)
a__ : Any = torch.clamp(2 * (images / 255 - 0.5) , min=-1.0 , max=1.0)
return images
| 99
|
import inspect
import unittest
import numpy as np
from tests.test_modeling_common import floats_tensor
from transformers import DetrConfig, MaskFormerConfig, SwinConfig, is_torch_available, is_vision_available
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import MaskFormerForInstanceSegmentation, MaskFormerModel
if is_vision_available():
from transformers import MaskFormerImageProcessor
if is_vision_available():
from PIL import Image
class __snake_case :
def __init__( self : List[str] , _snake_case : Union[str, Any] , _snake_case : List[str]=2 , _snake_case : Any=True , _snake_case : Any=False , _snake_case : List[str]=10 , _snake_case : Any=3 , _snake_case : Union[str, Any]=32 * 4 , _snake_case : List[Any]=32 * 6 , _snake_case : Tuple=4 , _snake_case : Dict=32 , ):
"""simple docstring"""
UpperCAmelCase_ = parent
UpperCAmelCase_ = batch_size
UpperCAmelCase_ = is_training
UpperCAmelCase_ = use_auxiliary_loss
UpperCAmelCase_ = num_queries
UpperCAmelCase_ = num_channels
UpperCAmelCase_ = min_size
UpperCAmelCase_ = max_size
UpperCAmelCase_ = num_labels
UpperCAmelCase_ = mask_feature_size
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = floats_tensor([self.batch_size, self.num_channels, self.min_size, self.max_size]).to(
_snake_case)
UpperCAmelCase_ = torch.ones([self.batch_size, self.min_size, self.max_size] , device=_snake_case)
UpperCAmelCase_ = (
torch.rand([self.batch_size, self.num_labels, self.min_size, self.max_size] , device=_snake_case) > 0.5
).float()
UpperCAmelCase_ = (torch.rand((self.batch_size, self.num_labels) , device=_snake_case) > 0.5).long()
UpperCAmelCase_ = self.get_config()
return config, pixel_values, pixel_mask, mask_labels, class_labels
def lowerCamelCase ( self : Any):
"""simple docstring"""
return MaskFormerConfig.from_backbone_and_decoder_configs(
backbone_config=SwinConfig(
depths=[1, 1, 1, 1] , ) , decoder_config=DetrConfig(
decoder_ffn_dim=128 , num_queries=self.num_queries , decoder_attention_heads=2 , d_model=self.mask_feature_size , ) , mask_feature_size=self.mask_feature_size , fpn_feature_size=self.mask_feature_size , num_channels=self.num_channels , num_labels=self.num_labels , )
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.prepare_config_and_inputs()
UpperCAmelCase_ = {'''pixel_values''': pixel_values, '''pixel_mask''': pixel_mask}
return config, inputs_dict
def lowerCamelCase ( self : str , _snake_case : List[Any] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = output.encoder_hidden_states
UpperCAmelCase_ = output.pixel_decoder_hidden_states
UpperCAmelCase_ = output.transformer_decoder_hidden_states
self.parent.assertTrue(len(_snake_case) , len(config.backbone_config.depths))
self.parent.assertTrue(len(_snake_case) , len(config.backbone_config.depths))
self.parent.assertTrue(len(_snake_case) , config.decoder_config.decoder_layers)
def lowerCamelCase ( self : Union[str, Any] , _snake_case : List[str] , _snake_case : int , _snake_case : Optional[Any] , _snake_case : str=False):
"""simple docstring"""
with torch.no_grad():
UpperCAmelCase_ = MaskFormerModel(config=_snake_case)
model.to(_snake_case)
model.eval()
UpperCAmelCase_ = model(pixel_values=_snake_case , pixel_mask=_snake_case)
UpperCAmelCase_ = model(_snake_case , output_hidden_states=_snake_case)
# the correct shape of output.transformer_decoder_hidden_states ensure the correcteness of the
# encoder and pixel decoder
self.parent.assertEqual(
output.transformer_decoder_last_hidden_state.shape , (self.batch_size, self.num_queries, self.mask_feature_size) , )
# let's ensure the other two hidden state exists
self.parent.assertTrue(output.pixel_decoder_last_hidden_state is not None)
self.parent.assertTrue(output.encoder_last_hidden_state is not None)
if output_hidden_states:
self.check_output_hidden_state(_snake_case , _snake_case)
def lowerCamelCase ( self : List[Any] , _snake_case : List[Any] , _snake_case : List[Any] , _snake_case : str , _snake_case : Optional[int] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerForInstanceSegmentation(config=_snake_case)
model.to(_snake_case)
model.eval()
def comm_check_on_output(_snake_case : Tuple):
# let's still check that all the required stuff is there
self.parent.assertTrue(result.transformer_decoder_last_hidden_state is not None)
self.parent.assertTrue(result.pixel_decoder_last_hidden_state is not None)
self.parent.assertTrue(result.encoder_last_hidden_state is not None)
# okay, now we need to check the logits shape
# due to the encoder compression, masks have a //4 spatial size
self.parent.assertEqual(
result.masks_queries_logits.shape , (self.batch_size, self.num_queries, self.min_size // 4, self.max_size // 4) , )
# + 1 for null class
self.parent.assertEqual(
result.class_queries_logits.shape , (self.batch_size, self.num_queries, self.num_labels + 1))
with torch.no_grad():
UpperCAmelCase_ = model(pixel_values=_snake_case , pixel_mask=_snake_case)
UpperCAmelCase_ = model(_snake_case)
comm_check_on_output(_snake_case)
UpperCAmelCase_ = model(
pixel_values=_snake_case , pixel_mask=_snake_case , mask_labels=_snake_case , class_labels=_snake_case)
comm_check_on_output(_snake_case)
self.parent.assertTrue(result.loss is not None)
self.parent.assertEqual(result.loss.shape , torch.Size([1]))
@require_torch
class __snake_case ( a , a , unittest.TestCase ):
UpperCAmelCase__ : Union[str, Any] = (MaskFormerModel, MaskFormerForInstanceSegmentation) if is_torch_available() else ()
UpperCAmelCase__ : Optional[Any] = (
{'''feature-extraction''': MaskFormerModel, '''image-segmentation''': MaskFormerForInstanceSegmentation}
if is_torch_available()
else {}
)
UpperCAmelCase__ : Dict = False
UpperCAmelCase__ : List[str] = False
UpperCAmelCase__ : Optional[Any] = False
UpperCAmelCase__ : Union[str, Any] = False
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerModelTester(self)
UpperCAmelCase_ = ConfigTester(self , config_class=_snake_case , has_text_modality=_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
self.config_tester.run_common_tests()
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskformer_model(_snake_case , **_snake_case , output_hidden_states=_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_maskformer_instance_segmentation_head_model(*_snake_case)
@unittest.skip(reason='''MaskFormer does not use inputs_embeds''')
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer does not have a get_input_embeddings method''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer is not a generative model''')
def lowerCamelCase ( self : str):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer does not use token embeddings''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
@require_torch_multi_gpu
@unittest.skip(
reason='''MaskFormer has some layers using `add_module` which doesn\'t work well with `nn.DataParallel`''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
pass
@unittest.skip('''Will be fixed soon by reducing the size of the model used for common tests.''')
def lowerCamelCase ( self : str):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ = model_class(_snake_case)
UpperCAmelCase_ = inspect.signature(model.forward)
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCAmelCase_ = [*signature.parameters.keys()]
UpperCAmelCase_ = ['''pixel_values''']
self.assertListEqual(arg_names[:1] , _snake_case)
@slow
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
for model_name in ["facebook/maskformer-swin-small-coco"]:
UpperCAmelCase_ = MaskFormerModel.from_pretrained(_snake_case)
self.assertIsNotNone(_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = (self.model_tester.min_size,) * 2
UpperCAmelCase_ = {
'''pixel_values''': torch.randn((2, 3, *size) , device=_snake_case),
'''mask_labels''': torch.randn((2, 10, *size) , device=_snake_case),
'''class_labels''': torch.zeros(2 , 10 , device=_snake_case).long(),
}
UpperCAmelCase_ = MaskFormerForInstanceSegmentation(MaskFormerConfig()).to(_snake_case)
UpperCAmelCase_ = model(**_snake_case)
self.assertTrue(outputs.loss is not None)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskformer_model(_snake_case , **_snake_case , output_hidden_states=_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ = model_class(_snake_case).to(_snake_case)
UpperCAmelCase_ = model(**_snake_case , output_attentions=_snake_case)
self.assertTrue(outputs.attentions is not None)
def lowerCamelCase ( self : int):
"""simple docstring"""
if not self.model_tester.is_training:
return
# only MaskFormerForInstanceSegmentation has the loss
UpperCAmelCase_ = self.all_model_classes[1]
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCAmelCase_ = model_class(_snake_case)
model.to(_snake_case)
model.train()
UpperCAmelCase_ = model(_snake_case , mask_labels=_snake_case , class_labels=_snake_case).loss
loss.backward()
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = self.all_model_classes[1]
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCAmelCase_ = True
UpperCAmelCase_ = True
UpperCAmelCase_ = model_class(_snake_case)
model.to(_snake_case)
model.train()
UpperCAmelCase_ = model(_snake_case , mask_labels=_snake_case , class_labels=_snake_case)
UpperCAmelCase_ = outputs.encoder_hidden_states[0]
encoder_hidden_states.retain_grad()
UpperCAmelCase_ = outputs.pixel_decoder_hidden_states[0]
pixel_decoder_hidden_states.retain_grad()
# we requires_grad=True in inputs_embeds (line 2152), the original implementation don't
UpperCAmelCase_ = outputs.transformer_decoder_hidden_states[0]
transformer_decoder_hidden_states.retain_grad()
UpperCAmelCase_ = outputs.attentions[0]
attentions.retain_grad()
outputs.loss.backward(retain_graph=_snake_case)
self.assertIsNotNone(encoder_hidden_states.grad)
self.assertIsNotNone(pixel_decoder_hidden_states.grad)
self.assertIsNotNone(transformer_decoder_hidden_states.grad)
self.assertIsNotNone(attentions.grad)
snake_case_ : Dict = 1e-4
def A () -> Union[str, Any]:
"""simple docstring"""
UpperCAmelCase_ = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
return image
@require_vision
@slow
class __snake_case ( unittest.TestCase ):
@cached_property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return (
MaskFormerImageProcessor.from_pretrained('''facebook/maskformer-swin-small-coco''')
if is_vision_available()
else None
)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerModel.from_pretrained('''facebook/maskformer-swin-small-coco''').to(_snake_case)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
UpperCAmelCase_ = torch.tensor(
[[-0.0_4_8_2, 0.9_2_2_8, 0.4_9_5_1], [-0.2_5_4_7, 0.8_0_1_7, 0.8_5_2_7], [-0.0_0_6_9, 0.3_3_8_5, -0.0_0_8_9]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.encoder_last_hidden_state[0, 0, :3, :3] , _snake_case , atol=_snake_case))
UpperCAmelCase_ = torch.tensor(
[[-0.8_4_2_2, -0.8_4_3_4, -0.9_7_1_8], [-1.0_1_4_4, -0.5_5_6_5, -0.4_1_9_5], [-1.0_0_3_8, -0.4_4_8_4, -0.1_9_6_1]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.pixel_decoder_last_hidden_state[0, 0, :3, :3] , _snake_case , atol=_snake_case))
UpperCAmelCase_ = torch.tensor(
[[0.2_8_5_2, -0.0_1_5_9, 0.9_7_3_5], [0.6_2_5_4, 0.1_8_5_8, 0.8_5_2_9], [-0.0_6_8_0, -0.4_1_1_6, 1.8_4_1_3]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.transformer_decoder_last_hidden_state[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-swin-small-coco''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
# masks_queries_logits
UpperCAmelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.decoder_config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) , )
UpperCAmelCase_ = [
[-1.3_7_3_7_1_2_4, -1.7_7_2_4_9_3_7, -1.9_3_6_4_2_3_3],
[-1.5_9_7_7_2_8_1, -1.9_8_6_7_9_3_9, -2.1_5_2_3_6_9_5],
[-1.5_7_9_5_3_9_8, -1.9_2_6_9_8_3_2, -2.0_9_3_9_4_2],
]
UpperCAmelCase_ = torch.tensor(_snake_case).to(_snake_case)
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _snake_case , atol=_snake_case))
# class_queries_logits
UpperCAmelCase_ = outputs.class_queries_logits
self.assertEqual(
class_queries_logits.shape , (1, model.config.decoder_config.num_queries, model.config.num_labels + 1))
UpperCAmelCase_ = torch.tensor(
[
[1.65_12e00, -5.25_72e00, -3.35_19e00],
[3.61_69e-02, -5.90_25e00, -2.93_13e00],
[1.07_66e-04, -7.76_30e00, -5.12_63e00],
]).to(_snake_case)
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-resnet101-coco-stuff''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
# masks_queries_logits
UpperCAmelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.decoder_config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) , )
UpperCAmelCase_ = [[-0.9_0_4_6, -2.6_3_6_6, -4.6_0_6_2], [-3.4_1_7_9, -5.7_8_9_0, -8.8_0_5_7], [-4.9_1_7_9, -7.6_5_6_0, -1_0.7_7_1_1]]
UpperCAmelCase_ = torch.tensor(_snake_case).to(_snake_case)
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _snake_case , atol=_snake_case))
# class_queries_logits
UpperCAmelCase_ = outputs.class_queries_logits
self.assertEqual(
class_queries_logits.shape , (1, model.config.decoder_config.num_queries, model.config.num_labels + 1))
UpperCAmelCase_ = torch.tensor(
[[4.7_1_8_8, -3.2_5_8_5, -2.8_8_5_7], [6.6_8_7_1, -2.9_1_8_1, -1.2_4_8_7], [7.2_4_4_9, -2.2_7_6_4, -2.1_8_7_4]]).to(_snake_case)
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-swin-small-coco''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = image_processor(
[np.zeros((3, 800, 1333)), np.zeros((3, 800, 1333))] , segmentation_maps=[np.zeros((384, 384)).astype(np.floataa), np.zeros((384, 384)).astype(np.floataa)] , return_tensors='''pt''' , )
UpperCAmelCase_ = inputs['''pixel_values'''].to(_snake_case)
UpperCAmelCase_ = [el.to(_snake_case) for el in inputs['''mask_labels''']]
UpperCAmelCase_ = [el.to(_snake_case) for el in inputs['''class_labels''']]
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
self.assertTrue(outputs.loss is not None)
| 51
| 0
|
"""simple docstring"""
def _lowerCAmelCase ( UpperCamelCase_ = 100 ):
__SCREAMING_SNAKE_CASE = 0
__SCREAMING_SNAKE_CASE = 0
for i in range(1 , n + 1 ):
sum_of_squares += i**2
sum_of_ints += i
return sum_of_ints**2 - sum_of_squares
if __name__ == "__main__":
print(F"""{solution() = }""")
| 100
|
import argparse
import pickle
import numpy as np
import torch
from torch import nn
from transformers import ReformerConfig, ReformerModelWithLMHead
from transformers.utils import logging
logging.set_verbosity_info()
def A (__A : Optional[int] , __A : int , __A : str=None ) -> List[Any]:
"""simple docstring"""
assert torch_layer.weight.shape == weight.shape, F"""{torch_layer} layer.weight does not match"""
UpperCAmelCase_ = nn.Parameter(__A )
if bias is not None:
assert torch_layer.bias.shape == bias.shape, F"""{torch_layer} layer.bias does not match"""
UpperCAmelCase_ = nn.Parameter(__A )
def A (__A : Tuple , __A : Dict , __A : str ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = np.asarray(weights[0] )
UpperCAmelCase_ = np.asarray(weights[1] )
UpperCAmelCase_ = np.asarray(weights[2] )
set_param(
torch_layer.self_attention.query_key , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.value , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.output.dense , torch.tensor(__A ).view(-1 , __A ).contiguous().transpose(0 , 1 ) , )
def A (__A : Optional[Any] , __A : Any , __A : List[Any] ) -> int:
"""simple docstring"""
UpperCAmelCase_ = np.asarray(weights[0] )
UpperCAmelCase_ = np.asarray(weights[1] )
UpperCAmelCase_ = np.asarray(weights[2] )
UpperCAmelCase_ = np.asarray(weights[3] )
set_param(
torch_layer.self_attention.query , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.key , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.value , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.output.dense , torch.tensor(__A ).view(-1 , __A ).contiguous().transpose(0 , 1 ) , )
def A (__A : int , __A : Union[str, Any] , __A : List[str] ) -> List[Any]:
"""simple docstring"""
UpperCAmelCase_ = weights[0][0][0]
UpperCAmelCase_ = np.asarray(layer_norm_a[0] )
UpperCAmelCase_ = np.asarray(layer_norm_a[1] )
set_param(
torch_block.attention.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# lsh weights + output
UpperCAmelCase_ = weights[0][1]
if len(__A ) < 4:
set_layer_weights_in_torch_lsh(__A , torch_block.attention , __A )
else:
set_layer_weights_in_torch_local(__A , torch_block.attention , __A )
# intermediate weighs
UpperCAmelCase_ = weights[2][0][1][2]
# Chunked Feed Forward
if len(__A ) == 4:
UpperCAmelCase_ = intermediate_weights[2]
# layernorm 2
UpperCAmelCase_ = np.asarray(intermediate_weights[0][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[0][1] )
set_param(
torch_block.feed_forward.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# intermediate dense
UpperCAmelCase_ = np.asarray(intermediate_weights[1][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[1][1] )
set_param(
torch_block.feed_forward.dense.dense , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
# intermediate out
UpperCAmelCase_ = np.asarray(intermediate_weights[4][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[4][1] )
set_param(
torch_block.feed_forward.output.dense , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
def A (__A : Optional[int] , __A : Tuple , __A : Any ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = torch_model.reformer
# word embeds
UpperCAmelCase_ = np.asarray(weights[1] )
set_param(
torch_model_reformer.embeddings.word_embeddings , torch.tensor(__A ) , )
if isinstance(weights[3] , __A ):
UpperCAmelCase_ = torch_model_reformer.embeddings.position_embeddings
for emb_idx in range(len(position_embeddings.weights ) ):
UpperCAmelCase_ = np.asarray(weights[3][emb_idx][0] )
assert (
position_embeddings.weights[emb_idx].shape == emb_weights.shape
), F"""{position_embeddings[emb_idx]} emb does not match"""
UpperCAmelCase_ = nn.Parameter(torch.tensor(__A ) )
UpperCAmelCase_ = weights[5]
assert len(torch_model_reformer.encoder.layers ) * 4 == len(
__A ), "HF and trax model do not have the same number of layers"
for layer_idx, layer in enumerate(torch_model_reformer.encoder.layers ):
UpperCAmelCase_ = trax_layer_weights[4 * layer_idx : 4 * (layer_idx + 1)]
set_block_weights_in_torch(__A , __A , __A )
# output layer norm
UpperCAmelCase_ = np.asarray(weights[7][0] )
UpperCAmelCase_ = np.asarray(weights[7][1] )
set_param(
torch_model_reformer.encoder.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# output embeddings
UpperCAmelCase_ = np.asarray(weights[9][0] )
UpperCAmelCase_ = np.asarray(weights[9][1] )
set_param(
torch_model.lm_head.decoder , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
def A (__A : Tuple , __A : int , __A : str ) -> Union[str, Any]:
"""simple docstring"""
UpperCAmelCase_ = ReformerConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = ReformerModelWithLMHead(__A )
with open(__A , '''rb''' ) as f:
UpperCAmelCase_ = pickle.load(__A )['''weights''']
set_model_weights_in_torch(__A , __A , config.hidden_size )
# Save pytorch-model
print(F"""Save PyTorch model to {pytorch_dump_path}""" )
torch.save(model.state_dict() , __A )
if __name__ == "__main__":
snake_case_ : Optional[Any] = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--trax_model_pkl_path", default=None, type=str, required=True, help="Path to the TensorFlow checkpoint path."
)
parser.add_argument(
"--config_file",
default=None,
type=str,
required=True,
help=(
"The config json file corresponding to the pre-trained Reformer model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--pytorch_dump_path", default=None, type=str, required=True, help="Path to the output PyTorch model."
)
snake_case_ : List[Any] = parser.parse_args()
convert_trax_checkpoint_to_pytorch(args.trax_model_pkl_path, args.config_file, args.pytorch_dump_path)
| 51
| 0
|
import importlib.metadata
import operator
import re
import sys
from typing import Optional
from packaging import version
lowercase__ :Dict = {
"<": operator.lt,
"<=": operator.le,
"==": operator.eq,
"!=": operator.ne,
">=": operator.ge,
">": operator.gt,
}
def UpperCamelCase ( lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ ):
'''simple docstring'''
if got_ver is None or want_ver is None:
raise ValueError(
f'Unable to compare versions for {requirement}: need={want_ver} found={got_ver}. This is unusual. Consider'
f' reinstalling {pkg}.' )
if not ops[op](version.parse(lowerCAmelCase__ ) , version.parse(lowerCAmelCase__ ) ):
raise ImportError(
f'{requirement} is required for a normal functioning of this module, but found {pkg}=={got_ver}.{hint}' )
def UpperCamelCase ( lowerCAmelCase__ , lowerCAmelCase__ = None ):
'''simple docstring'''
lowercase = f'\n{hint}' if hint is not None else ''''''
# non-versioned check
if re.match(R'''^[\w_\-\d]+$''' , lowerCAmelCase__ ):
lowercase , lowercase , lowercase = requirement, None, None
else:
lowercase = re.findall(R'''^([^!=<>\s]+)([\s!=<>]{1,2}.+)''' , lowerCAmelCase__ )
if not match:
raise ValueError(
'''requirement needs to be in the pip package format, .e.g., package_a==1.23, or package_b>=1.23, but'''
f' got {requirement}' )
lowercase , lowercase = match[0]
lowercase = want_full.split(''',''' ) # there could be multiple requirements
lowercase = {}
for w in want_range:
lowercase = re.findall(R'''^([\s!=<>]{1,2})(.+)''' , lowerCAmelCase__ )
if not match:
raise ValueError(
'''requirement needs to be in the pip package format, .e.g., package_a==1.23, or package_b>=1.23,'''
f' but got {requirement}' )
lowercase , lowercase = match[0]
lowercase = want_ver
if op not in ops:
raise ValueError(f'{requirement}: need one of {list(ops.keys() )}, but got {op}' )
# special case
if pkg == "python":
lowercase = '''.'''.join([str(lowerCAmelCase__ ) for x in sys.version_info[:3]] )
for op, want_ver in wanted.items():
_compare_versions(lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ )
return
# check if any version is installed
try:
lowercase = importlib.metadata.version(lowerCAmelCase__ )
except importlib.metadata.PackageNotFoundError:
raise importlib.metadata.PackageNotFoundError(
f'The \'{requirement}\' distribution was not found and is required by this application. {hint}' )
# check that the right version is installed if version number or a range was provided
if want_ver is not None:
for op, want_ver in wanted.items():
_compare_versions(lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ )
def UpperCamelCase ( lowerCAmelCase__ ):
'''simple docstring'''
lowercase = '''Try: pip install transformers -U or pip install -e \'.[dev]\' if you\'re working with git main'''
return require_version(lowerCAmelCase__ , lowerCAmelCase__ )
| 101
|
# This model implementation is heavily inspired by https://github.com/haofanwang/ControlNet-for-Diffusers/
import gc
import random
import tempfile
import unittest
import numpy as np
import torch
from PIL import Image
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
ControlNetModel,
DDIMScheduler,
StableDiffusionControlNetImgaImgPipeline,
UNetaDConditionModel,
)
from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_controlnet import MultiControlNetModel
from diffusers.utils import floats_tensor, load_image, load_numpy, randn_tensor, slow, torch_device
from diffusers.utils.import_utils import is_xformers_available
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import (
IMAGE_TO_IMAGE_IMAGE_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_PARAMS,
)
from ..test_pipelines_common import (
PipelineKarrasSchedulerTesterMixin,
PipelineLatentTesterMixin,
PipelineTesterMixin,
)
enable_full_determinism()
class __snake_case ( a , a , a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = StableDiffusionControlNetImgaImgPipeline
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''}
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
UpperCAmelCase__ : Union[str, Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS.union({'''control_image'''} )
UpperCAmelCase__ : Optional[Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS
def lowerCamelCase ( self : int):
"""simple docstring"""
torch.manual_seed(0)
UpperCAmelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , )
torch.manual_seed(0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
torch.manual_seed(0)
UpperCAmelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='''scaled_linear''' , clip_sample=_snake_case , set_alpha_to_one=_snake_case , )
torch.manual_seed(0)
UpperCAmelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , )
torch.manual_seed(0)
UpperCAmelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCAmelCase_ = CLIPTextModel(_snake_case)
UpperCAmelCase_ = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''')
UpperCAmelCase_ = {
'''unet''': unet,
'''controlnet''': controlnet,
'''scheduler''': scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def lowerCamelCase ( self : Union[str, Any] , _snake_case : Any , _snake_case : Dict=0):
"""simple docstring"""
if str(_snake_case).startswith('''mps'''):
UpperCAmelCase_ = torch.manual_seed(_snake_case)
else:
UpperCAmelCase_ = torch.Generator(device=_snake_case).manual_seed(_snake_case)
UpperCAmelCase_ = 2
UpperCAmelCase_ = randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , )
UpperCAmelCase_ = floats_tensor(control_image.shape , rng=random.Random(_snake_case)).to(_snake_case)
UpperCAmelCase_ = image.cpu().permute(0 , 2 , 3 , 1)[0]
UpperCAmelCase_ = Image.fromarray(np.uinta(_snake_case)).convert('''RGB''').resize((64, 64))
UpperCAmelCase_ = {
'''prompt''': '''A painting of a squirrel eating a burger''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
'''image''': image,
'''control_image''': control_image,
}
return inputs
def lowerCamelCase ( self : Any):
"""simple docstring"""
return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3)
@unittest.skipIf(
torch_device != '''cuda''' or not is_xformers_available() , reason='''XFormers attention is only available with CUDA and `xformers` installed''' , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
self._test_inference_batch_single_identical(expected_max_diff=2e-3)
class __snake_case ( a , a , unittest.TestCase ):
UpperCAmelCase__ : str = StableDiffusionControlNetImgaImgPipeline
UpperCAmelCase__ : List[str] = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''}
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
UpperCAmelCase__ : str = frozenset([] ) # TO_DO: add image_params once refactored VaeImageProcessor.preprocess
def lowerCamelCase ( self : str):
"""simple docstring"""
torch.manual_seed(0)
UpperCAmelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , )
torch.manual_seed(0)
def init_weights(_snake_case : Optional[int]):
if isinstance(_snake_case , torch.nn.Convad):
torch.nn.init.normal(m.weight)
m.bias.data.fill_(1.0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
controlneta.controlnet_down_blocks.apply(_snake_case)
torch.manual_seed(0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
controlneta.controlnet_down_blocks.apply(_snake_case)
torch.manual_seed(0)
UpperCAmelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='''scaled_linear''' , clip_sample=_snake_case , set_alpha_to_one=_snake_case , )
torch.manual_seed(0)
UpperCAmelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , )
torch.manual_seed(0)
UpperCAmelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCAmelCase_ = CLIPTextModel(_snake_case)
UpperCAmelCase_ = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''')
UpperCAmelCase_ = MultiControlNetModel([controlneta, controlneta])
UpperCAmelCase_ = {
'''unet''': unet,
'''controlnet''': controlnet,
'''scheduler''': scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def lowerCamelCase ( self : int , _snake_case : Union[str, Any] , _snake_case : str=0):
"""simple docstring"""
if str(_snake_case).startswith('''mps'''):
UpperCAmelCase_ = torch.manual_seed(_snake_case)
else:
UpperCAmelCase_ = torch.Generator(device=_snake_case).manual_seed(_snake_case)
UpperCAmelCase_ = 2
UpperCAmelCase_ = [
randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , ),
randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , ),
]
UpperCAmelCase_ = floats_tensor(control_image[0].shape , rng=random.Random(_snake_case)).to(_snake_case)
UpperCAmelCase_ = image.cpu().permute(0 , 2 , 3 , 1)[0]
UpperCAmelCase_ = Image.fromarray(np.uinta(_snake_case)).convert('''RGB''').resize((64, 64))
UpperCAmelCase_ = {
'''prompt''': '''A painting of a squirrel eating a burger''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
'''image''': image,
'''control_image''': control_image,
}
return inputs
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_dummy_components()
UpperCAmelCase_ = self.pipeline_class(**_snake_case)
pipe.to(_snake_case)
UpperCAmelCase_ = 1_0.0
UpperCAmelCase_ = 4
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case)[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=0.1 , control_guidance_end=0.2)[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=[0.1, 0.3] , control_guidance_end=[0.2, 0.7])[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=0.4 , control_guidance_end=[0.5, 0.8])[0]
# make sure that all outputs are different
assert np.sum(np.abs(output_a - output_a)) > 1e-3
assert np.sum(np.abs(output_a - output_a)) > 1e-3
assert np.sum(np.abs(output_a - output_a)) > 1e-3
def lowerCamelCase ( self : Dict):
"""simple docstring"""
return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3)
@unittest.skipIf(
torch_device != '''cuda''' or not is_xformers_available() , reason='''XFormers attention is only available with CUDA and `xformers` installed''' , )
def lowerCamelCase ( self : int):
"""simple docstring"""
self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3)
def lowerCamelCase ( self : int):
"""simple docstring"""
self._test_inference_batch_single_identical(expected_max_diff=2e-3)
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = self.get_dummy_components()
UpperCAmelCase_ = self.pipeline_class(**_snake_case)
pipe.to(_snake_case)
pipe.set_progress_bar_config(disable=_snake_case)
with tempfile.TemporaryDirectory() as tmpdir:
try:
# save_pretrained is not implemented for Multi-ControlNet
pipe.save_pretrained(_snake_case)
except NotImplementedError:
pass
@slow
@require_torch_gpu
class __snake_case ( unittest.TestCase ):
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = ControlNetModel.from_pretrained('''lllyasviel/sd-controlnet-canny''')
UpperCAmelCase_ = StableDiffusionControlNetImgaImgPipeline.from_pretrained(
'''runwayml/stable-diffusion-v1-5''' , safety_checker=_snake_case , controlnet=_snake_case)
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=_snake_case)
UpperCAmelCase_ = torch.Generator(device='''cpu''').manual_seed(0)
UpperCAmelCase_ = '''evil space-punk bird'''
UpperCAmelCase_ = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/bird_canny.png''').resize((512, 512))
UpperCAmelCase_ = load_image(
'''https://huggingface.co/lllyasviel/sd-controlnet-canny/resolve/main/images/bird.png''').resize((512, 512))
UpperCAmelCase_ = pipe(
_snake_case , _snake_case , control_image=_snake_case , generator=_snake_case , output_type='''np''' , num_inference_steps=50 , strength=0.6 , )
UpperCAmelCase_ = output.images[0]
assert image.shape == (512, 512, 3)
UpperCAmelCase_ = load_numpy(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/img2img.npy''')
assert np.abs(expected_image - image).max() < 9e-2
| 51
| 0
|
"""simple docstring"""
import argparse
import os
from io import BytesIO
from pathlib import Path
import requests
from clip_retrieval.clip_client import ClipClient
from PIL import Image
from tqdm import tqdm
def lowercase ( _snake_case : Optional[Any] , _snake_case : List[Any] , _snake_case : Union[str, Any] ) ->List[str]:
"""simple docstring"""
__snake_case : List[Any] = 1.5
__snake_case : List[str] = int(factor * num_class_images )
__snake_case : Tuple = ClipClient(
url='''https://knn.laion.ai/knn-service''' , indice_name='''laion_400m''' , num_images=_snake_case , aesthetic_weight=0.1 )
os.makedirs(f"""{class_data_dir}/images""" , exist_ok=_snake_case )
if len(list(Path(f"""{class_data_dir}/images""" ).iterdir() ) ) >= num_class_images:
return
while True:
__snake_case : Dict = client.query(text=_snake_case )
if len(_snake_case ) >= factor * num_class_images or num_images > 1e4:
break
else:
__snake_case : str = int(factor * num_images )
__snake_case : Union[str, Any] = ClipClient(
url='''https://knn.laion.ai/knn-service''' , indice_name='''laion_400m''' , num_images=_snake_case , aesthetic_weight=0.1 , )
__snake_case : Any = 0
__snake_case : List[str] = 0
__snake_case : Union[str, Any] = tqdm(desc='''downloading real regularization images''' , total=_snake_case )
with open(f"""{class_data_dir}/caption.txt""" , '''w''' ) as fa, open(f"""{class_data_dir}/urls.txt""" , '''w''' ) as fa, open(
f"""{class_data_dir}/images.txt""" , '''w''' ) as fa:
while total < num_class_images:
__snake_case : Tuple = class_images[count]
count += 1
try:
__snake_case : List[str] = requests.get(images['''url'''] )
if img.status_code == 200:
__snake_case : Optional[int] = Image.open(BytesIO(img.content ) )
with open(f"""{class_data_dir}/images/{total}.jpg""" , '''wb''' ) as f:
f.write(img.content )
fa.write(images['''caption'''] + '''\n''' )
fa.write(images['''url'''] + '''\n''' )
fa.write(f"""{class_data_dir}/images/{total}.jpg""" + '''\n''' )
total += 1
pbar.update(1 )
else:
continue
except Exception:
continue
return
def lowercase ( ) ->List[str]:
"""simple docstring"""
__snake_case : Any = argparse.ArgumentParser('''''' , add_help=_snake_case )
parser.add_argument('''--class_prompt''' , help='''text prompt to retrieve images''' , required=_snake_case , type=_snake_case )
parser.add_argument('''--class_data_dir''' , help='''path to save images''' , required=_snake_case , type=_snake_case )
parser.add_argument('''--num_class_images''' , help='''number of images to download''' , default=200 , type=_snake_case )
return parser.parse_args()
if __name__ == "__main__":
SCREAMING_SNAKE_CASE : List[Any] = parse_args()
retrieve(args.class_prompt, args.class_data_dir, args.num_class_images)
| 102
|
import random
import timeit
from functools import wraps
from typing import Callable, Optional
from ..configuration_utils import PretrainedConfig
from ..models.auto.modeling_tf_auto import TF_MODEL_MAPPING, TF_MODEL_WITH_LM_HEAD_MAPPING
from ..utils import is_pyanvml_available, is_tf_available, logging
from .benchmark_utils import (
Benchmark,
Memory,
MemorySummary,
measure_peak_memory_cpu,
start_memory_tracing,
stop_memory_tracing,
)
if is_tf_available():
import tensorflow as tf
from tensorflow.python.framework.errors_impl import ResourceExhaustedError
from .benchmark_args_tf import TensorFlowBenchmarkArguments
if is_pyanvml_available():
import pyanvml.pyanvml as nvml
snake_case_ : Tuple = logging.get_logger(__name__)
def A (__A : bool , __A : bool ) -> Optional[Any]:
"""simple docstring"""
def run_func(__A : Optional[Any] ):
@wraps(__A )
def run_in_eager_mode(*__A : Dict , **__A : List[Any] ):
return func(*__A , **__A )
@wraps(__A )
@tf.function(experimental_compile=__A )
def run_in_graph_mode(*__A : Optional[Any] , **__A : Any ):
return func(*__A , **__A )
if do_eager_mode is True:
if use_xla is not False:
raise ValueError(
'''Cannot run model in XLA, if `args.eager_mode` is set to `True`. Please set `args.eager_mode=False`.''' )
return run_in_eager_mode
else:
return run_in_graph_mode
return run_func
def A (__A : int , __A : int , __A : int ) -> ["tf.Tensor"]:
"""simple docstring"""
UpperCAmelCase_ = random.Random()
UpperCAmelCase_ = [rng.randint(0 , vocab_size - 1 ) for i in range(batch_size * sequence_length )]
return tf.constant(__A , shape=(batch_size, sequence_length) , dtype=tf.intaa )
class __snake_case ( a ):
UpperCAmelCase__ : TensorFlowBenchmarkArguments
UpperCAmelCase__ : PretrainedConfig
UpperCAmelCase__ : str = "TensorFlow"
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return tf.__version__
def lowerCamelCase ( self : Dict , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case)
return self._measure_speed(_inference)
def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case)
return self._measure_speed(_train)
def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
if self.args.is_gpu:
tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case)
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case)
return self._measure_memory(_inference)
def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
if self.args.is_gpu:
tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case)
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case)
return self._measure_memory(_train)
def lowerCamelCase ( self : Optional[int] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.config_dict[model_name]
if self.args.fpaa:
raise NotImplementedError('''Mixed precision is currently not supported.''')
UpperCAmelCase_ = (
hasattr(_snake_case , '''architectures''')
and isinstance(config.architectures , _snake_case)
and len(config.architectures) > 0
)
if not self.args.only_pretrain_model and has_model_class_in_config:
try:
UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model
UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class])
UpperCAmelCase_ = getattr(_snake_case , _snake_case)
UpperCAmelCase_ = model_cls(_snake_case)
except ImportError:
raise ImportError(
F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to"""
''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''')
else:
UpperCAmelCase_ = TF_MODEL_MAPPING[config.__class__](_snake_case)
# encoder-decoder has vocab size saved differently
UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size
UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_decoder_forward():
return model(_snake_case , decoder_input_ids=_snake_case , training=_snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_forward():
return model(_snake_case , training=_snake_case)
UpperCAmelCase_ = encoder_decoder_forward if config.is_encoder_decoder else encoder_forward
return _inference
def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.config_dict[model_name]
if self.args.eager_mode is not False:
raise ValueError('''Training cannot be done in eager mode. Please make sure that `args.eager_mode = False`.''')
if self.args.fpaa:
raise NotImplementedError('''Mixed precision is currently not supported.''')
UpperCAmelCase_ = (
hasattr(_snake_case , '''architectures''')
and isinstance(config.architectures , _snake_case)
and len(config.architectures) > 0
)
if not self.args.only_pretrain_model and has_model_class_in_config:
try:
UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model
UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class])
UpperCAmelCase_ = getattr(_snake_case , _snake_case)
UpperCAmelCase_ = model_cls(_snake_case)
except ImportError:
raise ImportError(
F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to"""
''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''')
else:
UpperCAmelCase_ = TF_MODEL_WITH_LM_HEAD_MAPPING[config.__class__](_snake_case)
# encoder-decoder has vocab size saved differently
UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size
UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_decoder_train():
UpperCAmelCase_ = model(_snake_case , decoder_input_ids=_snake_case , labels=_snake_case , training=_snake_case)[0]
UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables)
return gradients
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_train():
UpperCAmelCase_ = model(_snake_case , labels=_snake_case , training=_snake_case)[0]
UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables)
return gradients
UpperCAmelCase_ = encoder_decoder_train if config.is_encoder_decoder else encoder_train
return _train
def lowerCamelCase ( self : Any , _snake_case : Optional[Any]):
"""simple docstring"""
with self.args.strategy.scope():
try:
if self.args.is_tpu or self.args.use_xla:
# run additional 10 times to stabilize compilation for tpu
logger.info('''Do inference on TPU. Running model 5 times to stabilize compilation''')
timeit.repeat(_snake_case , repeat=1 , number=5)
# as written in https://docs.python.org/2/library/timeit.html#timeit.Timer.repeat, min should be taken rather than the average
UpperCAmelCase_ = timeit.repeat(
_snake_case , repeat=self.args.repeat , number=10 , )
return min(_snake_case) / 1_0.0
except ResourceExhaustedError as e:
self.print_fn(F"""Doesn't fit on GPU. {e}""")
def lowerCamelCase ( self : Dict , _snake_case : Callable[[], None]):
"""simple docstring"""
logger.info(
'''Note that TensorFlow allocates more memory than '''
'''it might need to speed up computation. '''
'''The memory reported here corresponds to the memory '''
'''reported by `nvidia-smi`, which can vary depending '''
'''on total available memory on the GPU that is used.''')
with self.args.strategy.scope():
try:
if self.args.trace_memory_line_by_line:
if not self.args.eager_mode:
raise ValueError(
'''`args.eager_mode` is set to `False`. Make sure to run model in eager mode to measure memory'''
''' consumption line by line.''')
UpperCAmelCase_ = start_memory_tracing('''transformers''')
if self.args.is_tpu:
# tpu
raise NotImplementedError(
'''Memory Benchmarking is currently not implemented for TPU. Please disable memory benchmarking'''
''' with `args.memory=False`''')
elif self.args.is_gpu:
# gpu
if not is_pyanvml_available():
logger.warning(
'''py3nvml not installed, we won\'t log GPU memory usage. '''
'''Install py3nvml (pip install py3nvml) to log information about GPU.''')
UpperCAmelCase_ = '''N/A'''
else:
logger.info(
'''Measuring total GPU usage on GPU device. Make sure to not have additional processes'''
''' running on the same GPU.''')
# init nvml
nvml.nvmlInit()
func()
UpperCAmelCase_ = nvml.nvmlDeviceGetHandleByIndex(self.args.device_idx)
UpperCAmelCase_ = nvml.nvmlDeviceGetMemoryInfo(_snake_case)
UpperCAmelCase_ = meminfo.used
UpperCAmelCase_ = Memory(_snake_case)
# shutdown nvml
nvml.nvmlShutdown()
else:
# cpu
if self.args.trace_memory_line_by_line:
logger.info(
'''When enabling line by line tracing, the max peak memory for CPU is inaccurate in'''
''' TensorFlow.''')
UpperCAmelCase_ = None
else:
UpperCAmelCase_ = measure_peak_memory_cpu(_snake_case)
UpperCAmelCase_ = Memory(_snake_case) if isinstance(_snake_case , _snake_case) else memory_bytes
if self.args.trace_memory_line_by_line:
UpperCAmelCase_ = stop_memory_tracing(_snake_case)
if memory is None:
UpperCAmelCase_ = summary.total
else:
UpperCAmelCase_ = None
return memory, summary
except ResourceExhaustedError as e:
self.print_fn(F"""Doesn't fit on GPU. {e}""")
return "N/A", None
| 51
| 0
|
from bisect import bisect
from itertools import accumulate
def UpperCamelCase( __UpperCamelCase : Optional[int] ,__UpperCamelCase : Optional[Any] ,__UpperCamelCase : Dict ,__UpperCamelCase : str ):
lowerCAmelCase_ : Union[str, Any] = sorted(zip(__UpperCamelCase ,__UpperCamelCase ) ,key=lambda __UpperCamelCase : x[0] / x[1] ,reverse=__UpperCamelCase )
lowerCAmelCase_ , lowerCAmelCase_ : Any = [i[0] for i in r], [i[1] for i in r]
lowerCAmelCase_ : Union[str, Any] = list(accumulate(__UpperCamelCase ) )
lowerCAmelCase_ : List[Any] = bisect(__UpperCamelCase ,__UpperCamelCase )
return (
0
if k == 0
else sum(vl[:k] ) + (w - acc[k - 1]) * (vl[k]) / (wt[k])
if k != n
else sum(vl[:k] )
)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 103
|
import hashlib
import unittest
from transformers import MODEL_FOR_DEPTH_ESTIMATION_MAPPING, is_torch_available, is_vision_available
from transformers.pipelines import DepthEstimationPipeline, pipeline
from transformers.testing_utils import (
is_pipeline_test,
nested_simplify,
require_tf,
require_timm,
require_torch,
require_vision,
slow,
)
from .test_pipelines_common import ANY
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
else:
class __snake_case :
@staticmethod
def lowerCamelCase ( *_snake_case : Optional[int] , **_snake_case : int):
"""simple docstring"""
pass
def A (__A : Image ) -> str:
"""simple docstring"""
UpperCAmelCase_ = hashlib.mda(image.tobytes() )
return m.hexdigest()
@is_pipeline_test
@require_vision
@require_timm
@require_torch
class __snake_case ( unittest.TestCase ):
UpperCAmelCase__ : Tuple = MODEL_FOR_DEPTH_ESTIMATION_MAPPING
def lowerCamelCase ( self : Tuple , _snake_case : Optional[Any] , _snake_case : Optional[Any] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = DepthEstimationPipeline(model=_snake_case , image_processor=_snake_case)
return depth_estimator, [
"./tests/fixtures/tests_samples/COCO/000000039769.png",
"./tests/fixtures/tests_samples/COCO/000000039769.png",
]
def lowerCamelCase ( self : str , _snake_case : Optional[int] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = depth_estimator('''./tests/fixtures/tests_samples/COCO/000000039769.png''')
self.assertEqual({'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)} , _snake_case)
import datasets
UpperCAmelCase_ = datasets.load_dataset('''hf-internal-testing/fixtures_image_utils''' , '''image''' , split='''test''')
UpperCAmelCase_ = depth_estimator(
[
Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png'''),
'''http://images.cocodataset.org/val2017/000000039769.jpg''',
# RGBA
dataset[0]['''file'''],
# LA
dataset[1]['''file'''],
# L
dataset[2]['''file'''],
])
self.assertEqual(
[
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
] , _snake_case , )
@require_tf
@unittest.skip('''Depth estimation is not implemented in TF''')
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
pass
@slow
@require_torch
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = '''Intel/dpt-large'''
UpperCAmelCase_ = pipeline('''depth-estimation''' , model=_snake_case)
UpperCAmelCase_ = depth_estimator('''http://images.cocodataset.org/val2017/000000039769.jpg''')
UpperCAmelCase_ = hashimage(outputs['''depth'''])
# This seems flaky.
# self.assertEqual(outputs["depth"], "1a39394e282e9f3b0741a90b9f108977")
self.assertEqual(nested_simplify(outputs['''predicted_depth'''].max().item()) , 2_9.3_0_4)
self.assertEqual(nested_simplify(outputs['''predicted_depth'''].min().item()) , 2.6_6_2)
@require_torch
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
self.skipTest('''There is not hf-internal-testing tiny model for either GLPN nor DPT''')
| 51
| 0
|
'''simple docstring'''
from typing import Optional
from .. import Features, NamedSplit
from ..packaged_modules.text.text import Text
from ..utils.typing import NestedDataStructureLike, PathLike
from .abc import AbstractDatasetReader
class lowercase_ (lowerCamelCase__ ):
"""simple docstring"""
def __init__( self : Optional[Any] ,lowercase__ : NestedDataStructureLike[PathLike] ,lowercase__ : Optional[NamedSplit] = None ,lowercase__ : Optional[Features] = None ,lowercase__ : str = None ,lowercase__ : bool = False ,lowercase__ : bool = False ,lowercase__ : Optional[int] = None ,**lowercase__ : Optional[int] ,):
super().__init__(
lowercase__ ,split=lowercase__ ,features=lowercase__ ,cache_dir=lowercase__ ,keep_in_memory=lowercase__ ,streaming=lowercase__ ,num_proc=lowercase__ ,**lowercase__ ,)
__lowercase = path_or_paths if isinstance(lowercase__ ,lowercase__ ) else {self.split: path_or_paths}
__lowercase = Text(
cache_dir=lowercase__ ,data_files=lowercase__ ,features=lowercase__ ,**lowercase__ ,)
def SCREAMING_SNAKE_CASE ( self : Tuple ):
# Build iterable dataset
if self.streaming:
__lowercase = self.builder.as_streaming_dataset(split=self.split )
# Build regular (map-style) dataset
else:
__lowercase = None
__lowercase = None
__lowercase = None
__lowercase = None
self.builder.download_and_prepare(
download_config=lowercase__ ,download_mode=lowercase__ ,verification_mode=lowercase__ ,base_path=lowercase__ ,num_proc=self.num_proc ,)
__lowercase = self.builder.as_dataset(
split=self.split ,verification_mode=lowercase__ ,in_memory=self.keep_in_memory )
return dataset
| 104
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : int = {
"configuration_deberta": ["DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP", "DebertaConfig", "DebertaOnnxConfig"],
"tokenization_deberta": ["DebertaTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : int = ["DebertaTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"DebertaForMaskedLM",
"DebertaForQuestionAnswering",
"DebertaForSequenceClassification",
"DebertaForTokenClassification",
"DebertaModel",
"DebertaPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFDebertaForMaskedLM",
"TFDebertaForQuestionAnswering",
"TFDebertaForSequenceClassification",
"TFDebertaForTokenClassification",
"TFDebertaModel",
"TFDebertaPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_deberta import DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP, DebertaConfig, DebertaOnnxConfig
from .tokenization_deberta import DebertaTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_deberta_fast import DebertaTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_deberta import (
DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
DebertaForMaskedLM,
DebertaForQuestionAnswering,
DebertaForSequenceClassification,
DebertaForTokenClassification,
DebertaModel,
DebertaPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_deberta import (
TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
TFDebertaForMaskedLM,
TFDebertaForQuestionAnswering,
TFDebertaForSequenceClassification,
TFDebertaForTokenClassification,
TFDebertaModel,
TFDebertaPreTrainedModel,
)
else:
import sys
snake_case_ : Dict = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
"""simple docstring"""
from graphs.minimum_spanning_tree_kruskal import kruskal
def _SCREAMING_SNAKE_CASE ( ) ->Dict:
'''simple docstring'''
a : Any = 9
a : int = [
[0, 1, 4],
[0, 7, 8],
[1, 2, 8],
[7, 8, 7],
[7, 6, 1],
[2, 8, 2],
[8, 6, 6],
[2, 3, 7],
[2, 5, 4],
[6, 5, 2],
[3, 5, 14],
[3, 4, 9],
[5, 4, 10],
[1, 7, 11],
]
a : Any = kruskal(_lowercase , _lowercase )
a : Optional[int] = [
[7, 6, 1],
[2, 8, 2],
[6, 5, 2],
[0, 1, 4],
[2, 5, 4],
[2, 3, 7],
[0, 7, 8],
[3, 4, 9],
]
assert sorted(_lowercase ) == sorted(_lowercase )
| 105
|
from typing import TYPE_CHECKING
from ...file_utils import _LazyModule, is_tokenizers_available, is_torch_available
from ...utils import OptionalDependencyNotAvailable
snake_case_ : Union[str, Any] = {"configuration_gpt_neox": ["GPT_NEOX_PRETRAINED_CONFIG_ARCHIVE_MAP", "GPTNeoXConfig"]}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Dict = ["GPTNeoXTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = [
"GPT_NEOX_PRETRAINED_MODEL_ARCHIVE_LIST",
"GPTNeoXForCausalLM",
"GPTNeoXForQuestionAnswering",
"GPTNeoXForSequenceClassification",
"GPTNeoXForTokenClassification",
"GPTNeoXLayer",
"GPTNeoXModel",
"GPTNeoXPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_gpt_neox import GPT_NEOX_PRETRAINED_CONFIG_ARCHIVE_MAP, GPTNeoXConfig
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_gpt_neox_fast import GPTNeoXTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_gpt_neox import (
GPT_NEOX_PRETRAINED_MODEL_ARCHIVE_LIST,
GPTNeoXForCausalLM,
GPTNeoXForQuestionAnswering,
GPTNeoXForSequenceClassification,
GPTNeoXForTokenClassification,
GPTNeoXLayer,
GPTNeoXModel,
GPTNeoXPreTrainedModel,
)
else:
import sys
snake_case_ : Optional[int] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
"""simple docstring"""
from graphs.minimum_spanning_tree_kruskal import kruskal
def __SCREAMING_SNAKE_CASE ( ):
lowerCAmelCase__ : Optional[Any] = 9
lowerCAmelCase__ : Any = [
[0, 1, 4],
[0, 7, 8],
[1, 2, 8],
[7, 8, 7],
[7, 6, 1],
[2, 8, 2],
[8, 6, 6],
[2, 3, 7],
[2, 5, 4],
[6, 5, 2],
[3, 5, 14],
[3, 4, 9],
[5, 4, 10],
[1, 7, 11],
]
lowerCAmelCase__ : Optional[int] = kruskal(A_ , A_ )
lowerCAmelCase__ : List[str] = [
[7, 6, 1],
[2, 8, 2],
[6, 5, 2],
[0, 1, 4],
[2, 5, 4],
[2, 3, 7],
[0, 7, 8],
[3, 4, 9],
]
assert sorted(A_ ) == sorted(A_ )
| 106
|
def A (__A : list , __A : int , __A : int = 0 , __A : int = 0 ) -> int:
"""simple docstring"""
UpperCAmelCase_ = right or len(__A ) - 1
if left > right:
return -1
elif list_data[left] == key:
return left
elif list_data[right] == key:
return right
else:
return search(__A , __A , left + 1 , right - 1 )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 51
| 0
|
import enum
import warnings
from ..tokenization_utils import TruncationStrategy
from ..utils import add_end_docstrings, is_tf_available, is_torch_available, logging
from .base import PIPELINE_INIT_ARGS, Pipeline
if is_tf_available():
import tensorflow as tf
from ..models.auto.modeling_tf_auto import TF_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING
if is_torch_available():
from ..models.auto.modeling_auto import MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING
__lowerCAmelCase : Any = logging.get_logger(__name__)
class snake_case__ (enum.Enum ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : Dict = 0
SCREAMING_SNAKE_CASE_ : Optional[Any] = 1
@add_end_docstrings(_UpperCamelCase )
class snake_case__ (_UpperCamelCase ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : int = """generated"""
def __init__( self : Dict , *__lowerCamelCase : List[Any] , **__lowerCamelCase : Union[str, Any] ) -> List[Any]:
super().__init__(*__lowerCamelCase , **__lowerCamelCase )
self.check_model_type(
TF_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING
if self.framework == "tf"
else MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING )
def __UpperCAmelCase ( self : List[Any] , __lowerCamelCase : List[str]=None , __lowerCamelCase : List[str]=None , __lowerCamelCase : Any=None , __lowerCamelCase : Union[str, Any]=None , __lowerCamelCase : Union[str, Any]=None , __lowerCamelCase : str=None , **__lowerCamelCase : str , ) -> Tuple:
a = {}
if truncation is not None:
a = truncation
a = generate_kwargs
a = {}
if return_tensors is not None and return_type is None:
a = ReturnType.TENSORS if return_tensors else ReturnType.TEXT
if return_type is not None:
a = return_type
if clean_up_tokenization_spaces is not None:
a = clean_up_tokenization_spaces
if stop_sequence is not None:
a = self.tokenizer.encode(__lowerCamelCase , add_special_tokens=__lowerCamelCase )
if len(__lowerCamelCase ) > 1:
warnings.warn(
"Stopping on a multiple token sequence is not yet supported on transformers. The first token of"
" the stop sequence will be used as the stop sequence string in the interim." )
a = stop_sequence_ids[0]
return preprocess_params, forward_params, postprocess_params
def __UpperCAmelCase ( self : List[Any] , __lowerCamelCase : int , __lowerCamelCase : int , __lowerCamelCase : int ) -> Union[str, Any]:
return True
def __UpperCAmelCase ( self : Optional[int] , *__lowerCamelCase : str , __lowerCamelCase : Optional[int] ) -> Union[str, Any]:
a = self.model.config.prefix if self.model.config.prefix is not None else ""
if isinstance(args[0] , __lowerCamelCase ):
if self.tokenizer.pad_token_id is None:
raise ValueError("Please make sure that the tokenizer has a pad_token_id when using a batch input" )
a = ([prefix + arg for arg in args[0]],)
a = True
elif isinstance(args[0] , __lowerCamelCase ):
a = (prefix + args[0],)
a = False
else:
raise ValueError(
f""" `args[0]`: {args[0]} have the wrong format. The should be either of type `str` or type `list`""" )
a = self.tokenizer(*__lowerCamelCase , padding=__lowerCamelCase , truncation=__lowerCamelCase , return_tensors=self.framework )
# This is produced by tokenizers but is an invalid generate kwargs
if "token_type_ids" in inputs:
del inputs["token_type_ids"]
return inputs
def __call__( self : Any , *__lowerCamelCase : List[Any] , **__lowerCamelCase : int ) -> Optional[int]:
a = super().__call__(*__lowerCamelCase , **__lowerCamelCase )
if (
isinstance(args[0] , __lowerCamelCase )
and all(isinstance(__lowerCamelCase , __lowerCamelCase ) for el in args[0] )
and all(len(__lowerCamelCase ) == 1 for res in result )
):
return [res[0] for res in result]
return result
def __UpperCAmelCase ( self : str , __lowerCamelCase : Any , __lowerCamelCase : List[Any]=TruncationStrategy.DO_NOT_TRUNCATE , **__lowerCamelCase : str ) -> Optional[int]:
a = self._parse_and_tokenize(__lowerCamelCase , truncation=__lowerCamelCase , **__lowerCamelCase )
return inputs
def __UpperCAmelCase ( self : List[Any] , __lowerCamelCase : Optional[int] , **__lowerCamelCase : Dict ) -> Optional[int]:
if self.framework == "pt":
a , a = model_inputs["input_ids"].shape
elif self.framework == "tf":
a , a = tf.shape(model_inputs["input_ids"] ).numpy()
a = generate_kwargs.get("min_length" , self.model.config.min_length )
a = generate_kwargs.get("max_length" , self.model.config.max_length )
self.check_inputs(__lowerCamelCase , generate_kwargs["min_length"] , generate_kwargs["max_length"] )
a = self.model.generate(**__lowerCamelCase , **__lowerCamelCase )
a = output_ids.shape[0]
if self.framework == "pt":
a = output_ids.reshape(__lowerCamelCase , out_b // in_b , *output_ids.shape[1:] )
elif self.framework == "tf":
a = tf.reshape(__lowerCamelCase , (in_b, out_b // in_b, *output_ids.shape[1:]) )
return {"output_ids": output_ids}
def __UpperCAmelCase ( self : Optional[int] , __lowerCamelCase : List[Any] , __lowerCamelCase : List[Any]=ReturnType.TEXT , __lowerCamelCase : List[Any]=False ) -> int:
a = []
for output_ids in model_outputs["output_ids"][0]:
if return_type == ReturnType.TENSORS:
a = {f"""{self.return_name}_token_ids""": output_ids}
elif return_type == ReturnType.TEXT:
a = {
f"""{self.return_name}_text""": self.tokenizer.decode(
__lowerCamelCase , skip_special_tokens=__lowerCamelCase , clean_up_tokenization_spaces=__lowerCamelCase , )
}
records.append(__lowerCamelCase )
return records
@add_end_docstrings(_UpperCamelCase )
class snake_case__ (_UpperCamelCase ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : Union[str, Any] = """summary"""
def __call__( self : Dict , *__lowerCamelCase : List[Any] , **__lowerCamelCase : str ) -> Optional[Any]:
return super().__call__(*__lowerCamelCase , **__lowerCamelCase )
def __UpperCAmelCase ( self : Tuple , __lowerCamelCase : int , __lowerCamelCase : int , __lowerCamelCase : int ) -> bool:
if max_length < min_length:
logger.warning(f"""Your min_length={min_length} must be inferior than your max_length={max_length}.""" )
if input_length < max_length:
logger.warning(
f"""Your max_length is set to {max_length}, but your input_length is only {input_length}. Since this is """
"a summarization task, where outputs shorter than the input are typically wanted, you might "
f"""consider decreasing max_length manually, e.g. summarizer('...', max_length={input_length//2})""" )
@add_end_docstrings(_UpperCamelCase )
class snake_case__ (_UpperCamelCase ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : Dict = """translation"""
def __UpperCAmelCase ( self : Optional[int] , __lowerCamelCase : int , __lowerCamelCase : int , __lowerCamelCase : int ) -> Dict:
if input_length > 0.9 * max_length:
logger.warning(
f"""Your input_length: {input_length} is bigger than 0.9 * max_length: {max_length}. You might consider """
"increasing your max_length manually, e.g. translator('...', max_length=400)" )
return True
def __UpperCAmelCase ( self : Union[str, Any] , *__lowerCamelCase : Optional[int] , __lowerCamelCase : Any=TruncationStrategy.DO_NOT_TRUNCATE , __lowerCamelCase : Optional[int]=None , __lowerCamelCase : str=None ) -> Optional[Any]:
if getattr(self.tokenizer , "_build_translation_inputs" , __lowerCamelCase ):
return self.tokenizer._build_translation_inputs(
*__lowerCamelCase , return_tensors=self.framework , truncation=__lowerCamelCase , src_lang=__lowerCamelCase , tgt_lang=__lowerCamelCase )
else:
return super()._parse_and_tokenize(*__lowerCamelCase , truncation=__lowerCamelCase )
def __UpperCAmelCase ( self : Optional[int] , __lowerCamelCase : Any=None , __lowerCamelCase : str=None , **__lowerCamelCase : Union[str, Any] ) -> List[Any]:
a , a , a = super()._sanitize_parameters(**__lowerCamelCase )
if src_lang is not None:
a = src_lang
if tgt_lang is not None:
a = tgt_lang
if src_lang is None and tgt_lang is None:
# Backward compatibility, direct arguments use is preferred.
a = kwargs.get("task" , self.task )
a = task.split("_" )
if task and len(__lowerCamelCase ) == 4:
# translation, XX, to YY
a = items[1]
a = items[3]
return preprocess_params, forward_params, postprocess_params
def __call__( self : List[str] , *__lowerCamelCase : int , **__lowerCamelCase : List[str] ) -> Dict:
return super().__call__(*__lowerCamelCase , **__lowerCamelCase )
| 107
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : int = logging.get_logger(__name__)
snake_case_ : str = {}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''llama'''
UpperCAmelCase__ : Dict = ['''past_key_values''']
def __init__( self : str , _snake_case : List[str]=32000 , _snake_case : int=4096 , _snake_case : List[str]=11008 , _snake_case : Optional[int]=32 , _snake_case : List[Any]=32 , _snake_case : Tuple=None , _snake_case : int="silu" , _snake_case : List[Any]=2048 , _snake_case : List[str]=0.0_2 , _snake_case : Any=1e-6 , _snake_case : List[str]=True , _snake_case : Optional[Any]=0 , _snake_case : Dict=1 , _snake_case : List[Any]=2 , _snake_case : str=1 , _snake_case : Union[str, Any]=False , _snake_case : str=None , **_snake_case : List[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = max_position_embeddings
UpperCAmelCase_ = hidden_size
UpperCAmelCase_ = intermediate_size
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
# for backward compatibility
if num_key_value_heads is None:
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = num_key_value_heads
UpperCAmelCase_ = hidden_act
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = rms_norm_eps
UpperCAmelCase_ = pretraining_tp
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = rope_scaling
self._rope_scaling_validation()
super().__init__(
pad_token_id=_snake_case , bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case , )
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
if self.rope_scaling is None:
return
if not isinstance(self.rope_scaling , _snake_case) or len(self.rope_scaling) != 2:
raise ValueError(
'''`rope_scaling` must be a dictionary with with two fields, `name` and `factor`, '''
F"""got {self.rope_scaling}""")
UpperCAmelCase_ = self.rope_scaling.get('''type''' , _snake_case)
UpperCAmelCase_ = self.rope_scaling.get('''factor''' , _snake_case)
if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic"]:
raise ValueError(
F"""`rope_scaling`'s name field must be one of ['linear', 'dynamic'], got {rope_scaling_type}""")
if rope_scaling_factor is None or not isinstance(_snake_case , _snake_case) or rope_scaling_factor <= 1.0:
raise ValueError(F"""`rope_scaling`'s factor field must be an float > 1, got {rope_scaling_factor}""")
| 51
| 0
|
"""simple docstring"""
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_tokenizers_available,
is_torch_available,
is_vision_available,
)
lowerCAmelCase__ = {
'''configuration_layoutlmv2''': ['''LAYOUTLMV2_PRETRAINED_CONFIG_ARCHIVE_MAP''', '''LayoutLMv2Config'''],
'''processing_layoutlmv2''': ['''LayoutLMv2Processor'''],
'''tokenization_layoutlmv2''': ['''LayoutLMv2Tokenizer'''],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowerCAmelCase__ = ['''LayoutLMv2TokenizerFast''']
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowerCAmelCase__ = ['''LayoutLMv2FeatureExtractor''']
lowerCAmelCase__ = ['''LayoutLMv2ImageProcessor''']
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowerCAmelCase__ = [
'''LAYOUTLMV2_PRETRAINED_MODEL_ARCHIVE_LIST''',
'''LayoutLMv2ForQuestionAnswering''',
'''LayoutLMv2ForSequenceClassification''',
'''LayoutLMv2ForTokenClassification''',
'''LayoutLMv2Layer''',
'''LayoutLMv2Model''',
'''LayoutLMv2PreTrainedModel''',
]
if TYPE_CHECKING:
from .configuration_layoutlmva import LAYOUTLMV2_PRETRAINED_CONFIG_ARCHIVE_MAP, LayoutLMvaConfig
from .processing_layoutlmva import LayoutLMvaProcessor
from .tokenization_layoutlmva import LayoutLMvaTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_layoutlmva_fast import LayoutLMvaTokenizerFast
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .feature_extraction_layoutlmva import LayoutLMvaFeatureExtractor, LayoutLMvaImageProcessor
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_layoutlmva import (
LAYOUTLMV2_PRETRAINED_MODEL_ARCHIVE_LIST,
LayoutLMvaForQuestionAnswering,
LayoutLMvaForSequenceClassification,
LayoutLMvaForTokenClassification,
LayoutLMvaLayer,
LayoutLMvaModel,
LayoutLMvaPreTrainedModel,
)
else:
import sys
lowerCAmelCase__ = _LazyModule(__name__, globals()['''__file__'''], _import_structure, module_spec=__spec__)
| 108
|
from collections import OrderedDict
from typing import Any, List, Mapping, Optional
from ... import PreTrainedTokenizer, TensorType, is_torch_available
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfigWithPast, PatchingSpec
from ...utils import logging
snake_case_ : List[str] = logging.get_logger(__name__)
snake_case_ : Tuple = {
"Salesforce/codegen-350M-nl": "https://huggingface.co/Salesforce/codegen-350M-nl/resolve/main/config.json",
"Salesforce/codegen-350M-multi": "https://huggingface.co/Salesforce/codegen-350M-multi/resolve/main/config.json",
"Salesforce/codegen-350M-mono": "https://huggingface.co/Salesforce/codegen-350M-mono/resolve/main/config.json",
"Salesforce/codegen-2B-nl": "https://huggingface.co/Salesforce/codegen-2B-nl/resolve/main/config.json",
"Salesforce/codegen-2B-multi": "https://huggingface.co/Salesforce/codegen-2B-multi/resolve/main/config.json",
"Salesforce/codegen-2B-mono": "https://huggingface.co/Salesforce/codegen-2B-mono/resolve/main/config.json",
"Salesforce/codegen-6B-nl": "https://huggingface.co/Salesforce/codegen-6B-nl/resolve/main/config.json",
"Salesforce/codegen-6B-multi": "https://huggingface.co/Salesforce/codegen-6B-multi/resolve/main/config.json",
"Salesforce/codegen-6B-mono": "https://huggingface.co/Salesforce/codegen-6B-mono/resolve/main/config.json",
"Salesforce/codegen-16B-nl": "https://huggingface.co/Salesforce/codegen-16B-nl/resolve/main/config.json",
"Salesforce/codegen-16B-multi": "https://huggingface.co/Salesforce/codegen-16B-multi/resolve/main/config.json",
"Salesforce/codegen-16B-mono": "https://huggingface.co/Salesforce/codegen-16B-mono/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''codegen'''
UpperCAmelCase__ : int = {
'''max_position_embeddings''': '''n_positions''',
'''hidden_size''': '''n_embd''',
'''num_attention_heads''': '''n_head''',
'''num_hidden_layers''': '''n_layer''',
}
def __init__( self : Union[str, Any] , _snake_case : Union[str, Any]=50400 , _snake_case : Optional[int]=2048 , _snake_case : Union[str, Any]=2048 , _snake_case : List[str]=4096 , _snake_case : Any=28 , _snake_case : List[str]=16 , _snake_case : int=64 , _snake_case : Tuple=None , _snake_case : Dict="gelu_new" , _snake_case : Union[str, Any]=0.0 , _snake_case : Optional[Any]=0.0 , _snake_case : List[Any]=0.0 , _snake_case : List[Any]=1e-5 , _snake_case : List[str]=0.0_2 , _snake_case : Optional[Any]=True , _snake_case : int=50256 , _snake_case : Tuple=50256 , _snake_case : int=False , **_snake_case : Any , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = n_ctx
UpperCAmelCase_ = n_positions
UpperCAmelCase_ = n_embd
UpperCAmelCase_ = n_layer
UpperCAmelCase_ = n_head
UpperCAmelCase_ = n_inner
UpperCAmelCase_ = rotary_dim
UpperCAmelCase_ = activation_function
UpperCAmelCase_ = resid_pdrop
UpperCAmelCase_ = embd_pdrop
UpperCAmelCase_ = attn_pdrop
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
super().__init__(
bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case)
class __snake_case ( a ):
def __init__( self : Tuple , _snake_case : PretrainedConfig , _snake_case : str = "default" , _snake_case : List[PatchingSpec] = None , _snake_case : bool = False , ):
"""simple docstring"""
super().__init__(_snake_case , task=_snake_case , patching_specs=_snake_case , use_past=_snake_case)
if not getattr(self._config , '''pad_token_id''' , _snake_case):
# TODO: how to do that better?
UpperCAmelCase_ = 0
@property
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = OrderedDict({'''input_ids''': {0: '''batch''', 1: '''sequence'''}})
if self.use_past:
self.fill_with_past_key_values_(_snake_case , direction='''inputs''')
UpperCAmelCase_ = {0: '''batch''', 1: '''past_sequence + sequence'''}
else:
UpperCAmelCase_ = {0: '''batch''', 1: '''sequence'''}
return common_inputs
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return self._config.n_layer
@property
def lowerCamelCase ( self : int):
"""simple docstring"""
return self._config.n_head
def lowerCamelCase ( self : Optional[int] , _snake_case : PreTrainedTokenizer , _snake_case : int = -1 , _snake_case : int = -1 , _snake_case : bool = False , _snake_case : Optional[TensorType] = None , ):
"""simple docstring"""
UpperCAmelCase_ = super(_snake_case , self).generate_dummy_inputs(
_snake_case , batch_size=_snake_case , seq_length=_snake_case , is_pair=_snake_case , framework=_snake_case)
# We need to order the input in the way they appears in the forward()
UpperCAmelCase_ = OrderedDict({'''input_ids''': common_inputs['''input_ids''']})
# Need to add the past_keys
if self.use_past:
if not is_torch_available():
raise ValueError('''Cannot generate dummy past_keys inputs without PyTorch installed.''')
else:
import torch
UpperCAmelCase_ , UpperCAmelCase_ = common_inputs['''input_ids'''].shape
# Not using the same length for past_key_values
UpperCAmelCase_ = seqlen + 2
UpperCAmelCase_ = (
batch,
self.num_attention_heads,
past_key_values_length,
self._config.hidden_size // self.num_attention_heads,
)
UpperCAmelCase_ = [
(torch.zeros(_snake_case), torch.zeros(_snake_case)) for _ in range(self.num_layers)
]
UpperCAmelCase_ = common_inputs['''attention_mask''']
if self.use_past:
UpperCAmelCase_ = ordered_inputs['''attention_mask'''].dtype
UpperCAmelCase_ = torch.cat(
[ordered_inputs['''attention_mask'''], torch.ones(_snake_case , _snake_case , dtype=_snake_case)] , dim=1)
return ordered_inputs
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return 13
| 51
| 0
|
"""simple docstring"""
from ...configuration_utils import PretrainedConfig
from ...utils import logging
A: Tuple = logging.get_logger(__name__)
A: Any = {
"unc-nlp/lxmert-base-uncased": "https://huggingface.co/unc-nlp/lxmert-base-uncased/resolve/main/config.json",
}
class SCREAMING_SNAKE_CASE__ ( UpperCAmelCase__ ):
__lowerCAmelCase : Union[str, Any] = 'lxmert'
__lowerCAmelCase : Dict = {}
def __init__( self , _SCREAMING_SNAKE_CASE=30522 , _SCREAMING_SNAKE_CASE=768 , _SCREAMING_SNAKE_CASE=12 , _SCREAMING_SNAKE_CASE=9500 , _SCREAMING_SNAKE_CASE=1600 , _SCREAMING_SNAKE_CASE=400 , _SCREAMING_SNAKE_CASE=3072 , _SCREAMING_SNAKE_CASE="gelu" , _SCREAMING_SNAKE_CASE=0.1 , _SCREAMING_SNAKE_CASE=0.1 , _SCREAMING_SNAKE_CASE=512 , _SCREAMING_SNAKE_CASE=2 , _SCREAMING_SNAKE_CASE=0.02 , _SCREAMING_SNAKE_CASE=1E-12 , _SCREAMING_SNAKE_CASE=9 , _SCREAMING_SNAKE_CASE=5 , _SCREAMING_SNAKE_CASE=5 , _SCREAMING_SNAKE_CASE=2048 , _SCREAMING_SNAKE_CASE=4 , _SCREAMING_SNAKE_CASE=6.67 , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=True , **_SCREAMING_SNAKE_CASE , ) -> str:
'''simple docstring'''
UpperCAmelCase : List[Any] = vocab_size
UpperCAmelCase : Dict = hidden_size
UpperCAmelCase : Any = num_attention_heads
UpperCAmelCase : str = hidden_act
UpperCAmelCase : Tuple = intermediate_size
UpperCAmelCase : List[str] = hidden_dropout_prob
UpperCAmelCase : Union[str, Any] = attention_probs_dropout_prob
UpperCAmelCase : str = max_position_embeddings
UpperCAmelCase : str = type_vocab_size
UpperCAmelCase : Optional[Any] = initializer_range
UpperCAmelCase : str = layer_norm_eps
UpperCAmelCase : int = num_qa_labels
UpperCAmelCase : Optional[Any] = num_object_labels
UpperCAmelCase : List[str] = num_attr_labels
UpperCAmelCase : Dict = l_layers
UpperCAmelCase : Optional[int] = x_layers
UpperCAmelCase : Optional[Any] = r_layers
UpperCAmelCase : int = visual_feat_dim
UpperCAmelCase : Tuple = visual_pos_dim
UpperCAmelCase : str = visual_loss_normalizer
UpperCAmelCase : int = task_matched
UpperCAmelCase : Dict = task_mask_lm
UpperCAmelCase : Union[str, Any] = task_obj_predict
UpperCAmelCase : str = task_qa
UpperCAmelCase : List[str] = visual_obj_loss
UpperCAmelCase : str = visual_attr_loss
UpperCAmelCase : Tuple = visual_feat_loss
UpperCAmelCase : Any = {"""vision""": r_layers, """cross_encoder""": x_layers, """language""": l_layers}
super().__init__(**_SCREAMING_SNAKE_CASE )
| 109
|
import os
import unittest
from transformers.models.phobert.tokenization_phobert import VOCAB_FILES_NAMES, PhobertTokenizer
from ...test_tokenization_common import TokenizerTesterMixin
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : Any = PhobertTokenizer
UpperCAmelCase__ : List[str] = False
def lowerCamelCase ( self : str):
"""simple docstring"""
super().setUp()
# Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt
UpperCAmelCase_ = ['''T@@''', '''i''', '''I''', '''R@@''', '''r''', '''e@@''']
UpperCAmelCase_ = dict(zip(_snake_case , range(len(_snake_case))))
UpperCAmelCase_ = ['''#version: 0.2''', '''l à</w>''']
UpperCAmelCase_ = {'''unk_token''': '''<unk>'''}
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''vocab_file'''])
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''merges_file'''])
with open(self.vocab_file , '''w''' , encoding='''utf-8''') as fp:
for token in vocab_tokens:
fp.write(F"""{token} {vocab_tokens[token]}\n""")
with open(self.merges_file , '''w''' , encoding='''utf-8''') as fp:
fp.write('''\n'''.join(_snake_case))
def lowerCamelCase ( self : int , **_snake_case : Any):
"""simple docstring"""
kwargs.update(self.special_tokens_map)
return PhobertTokenizer.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Optional[Any] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T<unk> i <unk> <unk> <unk> <unk> <unk> <unk> I Re<unk> e<unk> <unk> <unk> <unk>'''
return input_text, output_text
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = PhobertTokenizer(self.vocab_file , self.merges_file , **self.special_tokens_map)
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T@@ ô@@ i l@@ à V@@ i@@ n@@ A@@ I R@@ e@@ s@@ e@@ a@@ r@@ c@@ h'''.split()
UpperCAmelCase_ = tokenizer.tokenize(_snake_case)
print(_snake_case)
self.assertListEqual(_snake_case , _snake_case)
UpperCAmelCase_ = tokens + [tokenizer.unk_token]
UpperCAmelCase_ = [4, 3, 5, 3, 3, 3, 3, 3, 3, 6, 7, 9, 3, 9, 3, 3, 3, 3, 3]
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case) , _snake_case)
| 51
| 0
|
from __future__ import annotations
from collections.abc import Iterator
class _a :
def __init__( self: List[str] , UpperCamelCase_: int ) -> None:
"""simple docstring"""
lowercase__ = value
lowercase__ = None
lowercase__ = None
class _a :
def __init__( self: Union[str, Any] , UpperCamelCase_: Node ) -> None:
"""simple docstring"""
lowercase__ = tree
def lowerCamelCase_ ( self: Any , UpperCamelCase_: Node | None ) -> int:
"""simple docstring"""
if node is None:
return 0
return node.value + (
self.depth_first_search(node.left ) + self.depth_first_search(node.right )
)
def __iter__( self: List[str] ) -> Iterator[int]:
"""simple docstring"""
yield self.depth_first_search(self.tree )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 110
|
from typing import List, Optional, TypeVar
from .arrow_dataset import Dataset, _concatenate_map_style_datasets, _interleave_map_style_datasets
from .dataset_dict import DatasetDict, IterableDatasetDict
from .info import DatasetInfo
from .iterable_dataset import IterableDataset, _concatenate_iterable_datasets, _interleave_iterable_datasets
from .splits import NamedSplit
from .utils import logging
from .utils.py_utils import Literal
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[int] = TypeVar("DatasetType", Dataset, IterableDataset)
def A (__A : List[DatasetType] , __A : Optional[List[float]] = None , __A : Optional[int] = None , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : Literal["first_exhausted", "all_exhausted"] = "first_exhausted" , ) -> DatasetType:
"""simple docstring"""
from .arrow_dataset import Dataset
from .iterable_dataset import IterableDataset
if not datasets:
raise ValueError('''Unable to interleave an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if stopping_strategy not in ["first_exhausted", "all_exhausted"]:
raise ValueError(F"""{stopping_strategy} is not supported. Please enter a valid stopping_strategy.""" )
if dataset_type is Dataset:
return _interleave_map_style_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
else:
return _interleave_iterable_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
def A (__A : List[DatasetType] , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : int = 0 , ) -> DatasetType:
"""simple docstring"""
if not dsets:
raise ValueError('''Unable to concatenate an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if dataset_type is Dataset:
return _concatenate_map_style_datasets(__A , info=__A , split=__A , axis=__A )
else:
return _concatenate_iterable_datasets(__A , info=__A , split=__A , axis=__A )
| 51
| 0
|
"""simple docstring"""
import copy
from dataclasses import dataclass, field
from typing import ClassVar, Dict
from ..features import ClassLabel, Features, Value
from .base import TaskTemplate
@dataclass(frozen=A__ )
class _SCREAMING_SNAKE_CASE ( A__ ):
# `task` is not a ClassVar since we want it to be part of the `asdict` output for JSON serialization
UpperCAmelCase_ :str = field(default="text-classification" , metadata={"include_in_asdict_even_if_is_default": True} )
UpperCAmelCase_ :ClassVar[Features] = Features({"text": Value("string" )} )
UpperCAmelCase_ :ClassVar[Features] = Features({"labels": ClassLabel} )
UpperCAmelCase_ :str = "text"
UpperCAmelCase_ :str = "labels"
def __lowerCAmelCase ( self , __A ) -> List[Any]:
if self.label_column not in features:
raise ValueError(f"""Column {self.label_column} is not present in features.""" )
if not isinstance(features[self.label_column] , _snake_case ):
raise ValueError(f"""Column {self.label_column} is not a ClassLabel.""" )
lowerCAmelCase_ :List[str] = copy.deepcopy(self )
lowerCAmelCase_ :Any = self.label_schema.copy()
lowerCAmelCase_ :Any = features[self.label_column]
lowerCAmelCase_ :Optional[Any] = label_schema
return task_template
@property
def __lowerCAmelCase ( self ) -> Union[str, Any]:
return {
self.text_column: "text",
self.label_column: "labels",
}
| 84
|
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
snake_case_ : Optional[Any] = "pt"
elif is_tf_available():
snake_case_ : Union[str, Any] = "tf"
else:
snake_case_ : str = "jax"
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = ByTaTokenizer
UpperCAmelCase__ : int = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().setUp()
UpperCAmelCase_ = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname)
@cached_property
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
return ByTaTokenizer.from_pretrained('''google/byt5-small''')
def lowerCamelCase ( self : List[str] , **_snake_case : Union[str, Any]):
"""simple docstring"""
return self.tokenizer_class.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Dict , _snake_case : int , _snake_case : Tuple=False , _snake_case : Dict=20 , _snake_case : Optional[Any]=5):
"""simple docstring"""
UpperCAmelCase_ = []
for i in range(len(_snake_case)):
try:
UpperCAmelCase_ = tokenizer.decode([i] , clean_up_tokenization_spaces=_snake_case)
except UnicodeDecodeError:
pass
toks.append((i, tok))
UpperCAmelCase_ = list(filter(lambda _snake_case: re.match(r'''^[ a-zA-Z]+$''' , t[1]) , _snake_case))
UpperCAmelCase_ = list(filter(lambda _snake_case: [t[0]] == tokenizer.encode(t[1] , add_special_tokens=_snake_case) , _snake_case))
if max_length is not None and len(_snake_case) > max_length:
UpperCAmelCase_ = toks[:max_length]
if min_length is not None and len(_snake_case) < min_length and len(_snake_case) > 0:
while len(_snake_case) < min_length:
UpperCAmelCase_ = toks + toks
# toks_str = [t[1] for t in toks]
UpperCAmelCase_ = [t[0] for t in toks]
# Ensure consistency
UpperCAmelCase_ = tokenizer.decode(_snake_case , clean_up_tokenization_spaces=_snake_case)
if " " not in output_txt and len(_snake_case) > 1:
UpperCAmelCase_ = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=_snake_case)
+ ''' '''
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=_snake_case)
)
if with_prefix_space:
UpperCAmelCase_ = ''' ''' + output_txt
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
return output_txt, output_ids
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = tokenizer(['''hi</s>''', '''I went to the gym</s>''', '''</s>'''])
UpperCAmelCase_ = tokenizer(['''hi''', '''I went to the gym''', ''''''])
self.assertListEqual(batch_with_eos_added['''input_ids'''] , batch_without_eos_added['''input_ids'''])
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = '''Unicode €.'''
UpperCAmelCase_ = tokenizer(_snake_case)
UpperCAmelCase_ = [88, 113, 108, 102, 114, 103, 104, 35, 229, 133, 175, 49, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''Unicode €.</s>''')
UpperCAmelCase_ = tokenizer('''e è é ê ë''')
UpperCAmelCase_ = [104, 35, 198, 171, 35, 198, 172, 35, 198, 173, 35, 198, 174, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''e è é ê ë</s>''')
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('''e è é ê ë''')) , '''e è é ê ë</s>''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 1, 0]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
if FRAMEWORK != "jax":
UpperCAmelCase_ = list(batch.input_ids.numpy()[0])
else:
UpperCAmelCase_ = list(batch.input_ids.tolist()[0])
self.assertListEqual(_snake_case , _snake_case)
self.assertEqual((2, 37) , batch.input_ids.shape)
self.assertEqual((2, 37) , batch.attention_mask.shape)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
# check if input_ids are returned and no decoder_input_ids
self.assertIn('''input_ids''' , _snake_case)
self.assertIn('''attention_mask''' , _snake_case)
self.assertNotIn('''decoder_input_ids''' , _snake_case)
self.assertNotIn('''decoder_attention_mask''' , _snake_case)
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = [
'''Summary of the text.''',
'''Another summary.''',
]
UpperCAmelCase_ = tokenizer(
text_target=_snake_case , max_length=32 , padding='''max_length''' , truncation=_snake_case , return_tensors=_snake_case)
self.assertEqual(32 , targets['''input_ids'''].shape[1])
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization. </s>''']
UpperCAmelCase_ = ['''Summary of the text. </s>''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 35, 1]
UpperCAmelCase_ = [86, 120, 112, 112, 100, 117, 124, 35, 114, 105, 35, 119, 107, 104, 35, 119, 104, 123, 119, 49, 35, 1]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , text_target=_snake_case)
self.assertEqual(_snake_case , batch['''input_ids'''][0])
self.assertEqual(_snake_case , batch['''labels'''][0])
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
self.assertNotEqual(tokenizer.model_max_length , 42)
# Now let's start the test
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
shutil.rmtree(_snake_case)
UpperCAmelCase_ = self.get_tokenizers(model_max_length=42)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
tokenizer.add_tokens(['''bim''', '''bambam'''])
UpperCAmelCase_ = tokenizer.additional_special_tokens
additional_special_tokens.append('''new_additional_special_token''')
tokenizer.add_special_tokens({'''additional_special_tokens''': additional_special_tokens})
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
self.assertIn('''new_additional_special_token''' , after_tokenizer.additional_special_tokens)
self.assertEqual(after_tokenizer.model_max_length , 42)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case , model_max_length=43)
self.assertEqual(tokenizer.model_max_length , 43)
shutil.rmtree(_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
UpperCAmelCase_ = [F"""<extra_id_{i}>""" for i in range(125)]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , )
self.assertIn(
'''an_additional_special_token''' , tokenizer_without_change_in_init.additional_special_tokens)
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['''an_additional_special_token'''] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['''an_additional_special_token'''])) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
UpperCAmelCase_ = added_tokens_extra_ids + [AddedToken('''a_new_additional_special_token''' , lstrip=_snake_case)]
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , additional_special_tokens=_snake_case , )
self.assertIn('''a_new_additional_special_token''' , tokenizer.additional_special_tokens)
self.assertEqual(
['''a_new_additional_special_token'''] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['''a_new_additional_special_token'''])) , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer_class.from_pretrained(_snake_case)
self.assertTrue(tokenizer.decode([255]) == '''''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers(fast=_snake_case , do_lower_case=_snake_case)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = ['''t''', '''h''', '''i''', '''s''', ''' ''', '''i''', '''s''', ''' ''', '''a''', ''' ''', '''t''', '''e''', '''x''', '''t''', '''</s>''']
UpperCAmelCase_ = tokenizer.convert_tokens_to_string(_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = [
'''bos_token''',
'''eos_token''',
'''unk_token''',
'''sep_token''',
'''pad_token''',
'''cls_token''',
'''mask_token''',
]
UpperCAmelCase_ = 0
UpperCAmelCase_ = tokenizer.convert_ids_to_tokens(
_snake_case , skip_special_tokens=_snake_case)
for attr in attributes_list:
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , '''additional_special_tokens_ids''' , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [])
setattr(_snake_case , '''additional_special_tokens_ids''' , [token_id_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [token_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [token_id_to_test_setters])
| 51
| 0
|
'''simple docstring'''
from manim import *
class SCREAMING_SNAKE_CASE( A__ ):
"""simple docstring"""
def A ( self : str ) -> Any:
UpperCAmelCase : Union[str, Any] = Rectangle(height=0.5 , width=0.5 )
UpperCAmelCase : List[Any] = Rectangle(height=0.46 , width=0.46 ).set_stroke(width=0 )
UpperCAmelCase : Optional[Any] = [mem.copy() for i in range(6 )]
UpperCAmelCase : Tuple = [mem.copy() for i in range(6 )]
UpperCAmelCase : int = VGroup(*_snake_case ).arrange(_snake_case , buff=0 )
UpperCAmelCase : List[str] = VGroup(*_snake_case ).arrange(_snake_case , buff=0 )
UpperCAmelCase : Dict = VGroup(_snake_case , _snake_case ).arrange(_snake_case , buff=0 )
UpperCAmelCase : Union[str, Any] = Text('''CPU''' , font_size=24 )
UpperCAmelCase : Optional[int] = Group(_snake_case , _snake_case ).arrange(_snake_case , buff=0.5 , aligned_edge=_snake_case )
cpu.move_to([-2.5, -0.5, 0] )
self.add(_snake_case )
UpperCAmelCase : Optional[Any] = [mem.copy() for i in range(1 )]
UpperCAmelCase : List[str] = VGroup(*_snake_case ).arrange(_snake_case , buff=0 )
UpperCAmelCase : List[Any] = Text('''GPU''' , font_size=24 )
UpperCAmelCase : Optional[int] = Group(_snake_case , _snake_case ).arrange(_snake_case , buff=0.5 , aligned_edge=_snake_case )
gpu.align_to(_snake_case , _snake_case )
gpu.set_x(gpu.get_x() - 1 )
self.add(_snake_case )
UpperCAmelCase : List[Any] = [mem.copy() for i in range(6 )]
UpperCAmelCase : Union[str, Any] = VGroup(*_snake_case ).arrange(_snake_case , buff=0 )
UpperCAmelCase : List[str] = Text('''Model''' , font_size=24 )
UpperCAmelCase : List[str] = Group(_snake_case , _snake_case ).arrange(_snake_case , buff=0.5 , aligned_edge=_snake_case )
model.move_to([3, -1.0, 0] )
self.play(
Create(_snake_case , run_time=1 ) , Create(_snake_case , run_time=1 ) , Create(_snake_case , run_time=1 ) , )
UpperCAmelCase : Tuple = MarkupText(
F"""First, an empty model skeleton is loaded\ninto <span fgcolor='{YELLOW}'>memory</span> without using much RAM.""" , font_size=24 , )
UpperCAmelCase : str = Square(side_length=2.2 )
key.move_to([-5, 2, 0] )
UpperCAmelCase : int = MarkupText(
F"""<b>Key:</b>\n\n<span fgcolor='{YELLOW}'>●</span> Empty Model""" , font_size=18 , )
key_text.move_to([-5, 2.4, 0] )
step_a.move_to([2, 2, 0] )
self.play(Write(_snake_case , run_time=2.5 ) , Write(_snake_case ) , Write(_snake_case ) )
self.add(_snake_case )
UpperCAmelCase : Optional[int] = []
UpperCAmelCase : List[Any] = []
UpperCAmelCase : List[str] = []
for i, rect in enumerate(_snake_case ):
UpperCAmelCase : Tuple = Rectangle(height=0.46 , width=0.46 ).set_stroke(width=0.0 ).set_fill(_snake_case , opacity=0.7 )
cpu_target.move_to(_snake_case )
cpu_target.generate_target()
UpperCAmelCase : Dict = 0.46 / 4
UpperCAmelCase : Optional[int] = 0.46 / 3
if i == 0:
cpu_target.target.next_to(cpu_left_col_base[0].get_corner(DOWN + LEFT ) , buff=0.02 , direction=_snake_case )
cpu_target.target.set_x(cpu_target.target.get_x() + 0.1 )
elif i == 3:
cpu_target.target.next_to(cpu_targs[0].target , direction=_snake_case , buff=0.0 )
else:
cpu_target.target.next_to(cpu_targs[i - 1].target , direction=_snake_case , buff=0.0 )
cpu_targs.append(_snake_case )
first_animations.append(rect.animate(run_time=0.5 ).set_stroke(_snake_case ) )
second_animations.append(MoveToTarget(_snake_case , run_time=1.5 ) )
self.play(*_snake_case )
self.play(*_snake_case )
self.wait()
| 23
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_sentencepiece_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : Dict = {"configuration_mbart": ["MBART_PRETRAINED_CONFIG_ARCHIVE_MAP", "MBartConfig", "MBartOnnxConfig"]}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Tuple = ["MBartTokenizer"]
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = ["MBartTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[Any] = [
"MBART_PRETRAINED_MODEL_ARCHIVE_LIST",
"MBartForCausalLM",
"MBartForConditionalGeneration",
"MBartForQuestionAnswering",
"MBartForSequenceClassification",
"MBartModel",
"MBartPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TFMBartForConditionalGeneration",
"TFMBartModel",
"TFMBartPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"FlaxMBartForConditionalGeneration",
"FlaxMBartForQuestionAnswering",
"FlaxMBartForSequenceClassification",
"FlaxMBartModel",
"FlaxMBartPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_mbart import MBART_PRETRAINED_CONFIG_ARCHIVE_MAP, MBartConfig, MBartOnnxConfig
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart import MBartTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart_fast import MBartTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_mbart import (
MBART_PRETRAINED_MODEL_ARCHIVE_LIST,
MBartForCausalLM,
MBartForConditionalGeneration,
MBartForQuestionAnswering,
MBartForSequenceClassification,
MBartModel,
MBartPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_mbart import TFMBartForConditionalGeneration, TFMBartModel, TFMBartPreTrainedModel
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_mbart import (
FlaxMBartForConditionalGeneration,
FlaxMBartForQuestionAnswering,
FlaxMBartForSequenceClassification,
FlaxMBartModel,
FlaxMBartPreTrainedModel,
)
else:
import sys
snake_case_ : Any = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
import csv
from collections import defaultdict
from dataclasses import dataclass, field
from typing import List, Optional
import matplotlib.pyplot as plt
import numpy as np
from matplotlib.ticker import ScalarFormatter
from transformers import HfArgumentParser
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase=None , _UpperCAmelCase=None ) -> Dict:
return field(default_factory=lambda: default , metadata=__A )
@dataclass
class lowerCAmelCase :
UpperCAmelCase__ = field(
metadata={"""help""": """The csv file to plot."""}, )
UpperCAmelCase__ = field(
default=__UpperCamelCase, metadata={"""help""": """Whether to plot along batch size or sequence length. Defaults to sequence length."""}, )
UpperCAmelCase__ = field(
default=__UpperCamelCase, metadata={"""help""": """Whether the csv file has time results or memory results. Defaults to memory results."""}, )
UpperCAmelCase__ = field(
default=__UpperCamelCase, metadata={"""help""": """Disable logarithmic scale when plotting"""}, )
UpperCAmelCase__ = field(
default=__UpperCamelCase, metadata={
"""help""": """Whether the csv file has training results or inference results. Defaults to inference results."""
}, )
UpperCAmelCase__ = field(
default=__UpperCamelCase, metadata={"""help""": """Filename under which the plot will be saved. If unused no plot is saved."""}, )
UpperCAmelCase__ = list_field(
default=__UpperCamelCase, metadata={"""help""": """List of model names that are used instead of the ones in the csv file."""} )
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase ) -> Dict:
try:
int(__A )
return True
except ValueError:
return False
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase ) -> int:
try:
float(__A )
return True
except ValueError:
return False
class lowerCAmelCase :
def __init__( self : Optional[int] , UpperCAmelCase : Union[str, Any] ) -> Dict:
lowerCamelCase__ : List[Any] = args
lowerCamelCase__ : Optional[Any] = defaultdict(lambda: {"bsz": [], "seq_len": [], "result": {}} )
with open(self.args.csv_file , newline='' ) as csv_file:
lowerCamelCase__ : Any = csv.DictReader(_snake_case )
for row in reader:
lowerCamelCase__ : Any = row['model']
self.result_dict[model_name]["bsz"].append(int(row['batch_size'] ) )
self.result_dict[model_name]["seq_len"].append(int(row['sequence_length'] ) )
if can_convert_to_int(row['result'] ):
# value is not None
lowerCamelCase__ : Tuple = int(row['result'] )
elif can_convert_to_float(row['result'] ):
# value is not None
lowerCamelCase__ : List[Any] = float(row['result'] )
def A_ ( self : int ) -> Optional[Any]:
lowerCamelCase__ , lowerCamelCase__ : Tuple = plt.subplots()
lowerCamelCase__ : int = 'Time usage' if self.args.is_time else 'Memory usage'
lowerCamelCase__ : Dict = title_str + ' for training' if self.args.is_train else title_str + ' for inference'
if not self.args.no_log_scale:
# set logarithm scales
ax.set_xscale('log' )
ax.set_yscale('log' )
for axis in [ax.xaxis, ax.yaxis]:
axis.set_major_formatter(ScalarFormatter() )
for model_name_idx, model_name in enumerate(self.result_dict.keys() ):
lowerCamelCase__ : str = sorted(set(self.result_dict[model_name]['bsz'] ) )
lowerCamelCase__ : Optional[int] = sorted(set(self.result_dict[model_name]['seq_len'] ) )
lowerCamelCase__ : Optional[int] = self.result_dict[model_name]['result']
((lowerCamelCase__) , (lowerCamelCase__)) : int = (
(batch_sizes, sequence_lengths) if self.args.plot_along_batch else (sequence_lengths, batch_sizes)
)
lowerCamelCase__ : int = (
model_name if self.args.short_model_names is None else self.args.short_model_names[model_name_idx]
)
for inner_loop_value in inner_loop_array:
if self.args.plot_along_batch:
lowerCamelCase__ : Any = np.asarray(
[results[(x, inner_loop_value)] for x in x_axis_array if (x, inner_loop_value) in results] , dtype=_snake_case , )
else:
lowerCamelCase__ : Tuple = np.asarray(
[results[(inner_loop_value, x)] for x in x_axis_array if (inner_loop_value, x) in results] , dtype=np.floataa , )
((lowerCamelCase__) , (lowerCamelCase__)) : Optional[int] = (
('batch_size', 'len') if self.args.plot_along_batch else ('in #tokens', 'bsz')
)
lowerCamelCase__ : str = np.asarray(_snake_case , _snake_case )[: len(_snake_case )]
plt.scatter(
_snake_case , _snake_case , label=F"""{label_model_name} - {inner_loop_label}: {inner_loop_value}""" )
plt.plot(_snake_case , _snake_case , '--' )
title_str += F""" {label_model_name} vs."""
lowerCamelCase__ : Any = title_str[:-4]
lowerCamelCase__ : Union[str, Any] = 'Time in s' if self.args.is_time else 'Memory in MB'
# plot
plt.title(_snake_case )
plt.xlabel(_snake_case )
plt.ylabel(_snake_case )
plt.legend()
if self.args.figure_png_file is not None:
plt.savefig(self.args.figure_png_file )
else:
plt.show()
def SCREAMING_SNAKE_CASE ( ) -> Union[str, Any]:
lowerCamelCase__ : int = HfArgumentParser(__A )
lowerCamelCase__ : Any = parser.parse_args_into_dataclasses()[0]
lowerCamelCase__ : Optional[Any] = Plot(args=__A )
plot.plot()
if __name__ == "__main__":
main()
| 50
|
import warnings
from typing import List, Optional, Union
from ...image_utils import ImageInput
from ...processing_utils import ProcessorMixin
from ...tokenization_utils_base import BatchEncoding, PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy
from ...utils import TensorType
class __snake_case ( a ):
UpperCAmelCase__ : Dict = ['''image_processor''', '''tokenizer''']
UpperCAmelCase__ : Dict = '''FlavaImageProcessor'''
UpperCAmelCase__ : Dict = ('''BertTokenizer''', '''BertTokenizerFast''')
def __init__( self : Union[str, Any] , _snake_case : List[str]=None , _snake_case : str=None , **_snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = None
if "feature_extractor" in kwargs:
warnings.warn(
'''The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor`'''
''' instead.''' , _snake_case , )
UpperCAmelCase_ = kwargs.pop('''feature_extractor''')
UpperCAmelCase_ = image_processor if image_processor is not None else feature_extractor
if image_processor is None:
raise ValueError('''You need to specify an `image_processor`.''')
if tokenizer is None:
raise ValueError('''You need to specify a `tokenizer`.''')
super().__init__(_snake_case , _snake_case)
UpperCAmelCase_ = self.image_processor
def __call__( self : List[Any] , _snake_case : Optional[ImageInput] = None , _snake_case : Optional[Union[TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]]] = None , _snake_case : bool = True , _snake_case : Union[bool, str, PaddingStrategy] = False , _snake_case : Union[bool, str, TruncationStrategy] = False , _snake_case : Optional[int] = None , _snake_case : int = 0 , _snake_case : Optional[int] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = True , _snake_case : Optional[Union[str, TensorType]] = None , **_snake_case : Any , ):
"""simple docstring"""
if text is None and images is None:
raise ValueError('''You have to specify either text or images. Both cannot be none.''')
if text is not None:
UpperCAmelCase_ = self.tokenizer(
text=_snake_case , add_special_tokens=_snake_case , padding=_snake_case , truncation=_snake_case , max_length=_snake_case , stride=_snake_case , pad_to_multiple_of=_snake_case , return_token_type_ids=_snake_case , return_attention_mask=_snake_case , return_overflowing_tokens=_snake_case , return_special_tokens_mask=_snake_case , return_offsets_mapping=_snake_case , return_length=_snake_case , verbose=_snake_case , return_tensors=_snake_case , **_snake_case , )
if images is not None:
UpperCAmelCase_ = self.image_processor(
_snake_case , return_image_mask=_snake_case , return_codebook_pixels=_snake_case , return_tensors=_snake_case , **_snake_case , )
if text is not None and images is not None:
encoding.update(_snake_case)
return encoding
elif text is not None:
return encoding
else:
return BatchEncoding(data=dict(**_snake_case) , tensor_type=_snake_case)
def lowerCamelCase ( self : Any , *_snake_case : Optional[Any] , **_snake_case : int):
"""simple docstring"""
return self.tokenizer.batch_decode(*_snake_case , **_snake_case)
def lowerCamelCase ( self : Optional[int] , *_snake_case : int , **_snake_case : Dict):
"""simple docstring"""
return self.tokenizer.decode(*_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.tokenizer.model_input_names
UpperCAmelCase_ = self.image_processor.model_input_names
return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names))
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
warnings.warn(
'''`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.''' , _snake_case , )
return self.image_processor_class
@property
def lowerCamelCase ( self : Any):
"""simple docstring"""
warnings.warn(
'''`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.''' , _snake_case , )
return self.image_processor
| 51
| 0
|
import os
import socket
from contextlib import contextmanager
import torch
from ..commands.config.default import write_basic_config # noqa: F401
from ..state import PartialState
from .dataclasses import DistributedType
from .imports import is_deepspeed_available, is_tpu_available
from .transformer_engine import convert_model
from .versions import is_torch_version
if is_deepspeed_available():
from deepspeed import DeepSpeedEngine
if is_tpu_available(check_device=False):
import torch_xla.core.xla_model as xm
def lowerCAmelCase_ ( _lowercase : int) -> List[Any]:
"""simple docstring"""
if is_torch_version("""<""" , """2.0.0""") or not hasattr(__A , """_dynamo"""):
return False
return isinstance(__A , torch._dynamo.eval_frame.OptimizedModule)
def lowerCAmelCase_ ( _lowercase : List[Any] , _lowercase : bool = True) -> Tuple:
"""simple docstring"""
a__ : Any = (torch.nn.parallel.DistributedDataParallel, torch.nn.DataParallel)
a__ : Dict = is_compiled_module(__A)
if is_compiled:
a__ : str = model
a__ : Tuple = model._orig_mod
if is_deepspeed_available():
options += (DeepSpeedEngine,)
while isinstance(__A , __A):
a__ : List[Any] = model.module
if not keep_fpaa_wrapper:
a__ : int = getattr(__A , """forward""")
a__ : Dict = model.__dict__.pop("""_original_forward""" , __A)
if original_forward is not None:
while hasattr(__A , """__wrapped__"""):
a__ : Union[str, Any] = forward.__wrapped__
if forward == original_forward:
break
a__ : Optional[Any] = forward
if getattr(__A , """_converted_to_transformer_engine""" , __A):
convert_model(__A , to_transformer_engine=__A)
if is_compiled:
a__ : int = model
a__ : Optional[Any] = compiled_model
return model
def lowerCAmelCase_ ( ) -> List[str]:
"""simple docstring"""
PartialState().wait_for_everyone()
def lowerCAmelCase_ ( _lowercase : Dict , _lowercase : Union[str, Any]) -> List[str]:
"""simple docstring"""
if PartialState().distributed_type == DistributedType.TPU:
xm.save(__A , __A)
elif PartialState().local_process_index == 0:
torch.save(__A , __A)
@contextmanager
def lowerCAmelCase_ ( **_lowercase : Union[str, Any]) -> Dict:
"""simple docstring"""
for key, value in kwargs.items():
a__ : int = str(__A)
yield
for key in kwargs:
if key.upper() in os.environ:
del os.environ[key.upper()]
def lowerCAmelCase_ ( _lowercase : List[Any]) -> Any:
"""simple docstring"""
if not hasattr(__A , """__qualname__""") and not hasattr(__A , """__name__"""):
a__ : int = getattr(__A , """__class__""" , __A)
if hasattr(__A , """__qualname__"""):
return obj.__qualname__
if hasattr(__A , """__name__"""):
return obj.__name__
return str(__A)
def lowerCAmelCase_ ( _lowercase : Tuple , _lowercase : str) -> List[str]:
"""simple docstring"""
for key, value in source.items():
if isinstance(__A , __A):
a__ : Dict = destination.setdefault(__A , {})
merge_dicts(__A , __A)
else:
a__ : Dict = value
return destination
def lowerCAmelCase_ ( _lowercase : int = None) -> bool:
"""simple docstring"""
if port is None:
a__ : Any = 2_9500
with socket.socket(socket.AF_INET , socket.SOCK_STREAM) as s:
return s.connect_ex(("""localhost""", port)) == 0
| 170
|
from diffusers.utils.testing_utils import require_onnxruntime
@require_onnxruntime
class __snake_case :
pass
| 51
| 0
|
import warnings
from functools import wraps
from typing import Callable
def A_ ( snake_case : Callable ) -> Callable:
'''simple docstring'''
@wraps(__A )
def _inner_fn(*snake_case : Dict , **snake_case : int ):
warnings.warn(
(f"'{fn.__name__}' is experimental and might be subject to breaking changes in the future.") , __A , )
return fn(*__A , **__A )
return _inner_fn
| 328
|
import argparse
import os
import pickle
import sys
import torch
from transformers import TransfoXLConfig, TransfoXLLMHeadModel, load_tf_weights_in_transfo_xl
from transformers.models.transfo_xl import tokenization_transfo_xl as data_utils
from transformers.models.transfo_xl.tokenization_transfo_xl import CORPUS_NAME, VOCAB_FILES_NAMES
from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging
logging.set_verbosity_info()
# We do this to be able to load python 2 datasets pickles
# See e.g. https://stackoverflow.com/questions/2121874/python-pickling-after-changing-a-modules-directory/2121918#2121918
snake_case_ : List[Any] = data_utils.TransfoXLTokenizer
snake_case_ : int = data_utils.TransfoXLCorpus
snake_case_ : List[Any] = data_utils
snake_case_ : int = data_utils
def A (__A : Dict , __A : List[Any] , __A : Union[str, Any] , __A : Tuple ) -> Union[str, Any]:
"""simple docstring"""
if transfo_xl_dataset_file:
# Convert a pre-processed corpus (see original TensorFlow repo)
with open(__A , '''rb''' ) as fp:
UpperCAmelCase_ = pickle.load(__A , encoding='''latin1''' )
# Save vocabulary and dataset cache as Dictionaries (should be better than pickles for the long-term)
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + VOCAB_FILES_NAMES['''pretrained_vocab_file''']
print(F"""Save vocabulary to {pytorch_vocab_dump_path}""" )
UpperCAmelCase_ = corpus.vocab.__dict__
torch.save(__A , __A )
UpperCAmelCase_ = corpus.__dict__
corpus_dict_no_vocab.pop('''vocab''' , __A )
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + CORPUS_NAME
print(F"""Save dataset to {pytorch_dataset_dump_path}""" )
torch.save(__A , __A )
if tf_checkpoint_path:
# Convert a pre-trained TensorFlow model
UpperCAmelCase_ = os.path.abspath(__A )
UpperCAmelCase_ = os.path.abspath(__A )
print(F"""Converting Transformer XL checkpoint from {tf_path} with config at {config_path}.""" )
# Initialise PyTorch model
if transfo_xl_config_file == "":
UpperCAmelCase_ = TransfoXLConfig()
else:
UpperCAmelCase_ = TransfoXLConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = TransfoXLLMHeadModel(__A )
UpperCAmelCase_ = load_tf_weights_in_transfo_xl(__A , __A , __A )
# Save pytorch-model
UpperCAmelCase_ = os.path.join(__A , __A )
UpperCAmelCase_ = os.path.join(__A , __A )
print(F"""Save PyTorch model to {os.path.abspath(__A )}""" )
torch.save(model.state_dict() , __A )
print(F"""Save configuration file to {os.path.abspath(__A )}""" )
with open(__A , '''w''' , encoding='''utf-8''' ) as f:
f.write(config.to_json_string() )
if __name__ == "__main__":
snake_case_ : List[str] = argparse.ArgumentParser()
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
type=str,
required=True,
help="Path to the folder to store the PyTorch model or dataset/vocab.",
)
parser.add_argument(
"--tf_checkpoint_path",
default="",
type=str,
help="An optional path to a TensorFlow checkpoint path to be converted.",
)
parser.add_argument(
"--transfo_xl_config_file",
default="",
type=str,
help=(
"An optional config json file corresponding to the pre-trained BERT model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--transfo_xl_dataset_file",
default="",
type=str,
help="An optional dataset file to be converted in a vocabulary.",
)
snake_case_ : int = parser.parse_args()
convert_transfo_xl_checkpoint_to_pytorch(
args.tf_checkpoint_path,
args.transfo_xl_config_file,
args.pytorch_dump_folder_path,
args.transfo_xl_dataset_file,
)
| 51
| 0
|
'''simple docstring'''
import argparse
import pickle
import numpy as np
import torch
from torch import nn
from transformers import ReformerConfig, ReformerModelWithLMHead
from transformers.utils import logging
logging.set_verbosity_info()
def UpperCAmelCase__ ( UpperCAmelCase__, UpperCAmelCase__, UpperCAmelCase__=None ) -> List[Any]:
assert torch_layer.weight.shape == weight.shape, F'''{torch_layer} layer.weight does not match'''
A_ = nn.Parameter(__A )
if bias is not None:
assert torch_layer.bias.shape == bias.shape, F'''{torch_layer} layer.bias does not match'''
A_ = nn.Parameter(__A )
def UpperCAmelCase__ ( UpperCAmelCase__, UpperCAmelCase__, UpperCAmelCase__ ) -> Tuple:
A_ = np.asarray(weights[0] )
A_ = np.asarray(weights[1] )
A_ = np.asarray(weights[2] )
set_param(
torch_layer.self_attention.query_key, torch.tensor(__A ).transpose(1, 2 ).contiguous().view(-1, __A ), )
set_param(
torch_layer.self_attention.value, torch.tensor(__A ).transpose(1, 2 ).contiguous().view(-1, __A ), )
set_param(
torch_layer.output.dense, torch.tensor(__A ).view(-1, __A ).contiguous().transpose(0, 1 ), )
def UpperCAmelCase__ ( UpperCAmelCase__, UpperCAmelCase__, UpperCAmelCase__ ) -> int:
A_ = np.asarray(weights[0] )
A_ = np.asarray(weights[1] )
A_ = np.asarray(weights[2] )
A_ = np.asarray(weights[3] )
set_param(
torch_layer.self_attention.query, torch.tensor(__A ).transpose(1, 2 ).contiguous().view(-1, __A ), )
set_param(
torch_layer.self_attention.key, torch.tensor(__A ).transpose(1, 2 ).contiguous().view(-1, __A ), )
set_param(
torch_layer.self_attention.value, torch.tensor(__A ).transpose(1, 2 ).contiguous().view(-1, __A ), )
set_param(
torch_layer.output.dense, torch.tensor(__A ).view(-1, __A ).contiguous().transpose(0, 1 ), )
def UpperCAmelCase__ ( UpperCAmelCase__, UpperCAmelCase__, UpperCAmelCase__ ) -> List[Any]:
A_ = weights[0][0][0]
A_ = np.asarray(layer_norm_a[0] )
A_ = np.asarray(layer_norm_a[1] )
set_param(
torch_block.attention.layer_norm, torch.tensor(__A ), torch.tensor(__A ), )
# lsh weights + output
A_ = weights[0][1]
if len(__A ) < 4:
set_layer_weights_in_torch_lsh(__A, torch_block.attention, __A )
else:
set_layer_weights_in_torch_local(__A, torch_block.attention, __A )
# intermediate weighs
A_ = weights[2][0][1][2]
# Chunked Feed Forward
if len(__A ) == 4:
A_ = intermediate_weights[2]
# layernorm 2
A_ = np.asarray(intermediate_weights[0][0] )
A_ = np.asarray(intermediate_weights[0][1] )
set_param(
torch_block.feed_forward.layer_norm, torch.tensor(__A ), torch.tensor(__A ), )
# intermediate dense
A_ = np.asarray(intermediate_weights[1][0] )
A_ = np.asarray(intermediate_weights[1][1] )
set_param(
torch_block.feed_forward.dense.dense, torch.tensor(__A ).transpose(0, 1 ).contiguous(), torch.tensor(__A ), )
# intermediate out
A_ = np.asarray(intermediate_weights[4][0] )
A_ = np.asarray(intermediate_weights[4][1] )
set_param(
torch_block.feed_forward.output.dense, torch.tensor(__A ).transpose(0, 1 ).contiguous(), torch.tensor(__A ), )
def UpperCAmelCase__ ( UpperCAmelCase__, UpperCAmelCase__, UpperCAmelCase__ ) -> Tuple:
A_ = torch_model.reformer
# word embeds
A_ = np.asarray(weights[1] )
set_param(
torch_model_reformer.embeddings.word_embeddings, torch.tensor(__A ), )
if isinstance(weights[3], __A ):
A_ = torch_model_reformer.embeddings.position_embeddings
for emb_idx in range(len(position_embeddings.weights ) ):
A_ = np.asarray(weights[3][emb_idx][0] )
assert (
position_embeddings.weights[emb_idx].shape == emb_weights.shape
), F'''{position_embeddings[emb_idx]} emb does not match'''
A_ = nn.Parameter(torch.tensor(__A ) )
A_ = weights[5]
assert len(torch_model_reformer.encoder.layers ) * 4 == len(
__A ), "HF and trax model do not have the same number of layers"
for layer_idx, layer in enumerate(torch_model_reformer.encoder.layers ):
A_ = trax_layer_weights[4 * layer_idx : 4 * (layer_idx + 1)]
set_block_weights_in_torch(__A, __A, __A )
# output layer norm
A_ = np.asarray(weights[7][0] )
A_ = np.asarray(weights[7][1] )
set_param(
torch_model_reformer.encoder.layer_norm, torch.tensor(__A ), torch.tensor(__A ), )
# output embeddings
A_ = np.asarray(weights[9][0] )
A_ = np.asarray(weights[9][1] )
set_param(
torch_model.lm_head.decoder, torch.tensor(__A ).transpose(0, 1 ).contiguous(), torch.tensor(__A ), )
def UpperCAmelCase__ ( UpperCAmelCase__, UpperCAmelCase__, UpperCAmelCase__ ) -> Union[str, Any]:
A_ = ReformerConfig.from_json_file(__A )
print(F'''Building PyTorch model from configuration: {config}''' )
A_ = ReformerModelWithLMHead(__A )
with open(__A, """rb""" ) as f:
A_ = pickle.load(__A )["""weights"""]
set_model_weights_in_torch(__A, __A, config.hidden_size )
# Save pytorch-model
print(F'''Save PyTorch model to {pytorch_dump_path}''' )
torch.save(model.state_dict(), __A )
if __name__ == "__main__":
__lowerCamelCase = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'''--trax_model_pkl_path''', default=None, type=str, required=True, help='''Path to the TensorFlow checkpoint path.'''
)
parser.add_argument(
'''--config_file''',
default=None,
type=str,
required=True,
help=(
'''The config json file corresponding to the pre-trained Reformer model. \n'''
'''This specifies the model architecture.'''
),
)
parser.add_argument(
'''--pytorch_dump_path''', default=None, type=str, required=True, help='''Path to the output PyTorch model.'''
)
__lowerCamelCase = parser.parse_args()
convert_trax_checkpoint_to_pytorch(args.trax_model_pkl_path, args.config_file, args.pytorch_dump_path)
| 162
|
from typing import Optional, Tuple, Union
import torch
from einops import rearrange, reduce
from diffusers import DDIMScheduler, DDPMScheduler, DiffusionPipeline, ImagePipelineOutput, UNetaDConditionModel
from diffusers.schedulers.scheduling_ddim import DDIMSchedulerOutput
from diffusers.schedulers.scheduling_ddpm import DDPMSchedulerOutput
snake_case_ : List[str] = 8
def A (__A : Union[str, Any] , __A : List[Any]=BITS ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x * 255).int().clamp(0 , 255 )
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b c h w -> b c 1 h w''' )
UpperCAmelCase_ = ((x & mask) != 0).float()
UpperCAmelCase_ = rearrange(__A , '''b c d h w -> b (c d) h w''' )
UpperCAmelCase_ = bits * 2 - 1
return bits
def A (__A : Dict , __A : Tuple=BITS ) -> List[str]:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x > 0).int()
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A , dtype=torch.intaa )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b (c d) h w -> b c d h w''' , d=8 )
UpperCAmelCase_ = reduce(x * mask , '''b c d h w -> b c h w''' , '''sum''' )
return (dec / 255).clamp(0.0 , 1.0 )
def A (self : List[Any] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : float = 0.0 , __A : bool = True , __A : Tuple=None , __A : bool = True , ) -> Union[DDIMSchedulerOutput, Tuple]:
"""simple docstring"""
if self.num_inference_steps is None:
raise ValueError(
'''Number of inference steps is \'None\', you need to run \'set_timesteps\' after creating the scheduler''' )
# See formulas (12) and (16) of DDIM paper https://arxiv.org/pdf/2010.02502.pdf
# Ideally, read DDIM paper in-detail understanding
# Notation (<variable name> -> <name in paper>
# - pred_noise_t -> e_theta(x_t, t)
# - pred_original_sample -> f_theta(x_t, t) or x_0
# - std_dev_t -> sigma_t
# - eta -> η
# - pred_sample_direction -> "direction pointing to x_t"
# - pred_prev_sample -> "x_t-1"
# 1. get previous step value (=t-1)
UpperCAmelCase_ = timestep - self.config.num_train_timesteps // self.num_inference_steps
# 2. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[timestep]
UpperCAmelCase_ = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
UpperCAmelCase_ = 1 - alpha_prod_t
# 3. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
# 4. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 5. compute variance: "sigma_t(η)" -> see formula (16)
# σ_t = sqrt((1 − α_t−1)/(1 − α_t)) * sqrt(1 − α_t/α_t−1)
UpperCAmelCase_ = self._get_variance(__A , __A )
UpperCAmelCase_ = eta * variance ** 0.5
if use_clipped_model_output:
# the model_output is always re-derived from the clipped x_0 in Glide
UpperCAmelCase_ = (sample - alpha_prod_t ** 0.5 * pred_original_sample) / beta_prod_t ** 0.5
# 6. compute "direction pointing to x_t" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (1 - alpha_prod_t_prev - std_dev_t**2) ** 0.5 * model_output
# 7. compute x_t without "random noise" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = alpha_prod_t_prev ** 0.5 * pred_original_sample + pred_sample_direction
if eta > 0:
# randn_like does not support generator https://github.com/pytorch/pytorch/issues/27072
UpperCAmelCase_ = model_output.device if torch.is_tensor(__A ) else '''cpu'''
UpperCAmelCase_ = torch.randn(model_output.shape , dtype=model_output.dtype , generator=__A ).to(__A )
UpperCAmelCase_ = self._get_variance(__A , __A ) ** 0.5 * eta * noise
UpperCAmelCase_ = prev_sample + variance
if not return_dict:
return (prev_sample,)
return DDIMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
def A (self : Optional[int] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : int="epsilon" , __A : Optional[Any]=None , __A : bool = True , ) -> Union[DDPMSchedulerOutput, Tuple]:
"""simple docstring"""
UpperCAmelCase_ = timestep
if model_output.shape[1] == sample.shape[1] * 2 and self.variance_type in ["learned", "learned_range"]:
UpperCAmelCase_ , UpperCAmelCase_ = torch.split(__A , sample.shape[1] , dim=1 )
else:
UpperCAmelCase_ = None
# 1. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[t]
UpperCAmelCase_ = self.alphas_cumprod[t - 1] if t > 0 else self.one
UpperCAmelCase_ = 1 - alpha_prod_t
UpperCAmelCase_ = 1 - alpha_prod_t_prev
# 2. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
if prediction_type == "epsilon":
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
elif prediction_type == "sample":
UpperCAmelCase_ = model_output
else:
raise ValueError(F"""Unsupported prediction_type {prediction_type}.""" )
# 3. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = (alpha_prod_t_prev ** 0.5 * self.betas[t]) / beta_prod_t
UpperCAmelCase_ = self.alphas[t] ** 0.5 * beta_prod_t_prev / beta_prod_t
# 5. Compute predicted previous sample µ_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * sample
# 6. Add noise
UpperCAmelCase_ = 0
if t > 0:
UpperCAmelCase_ = torch.randn(
model_output.size() , dtype=model_output.dtype , layout=model_output.layout , generator=__A ).to(model_output.device )
UpperCAmelCase_ = (self._get_variance(__A , predicted_variance=__A ) ** 0.5) * noise
UpperCAmelCase_ = pred_prev_sample + variance
if not return_dict:
return (pred_prev_sample,)
return DDPMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
class __snake_case ( a ):
def __init__( self : Union[str, Any] , _snake_case : UNetaDConditionModel , _snake_case : Union[DDIMScheduler, DDPMScheduler] , _snake_case : Optional[float] = 1.0 , ):
"""simple docstring"""
super().__init__()
UpperCAmelCase_ = bit_scale
UpperCAmelCase_ = (
ddim_bit_scheduler_step if isinstance(_snake_case , _snake_case) else ddpm_bit_scheduler_step
)
self.register_modules(unet=_snake_case , scheduler=_snake_case)
@torch.no_grad()
def __call__( self : Union[str, Any] , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 50 , _snake_case : Optional[torch.Generator] = None , _snake_case : Optional[int] = 1 , _snake_case : Optional[str] = "pil" , _snake_case : bool = True , **_snake_case : Optional[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = torch.randn(
(batch_size, self.unet.config.in_channels, height, width) , generator=_snake_case , )
UpperCAmelCase_ = decimal_to_bits(_snake_case) * self.bit_scale
UpperCAmelCase_ = latents.to(self.device)
self.scheduler.set_timesteps(_snake_case)
for t in self.progress_bar(self.scheduler.timesteps):
# predict the noise residual
UpperCAmelCase_ = self.unet(_snake_case , _snake_case).sample
# compute the previous noisy sample x_t -> x_t-1
UpperCAmelCase_ = self.scheduler.step(_snake_case , _snake_case , _snake_case).prev_sample
UpperCAmelCase_ = bits_to_decimal(_snake_case)
if output_type == "pil":
UpperCAmelCase_ = self.numpy_to_pil(_snake_case)
if not return_dict:
return (image,)
return ImagePipelineOutput(images=_snake_case)
| 51
| 0
|
'''simple docstring'''
from ...configuration_utils import PretrainedConfig
from ...utils import logging
__lowercase : List[Any] = logging.get_logger(__name__)
__lowercase : int = {
"transfo-xl-wt103": "https://huggingface.co/transfo-xl-wt103/resolve/main/config.json",
}
class __lowercase ( _lowercase ):
lowerCamelCase : Optional[int] = '''transfo-xl'''
lowerCamelCase : Optional[Any] = ['''mems''']
lowerCamelCase : Optional[Any] = {
'''n_token''': '''vocab_size''',
'''hidden_size''': '''d_model''',
'''num_attention_heads''': '''n_head''',
'''num_hidden_layers''': '''n_layer''',
}
def __init__(self , A=2_6_7_7_3_5 , A=[2_0_0_0_0, 4_0_0_0_0, 2_0_0_0_0_0] , A=1_0_2_4 , A=1_0_2_4 , A=1_6 , A=6_4 , A=4_0_9_6 , A=4 , A=False , A=1_8 , A=1_6_0_0 , A=1_0_0_0 , A=True , A=True , A=0 , A=-1 , A=True , A=0.1 , A=0.0 , A=True , A="normal" , A=0.01 , A=0.01 , A=0.02 , A=1E-5 , A=0 , **A , ):
lowerCamelCase_ : int = vocab_size
lowerCamelCase_ : Union[str, Any] = []
self.cutoffs.extend(_snake_case )
if proj_share_all_but_first:
lowerCamelCase_ : Optional[Any] = [False] + [True] * len(self.cutoffs )
else:
lowerCamelCase_ : int = [False] + [False] * len(self.cutoffs )
lowerCamelCase_ : Union[str, Any] = d_model
lowerCamelCase_ : Tuple = d_embed
lowerCamelCase_ : int = d_head
lowerCamelCase_ : List[str] = d_inner
lowerCamelCase_ : int = div_val
lowerCamelCase_ : Optional[int] = pre_lnorm
lowerCamelCase_ : int = n_layer
lowerCamelCase_ : Any = n_head
lowerCamelCase_ : Tuple = mem_len
lowerCamelCase_ : Optional[Any] = same_length
lowerCamelCase_ : Optional[int] = attn_type
lowerCamelCase_ : Optional[int] = clamp_len
lowerCamelCase_ : Optional[int] = sample_softmax
lowerCamelCase_ : Union[str, Any] = adaptive
lowerCamelCase_ : int = dropout
lowerCamelCase_ : Optional[int] = dropatt
lowerCamelCase_ : Tuple = untie_r
lowerCamelCase_ : Any = init
lowerCamelCase_ : Dict = init_range
lowerCamelCase_ : Optional[int] = proj_init_std
lowerCamelCase_ : Union[str, Any] = init_std
lowerCamelCase_ : Tuple = layer_norm_epsilon
super().__init__(eos_token_id=_snake_case , **_snake_case )
@property
def UpperCAmelCase__ (self ):
logger.info(F"""The model {self.model_type} is one of the few models that has no sequence length limit.""" )
return -1
@max_position_embeddings.setter
def UpperCAmelCase__ (self , A ):
raise NotImplementedError(
F"""The model {self.model_type} is one of the few models that has no sequence length limit.""" )
| 318
|
snake_case_ : Dict = {
"Pillow": "Pillow",
"accelerate": "accelerate>=0.11.0",
"compel": "compel==0.1.8",
"black": "black~=23.1",
"datasets": "datasets",
"filelock": "filelock",
"flax": "flax>=0.4.1",
"hf-doc-builder": "hf-doc-builder>=0.3.0",
"huggingface-hub": "huggingface-hub>=0.13.2",
"requests-mock": "requests-mock==1.10.0",
"importlib_metadata": "importlib_metadata",
"invisible-watermark": "invisible-watermark",
"isort": "isort>=5.5.4",
"jax": "jax>=0.2.8,!=0.3.2",
"jaxlib": "jaxlib>=0.1.65",
"Jinja2": "Jinja2",
"k-diffusion": "k-diffusion>=0.0.12",
"torchsde": "torchsde",
"note_seq": "note_seq",
"librosa": "librosa",
"numpy": "numpy",
"omegaconf": "omegaconf",
"parameterized": "parameterized",
"protobuf": "protobuf>=3.20.3,<4",
"pytest": "pytest",
"pytest-timeout": "pytest-timeout",
"pytest-xdist": "pytest-xdist",
"ruff": "ruff>=0.0.241",
"safetensors": "safetensors",
"sentencepiece": "sentencepiece>=0.1.91,!=0.1.92",
"scipy": "scipy",
"onnx": "onnx",
"regex": "regex!=2019.12.17",
"requests": "requests",
"tensorboard": "tensorboard",
"torch": "torch>=1.4",
"torchvision": "torchvision",
"transformers": "transformers>=4.25.1",
"urllib3": "urllib3<=2.0.0",
}
| 51
| 0
|
from __future__ import annotations
import unittest
from transformers import is_tf_available
from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor, random_attention_mask
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import numpy as np
import tensorflow as tf
from transformers import (
TF_FLAUBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
FlaubertConfig,
TFFlaubertForMultipleChoice,
TFFlaubertForQuestionAnsweringSimple,
TFFlaubertForSequenceClassification,
TFFlaubertForTokenClassification,
TFFlaubertModel,
TFFlaubertWithLMHeadModel,
)
class __lowerCamelCase :
def __init__( self: List[str],A_: Tuple,):
'''simple docstring'''
__UpperCamelCase = parent
__UpperCamelCase = 13
__UpperCamelCase = 7
__UpperCamelCase = True
__UpperCamelCase = True
__UpperCamelCase = True
__UpperCamelCase = True
__UpperCamelCase = True
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = False
__UpperCamelCase = 2
__UpperCamelCase = 99
__UpperCamelCase = 0
__UpperCamelCase = 32
__UpperCamelCase = 2
__UpperCamelCase = 4
__UpperCamelCase = 0.1
__UpperCamelCase = 0.1
__UpperCamelCase = 512
__UpperCamelCase = 16
__UpperCamelCase = 2
__UpperCamelCase = 0.0_2
__UpperCamelCase = 3
__UpperCamelCase = 4
__UpperCamelCase = 'last'
__UpperCamelCase = True
__UpperCamelCase = None
__UpperCamelCase = 0
def snake_case_ ( self: int ):
'''simple docstring'''
__UpperCamelCase = ids_tensor([self.batch_size, self.seq_length],self.vocab_size )
__UpperCamelCase = random_attention_mask([self.batch_size, self.seq_length],dtype=tf.floataa )
__UpperCamelCase = None
if self.use_input_lengths:
__UpperCamelCase = (
ids_tensor([self.batch_size],vocab_size=2 ) + self.seq_length - 2
) # small variation of seq_length
__UpperCamelCase = None
if self.use_token_type_ids:
__UpperCamelCase = ids_tensor([self.batch_size, self.seq_length],self.n_langs )
__UpperCamelCase = None
__UpperCamelCase = None
__UpperCamelCase = None
if self.use_labels:
__UpperCamelCase = ids_tensor([self.batch_size],self.type_sequence_label_size )
__UpperCamelCase = ids_tensor([self.batch_size, self.seq_length],self.num_labels )
__UpperCamelCase = ids_tensor([self.batch_size],2,dtype=tf.floataa )
__UpperCamelCase = ids_tensor([self.batch_size],self.num_choices )
__UpperCamelCase = FlaubertConfig(
vocab_size=self.vocab_size,n_special=self.n_special,emb_dim=self.hidden_size,n_layers=self.num_hidden_layers,n_heads=self.num_attention_heads,dropout=self.hidden_dropout_prob,attention_dropout=self.attention_probs_dropout_prob,gelu_activation=self.gelu_activation,sinusoidal_embeddings=self.sinusoidal_embeddings,asm=self.asm,causal=self.causal,n_langs=self.n_langs,max_position_embeddings=self.max_position_embeddings,initializer_range=self.initializer_range,summary_type=self.summary_type,use_proj=self.use_proj,bos_token_id=self.bos_token_id,)
return (
config,
input_ids,
token_type_ids,
input_lengths,
sequence_labels,
token_labels,
is_impossible_labels,
choice_labels,
input_mask,
)
def snake_case_ ( self: Tuple,A_: int,A_: Tuple,A_: Optional[Any],A_: Union[str, Any],A_: Optional[Any],A_: Any,A_: List[Any],A_: Any,A_: Tuple,):
'''simple docstring'''
__UpperCamelCase = TFFlaubertModel(config=_snake_case )
__UpperCamelCase = {'input_ids': input_ids, 'lengths': input_lengths, 'langs': token_type_ids}
__UpperCamelCase = model(_snake_case )
__UpperCamelCase = [input_ids, input_mask]
__UpperCamelCase = model(_snake_case )
self.parent.assertEqual(result.last_hidden_state.shape,(self.batch_size, self.seq_length, self.hidden_size) )
def snake_case_ ( self: Union[str, Any],A_: List[Any],A_: Optional[int],A_: str,A_: Optional[int],A_: int,A_: Optional[Any],A_: Tuple,A_: Optional[int],A_: Union[str, Any],):
'''simple docstring'''
__UpperCamelCase = TFFlaubertWithLMHeadModel(_snake_case )
__UpperCamelCase = {'input_ids': input_ids, 'lengths': input_lengths, 'langs': token_type_ids}
__UpperCamelCase = model(_snake_case )
self.parent.assertEqual(result.logits.shape,(self.batch_size, self.seq_length, self.vocab_size) )
def snake_case_ ( self: str,A_: Optional[int],A_: Optional[int],A_: Optional[int],A_: Dict,A_: str,A_: Any,A_: List[str],A_: str,A_: Optional[int],):
'''simple docstring'''
__UpperCamelCase = TFFlaubertForQuestionAnsweringSimple(_snake_case )
__UpperCamelCase = {'input_ids': input_ids, 'lengths': input_lengths}
__UpperCamelCase = model(_snake_case )
self.parent.assertEqual(result.start_logits.shape,(self.batch_size, self.seq_length) )
self.parent.assertEqual(result.end_logits.shape,(self.batch_size, self.seq_length) )
def snake_case_ ( self: List[str],A_: List[Any],A_: List[str],A_: Optional[Any],A_: List[Any],A_: int,A_: Any,A_: Optional[Any],A_: Dict,A_: int,):
'''simple docstring'''
__UpperCamelCase = TFFlaubertForSequenceClassification(_snake_case )
__UpperCamelCase = {'input_ids': input_ids, 'lengths': input_lengths}
__UpperCamelCase = model(_snake_case )
self.parent.assertEqual(result.logits.shape,(self.batch_size, self.type_sequence_label_size) )
def snake_case_ ( self: int,A_: Any,A_: List[str],A_: Optional[int],A_: Any,A_: Dict,A_: List[Any],A_: Any,A_: Union[str, Any],A_: Union[str, Any],):
'''simple docstring'''
__UpperCamelCase = self.num_labels
__UpperCamelCase = TFFlaubertForTokenClassification(config=_snake_case )
__UpperCamelCase = {'input_ids': input_ids, 'attention_mask': input_mask, 'token_type_ids': token_type_ids}
__UpperCamelCase = model(_snake_case )
self.parent.assertEqual(result.logits.shape,(self.batch_size, self.seq_length, self.num_labels) )
def snake_case_ ( self: List[str],A_: Any,A_: Union[str, Any],A_: Tuple,A_: Optional[int],A_: Dict,A_: Tuple,A_: int,A_: List[str],A_: Tuple,):
'''simple docstring'''
__UpperCamelCase = self.num_choices
__UpperCamelCase = TFFlaubertForMultipleChoice(config=_snake_case )
__UpperCamelCase = tf.tile(tf.expand_dims(_snake_case,1 ),(1, self.num_choices, 1) )
__UpperCamelCase = tf.tile(tf.expand_dims(_snake_case,1 ),(1, self.num_choices, 1) )
__UpperCamelCase = tf.tile(tf.expand_dims(_snake_case,1 ),(1, self.num_choices, 1) )
__UpperCamelCase = {
'input_ids': multiple_choice_inputs_ids,
'attention_mask': multiple_choice_input_mask,
'token_type_ids': multiple_choice_token_type_ids,
}
__UpperCamelCase = model(_snake_case )
self.parent.assertEqual(result.logits.shape,(self.batch_size, self.num_choices) )
def snake_case_ ( self: int ):
'''simple docstring'''
__UpperCamelCase = self.prepare_config_and_inputs()
(
(
__UpperCamelCase
), (
__UpperCamelCase
), (
__UpperCamelCase
), (
__UpperCamelCase
), (
__UpperCamelCase
), (
__UpperCamelCase
), (
__UpperCamelCase
), (
__UpperCamelCase
), (
__UpperCamelCase
),
) = config_and_inputs
__UpperCamelCase = {
'input_ids': input_ids,
'token_type_ids': token_type_ids,
'langs': token_type_ids,
'lengths': input_lengths,
}
return config, inputs_dict
@require_tf
class __lowerCamelCase (_a , _a , unittest.TestCase ):
_lowercase = (
(
TFFlaubertModel,
TFFlaubertWithLMHeadModel,
TFFlaubertForSequenceClassification,
TFFlaubertForQuestionAnsweringSimple,
TFFlaubertForTokenClassification,
TFFlaubertForMultipleChoice,
)
if is_tf_available()
else ()
)
_lowercase = (
(TFFlaubertWithLMHeadModel,) if is_tf_available() else ()
) # TODO (PVP): Check other models whether language generation is also applicable
_lowercase = (
{
'''feature-extraction''': TFFlaubertModel,
'''fill-mask''': TFFlaubertWithLMHeadModel,
'''question-answering''': TFFlaubertForQuestionAnsweringSimple,
'''text-classification''': TFFlaubertForSequenceClassification,
'''token-classification''': TFFlaubertForTokenClassification,
'''zero-shot''': TFFlaubertForSequenceClassification,
}
if is_tf_available()
else {}
)
_lowercase = False
_lowercase = False
def snake_case_ ( self: Union[str, Any],A_: Optional[Any],A_: int,A_: Tuple,A_: Dict,A_: List[Any] ):
'''simple docstring'''
if (
pipeline_test_casse_name == "QAPipelineTests"
and tokenizer_name is not None
and not tokenizer_name.endswith('Fast' )
):
# `QAPipelineTests` fails for a few models when the slower tokenizer are used.
# (The slower tokenizers were never used for pipeline tests before the pipeline testing rework)
# TODO: check (and possibly fix) the `QAPipelineTests` with slower tokenizer
return True
return False
def snake_case_ ( self: Dict ):
'''simple docstring'''
__UpperCamelCase = TFFlaubertModelTester(self )
__UpperCamelCase = ConfigTester(self,config_class=_snake_case,emb_dim=37 )
def snake_case_ ( self: int ):
'''simple docstring'''
self.config_tester.run_common_tests()
def snake_case_ ( self: int ):
'''simple docstring'''
__UpperCamelCase = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_flaubert_model(*_snake_case )
def snake_case_ ( self: List[str] ):
'''simple docstring'''
__UpperCamelCase = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_flaubert_lm_head(*_snake_case )
def snake_case_ ( self: Optional[Any] ):
'''simple docstring'''
__UpperCamelCase = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_flaubert_qa(*_snake_case )
def snake_case_ ( self: Any ):
'''simple docstring'''
__UpperCamelCase = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_flaubert_sequence_classif(*_snake_case )
def snake_case_ ( self: int ):
'''simple docstring'''
__UpperCamelCase = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_flaubert_for_token_classification(*_snake_case )
def snake_case_ ( self: Optional[Any] ):
'''simple docstring'''
__UpperCamelCase = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_flaubert_for_multiple_choice(*_snake_case )
@slow
def snake_case_ ( self: Optional[Any] ):
'''simple docstring'''
for model_name in TF_FLAUBERT_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
__UpperCamelCase = TFFlaubertModel.from_pretrained(_snake_case )
self.assertIsNotNone(_snake_case )
@require_tf
@require_sentencepiece
@require_tokenizers
class __lowerCamelCase (unittest.TestCase ):
@slow
def snake_case_ ( self: int ):
'''simple docstring'''
__UpperCamelCase = TFFlaubertModel.from_pretrained('jplu/tf-flaubert-small-cased' )
__UpperCamelCase = tf.convert_to_tensor(
[[0, 158, 735, 2592, 1424, 6727, 82, 1]],dtype=tf.intaa,) # "J'aime flaubert !"
__UpperCamelCase = model(_snake_case )[0]
__UpperCamelCase = tf.TensorShape((1, 8, 512) )
self.assertEqual(output.shape,_snake_case )
# compare the actual values for a slice.
__UpperCamelCase = tf.convert_to_tensor(
[
[
[-1.8_7_6_8_7_7_3, -1.5_6_6_5_5_5, 0.2_7_0_7_2_4_1_8],
[-1.6_9_2_0_0_3_8, -0.5_8_7_3_5_0_5, 1.9_3_2_9_5_9_9],
[-2.9_5_6_3_9_8_5, -1.6_9_9_3_8_3_5, 1.7_9_7_2_0_5_2],
]
],dtype=tf.floataa,)
self.assertTrue(np.allclose(output[:, :3, :3].numpy(),expected_slice.numpy(),atol=1E-4 ) )
| 310
|
from datetime import datetime
import requests
def A (__A : str ) -> bytes:
"""simple docstring"""
UpperCAmelCase_ = '''https://downloadgram.net/wp-json/wppress/video-downloader/video?url='''
UpperCAmelCase_ = requests.get(base_url + url ).json()[0]['''urls'''][0]['''src''']
return requests.get(__A ).content
if __name__ == "__main__":
snake_case_ : Optional[Any] = input("Enter Video/IGTV url: ").strip()
snake_case_ : Any = f"{datetime.now():%Y-%m-%d_%H:%M:%S}.mp4"
with open(file_name, "wb") as fp:
fp.write(download_video(url))
print(f"Done. Video saved to disk as {file_name}.")
| 51
| 0
|
"""simple docstring"""
def __lowerCamelCase ( __UpperCamelCase , __UpperCamelCase = False ) -> bool:
"""simple docstring"""
if n == 2:
return True
if not n % 2 or n < 2:
return False
if n > 5 and n % 10 not in (1, 3, 7, 9): # can quickly check last digit
return False
if n > 3317044064679887385961981 and not allow_probable:
raise ValueError(
"Warning: upper bound of deterministic test is exceeded. "
"Pass allow_probable=True to allow probabilistic test. "
"A return value of True indicates a probable prime." )
# array bounds provided by analysis
lowerCAmelCase_ : Optional[int] = [
2047,
1373653,
25326001,
3215031751,
2152302898747,
3474749660383,
341550071728321,
1,
3825123056546413051,
1,
1,
318665857834031151167461,
3317044064679887385961981,
]
lowerCAmelCase_ : Union[str, Any] = [2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41]
for idx, _p in enumerate(__A , 1 ):
if n < _p:
# then we have our last prime to check
lowerCAmelCase_ : int = primes[:idx]
break
lowerCAmelCase_ , lowerCAmelCase_ : Any = n - 1, 0
# break up n -1 into a power of 2 (s) and
# remaining odd component
# essentially, solve for d * 2 ** s == n - 1
while d % 2 == 0:
d //= 2
s += 1
for prime in plist:
lowerCAmelCase_ : Any = False
for r in range(__A ):
lowerCAmelCase_ : Optional[Any] = pow(__A , d * 2**r , __A )
# see article for analysis explanation for m
if (r == 0 and m == 1) or ((m + 1) % n == 0):
lowerCAmelCase_ : Optional[int] = True
# this loop will not determine compositeness
break
if pr:
continue
# if pr is False, then the above loop never evaluated to true,
# and the n MUST be composite
return False
return True
def __lowerCamelCase ( ) -> None:
"""simple docstring"""
assert not miller_rabin(561 )
assert miller_rabin(563 )
# 2047
assert not miller_rabin(838201 )
assert miller_rabin(838207 )
# 1_373_653
assert not miller_rabin(17316001 )
assert miller_rabin(17316017 )
# 25_326_001
assert not miller_rabin(3078386641 )
assert miller_rabin(3078386653 )
# 3_215_031_751
assert not miller_rabin(1713045574801 )
assert miller_rabin(1713045574819 )
# 2_152_302_898_747
assert not miller_rabin(2779799728307 )
assert miller_rabin(2779799728327 )
# 3_474_749_660_383
assert not miller_rabin(113850023909441 )
assert miller_rabin(113850023909527 )
# 341_550_071_728_321
assert not miller_rabin(1275041018848804351 )
assert miller_rabin(1275041018848804391 )
# 3_825_123_056_546_413_051
assert not miller_rabin(79666464458507787791867 )
assert miller_rabin(79666464458507787791951 )
# 318_665_857_834_031_151_167_461
assert not miller_rabin(552840677446647897660333 )
assert miller_rabin(552840677446647897660359 )
# 3_317_044_064_679_887_385_961_981
# upper limit for probabilistic test
if __name__ == "__main__":
test_miller_rabin()
| 241
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[Any] = {
"tiiuae/falcon-40b": "https://huggingface.co/tiiuae/falcon-40b/resolve/main/config.json",
"tiiuae/falcon-7b": "https://huggingface.co/tiiuae/falcon-7b/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : Optional[Any] = '''falcon'''
UpperCAmelCase__ : List[Any] = ['''past_key_values''']
def __init__( self : Union[str, Any] , _snake_case : List[str]=65024 , _snake_case : int=4544 , _snake_case : int=32 , _snake_case : Any=71 , _snake_case : int=1e-5 , _snake_case : Dict=0.0_2 , _snake_case : int=True , _snake_case : List[Any]=0.0 , _snake_case : Tuple=0.0 , _snake_case : int=None , _snake_case : Tuple=False , _snake_case : Any=False , _snake_case : str=True , _snake_case : Any=True , _snake_case : List[str]=False , _snake_case : Tuple=11 , _snake_case : Dict=11 , **_snake_case : Optional[int] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
# Backward compatibility with n_embed kwarg
UpperCAmelCase_ = kwargs.pop('''n_embed''' , _snake_case)
UpperCAmelCase_ = hidden_size if n_embed is None else n_embed
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = hidden_dropout
UpperCAmelCase_ = attention_dropout
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
UpperCAmelCase_ = num_attention_heads if num_kv_heads is None else num_kv_heads
UpperCAmelCase_ = alibi
UpperCAmelCase_ = new_decoder_architecture
UpperCAmelCase_ = multi_query # Ignored when new_decoder_architecture is True
UpperCAmelCase_ = parallel_attn
UpperCAmelCase_ = bias
super().__init__(bos_token_id=_snake_case , eos_token_id=_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
return self.hidden_size // self.num_attention_heads
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return not self.alibi
| 51
| 0
|
'''simple docstring'''
from __future__ import annotations
import numpy as np
from numpy import floataa
from numpy.typing import NDArray
def lowerCAmelCase (__A , __A , __A , __A , ):
"""simple docstring"""
_a , _a = coefficient_matrix.shape
_a , _a = constant_matrix.shape
if rowsa != colsa:
_a = F'''Coefficient matrix dimensions must be nxn but received {rowsa}x{colsa}'''
raise ValueError(__A)
if colsa != 1:
_a = F'''Constant matrix must be nx1 but received {rowsa}x{colsa}'''
raise ValueError(__A)
if rowsa != rowsa:
_a = (
'''Coefficient and constant matrices dimensions must be nxn and nx1 but '''
F'''received {rowsa}x{colsa} and {rowsa}x{colsa}'''
)
raise ValueError(__A)
if len(__A) != rowsa:
_a = (
'''Number of initial values must be equal to number of rows in coefficient '''
F'''matrix but received {len(__A)} and {rowsa}'''
)
raise ValueError(__A)
if iterations <= 0:
raise ValueError('''Iterations must be at least 1''')
_a = np.concatenate(
(coefficient_matrix, constant_matrix) , axis=1)
_a , _a = table.shape
strictly_diagonally_dominant(__A)
# Iterates the whole matrix for given number of times
for _ in range(__A):
_a = []
for row in range(__A):
_a = 0
for col in range(__A):
if col == row:
_a = table[row][col]
elif col == cols - 1:
_a = table[row][col]
else:
temp += (-1) * table[row][col] * init_val[col]
_a = (temp + val) / denom
new_val.append(__A)
_a = new_val
return [float(__A) for i in new_val]
def lowerCAmelCase (__A):
"""simple docstring"""
_a , _a = table.shape
_a = True
for i in range(0 , __A):
_a = 0
for j in range(0 , cols - 1):
if i == j:
continue
else:
total += table[i][j]
if table[i][i] <= total:
raise ValueError('''Coefficient matrix is not strictly diagonally dominant''')
return is_diagonally_dominant
# Test Cases
if __name__ == "__main__":
import doctest
doctest.testmod()
| 211
|
from __future__ import annotations
import time
from math import sqrt
# 1 for manhattan, 0 for euclidean
snake_case_ : str = 0
snake_case_ : Union[str, Any] = [
[0, 0, 0, 0, 0, 0, 0],
[0, 1, 0, 0, 0, 0, 0], # 0 are free path whereas 1's are obstacles
[0, 0, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 0],
[1, 0, 1, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 0, 0],
]
snake_case_ : str = [[-1, 0], [0, -1], [1, 0], [0, 1]] # up, left, down, right
snake_case_ : List[Any] = tuple[int, int]
class __snake_case :
def __init__( self : Any , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : int , _snake_case : Node | None , ):
"""simple docstring"""
UpperCAmelCase_ = pos_x
UpperCAmelCase_ = pos_y
UpperCAmelCase_ = (pos_y, pos_x)
UpperCAmelCase_ = goal_x
UpperCAmelCase_ = goal_y
UpperCAmelCase_ = g_cost
UpperCAmelCase_ = parent
UpperCAmelCase_ = self.calculate_heuristic()
UpperCAmelCase_ = self.g_cost + self.h_cost
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = self.pos_x - self.goal_x
UpperCAmelCase_ = self.pos_y - self.goal_y
if HEURISTIC == 1:
return abs(_snake_case) + abs(_snake_case)
else:
return sqrt(dy**2 + dx**2)
def __lt__( self : Union[str, Any] , _snake_case : Node):
"""simple docstring"""
return self.f_cost < other.f_cost
class __snake_case :
def __init__( self : str , _snake_case : TPosition , _snake_case : TPosition):
"""simple docstring"""
UpperCAmelCase_ = Node(start[1] , start[0] , goal[1] , goal[0] , 0 , _snake_case)
UpperCAmelCase_ = Node(goal[1] , goal[0] , goal[1] , goal[0] , 99999 , _snake_case)
UpperCAmelCase_ = [self.start]
UpperCAmelCase_ = []
UpperCAmelCase_ = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
while self.open_nodes:
# Open Nodes are sorted using __lt__
self.open_nodes.sort()
UpperCAmelCase_ = self.open_nodes.pop(0)
if current_node.pos == self.target.pos:
return self.retrace_path(_snake_case)
self.closed_nodes.append(_snake_case)
UpperCAmelCase_ = self.get_successors(_snake_case)
for child_node in successors:
if child_node in self.closed_nodes:
continue
if child_node not in self.open_nodes:
self.open_nodes.append(_snake_case)
else:
# retrieve the best current path
UpperCAmelCase_ = self.open_nodes.pop(self.open_nodes.index(_snake_case))
if child_node.g_cost < better_node.g_cost:
self.open_nodes.append(_snake_case)
else:
self.open_nodes.append(_snake_case)
return [self.start.pos]
def lowerCamelCase ( self : Tuple , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = []
for action in delta:
UpperCAmelCase_ = parent.pos_x + action[1]
UpperCAmelCase_ = parent.pos_y + action[0]
if not (0 <= pos_x <= len(grid[0]) - 1 and 0 <= pos_y <= len(_snake_case) - 1):
continue
if grid[pos_y][pos_x] != 0:
continue
successors.append(
Node(
_snake_case , _snake_case , self.target.pos_y , self.target.pos_x , parent.g_cost + 1 , _snake_case , ))
return successors
def lowerCamelCase ( self : Any , _snake_case : Node | None):
"""simple docstring"""
UpperCAmelCase_ = node
UpperCAmelCase_ = []
while current_node is not None:
path.append((current_node.pos_y, current_node.pos_x))
UpperCAmelCase_ = current_node.parent
path.reverse()
return path
class __snake_case :
def __init__( self : Any , _snake_case : TPosition , _snake_case : TPosition):
"""simple docstring"""
UpperCAmelCase_ = AStar(_snake_case , _snake_case)
UpperCAmelCase_ = AStar(_snake_case , _snake_case)
UpperCAmelCase_ = False
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
while self.fwd_astar.open_nodes or self.bwd_astar.open_nodes:
self.fwd_astar.open_nodes.sort()
self.bwd_astar.open_nodes.sort()
UpperCAmelCase_ = self.fwd_astar.open_nodes.pop(0)
UpperCAmelCase_ = self.bwd_astar.open_nodes.pop(0)
if current_bwd_node.pos == current_fwd_node.pos:
return self.retrace_bidirectional_path(
_snake_case , _snake_case)
self.fwd_astar.closed_nodes.append(_snake_case)
self.bwd_astar.closed_nodes.append(_snake_case)
UpperCAmelCase_ = current_bwd_node
UpperCAmelCase_ = current_fwd_node
UpperCAmelCase_ = {
self.fwd_astar: self.fwd_astar.get_successors(_snake_case),
self.bwd_astar: self.bwd_astar.get_successors(_snake_case),
}
for astar in [self.fwd_astar, self.bwd_astar]:
for child_node in successors[astar]:
if child_node in astar.closed_nodes:
continue
if child_node not in astar.open_nodes:
astar.open_nodes.append(_snake_case)
else:
# retrieve the best current path
UpperCAmelCase_ = astar.open_nodes.pop(
astar.open_nodes.index(_snake_case))
if child_node.g_cost < better_node.g_cost:
astar.open_nodes.append(_snake_case)
else:
astar.open_nodes.append(_snake_case)
return [self.fwd_astar.start.pos]
def lowerCamelCase ( self : int , _snake_case : Node , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = self.fwd_astar.retrace_path(_snake_case)
UpperCAmelCase_ = self.bwd_astar.retrace_path(_snake_case)
bwd_path.pop()
bwd_path.reverse()
UpperCAmelCase_ = fwd_path + bwd_path
return path
if __name__ == "__main__":
# all coordinates are given in format [y,x]
snake_case_ : Any = (0, 0)
snake_case_ : Union[str, Any] = (len(grid) - 1, len(grid[0]) - 1)
for elem in grid:
print(elem)
snake_case_ : str = time.time()
snake_case_ : List[str] = AStar(init, goal)
snake_case_ : Optional[int] = a_star.search()
snake_case_ : Optional[Any] = time.time() - start_time
print(f"AStar execution time = {end_time:f} seconds")
snake_case_ : int = time.time()
snake_case_ : Dict = BidirectionalAStar(init, goal)
snake_case_ : str = time.time() - bd_start_time
print(f"BidirectionalAStar execution time = {bd_end_time:f} seconds")
| 51
| 0
|
from argparse import ArgumentParser
from . import BaseTransformersCLICommand
def _snake_case ( lowerCAmelCase : Any ):
"""simple docstring"""
return DownloadCommand(args.model , args.cache_dir , args.force , args.trust_remote_code )
class a__ ( A__ ):
@staticmethod
def __UpperCamelCase ( _A : ArgumentParser ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : Optional[Any] = parser.add_parser("download" )
download_parser.add_argument(
"--cache-dir",type=_snake_case,default=_snake_case,help="Path to location to store the models" )
download_parser.add_argument(
"--force",action="store_true",help="Force the model to be download even if already in cache-dir" )
download_parser.add_argument(
"--trust-remote-code",action="store_true",help="Whether or not to allow for custom models defined on the Hub in their own modeling files. Use only if you\'ve reviewed the code as it will execute on your local machine",)
download_parser.add_argument("model",type=_snake_case,help="Name of the model to download" )
download_parser.set_defaults(func=_snake_case )
def __init__( self : Tuple,_A : str,_A : str,_A : bool,_A : bool ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : Union[str, Any] = model
SCREAMING_SNAKE_CASE_ : List[str] = cache
SCREAMING_SNAKE_CASE_ : Union[str, Any] = force
SCREAMING_SNAKE_CASE_ : Union[str, Any] = trust_remote_code
def __UpperCamelCase ( self : Union[str, Any] ):
"""simple docstring"""
from ..models.auto import AutoModel, AutoTokenizer
AutoModel.from_pretrained(
self._model,cache_dir=self._cache,force_download=self._force,trust_remote_code=self._trust_remote_code )
AutoTokenizer.from_pretrained(
self._model,cache_dir=self._cache,force_download=self._force,trust_remote_code=self._trust_remote_code )
| 18
|
import inspect
import unittest
import numpy as np
from tests.test_modeling_common import floats_tensor
from transformers import DetrConfig, MaskFormerConfig, SwinConfig, is_torch_available, is_vision_available
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import MaskFormerForInstanceSegmentation, MaskFormerModel
if is_vision_available():
from transformers import MaskFormerImageProcessor
if is_vision_available():
from PIL import Image
class __snake_case :
def __init__( self : List[str] , _snake_case : Union[str, Any] , _snake_case : List[str]=2 , _snake_case : Any=True , _snake_case : Any=False , _snake_case : List[str]=10 , _snake_case : Any=3 , _snake_case : Union[str, Any]=32 * 4 , _snake_case : List[Any]=32 * 6 , _snake_case : Tuple=4 , _snake_case : Dict=32 , ):
"""simple docstring"""
UpperCAmelCase_ = parent
UpperCAmelCase_ = batch_size
UpperCAmelCase_ = is_training
UpperCAmelCase_ = use_auxiliary_loss
UpperCAmelCase_ = num_queries
UpperCAmelCase_ = num_channels
UpperCAmelCase_ = min_size
UpperCAmelCase_ = max_size
UpperCAmelCase_ = num_labels
UpperCAmelCase_ = mask_feature_size
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = floats_tensor([self.batch_size, self.num_channels, self.min_size, self.max_size]).to(
_snake_case)
UpperCAmelCase_ = torch.ones([self.batch_size, self.min_size, self.max_size] , device=_snake_case)
UpperCAmelCase_ = (
torch.rand([self.batch_size, self.num_labels, self.min_size, self.max_size] , device=_snake_case) > 0.5
).float()
UpperCAmelCase_ = (torch.rand((self.batch_size, self.num_labels) , device=_snake_case) > 0.5).long()
UpperCAmelCase_ = self.get_config()
return config, pixel_values, pixel_mask, mask_labels, class_labels
def lowerCamelCase ( self : Any):
"""simple docstring"""
return MaskFormerConfig.from_backbone_and_decoder_configs(
backbone_config=SwinConfig(
depths=[1, 1, 1, 1] , ) , decoder_config=DetrConfig(
decoder_ffn_dim=128 , num_queries=self.num_queries , decoder_attention_heads=2 , d_model=self.mask_feature_size , ) , mask_feature_size=self.mask_feature_size , fpn_feature_size=self.mask_feature_size , num_channels=self.num_channels , num_labels=self.num_labels , )
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.prepare_config_and_inputs()
UpperCAmelCase_ = {'''pixel_values''': pixel_values, '''pixel_mask''': pixel_mask}
return config, inputs_dict
def lowerCamelCase ( self : str , _snake_case : List[Any] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = output.encoder_hidden_states
UpperCAmelCase_ = output.pixel_decoder_hidden_states
UpperCAmelCase_ = output.transformer_decoder_hidden_states
self.parent.assertTrue(len(_snake_case) , len(config.backbone_config.depths))
self.parent.assertTrue(len(_snake_case) , len(config.backbone_config.depths))
self.parent.assertTrue(len(_snake_case) , config.decoder_config.decoder_layers)
def lowerCamelCase ( self : Union[str, Any] , _snake_case : List[str] , _snake_case : int , _snake_case : Optional[Any] , _snake_case : str=False):
"""simple docstring"""
with torch.no_grad():
UpperCAmelCase_ = MaskFormerModel(config=_snake_case)
model.to(_snake_case)
model.eval()
UpperCAmelCase_ = model(pixel_values=_snake_case , pixel_mask=_snake_case)
UpperCAmelCase_ = model(_snake_case , output_hidden_states=_snake_case)
# the correct shape of output.transformer_decoder_hidden_states ensure the correcteness of the
# encoder and pixel decoder
self.parent.assertEqual(
output.transformer_decoder_last_hidden_state.shape , (self.batch_size, self.num_queries, self.mask_feature_size) , )
# let's ensure the other two hidden state exists
self.parent.assertTrue(output.pixel_decoder_last_hidden_state is not None)
self.parent.assertTrue(output.encoder_last_hidden_state is not None)
if output_hidden_states:
self.check_output_hidden_state(_snake_case , _snake_case)
def lowerCamelCase ( self : List[Any] , _snake_case : List[Any] , _snake_case : List[Any] , _snake_case : str , _snake_case : Optional[int] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerForInstanceSegmentation(config=_snake_case)
model.to(_snake_case)
model.eval()
def comm_check_on_output(_snake_case : Tuple):
# let's still check that all the required stuff is there
self.parent.assertTrue(result.transformer_decoder_last_hidden_state is not None)
self.parent.assertTrue(result.pixel_decoder_last_hidden_state is not None)
self.parent.assertTrue(result.encoder_last_hidden_state is not None)
# okay, now we need to check the logits shape
# due to the encoder compression, masks have a //4 spatial size
self.parent.assertEqual(
result.masks_queries_logits.shape , (self.batch_size, self.num_queries, self.min_size // 4, self.max_size // 4) , )
# + 1 for null class
self.parent.assertEqual(
result.class_queries_logits.shape , (self.batch_size, self.num_queries, self.num_labels + 1))
with torch.no_grad():
UpperCAmelCase_ = model(pixel_values=_snake_case , pixel_mask=_snake_case)
UpperCAmelCase_ = model(_snake_case)
comm_check_on_output(_snake_case)
UpperCAmelCase_ = model(
pixel_values=_snake_case , pixel_mask=_snake_case , mask_labels=_snake_case , class_labels=_snake_case)
comm_check_on_output(_snake_case)
self.parent.assertTrue(result.loss is not None)
self.parent.assertEqual(result.loss.shape , torch.Size([1]))
@require_torch
class __snake_case ( a , a , unittest.TestCase ):
UpperCAmelCase__ : Union[str, Any] = (MaskFormerModel, MaskFormerForInstanceSegmentation) if is_torch_available() else ()
UpperCAmelCase__ : Optional[Any] = (
{'''feature-extraction''': MaskFormerModel, '''image-segmentation''': MaskFormerForInstanceSegmentation}
if is_torch_available()
else {}
)
UpperCAmelCase__ : Dict = False
UpperCAmelCase__ : List[str] = False
UpperCAmelCase__ : Optional[Any] = False
UpperCAmelCase__ : Union[str, Any] = False
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerModelTester(self)
UpperCAmelCase_ = ConfigTester(self , config_class=_snake_case , has_text_modality=_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
self.config_tester.run_common_tests()
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskformer_model(_snake_case , **_snake_case , output_hidden_states=_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_maskformer_instance_segmentation_head_model(*_snake_case)
@unittest.skip(reason='''MaskFormer does not use inputs_embeds''')
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer does not have a get_input_embeddings method''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer is not a generative model''')
def lowerCamelCase ( self : str):
"""simple docstring"""
pass
@unittest.skip(reason='''MaskFormer does not use token embeddings''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
@require_torch_multi_gpu
@unittest.skip(
reason='''MaskFormer has some layers using `add_module` which doesn\'t work well with `nn.DataParallel`''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
pass
@unittest.skip('''Will be fixed soon by reducing the size of the model used for common tests.''')
def lowerCamelCase ( self : str):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ = model_class(_snake_case)
UpperCAmelCase_ = inspect.signature(model.forward)
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCAmelCase_ = [*signature.parameters.keys()]
UpperCAmelCase_ = ['''pixel_values''']
self.assertListEqual(arg_names[:1] , _snake_case)
@slow
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
for model_name in ["facebook/maskformer-swin-small-coco"]:
UpperCAmelCase_ = MaskFormerModel.from_pretrained(_snake_case)
self.assertIsNotNone(_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = (self.model_tester.min_size,) * 2
UpperCAmelCase_ = {
'''pixel_values''': torch.randn((2, 3, *size) , device=_snake_case),
'''mask_labels''': torch.randn((2, 10, *size) , device=_snake_case),
'''class_labels''': torch.zeros(2 , 10 , device=_snake_case).long(),
}
UpperCAmelCase_ = MaskFormerForInstanceSegmentation(MaskFormerConfig()).to(_snake_case)
UpperCAmelCase_ = model(**_snake_case)
self.assertTrue(outputs.loss is not None)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskformer_model(_snake_case , **_snake_case , output_hidden_states=_snake_case)
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ = model_class(_snake_case).to(_snake_case)
UpperCAmelCase_ = model(**_snake_case , output_attentions=_snake_case)
self.assertTrue(outputs.attentions is not None)
def lowerCamelCase ( self : int):
"""simple docstring"""
if not self.model_tester.is_training:
return
# only MaskFormerForInstanceSegmentation has the loss
UpperCAmelCase_ = self.all_model_classes[1]
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCAmelCase_ = model_class(_snake_case)
model.to(_snake_case)
model.train()
UpperCAmelCase_ = model(_snake_case , mask_labels=_snake_case , class_labels=_snake_case).loss
loss.backward()
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = self.all_model_classes[1]
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCAmelCase_ = True
UpperCAmelCase_ = True
UpperCAmelCase_ = model_class(_snake_case)
model.to(_snake_case)
model.train()
UpperCAmelCase_ = model(_snake_case , mask_labels=_snake_case , class_labels=_snake_case)
UpperCAmelCase_ = outputs.encoder_hidden_states[0]
encoder_hidden_states.retain_grad()
UpperCAmelCase_ = outputs.pixel_decoder_hidden_states[0]
pixel_decoder_hidden_states.retain_grad()
# we requires_grad=True in inputs_embeds (line 2152), the original implementation don't
UpperCAmelCase_ = outputs.transformer_decoder_hidden_states[0]
transformer_decoder_hidden_states.retain_grad()
UpperCAmelCase_ = outputs.attentions[0]
attentions.retain_grad()
outputs.loss.backward(retain_graph=_snake_case)
self.assertIsNotNone(encoder_hidden_states.grad)
self.assertIsNotNone(pixel_decoder_hidden_states.grad)
self.assertIsNotNone(transformer_decoder_hidden_states.grad)
self.assertIsNotNone(attentions.grad)
snake_case_ : Dict = 1e-4
def A () -> Union[str, Any]:
"""simple docstring"""
UpperCAmelCase_ = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
return image
@require_vision
@slow
class __snake_case ( unittest.TestCase ):
@cached_property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return (
MaskFormerImageProcessor.from_pretrained('''facebook/maskformer-swin-small-coco''')
if is_vision_available()
else None
)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = MaskFormerModel.from_pretrained('''facebook/maskformer-swin-small-coco''').to(_snake_case)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
UpperCAmelCase_ = torch.tensor(
[[-0.0_4_8_2, 0.9_2_2_8, 0.4_9_5_1], [-0.2_5_4_7, 0.8_0_1_7, 0.8_5_2_7], [-0.0_0_6_9, 0.3_3_8_5, -0.0_0_8_9]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.encoder_last_hidden_state[0, 0, :3, :3] , _snake_case , atol=_snake_case))
UpperCAmelCase_ = torch.tensor(
[[-0.8_4_2_2, -0.8_4_3_4, -0.9_7_1_8], [-1.0_1_4_4, -0.5_5_6_5, -0.4_1_9_5], [-1.0_0_3_8, -0.4_4_8_4, -0.1_9_6_1]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.pixel_decoder_last_hidden_state[0, 0, :3, :3] , _snake_case , atol=_snake_case))
UpperCAmelCase_ = torch.tensor(
[[0.2_8_5_2, -0.0_1_5_9, 0.9_7_3_5], [0.6_2_5_4, 0.1_8_5_8, 0.8_5_2_9], [-0.0_6_8_0, -0.4_1_1_6, 1.8_4_1_3]]).to(_snake_case)
self.assertTrue(
torch.allclose(
outputs.transformer_decoder_last_hidden_state[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-swin-small-coco''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
# masks_queries_logits
UpperCAmelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.decoder_config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) , )
UpperCAmelCase_ = [
[-1.3_7_3_7_1_2_4, -1.7_7_2_4_9_3_7, -1.9_3_6_4_2_3_3],
[-1.5_9_7_7_2_8_1, -1.9_8_6_7_9_3_9, -2.1_5_2_3_6_9_5],
[-1.5_7_9_5_3_9_8, -1.9_2_6_9_8_3_2, -2.0_9_3_9_4_2],
]
UpperCAmelCase_ = torch.tensor(_snake_case).to(_snake_case)
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _snake_case , atol=_snake_case))
# class_queries_logits
UpperCAmelCase_ = outputs.class_queries_logits
self.assertEqual(
class_queries_logits.shape , (1, model.config.decoder_config.num_queries, model.config.num_labels + 1))
UpperCAmelCase_ = torch.tensor(
[
[1.65_12e00, -5.25_72e00, -3.35_19e00],
[3.61_69e-02, -5.90_25e00, -2.93_13e00],
[1.07_66e-04, -7.76_30e00, -5.12_63e00],
]).to(_snake_case)
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-resnet101-coco-stuff''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = prepare_img()
UpperCAmelCase_ = image_processor(_snake_case , return_tensors='''pt''').to(_snake_case)
UpperCAmelCase_ = inputs['''pixel_values'''].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0)
# check size
self.assertEqual(_snake_case , (1, 3, 800, 1088))
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
# masks_queries_logits
UpperCAmelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.decoder_config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) , )
UpperCAmelCase_ = [[-0.9_0_4_6, -2.6_3_6_6, -4.6_0_6_2], [-3.4_1_7_9, -5.7_8_9_0, -8.8_0_5_7], [-4.9_1_7_9, -7.6_5_6_0, -1_0.7_7_1_1]]
UpperCAmelCase_ = torch.tensor(_snake_case).to(_snake_case)
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _snake_case , atol=_snake_case))
# class_queries_logits
UpperCAmelCase_ = outputs.class_queries_logits
self.assertEqual(
class_queries_logits.shape , (1, model.config.decoder_config.num_queries, model.config.num_labels + 1))
UpperCAmelCase_ = torch.tensor(
[[4.7_1_8_8, -3.2_5_8_5, -2.8_8_5_7], [6.6_8_7_1, -2.9_1_8_1, -1.2_4_8_7], [7.2_4_4_9, -2.2_7_6_4, -2.1_8_7_4]]).to(_snake_case)
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _snake_case , atol=_snake_case))
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = (
MaskFormerForInstanceSegmentation.from_pretrained('''facebook/maskformer-swin-small-coco''')
.to(_snake_case)
.eval()
)
UpperCAmelCase_ = self.default_image_processor
UpperCAmelCase_ = image_processor(
[np.zeros((3, 800, 1333)), np.zeros((3, 800, 1333))] , segmentation_maps=[np.zeros((384, 384)).astype(np.floataa), np.zeros((384, 384)).astype(np.floataa)] , return_tensors='''pt''' , )
UpperCAmelCase_ = inputs['''pixel_values'''].to(_snake_case)
UpperCAmelCase_ = [el.to(_snake_case) for el in inputs['''mask_labels''']]
UpperCAmelCase_ = [el.to(_snake_case) for el in inputs['''class_labels''']]
with torch.no_grad():
UpperCAmelCase_ = model(**_snake_case)
self.assertTrue(outputs.loss is not None)
| 51
| 0
|
"""simple docstring"""
import doctest
import glob
import importlib
import inspect
import os
import re
from contextlib import contextmanager
from functools import wraps
from unittest.mock import patch
import numpy as np
import pytest
from absl.testing import parameterized
import datasets
from datasets import load_metric
from .utils import for_all_test_methods, local, slow
# mark all tests as integration
__UpperCAmelCase = pytest.mark.integration
__UpperCAmelCase = {"comet"}
__UpperCAmelCase = importlib.util.find_spec('fairseq') is not None
__UpperCAmelCase = {"code_eval"}
__UpperCAmelCase = os.name == "nt"
__UpperCAmelCase = {"bertscore", "frugalscore", "perplexity"}
__UpperCAmelCase = importlib.util.find_spec('transformers') is not None
def _snake_case ( lowercase__ : Dict ) -> List[Any]:
'''simple docstring'''
@wraps(__A )
def wrapper(self : Optional[int] , lowercase__ : Tuple ):
if not _has_fairseq and metric_name in REQUIRE_FAIRSEQ:
self.skipTest("""\"test requires Fairseq\"""" )
else:
test_case(self , __A )
return wrapper
def _snake_case ( lowercase__ : Tuple ) -> List[str]:
'''simple docstring'''
@wraps(__A )
def wrapper(self : str , lowercase__ : int ):
if not _has_transformers and metric_name in REQUIRE_TRANSFORMERS:
self.skipTest("""\"test requires transformers\"""" )
else:
test_case(self , __A )
return wrapper
def _snake_case ( lowercase__ : int ) -> List[Any]:
'''simple docstring'''
@wraps(__A )
def wrapper(self : List[Any] , lowercase__ : str ):
if _on_windows and metric_name in UNSUPPORTED_ON_WINDOWS:
self.skipTest("""\"test not supported on Windows\"""" )
else:
test_case(self , __A )
return wrapper
def _snake_case ( ) -> int:
'''simple docstring'''
lowerCAmelCase_ :Union[str, Any] = [metric_dir.split(os.sep )[-2] for metric_dir in glob.glob("""./metrics/*/""" )]
return [{"testcase_name": x, "metric_name": x} for x in metrics if x != "gleu"] # gleu is unfinished
@parameterized.named_parameters(get_local_metric_names() )
@for_all_test_methods(
A__ , A__ , A__ )
@local
class _SCREAMING_SNAKE_CASE ( parameterized.TestCase ):
UpperCAmelCase_ :List[Any] = {}
UpperCAmelCase_ :List[str] = None
@pytest.mark.filterwarnings("""ignore:metric_module_factory is deprecated:FutureWarning""" )
@pytest.mark.filterwarnings("""ignore:load_metric is deprecated:FutureWarning""" )
def __lowerCAmelCase ( self , __A ) -> Union[str, Any]:
lowerCAmelCase_ :Any = """[...]"""
lowerCAmelCase_ :int = importlib.import_module(
datasets.load.metric_module_factory(os.path.join("""metrics""" , _snake_case ) ).module_path )
lowerCAmelCase_ :List[Any] = datasets.load.import_main_class(metric_module.__name__ , dataset=_snake_case )
# check parameters
lowerCAmelCase_ :Optional[int] = inspect.signature(metric._compute ).parameters
self.assertTrue(all(p.kind != p.VAR_KEYWORD for p in parameters.values() ) ) # no **kwargs
# run doctest
with self.patch_intensive_calls(_snake_case , metric_module.__name__ ):
with self.use_local_metrics():
try:
lowerCAmelCase_ :List[str] = doctest.testmod(_snake_case , verbose=_snake_case , raise_on_error=_snake_case )
except doctest.UnexpectedException as e:
raise e.exc_info[1] # raise the exception that doctest caught
self.assertEqual(results.failed , 0 )
self.assertGreater(results.attempted , 1 )
@slow
def __lowerCAmelCase ( self , __A ) -> Tuple:
lowerCAmelCase_ :Dict = """[...]"""
lowerCAmelCase_ :List[str] = importlib.import_module(
datasets.load.metric_module_factory(os.path.join("""metrics""" , _snake_case ) ).module_path )
# run doctest
with self.use_local_metrics():
lowerCAmelCase_ :List[str] = doctest.testmod(_snake_case , verbose=_snake_case , raise_on_error=_snake_case )
self.assertEqual(results.failed , 0 )
self.assertGreater(results.attempted , 1 )
@contextmanager
def __lowerCAmelCase ( self , __A , __A ) -> Any:
if metric_name in self.INTENSIVE_CALLS_PATCHER:
with self.INTENSIVE_CALLS_PATCHER[metric_name](_snake_case ):
yield
else:
yield
@contextmanager
def __lowerCAmelCase ( self ) -> Tuple:
def load_local_metric(__A , *__A , **__A ):
return load_metric(os.path.join("""metrics""" , _snake_case ) , *_snake_case , **_snake_case )
with patch("""datasets.load_metric""" ) as mock_load_metric:
lowerCAmelCase_ :List[str] = load_local_metric
yield
@classmethod
def __lowerCAmelCase ( cls , __A ) -> Dict:
def wrapper(__A ):
lowerCAmelCase_ :List[str] = contextmanager(_snake_case )
lowerCAmelCase_ :Union[str, Any] = patcher
return patcher
return wrapper
@LocalMetricTest.register_intensive_calls_patcher("""bleurt""" )
def _snake_case ( lowercase__ : Any ) -> List[str]:
'''simple docstring'''
import tensorflow.compat.va as tf
from bleurt.score import Predictor
tf.flags.DEFINE_string("""sv""" , """""" , """""" ) # handle pytest cli flags
class _SCREAMING_SNAKE_CASE ( A__ ):
def __lowerCAmelCase ( self , __A ) -> Union[str, Any]:
assert len(input_dict["""input_ids"""] ) == 2
return np.array([1.0_3, 1.0_4] )
# mock predict_fn which is supposed to do a forward pass with a bleurt model
with patch("""bleurt.score._create_predictor""" ) as mock_create_predictor:
lowerCAmelCase_ :str = MockedPredictor()
yield
@LocalMetricTest.register_intensive_calls_patcher("""bertscore""" )
def _snake_case ( lowercase__ : int ) -> Tuple:
'''simple docstring'''
import torch
def bert_cos_score_idf(lowercase__ : str , lowercase__ : Optional[Any] , *lowercase__ : List[Any] , **lowercase__ : List[str] ):
return torch.tensor([[1.0, 1.0, 1.0]] * len(__A ) )
# mock get_model which is supposed to do download a bert model
# mock bert_cos_score_idf which is supposed to do a forward pass with a bert model
with patch("""bert_score.scorer.get_model""" ), patch(
"""bert_score.scorer.bert_cos_score_idf""" ) as mock_bert_cos_score_idf:
lowerCAmelCase_ :Dict = bert_cos_score_idf
yield
@LocalMetricTest.register_intensive_calls_patcher("""comet""" )
def _snake_case ( lowercase__ : List[str] ) -> Optional[Any]:
'''simple docstring'''
def load_from_checkpoint(lowercase__ : str ):
class _SCREAMING_SNAKE_CASE :
def __lowerCAmelCase ( self , __A , *__A , **__A ) -> int:
assert len(_snake_case ) == 2
lowerCAmelCase_ :Optional[int] = [0.1_9, 0.9_2]
return scores, sum(_snake_case ) / len(_snake_case )
return Model()
# mock load_from_checkpoint which is supposed to do download a bert model
# mock load_from_checkpoint which is supposed to do download a bert model
with patch("""comet.download_model""" ) as mock_download_model:
lowerCAmelCase_ :List[Any] = None
with patch("""comet.load_from_checkpoint""" ) as mock_load_from_checkpoint:
lowerCAmelCase_ :Optional[int] = load_from_checkpoint
yield
def _snake_case ( ) -> int:
'''simple docstring'''
lowerCAmelCase_ :str = load_metric(os.path.join("""metrics""" , """seqeval""" ) )
lowerCAmelCase_ :Any = """ERROR"""
lowerCAmelCase_ :List[Any] = f"""Scheme should be one of [IOB1, IOB2, IOE1, IOE2, IOBES, BILOU], got {wrong_scheme}"""
with pytest.raises(__A , match=re.escape(__A ) ):
metric.compute(predictions=[] , references=[] , scheme=__A )
| 84
|
import argparse
import pickle
import numpy as np
import torch
from torch import nn
from transformers import ReformerConfig, ReformerModelWithLMHead
from transformers.utils import logging
logging.set_verbosity_info()
def A (__A : Optional[int] , __A : int , __A : str=None ) -> List[Any]:
"""simple docstring"""
assert torch_layer.weight.shape == weight.shape, F"""{torch_layer} layer.weight does not match"""
UpperCAmelCase_ = nn.Parameter(__A )
if bias is not None:
assert torch_layer.bias.shape == bias.shape, F"""{torch_layer} layer.bias does not match"""
UpperCAmelCase_ = nn.Parameter(__A )
def A (__A : Tuple , __A : Dict , __A : str ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = np.asarray(weights[0] )
UpperCAmelCase_ = np.asarray(weights[1] )
UpperCAmelCase_ = np.asarray(weights[2] )
set_param(
torch_layer.self_attention.query_key , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.value , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.output.dense , torch.tensor(__A ).view(-1 , __A ).contiguous().transpose(0 , 1 ) , )
def A (__A : Optional[Any] , __A : Any , __A : List[Any] ) -> int:
"""simple docstring"""
UpperCAmelCase_ = np.asarray(weights[0] )
UpperCAmelCase_ = np.asarray(weights[1] )
UpperCAmelCase_ = np.asarray(weights[2] )
UpperCAmelCase_ = np.asarray(weights[3] )
set_param(
torch_layer.self_attention.query , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.key , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.self_attention.value , torch.tensor(__A ).transpose(1 , 2 ).contiguous().view(-1 , __A ) , )
set_param(
torch_layer.output.dense , torch.tensor(__A ).view(-1 , __A ).contiguous().transpose(0 , 1 ) , )
def A (__A : int , __A : Union[str, Any] , __A : List[str] ) -> List[Any]:
"""simple docstring"""
UpperCAmelCase_ = weights[0][0][0]
UpperCAmelCase_ = np.asarray(layer_norm_a[0] )
UpperCAmelCase_ = np.asarray(layer_norm_a[1] )
set_param(
torch_block.attention.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# lsh weights + output
UpperCAmelCase_ = weights[0][1]
if len(__A ) < 4:
set_layer_weights_in_torch_lsh(__A , torch_block.attention , __A )
else:
set_layer_weights_in_torch_local(__A , torch_block.attention , __A )
# intermediate weighs
UpperCAmelCase_ = weights[2][0][1][2]
# Chunked Feed Forward
if len(__A ) == 4:
UpperCAmelCase_ = intermediate_weights[2]
# layernorm 2
UpperCAmelCase_ = np.asarray(intermediate_weights[0][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[0][1] )
set_param(
torch_block.feed_forward.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# intermediate dense
UpperCAmelCase_ = np.asarray(intermediate_weights[1][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[1][1] )
set_param(
torch_block.feed_forward.dense.dense , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
# intermediate out
UpperCAmelCase_ = np.asarray(intermediate_weights[4][0] )
UpperCAmelCase_ = np.asarray(intermediate_weights[4][1] )
set_param(
torch_block.feed_forward.output.dense , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
def A (__A : Optional[int] , __A : Tuple , __A : Any ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = torch_model.reformer
# word embeds
UpperCAmelCase_ = np.asarray(weights[1] )
set_param(
torch_model_reformer.embeddings.word_embeddings , torch.tensor(__A ) , )
if isinstance(weights[3] , __A ):
UpperCAmelCase_ = torch_model_reformer.embeddings.position_embeddings
for emb_idx in range(len(position_embeddings.weights ) ):
UpperCAmelCase_ = np.asarray(weights[3][emb_idx][0] )
assert (
position_embeddings.weights[emb_idx].shape == emb_weights.shape
), F"""{position_embeddings[emb_idx]} emb does not match"""
UpperCAmelCase_ = nn.Parameter(torch.tensor(__A ) )
UpperCAmelCase_ = weights[5]
assert len(torch_model_reformer.encoder.layers ) * 4 == len(
__A ), "HF and trax model do not have the same number of layers"
for layer_idx, layer in enumerate(torch_model_reformer.encoder.layers ):
UpperCAmelCase_ = trax_layer_weights[4 * layer_idx : 4 * (layer_idx + 1)]
set_block_weights_in_torch(__A , __A , __A )
# output layer norm
UpperCAmelCase_ = np.asarray(weights[7][0] )
UpperCAmelCase_ = np.asarray(weights[7][1] )
set_param(
torch_model_reformer.encoder.layer_norm , torch.tensor(__A ) , torch.tensor(__A ) , )
# output embeddings
UpperCAmelCase_ = np.asarray(weights[9][0] )
UpperCAmelCase_ = np.asarray(weights[9][1] )
set_param(
torch_model.lm_head.decoder , torch.tensor(__A ).transpose(0 , 1 ).contiguous() , torch.tensor(__A ) , )
def A (__A : Tuple , __A : int , __A : str ) -> Union[str, Any]:
"""simple docstring"""
UpperCAmelCase_ = ReformerConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = ReformerModelWithLMHead(__A )
with open(__A , '''rb''' ) as f:
UpperCAmelCase_ = pickle.load(__A )['''weights''']
set_model_weights_in_torch(__A , __A , config.hidden_size )
# Save pytorch-model
print(F"""Save PyTorch model to {pytorch_dump_path}""" )
torch.save(model.state_dict() , __A )
if __name__ == "__main__":
snake_case_ : Optional[Any] = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--trax_model_pkl_path", default=None, type=str, required=True, help="Path to the TensorFlow checkpoint path."
)
parser.add_argument(
"--config_file",
default=None,
type=str,
required=True,
help=(
"The config json file corresponding to the pre-trained Reformer model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--pytorch_dump_path", default=None, type=str, required=True, help="Path to the output PyTorch model."
)
snake_case_ : List[Any] = parser.parse_args()
convert_trax_checkpoint_to_pytorch(args.trax_model_pkl_path, args.config_file, args.pytorch_dump_path)
| 51
| 0
|
'''simple docstring'''
import json
import os
import pickle
import shutil
import tempfile
from unittest import TestCase
from unittest.mock import patch
import numpy as np
from datasets import Dataset
from transformers import is_faiss_available
from transformers.models.bart.configuration_bart import BartConfig
from transformers.models.bart.tokenization_bart import BartTokenizer
from transformers.models.bert.tokenization_bert import VOCAB_FILES_NAMES as DPR_VOCAB_FILES_NAMES
from transformers.models.dpr.configuration_dpr import DPRConfig
from transformers.models.dpr.tokenization_dpr import DPRContextEncoderTokenizer, DPRQuestionEncoderTokenizer
from transformers.models.rag.configuration_rag import RagConfig
from transformers.models.rag.retrieval_rag import CustomHFIndex, RagRetriever
from transformers.models.roberta.tokenization_roberta import VOCAB_FILES_NAMES as BART_VOCAB_FILES_NAMES
from transformers.testing_utils import require_faiss, require_sentencepiece, require_tokenizers, require_torch
if is_faiss_available():
import faiss
@require_faiss
class SCREAMING_SNAKE_CASE( A__ ):
"""simple docstring"""
def A ( self : Tuple ) -> Any:
UpperCAmelCase : Any = tempfile.mkdtemp()
UpperCAmelCase : Any = 8
# DPR tok
UpperCAmelCase : Optional[int] = [
'''[UNK]''',
'''[CLS]''',
'''[SEP]''',
'''[PAD]''',
'''[MASK]''',
'''want''',
'''##want''',
'''##ed''',
'''wa''',
'''un''',
'''runn''',
'''##ing''',
''',''',
'''low''',
'''lowest''',
]
UpperCAmelCase : Union[str, Any] = os.path.join(self.tmpdirname , '''dpr_tokenizer''' )
os.makedirs(_snake_case , exist_ok=_snake_case )
UpperCAmelCase : List[str] = os.path.join(_snake_case , DPR_VOCAB_FILES_NAMES['''vocab_file'''] )
with open(self.vocab_file , '''w''' , encoding='''utf-8''' ) as vocab_writer:
vocab_writer.write(''''''.join([x + '''\n''' for x in vocab_tokens] ) )
# BART tok
UpperCAmelCase : Dict = [
'''l''',
'''o''',
'''w''',
'''e''',
'''r''',
'''s''',
'''t''',
'''i''',
'''d''',
'''n''',
'''\u0120''',
'''\u0120l''',
'''\u0120n''',
'''\u0120lo''',
'''\u0120low''',
'''er''',
'''\u0120lowest''',
'''\u0120newer''',
'''\u0120wider''',
'''<unk>''',
]
UpperCAmelCase : Optional[Any] = dict(zip(_snake_case , range(len(_snake_case ) ) ) )
UpperCAmelCase : Any = ['''#version: 0.2''', '''\u0120 l''', '''\u0120l o''', '''\u0120lo w''', '''e r''', '''''']
UpperCAmelCase : Any = {'''unk_token''': '''<unk>'''}
UpperCAmelCase : List[str] = os.path.join(self.tmpdirname , '''bart_tokenizer''' )
os.makedirs(_snake_case , exist_ok=_snake_case )
UpperCAmelCase : str = os.path.join(_snake_case , BART_VOCAB_FILES_NAMES['''vocab_file'''] )
UpperCAmelCase : Optional[int] = os.path.join(_snake_case , BART_VOCAB_FILES_NAMES['''merges_file'''] )
with open(self.vocab_file , '''w''' , encoding='''utf-8''' ) as fp:
fp.write(json.dumps(_snake_case ) + '''\n''' )
with open(self.merges_file , '''w''' , encoding='''utf-8''' ) as fp:
fp.write('''\n'''.join(_snake_case ) )
def A ( self : str ) -> Dict:
return DPRQuestionEncoderTokenizer.from_pretrained(os.path.join(self.tmpdirname , '''dpr_tokenizer''' ) )
def A ( self : int ) -> Any:
return DPRContextEncoderTokenizer.from_pretrained(os.path.join(self.tmpdirname , '''dpr_tokenizer''' ) )
def A ( self : Dict ) -> Union[str, Any]:
return BartTokenizer.from_pretrained(os.path.join(self.tmpdirname , '''bart_tokenizer''' ) )
def A ( self : Dict ) -> List[Any]:
shutil.rmtree(self.tmpdirname )
def A ( self : int ) -> Union[str, Any]:
UpperCAmelCase : Union[str, Any] = Dataset.from_dict(
{
'''id''': ['''0''', '''1'''],
'''text''': ['''foo''', '''bar'''],
'''title''': ['''Foo''', '''Bar'''],
'''embeddings''': [np.ones(self.retrieval_vector_size ), 2 * np.ones(self.retrieval_vector_size )],
} )
dataset.add_faiss_index('''embeddings''' , string_factory='''Flat''' , metric_type=faiss.METRIC_INNER_PRODUCT )
return dataset
def A ( self : Optional[Any] ) -> Tuple:
UpperCAmelCase : Any = self.get_dummy_dataset()
UpperCAmelCase : Optional[Any] = RagConfig(
retrieval_vector_size=self.retrieval_vector_size , question_encoder=DPRConfig().to_dict() , generator=BartConfig().to_dict() , )
with patch('''transformers.models.rag.retrieval_rag.load_dataset''' ) as mock_load_dataset:
UpperCAmelCase : Optional[Any] = dataset
UpperCAmelCase : Optional[Any] = RagRetriever(
_snake_case , question_encoder_tokenizer=self.get_dpr_tokenizer() , generator_tokenizer=self.get_bart_tokenizer() , )
return retriever
def A ( self : List[str] , __snake_case : bool ) -> Dict:
UpperCAmelCase : List[Any] = self.get_dummy_dataset()
UpperCAmelCase : Optional[int] = RagConfig(
retrieval_vector_size=self.retrieval_vector_size , question_encoder=DPRConfig().to_dict() , generator=BartConfig().to_dict() , index_name='''custom''' , )
if from_disk:
UpperCAmelCase : Union[str, Any] = os.path.join(self.tmpdirname , '''dataset''' )
UpperCAmelCase : Tuple = os.path.join(self.tmpdirname , '''index.faiss''' )
dataset.get_index('''embeddings''' ).save(os.path.join(self.tmpdirname , '''index.faiss''' ) )
dataset.drop_index('''embeddings''' )
dataset.save_to_disk(os.path.join(self.tmpdirname , '''dataset''' ) )
del dataset
UpperCAmelCase : List[str] = RagRetriever(
_snake_case , question_encoder_tokenizer=self.get_dpr_tokenizer() , generator_tokenizer=self.get_bart_tokenizer() , )
else:
UpperCAmelCase : Optional[Any] = RagRetriever(
_snake_case , question_encoder_tokenizer=self.get_dpr_tokenizer() , generator_tokenizer=self.get_bart_tokenizer() , index=CustomHFIndex(config.retrieval_vector_size , _snake_case ) , )
return retriever
def A ( self : Optional[int] ) -> Any:
UpperCAmelCase : Any = Dataset.from_dict(
{
'''id''': ['''0''', '''1'''],
'''text''': ['''foo''', '''bar'''],
'''title''': ['''Foo''', '''Bar'''],
'''embeddings''': [np.ones(self.retrieval_vector_size + 1 ), 2 * np.ones(self.retrieval_vector_size + 1 )],
} )
dataset.add_faiss_index('''embeddings''' , string_factory='''Flat''' , metric_type=faiss.METRIC_INNER_PRODUCT )
UpperCAmelCase : Tuple = os.path.join(self.tmpdirname , '''hf_bert_base.hnswSQ8_correct_phi_128.c_index''' )
dataset.save_faiss_index('''embeddings''' , index_file_name + '''.index.dpr''' )
pickle.dump(dataset['''id'''] , open(index_file_name + '''.index_meta.dpr''' , '''wb''' ) )
UpperCAmelCase : int = os.path.join(self.tmpdirname , '''psgs_w100.tsv.pkl''' )
UpperCAmelCase : Optional[int] = {sample['''id''']: [sample['''text'''], sample['''title''']] for sample in dataset}
pickle.dump(_snake_case , open(_snake_case , '''wb''' ) )
UpperCAmelCase : str = RagConfig(
retrieval_vector_size=self.retrieval_vector_size , question_encoder=DPRConfig().to_dict() , generator=BartConfig().to_dict() , index_name='''legacy''' , index_path=self.tmpdirname , )
UpperCAmelCase : int = RagRetriever(
_snake_case , question_encoder_tokenizer=self.get_dpr_tokenizer() , generator_tokenizer=self.get_bart_tokenizer() )
return retriever
def A ( self : Optional[Any] ) -> Tuple:
UpperCAmelCase : List[str] = 1
UpperCAmelCase : Tuple = self.get_dummy_canonical_hf_index_retriever()
UpperCAmelCase : Tuple = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase , UpperCAmelCase , UpperCAmelCase : List[str] = retriever.retrieve(_snake_case , n_docs=_snake_case )
self.assertEqual(retrieved_doc_embeds.shape , (2, n_docs, self.retrieval_vector_size) )
self.assertEqual(len(_snake_case ) , 2 )
self.assertEqual(sorted(doc_dicts[0] ) , ['''embeddings''', '''id''', '''text''', '''title'''] )
self.assertEqual(len(doc_dicts[0]['''id'''] ) , _snake_case )
self.assertEqual(doc_dicts[0]['''id'''][0] , '''1''' ) # max inner product is reached with second doc
self.assertEqual(doc_dicts[1]['''id'''][0] , '''0''' ) # max inner product is reached with first doc
self.assertListEqual(doc_ids.tolist() , [[1], [0]] )
def A ( self : List[Any] ) -> str:
UpperCAmelCase : Optional[int] = self.get_dummy_canonical_hf_index_retriever()
with tempfile.TemporaryDirectory() as tmp_dirname:
with patch('''transformers.models.rag.retrieval_rag.load_dataset''' ) as mock_load_dataset:
UpperCAmelCase : str = self.get_dummy_dataset()
retriever.save_pretrained(_snake_case )
UpperCAmelCase : Dict = RagRetriever.from_pretrained(_snake_case )
self.assertIsInstance(_snake_case , _snake_case )
UpperCAmelCase : int = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase : Optional[int] = retriever.retrieve(_snake_case , n_docs=1 )
self.assertTrue(out is not None )
def A ( self : Optional[int] ) -> Union[str, Any]:
UpperCAmelCase : List[str] = 1
UpperCAmelCase : Optional[Any] = self.get_dummy_custom_hf_index_retriever(from_disk=_snake_case )
UpperCAmelCase : Any = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase , UpperCAmelCase , UpperCAmelCase : Optional[int] = retriever.retrieve(_snake_case , n_docs=_snake_case )
self.assertEqual(retrieved_doc_embeds.shape , (2, n_docs, self.retrieval_vector_size) )
self.assertEqual(len(_snake_case ) , 2 )
self.assertEqual(sorted(doc_dicts[0] ) , ['''embeddings''', '''id''', '''text''', '''title'''] )
self.assertEqual(len(doc_dicts[0]['''id'''] ) , _snake_case )
self.assertEqual(doc_dicts[0]['''id'''][0] , '''1''' ) # max inner product is reached with second doc
self.assertEqual(doc_dicts[1]['''id'''][0] , '''0''' ) # max inner product is reached with first doc
self.assertListEqual(doc_ids.tolist() , [[1], [0]] )
def A ( self : Dict ) -> Optional[Any]:
UpperCAmelCase : int = self.get_dummy_custom_hf_index_retriever(from_disk=_snake_case )
with tempfile.TemporaryDirectory() as tmp_dirname:
retriever.save_pretrained(_snake_case )
UpperCAmelCase : List[str] = RagRetriever.from_pretrained(_snake_case )
self.assertIsInstance(_snake_case , _snake_case )
UpperCAmelCase : Dict = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase : str = retriever.retrieve(_snake_case , n_docs=1 )
self.assertTrue(out is not None )
def A ( self : List[str] ) -> List[str]:
UpperCAmelCase : Dict = 1
UpperCAmelCase : Union[str, Any] = self.get_dummy_custom_hf_index_retriever(from_disk=_snake_case )
UpperCAmelCase : Optional[int] = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase , UpperCAmelCase , UpperCAmelCase : str = retriever.retrieve(_snake_case , n_docs=_snake_case )
self.assertEqual(retrieved_doc_embeds.shape , (2, n_docs, self.retrieval_vector_size) )
self.assertEqual(len(_snake_case ) , 2 )
self.assertEqual(sorted(doc_dicts[0] ) , ['''embeddings''', '''id''', '''text''', '''title'''] )
self.assertEqual(len(doc_dicts[0]['''id'''] ) , _snake_case )
self.assertEqual(doc_dicts[0]['''id'''][0] , '''1''' ) # max inner product is reached with second doc
self.assertEqual(doc_dicts[1]['''id'''][0] , '''0''' ) # max inner product is reached with first doc
self.assertListEqual(doc_ids.tolist() , [[1], [0]] )
def A ( self : Optional[int] ) -> List[Any]:
UpperCAmelCase : Tuple = self.get_dummy_custom_hf_index_retriever(from_disk=_snake_case )
with tempfile.TemporaryDirectory() as tmp_dirname:
retriever.save_pretrained(_snake_case )
UpperCAmelCase : List[Any] = RagRetriever.from_pretrained(_snake_case )
self.assertIsInstance(_snake_case , _snake_case )
UpperCAmelCase : Dict = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase : Optional[Any] = retriever.retrieve(_snake_case , n_docs=1 )
self.assertTrue(out is not None )
def A ( self : Optional[Any] ) -> List[str]:
UpperCAmelCase : Tuple = 1
UpperCAmelCase : Any = self.get_dummy_legacy_index_retriever()
UpperCAmelCase : List[str] = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase , UpperCAmelCase , UpperCAmelCase : Optional[int] = retriever.retrieve(_snake_case , n_docs=_snake_case )
self.assertEqual(retrieved_doc_embeds.shape , (2, n_docs, self.retrieval_vector_size) )
self.assertEqual(len(_snake_case ) , 2 )
self.assertEqual(sorted(doc_dicts[0] ) , ['''text''', '''title'''] )
self.assertEqual(len(doc_dicts[0]['''text'''] ) , _snake_case )
self.assertEqual(doc_dicts[0]['''text'''][0] , '''bar''' ) # max inner product is reached with second doc
self.assertEqual(doc_dicts[1]['''text'''][0] , '''foo''' ) # max inner product is reached with first doc
self.assertListEqual(doc_ids.tolist() , [[1], [0]] )
def A ( self : List[Any] ) -> Tuple:
UpperCAmelCase : Dict = self.get_dummy_legacy_index_retriever()
with tempfile.TemporaryDirectory() as tmp_dirname:
retriever.save_pretrained(_snake_case )
UpperCAmelCase : List[Any] = RagRetriever.from_pretrained(_snake_case )
self.assertIsInstance(_snake_case , _snake_case )
UpperCAmelCase : Tuple = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase : Dict = retriever.retrieve(_snake_case , n_docs=1 )
self.assertTrue(out is not None )
@require_torch
@require_tokenizers
@require_sentencepiece
def A ( self : Union[str, Any] ) -> List[Any]:
import torch
UpperCAmelCase : str = 1
UpperCAmelCase : int = self.get_dummy_canonical_hf_index_retriever()
UpperCAmelCase : Optional[int] = [[5, 7], [10, 11]]
UpperCAmelCase : Union[str, Any] = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase : List[Any] = retriever(_snake_case , _snake_case , prefix=retriever.config.generator.prefix , n_docs=_snake_case )
UpperCAmelCase , UpperCAmelCase , UpperCAmelCase : Any = (
out['''context_input_ids'''],
out['''context_attention_mask'''],
out['''retrieved_doc_embeds'''],
)
self.assertEqual(retrieved_doc_embeds.shape , (2, n_docs, self.retrieval_vector_size) )
self.assertIsInstance(_snake_case , _snake_case )
self.assertIsInstance(_snake_case , _snake_case )
self.assertIsInstance(_snake_case , np.ndarray )
UpperCAmelCase : Optional[Any] = retriever(
_snake_case , _snake_case , prefix=retriever.config.generator.prefix , n_docs=_snake_case , return_tensors='''pt''' , )
UpperCAmelCase , UpperCAmelCase , UpperCAmelCase , UpperCAmelCase : Dict = ( # noqa: F841
out['''context_input_ids'''],
out['''context_attention_mask'''],
out['''retrieved_doc_embeds'''],
out['''doc_ids'''],
)
self.assertEqual(retrieved_doc_embeds.shape , (2, n_docs, self.retrieval_vector_size) )
self.assertIsInstance(_snake_case , torch.Tensor )
self.assertIsInstance(_snake_case , torch.Tensor )
self.assertIsInstance(_snake_case , torch.Tensor )
@require_torch
@require_tokenizers
@require_sentencepiece
def A ( self : Tuple ) -> int:
UpperCAmelCase : Dict = self.get_dpr_ctx_encoder_tokenizer()
UpperCAmelCase : int = 1
UpperCAmelCase : Union[str, Any] = self.get_dummy_custom_hf_index_retriever(from_disk=_snake_case )
retriever.set_ctx_encoder_tokenizer(_snake_case )
UpperCAmelCase : str = [[5, 7], [10, 11]]
UpperCAmelCase : Dict = np.array(
[np.ones(self.retrieval_vector_size ), -np.ones(self.retrieval_vector_size )] , dtype=np.floataa )
UpperCAmelCase : Optional[int] = retriever(_snake_case , _snake_case , prefix=retriever.config.generator.prefix , n_docs=_snake_case )
self.assertEqual(
len(_snake_case ) , 6 ) # check whether the retriever output consist of 6 attributes including tokenized docs
self.assertEqual(
all(k in out for k in ('''tokenized_doc_ids''', '''tokenized_doc_attention_mask''') ) , _snake_case ) # check for doc token related keys in dictionary.
| 23
|
# This model implementation is heavily inspired by https://github.com/haofanwang/ControlNet-for-Diffusers/
import gc
import random
import tempfile
import unittest
import numpy as np
import torch
from PIL import Image
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
ControlNetModel,
DDIMScheduler,
StableDiffusionControlNetImgaImgPipeline,
UNetaDConditionModel,
)
from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_controlnet import MultiControlNetModel
from diffusers.utils import floats_tensor, load_image, load_numpy, randn_tensor, slow, torch_device
from diffusers.utils.import_utils import is_xformers_available
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import (
IMAGE_TO_IMAGE_IMAGE_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_PARAMS,
)
from ..test_pipelines_common import (
PipelineKarrasSchedulerTesterMixin,
PipelineLatentTesterMixin,
PipelineTesterMixin,
)
enable_full_determinism()
class __snake_case ( a , a , a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = StableDiffusionControlNetImgaImgPipeline
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''}
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
UpperCAmelCase__ : Union[str, Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS.union({'''control_image'''} )
UpperCAmelCase__ : Optional[Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS
def lowerCamelCase ( self : int):
"""simple docstring"""
torch.manual_seed(0)
UpperCAmelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , )
torch.manual_seed(0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
torch.manual_seed(0)
UpperCAmelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='''scaled_linear''' , clip_sample=_snake_case , set_alpha_to_one=_snake_case , )
torch.manual_seed(0)
UpperCAmelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , )
torch.manual_seed(0)
UpperCAmelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCAmelCase_ = CLIPTextModel(_snake_case)
UpperCAmelCase_ = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''')
UpperCAmelCase_ = {
'''unet''': unet,
'''controlnet''': controlnet,
'''scheduler''': scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def lowerCamelCase ( self : Union[str, Any] , _snake_case : Any , _snake_case : Dict=0):
"""simple docstring"""
if str(_snake_case).startswith('''mps'''):
UpperCAmelCase_ = torch.manual_seed(_snake_case)
else:
UpperCAmelCase_ = torch.Generator(device=_snake_case).manual_seed(_snake_case)
UpperCAmelCase_ = 2
UpperCAmelCase_ = randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , )
UpperCAmelCase_ = floats_tensor(control_image.shape , rng=random.Random(_snake_case)).to(_snake_case)
UpperCAmelCase_ = image.cpu().permute(0 , 2 , 3 , 1)[0]
UpperCAmelCase_ = Image.fromarray(np.uinta(_snake_case)).convert('''RGB''').resize((64, 64))
UpperCAmelCase_ = {
'''prompt''': '''A painting of a squirrel eating a burger''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
'''image''': image,
'''control_image''': control_image,
}
return inputs
def lowerCamelCase ( self : Any):
"""simple docstring"""
return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3)
@unittest.skipIf(
torch_device != '''cuda''' or not is_xformers_available() , reason='''XFormers attention is only available with CUDA and `xformers` installed''' , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
self._test_inference_batch_single_identical(expected_max_diff=2e-3)
class __snake_case ( a , a , unittest.TestCase ):
UpperCAmelCase__ : str = StableDiffusionControlNetImgaImgPipeline
UpperCAmelCase__ : List[str] = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''}
UpperCAmelCase__ : Optional[Any] = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
UpperCAmelCase__ : str = frozenset([] ) # TO_DO: add image_params once refactored VaeImageProcessor.preprocess
def lowerCamelCase ( self : str):
"""simple docstring"""
torch.manual_seed(0)
UpperCAmelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , up_block_types=('''CrossAttnUpBlock2D''', '''UpBlock2D''') , cross_attention_dim=32 , )
torch.manual_seed(0)
def init_weights(_snake_case : Optional[int]):
if isinstance(_snake_case , torch.nn.Convad):
torch.nn.init.normal(m.weight)
m.bias.data.fill_(1.0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
controlneta.controlnet_down_blocks.apply(_snake_case)
torch.manual_seed(0)
UpperCAmelCase_ = ControlNetModel(
block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=('''DownBlock2D''', '''CrossAttnDownBlock2D''') , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , )
controlneta.controlnet_down_blocks.apply(_snake_case)
torch.manual_seed(0)
UpperCAmelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='''scaled_linear''' , clip_sample=_snake_case , set_alpha_to_one=_snake_case , )
torch.manual_seed(0)
UpperCAmelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['''DownEncoderBlock2D''', '''DownEncoderBlock2D'''] , up_block_types=['''UpDecoderBlock2D''', '''UpDecoderBlock2D'''] , latent_channels=4 , )
torch.manual_seed(0)
UpperCAmelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCAmelCase_ = CLIPTextModel(_snake_case)
UpperCAmelCase_ = CLIPTokenizer.from_pretrained('''hf-internal-testing/tiny-random-clip''')
UpperCAmelCase_ = MultiControlNetModel([controlneta, controlneta])
UpperCAmelCase_ = {
'''unet''': unet,
'''controlnet''': controlnet,
'''scheduler''': scheduler,
'''vae''': vae,
'''text_encoder''': text_encoder,
'''tokenizer''': tokenizer,
'''safety_checker''': None,
'''feature_extractor''': None,
}
return components
def lowerCamelCase ( self : int , _snake_case : Union[str, Any] , _snake_case : str=0):
"""simple docstring"""
if str(_snake_case).startswith('''mps'''):
UpperCAmelCase_ = torch.manual_seed(_snake_case)
else:
UpperCAmelCase_ = torch.Generator(device=_snake_case).manual_seed(_snake_case)
UpperCAmelCase_ = 2
UpperCAmelCase_ = [
randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , ),
randn_tensor(
(1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=_snake_case , device=torch.device(_snake_case) , ),
]
UpperCAmelCase_ = floats_tensor(control_image[0].shape , rng=random.Random(_snake_case)).to(_snake_case)
UpperCAmelCase_ = image.cpu().permute(0 , 2 , 3 , 1)[0]
UpperCAmelCase_ = Image.fromarray(np.uinta(_snake_case)).convert('''RGB''').resize((64, 64))
UpperCAmelCase_ = {
'''prompt''': '''A painting of a squirrel eating a burger''',
'''generator''': generator,
'''num_inference_steps''': 2,
'''guidance_scale''': 6.0,
'''output_type''': '''numpy''',
'''image''': image,
'''control_image''': control_image,
}
return inputs
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_dummy_components()
UpperCAmelCase_ = self.pipeline_class(**_snake_case)
pipe.to(_snake_case)
UpperCAmelCase_ = 1_0.0
UpperCAmelCase_ = 4
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case)[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=0.1 , control_guidance_end=0.2)[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=[0.1, 0.3] , control_guidance_end=[0.2, 0.7])[0]
UpperCAmelCase_ = self.get_dummy_inputs(_snake_case)
UpperCAmelCase_ = steps
UpperCAmelCase_ = scale
UpperCAmelCase_ = pipe(**_snake_case , control_guidance_start=0.4 , control_guidance_end=[0.5, 0.8])[0]
# make sure that all outputs are different
assert np.sum(np.abs(output_a - output_a)) > 1e-3
assert np.sum(np.abs(output_a - output_a)) > 1e-3
assert np.sum(np.abs(output_a - output_a)) > 1e-3
def lowerCamelCase ( self : Dict):
"""simple docstring"""
return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3)
@unittest.skipIf(
torch_device != '''cuda''' or not is_xformers_available() , reason='''XFormers attention is only available with CUDA and `xformers` installed''' , )
def lowerCamelCase ( self : int):
"""simple docstring"""
self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3)
def lowerCamelCase ( self : int):
"""simple docstring"""
self._test_inference_batch_single_identical(expected_max_diff=2e-3)
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = self.get_dummy_components()
UpperCAmelCase_ = self.pipeline_class(**_snake_case)
pipe.to(_snake_case)
pipe.set_progress_bar_config(disable=_snake_case)
with tempfile.TemporaryDirectory() as tmpdir:
try:
# save_pretrained is not implemented for Multi-ControlNet
pipe.save_pretrained(_snake_case)
except NotImplementedError:
pass
@slow
@require_torch_gpu
class __snake_case ( unittest.TestCase ):
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
UpperCAmelCase_ = ControlNetModel.from_pretrained('''lllyasviel/sd-controlnet-canny''')
UpperCAmelCase_ = StableDiffusionControlNetImgaImgPipeline.from_pretrained(
'''runwayml/stable-diffusion-v1-5''' , safety_checker=_snake_case , controlnet=_snake_case)
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=_snake_case)
UpperCAmelCase_ = torch.Generator(device='''cpu''').manual_seed(0)
UpperCAmelCase_ = '''evil space-punk bird'''
UpperCAmelCase_ = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/bird_canny.png''').resize((512, 512))
UpperCAmelCase_ = load_image(
'''https://huggingface.co/lllyasviel/sd-controlnet-canny/resolve/main/images/bird.png''').resize((512, 512))
UpperCAmelCase_ = pipe(
_snake_case , _snake_case , control_image=_snake_case , generator=_snake_case , output_type='''np''' , num_inference_steps=50 , strength=0.6 , )
UpperCAmelCase_ = output.images[0]
assert image.shape == (512, 512, 3)
UpperCAmelCase_ = load_numpy(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/img2img.npy''')
assert np.abs(expected_image - image).max() < 9e-2
| 51
| 0
|
import argparse
import json
from pathlib import Path
import requests
import timm
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import DeiTImageProcessor, ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
from transformers.utils import logging
logging.set_verbosity_info()
_UpperCAmelCase : Dict = logging.get_logger(__name__)
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase=False ) -> Union[str, Any]:
lowerCamelCase__ : Tuple = []
for i in range(config.num_hidden_layers ):
# encoder layers: output projection, 2 feedforward neural networks and 2 layernorms
rename_keys.append((F"""blocks.{i}.norm1.weight""", F"""vit.encoder.layer.{i}.layernorm_before.weight""") )
rename_keys.append((F"""blocks.{i}.norm1.bias""", F"""vit.encoder.layer.{i}.layernorm_before.bias""") )
rename_keys.append((F"""blocks.{i}.attn.proj.weight""", F"""vit.encoder.layer.{i}.attention.output.dense.weight""") )
rename_keys.append((F"""blocks.{i}.attn.proj.bias""", F"""vit.encoder.layer.{i}.attention.output.dense.bias""") )
rename_keys.append((F"""blocks.{i}.norm2.weight""", F"""vit.encoder.layer.{i}.layernorm_after.weight""") )
rename_keys.append((F"""blocks.{i}.norm2.bias""", F"""vit.encoder.layer.{i}.layernorm_after.bias""") )
rename_keys.append((F"""blocks.{i}.mlp.fc1.weight""", F"""vit.encoder.layer.{i}.intermediate.dense.weight""") )
rename_keys.append((F"""blocks.{i}.mlp.fc1.bias""", F"""vit.encoder.layer.{i}.intermediate.dense.bias""") )
rename_keys.append((F"""blocks.{i}.mlp.fc2.weight""", F"""vit.encoder.layer.{i}.output.dense.weight""") )
rename_keys.append((F"""blocks.{i}.mlp.fc2.bias""", F"""vit.encoder.layer.{i}.output.dense.bias""") )
# projection layer + position embeddings
rename_keys.extend(
[
('cls_token', 'vit.embeddings.cls_token'),
('patch_embed.proj.weight', 'vit.embeddings.patch_embeddings.projection.weight'),
('patch_embed.proj.bias', 'vit.embeddings.patch_embeddings.projection.bias'),
('pos_embed', 'vit.embeddings.position_embeddings'),
] )
if base_model:
# layernorm + pooler
rename_keys.extend(
[
('norm.weight', 'layernorm.weight'),
('norm.bias', 'layernorm.bias'),
('pre_logits.fc.weight', 'pooler.dense.weight'),
('pre_logits.fc.bias', 'pooler.dense.bias'),
] )
# if just the base model, we should remove "vit" from all keys that start with "vit"
lowerCamelCase__ : Any = [(pair[0], pair[1][4:]) if pair[1].startswith('vit' ) else pair for pair in rename_keys]
else:
# layernorm + classification head
rename_keys.extend(
[
('norm.weight', 'vit.layernorm.weight'),
('norm.bias', 'vit.layernorm.bias'),
('head.weight', 'classifier.weight'),
('head.bias', 'classifier.bias'),
] )
return rename_keys
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase=False ) -> Tuple:
for i in range(config.num_hidden_layers ):
if base_model:
lowerCamelCase__ : Optional[int] = ''
else:
lowerCamelCase__ : List[str] = 'vit.'
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
lowerCamelCase__ : str = state_dict.pop(F"""blocks.{i}.attn.qkv.weight""" )
lowerCamelCase__ : List[str] = state_dict.pop(F"""blocks.{i}.attn.qkv.bias""" )
# next, add query, keys and values (in that order) to the state dict
lowerCamelCase__ : int = in_proj_weight[
: config.hidden_size, :
]
lowerCamelCase__ : str = in_proj_bias[: config.hidden_size]
lowerCamelCase__ : int = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
lowerCamelCase__ : Any = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
lowerCamelCase__ : str = in_proj_weight[
-config.hidden_size :, :
]
lowerCamelCase__ : List[Any] = in_proj_bias[-config.hidden_size :]
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase ) -> Union[str, Any]:
lowerCamelCase__ : int = ['head.weight', 'head.bias']
for k in ignore_keys:
state_dict.pop(__A , __A )
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> int:
lowerCamelCase__ : Union[str, Any] = dct.pop(__A )
lowerCamelCase__ : List[str] = val
def SCREAMING_SNAKE_CASE ( ) -> List[Any]:
lowerCamelCase__ : List[str] = 'http://images.cocodataset.org/val2017/000000039769.jpg'
lowerCamelCase__ : Optional[Any] = Image.open(requests.get(__A , stream=__A ).raw )
return im
@torch.no_grad()
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
lowerCamelCase__ : Dict = ViTConfig()
lowerCamelCase__ : Union[str, Any] = False
# dataset (ImageNet-21k only or also fine-tuned on ImageNet 2012), patch_size and image_size
if vit_name[-5:] == "in21k":
lowerCamelCase__ : Dict = True
lowerCamelCase__ : Optional[Any] = int(vit_name[-12:-10] )
lowerCamelCase__ : Optional[Any] = int(vit_name[-9:-6] )
else:
lowerCamelCase__ : Optional[int] = 1000
lowerCamelCase__ : str = 'huggingface/label-files'
lowerCamelCase__ : List[Any] = 'imagenet-1k-id2label.json'
lowerCamelCase__ : List[Any] = json.load(open(hf_hub_download(__A , __A , repo_type='dataset' ) , 'r' ) )
lowerCamelCase__ : Dict = {int(__A ): v for k, v in idalabel.items()}
lowerCamelCase__ : Optional[int] = idalabel
lowerCamelCase__ : Optional[Any] = {v: k for k, v in idalabel.items()}
lowerCamelCase__ : Optional[int] = int(vit_name[-6:-4] )
lowerCamelCase__ : Optional[int] = int(vit_name[-3:] )
# size of the architecture
if "deit" in vit_name:
if vit_name[9:].startswith('tiny' ):
lowerCamelCase__ : Any = 192
lowerCamelCase__ : str = 768
lowerCamelCase__ : Tuple = 12
lowerCamelCase__ : Any = 3
elif vit_name[9:].startswith('small' ):
lowerCamelCase__ : Dict = 384
lowerCamelCase__ : int = 1536
lowerCamelCase__ : Tuple = 12
lowerCamelCase__ : List[Any] = 6
else:
pass
else:
if vit_name[4:].startswith('small' ):
lowerCamelCase__ : Optional[int] = 768
lowerCamelCase__ : Tuple = 2304
lowerCamelCase__ : List[str] = 8
lowerCamelCase__ : List[str] = 8
elif vit_name[4:].startswith('base' ):
pass
elif vit_name[4:].startswith('large' ):
lowerCamelCase__ : int = 1024
lowerCamelCase__ : Any = 4096
lowerCamelCase__ : Tuple = 24
lowerCamelCase__ : int = 16
elif vit_name[4:].startswith('huge' ):
lowerCamelCase__ : Any = 1280
lowerCamelCase__ : int = 5120
lowerCamelCase__ : str = 32
lowerCamelCase__ : List[Any] = 16
# load original model from timm
lowerCamelCase__ : List[str] = timm.create_model(__A , pretrained=__A )
timm_model.eval()
# load state_dict of original model, remove and rename some keys
lowerCamelCase__ : int = timm_model.state_dict()
if base_model:
remove_classification_head_(__A )
lowerCamelCase__ : Optional[int] = create_rename_keys(__A , __A )
for src, dest in rename_keys:
rename_key(__A , __A , __A )
read_in_q_k_v(__A , __A , __A )
# load HuggingFace model
if vit_name[-5:] == "in21k":
lowerCamelCase__ : Tuple = ViTModel(__A ).eval()
else:
lowerCamelCase__ : List[str] = ViTForImageClassification(__A ).eval()
model.load_state_dict(__A )
# Check outputs on an image, prepared by ViTImageProcessor/DeiTImageProcessor
if "deit" in vit_name:
lowerCamelCase__ : Union[str, Any] = DeiTImageProcessor(size=config.image_size )
else:
lowerCamelCase__ : Dict = ViTImageProcessor(size=config.image_size )
lowerCamelCase__ : Dict = image_processor(images=prepare_img() , return_tensors='pt' )
lowerCamelCase__ : Optional[int] = encoding['pixel_values']
lowerCamelCase__ : List[Any] = model(__A )
if base_model:
lowerCamelCase__ : Union[str, Any] = timm_model.forward_features(__A )
assert timm_pooled_output.shape == outputs.pooler_output.shape
assert torch.allclose(__A , outputs.pooler_output , atol=1e-3 )
else:
lowerCamelCase__ : Any = timm_model(__A )
assert timm_logits.shape == outputs.logits.shape
assert torch.allclose(__A , outputs.logits , atol=1e-3 )
Path(__A ).mkdir(exist_ok=__A )
print(F"""Saving model {vit_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(__A )
print(F"""Saving image processor to {pytorch_dump_folder_path}""" )
image_processor.save_pretrained(__A )
if __name__ == "__main__":
_UpperCAmelCase : Tuple = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--vit_name""",
default="""vit_base_patch16_224""",
type=str,
help="""Name of the ViT timm model you'd like to convert.""",
)
parser.add_argument(
"""--pytorch_dump_folder_path""", default=None, type=str, help="""Path to the output PyTorch model directory."""
)
_UpperCAmelCase : List[str] = parser.parse_args()
convert_vit_checkpoint(args.vit_name, args.pytorch_dump_folder_path)
| 50
|
import random
import timeit
from functools import wraps
from typing import Callable, Optional
from ..configuration_utils import PretrainedConfig
from ..models.auto.modeling_tf_auto import TF_MODEL_MAPPING, TF_MODEL_WITH_LM_HEAD_MAPPING
from ..utils import is_pyanvml_available, is_tf_available, logging
from .benchmark_utils import (
Benchmark,
Memory,
MemorySummary,
measure_peak_memory_cpu,
start_memory_tracing,
stop_memory_tracing,
)
if is_tf_available():
import tensorflow as tf
from tensorflow.python.framework.errors_impl import ResourceExhaustedError
from .benchmark_args_tf import TensorFlowBenchmarkArguments
if is_pyanvml_available():
import pyanvml.pyanvml as nvml
snake_case_ : Tuple = logging.get_logger(__name__)
def A (__A : bool , __A : bool ) -> Optional[Any]:
"""simple docstring"""
def run_func(__A : Optional[Any] ):
@wraps(__A )
def run_in_eager_mode(*__A : Dict , **__A : List[Any] ):
return func(*__A , **__A )
@wraps(__A )
@tf.function(experimental_compile=__A )
def run_in_graph_mode(*__A : Optional[Any] , **__A : Any ):
return func(*__A , **__A )
if do_eager_mode is True:
if use_xla is not False:
raise ValueError(
'''Cannot run model in XLA, if `args.eager_mode` is set to `True`. Please set `args.eager_mode=False`.''' )
return run_in_eager_mode
else:
return run_in_graph_mode
return run_func
def A (__A : int , __A : int , __A : int ) -> ["tf.Tensor"]:
"""simple docstring"""
UpperCAmelCase_ = random.Random()
UpperCAmelCase_ = [rng.randint(0 , vocab_size - 1 ) for i in range(batch_size * sequence_length )]
return tf.constant(__A , shape=(batch_size, sequence_length) , dtype=tf.intaa )
class __snake_case ( a ):
UpperCAmelCase__ : TensorFlowBenchmarkArguments
UpperCAmelCase__ : PretrainedConfig
UpperCAmelCase__ : str = "TensorFlow"
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return tf.__version__
def lowerCamelCase ( self : Dict , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case)
return self._measure_speed(_inference)
def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case)
return self._measure_speed(_train)
def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
if self.args.is_gpu:
tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case)
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case)
return self._measure_memory(_inference)
def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
if self.args.is_gpu:
tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case)
UpperCAmelCase_ = self.args.strategy
if strategy is None:
raise ValueError('''A device strategy has to be initialized before using TensorFlow.''')
UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case)
return self._measure_memory(_train)
def lowerCamelCase ( self : Optional[int] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.config_dict[model_name]
if self.args.fpaa:
raise NotImplementedError('''Mixed precision is currently not supported.''')
UpperCAmelCase_ = (
hasattr(_snake_case , '''architectures''')
and isinstance(config.architectures , _snake_case)
and len(config.architectures) > 0
)
if not self.args.only_pretrain_model and has_model_class_in_config:
try:
UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model
UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class])
UpperCAmelCase_ = getattr(_snake_case , _snake_case)
UpperCAmelCase_ = model_cls(_snake_case)
except ImportError:
raise ImportError(
F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to"""
''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''')
else:
UpperCAmelCase_ = TF_MODEL_MAPPING[config.__class__](_snake_case)
# encoder-decoder has vocab size saved differently
UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size
UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_decoder_forward():
return model(_snake_case , decoder_input_ids=_snake_case , training=_snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_forward():
return model(_snake_case , training=_snake_case)
UpperCAmelCase_ = encoder_decoder_forward if config.is_encoder_decoder else encoder_forward
return _inference
def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.config_dict[model_name]
if self.args.eager_mode is not False:
raise ValueError('''Training cannot be done in eager mode. Please make sure that `args.eager_mode = False`.''')
if self.args.fpaa:
raise NotImplementedError('''Mixed precision is currently not supported.''')
UpperCAmelCase_ = (
hasattr(_snake_case , '''architectures''')
and isinstance(config.architectures , _snake_case)
and len(config.architectures) > 0
)
if not self.args.only_pretrain_model and has_model_class_in_config:
try:
UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model
UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class])
UpperCAmelCase_ = getattr(_snake_case , _snake_case)
UpperCAmelCase_ = model_cls(_snake_case)
except ImportError:
raise ImportError(
F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to"""
''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''')
else:
UpperCAmelCase_ = TF_MODEL_WITH_LM_HEAD_MAPPING[config.__class__](_snake_case)
# encoder-decoder has vocab size saved differently
UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size
UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case)
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_decoder_train():
UpperCAmelCase_ = model(_snake_case , decoder_input_ids=_snake_case , labels=_snake_case , training=_snake_case)[0]
UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables)
return gradients
@run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla)
def encoder_train():
UpperCAmelCase_ = model(_snake_case , labels=_snake_case , training=_snake_case)[0]
UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables)
return gradients
UpperCAmelCase_ = encoder_decoder_train if config.is_encoder_decoder else encoder_train
return _train
def lowerCamelCase ( self : Any , _snake_case : Optional[Any]):
"""simple docstring"""
with self.args.strategy.scope():
try:
if self.args.is_tpu or self.args.use_xla:
# run additional 10 times to stabilize compilation for tpu
logger.info('''Do inference on TPU. Running model 5 times to stabilize compilation''')
timeit.repeat(_snake_case , repeat=1 , number=5)
# as written in https://docs.python.org/2/library/timeit.html#timeit.Timer.repeat, min should be taken rather than the average
UpperCAmelCase_ = timeit.repeat(
_snake_case , repeat=self.args.repeat , number=10 , )
return min(_snake_case) / 1_0.0
except ResourceExhaustedError as e:
self.print_fn(F"""Doesn't fit on GPU. {e}""")
def lowerCamelCase ( self : Dict , _snake_case : Callable[[], None]):
"""simple docstring"""
logger.info(
'''Note that TensorFlow allocates more memory than '''
'''it might need to speed up computation. '''
'''The memory reported here corresponds to the memory '''
'''reported by `nvidia-smi`, which can vary depending '''
'''on total available memory on the GPU that is used.''')
with self.args.strategy.scope():
try:
if self.args.trace_memory_line_by_line:
if not self.args.eager_mode:
raise ValueError(
'''`args.eager_mode` is set to `False`. Make sure to run model in eager mode to measure memory'''
''' consumption line by line.''')
UpperCAmelCase_ = start_memory_tracing('''transformers''')
if self.args.is_tpu:
# tpu
raise NotImplementedError(
'''Memory Benchmarking is currently not implemented for TPU. Please disable memory benchmarking'''
''' with `args.memory=False`''')
elif self.args.is_gpu:
# gpu
if not is_pyanvml_available():
logger.warning(
'''py3nvml not installed, we won\'t log GPU memory usage. '''
'''Install py3nvml (pip install py3nvml) to log information about GPU.''')
UpperCAmelCase_ = '''N/A'''
else:
logger.info(
'''Measuring total GPU usage on GPU device. Make sure to not have additional processes'''
''' running on the same GPU.''')
# init nvml
nvml.nvmlInit()
func()
UpperCAmelCase_ = nvml.nvmlDeviceGetHandleByIndex(self.args.device_idx)
UpperCAmelCase_ = nvml.nvmlDeviceGetMemoryInfo(_snake_case)
UpperCAmelCase_ = meminfo.used
UpperCAmelCase_ = Memory(_snake_case)
# shutdown nvml
nvml.nvmlShutdown()
else:
# cpu
if self.args.trace_memory_line_by_line:
logger.info(
'''When enabling line by line tracing, the max peak memory for CPU is inaccurate in'''
''' TensorFlow.''')
UpperCAmelCase_ = None
else:
UpperCAmelCase_ = measure_peak_memory_cpu(_snake_case)
UpperCAmelCase_ = Memory(_snake_case) if isinstance(_snake_case , _snake_case) else memory_bytes
if self.args.trace_memory_line_by_line:
UpperCAmelCase_ = stop_memory_tracing(_snake_case)
if memory is None:
UpperCAmelCase_ = summary.total
else:
UpperCAmelCase_ = None
return memory, summary
except ResourceExhaustedError as e:
self.print_fn(F"""Doesn't fit on GPU. {e}""")
return "N/A", None
| 51
| 0
|
import inspect
from typing import Callable, List, Optional, Union
import torch
from transformers import (
CLIPImageProcessor,
CLIPTextModel,
CLIPTokenizer,
WhisperForConditionalGeneration,
WhisperProcessor,
)
from diffusers import (
AutoencoderKL,
DDIMScheduler,
DiffusionPipeline,
LMSDiscreteScheduler,
PNDMScheduler,
UNetaDConditionModel,
)
from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion import StableDiffusionPipelineOutput
from diffusers.pipelines.stable_diffusion.safety_checker import StableDiffusionSafetyChecker
from diffusers.utils import logging
_lowercase : str =logging.get_logger(__name__) # pylint: disable=invalid-name
class snake_case__ (A__ ):
"""simple docstring"""
def __init__( self , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , ) -> Optional[int]:
"""simple docstring"""
super().__init__()
if safety_checker is None:
logger.warning(
F'''You have disabled the safety checker for {self.__class__} by passing `safety_checker=None`. Ensure'''
""" that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered"""
""" results in services or applications open to the public. Both the diffusers team and Hugging Face"""
""" strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling"""
""" it only for use-cases that involve analyzing network behavior or auditing its results. For more"""
""" information, please have a look at https://github.com/huggingface/diffusers/pull/254 .""" )
self.register_modules(
speech_model=_snake_case , speech_processor=_snake_case , vae=_snake_case , text_encoder=_snake_case , tokenizer=_snake_case , unet=_snake_case , scheduler=_snake_case , feature_extractor=_snake_case , )
def SCREAMING_SNAKE_CASE__( self , __lowercase = "auto" ) -> Optional[Any]:
"""simple docstring"""
if slice_size == "auto":
a__ : int = self.unet.config.attention_head_dim // 2
self.unet.set_attention_slice(_snake_case )
def SCREAMING_SNAKE_CASE__( self ) -> Union[str, Any]:
"""simple docstring"""
self.enable_attention_slicing(_snake_case )
@torch.no_grad()
def __call__( self , __lowercase , __lowercase=1_6_0_0_0 , __lowercase = 5_1_2 , __lowercase = 5_1_2 , __lowercase = 5_0 , __lowercase = 7.5 , __lowercase = None , __lowercase = 1 , __lowercase = 0.0 , __lowercase = None , __lowercase = None , __lowercase = "pil" , __lowercase = True , __lowercase = None , __lowercase = 1 , **__lowercase , ) -> Optional[Any]:
"""simple docstring"""
a__ : Any = self.speech_processor.feature_extractor(
_snake_case , return_tensors="""pt""" , sampling_rate=_snake_case ).input_features.to(self.device )
a__ : Any = self.speech_model.generate(_snake_case , max_length=4_8_0_0_0_0 )
a__ : Optional[int] = self.speech_processor.tokenizer.batch_decode(_snake_case , skip_special_tokens=_snake_case , normalize=_snake_case )[
0
]
if isinstance(_snake_case , _snake_case ):
a__ : Optional[Any] = 1
elif isinstance(_snake_case , _snake_case ):
a__ : Optional[int] = len(_snake_case )
else:
raise ValueError(F'''`prompt` has to be of type `str` or `list` but is {type(_snake_case )}''' )
if height % 8 != 0 or width % 8 != 0:
raise ValueError(F'''`height` and `width` have to be divisible by 8 but are {height} and {width}.''' )
if (callback_steps is None) or (
callback_steps is not None and (not isinstance(_snake_case , _snake_case ) or callback_steps <= 0)
):
raise ValueError(
F'''`callback_steps` has to be a positive integer but is {callback_steps} of type'''
F''' {type(_snake_case )}.''' )
# get prompt text embeddings
a__ : Tuple = self.tokenizer(
_snake_case , padding="""max_length""" , max_length=self.tokenizer.model_max_length , return_tensors="""pt""" , )
a__ : Union[str, Any] = text_inputs.input_ids
if text_input_ids.shape[-1] > self.tokenizer.model_max_length:
a__ : Optional[Any] = self.tokenizer.batch_decode(text_input_ids[:, self.tokenizer.model_max_length :] )
logger.warning(
"""The following part of your input was truncated because CLIP can only handle sequences up to"""
F''' {self.tokenizer.model_max_length} tokens: {removed_text}''' )
a__ : Any = text_input_ids[:, : self.tokenizer.model_max_length]
a__ : Tuple = self.text_encoder(text_input_ids.to(self.device ) )[0]
# duplicate text embeddings for each generation per prompt, using mps friendly method
a__ , a__ , a__ : List[Any] = text_embeddings.shape
a__ : Tuple = text_embeddings.repeat(1 , _snake_case , 1 )
a__ : Dict = text_embeddings.view(bs_embed * num_images_per_prompt , _snake_case , -1 )
# here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
# of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
# corresponds to doing no classifier free guidance.
a__ : Optional[Any] = guidance_scale > 1.0
# get unconditional embeddings for classifier free guidance
if do_classifier_free_guidance:
a__ : Optional[Any] = 4_2
if negative_prompt is None:
a__ : Dict = [""""""] * batch_size
elif type(_snake_case ) is not type(_snake_case ):
raise TypeError(
F'''`negative_prompt` should be the same type to `prompt`, but got {type(_snake_case )} !='''
F''' {type(_snake_case )}.''' )
elif isinstance(_snake_case , _snake_case ):
a__ : Dict = [negative_prompt]
elif batch_size != len(_snake_case ):
raise ValueError(
F'''`negative_prompt`: {negative_prompt} has batch size {len(_snake_case )}, but `prompt`:'''
F''' {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches'''
""" the batch size of `prompt`.""" )
else:
a__ : Tuple = negative_prompt
a__ : int = text_input_ids.shape[-1]
a__ : List[Any] = self.tokenizer(
_snake_case , padding="""max_length""" , max_length=_snake_case , truncation=_snake_case , return_tensors="""pt""" , )
a__ : Tuple = self.text_encoder(uncond_input.input_ids.to(self.device ) )[0]
# duplicate unconditional embeddings for each generation per prompt, using mps friendly method
a__ : Union[str, Any] = uncond_embeddings.shape[1]
a__ : Union[str, Any] = uncond_embeddings.repeat(1 , _snake_case , 1 )
a__ : Any = uncond_embeddings.view(batch_size * num_images_per_prompt , _snake_case , -1 )
# For classifier free guidance, we need to do two forward passes.
# Here we concatenate the unconditional and text embeddings into a single batch
# to avoid doing two forward passes
a__ : Optional[Any] = torch.cat([uncond_embeddings, text_embeddings] )
# get the initial random noise unless the user supplied it
# Unlike in other pipelines, latents need to be generated in the target device
# for 1-to-1 results reproducibility with the CompVis implementation.
# However this currently doesn't work in `mps`.
a__ : Optional[int] = (batch_size * num_images_per_prompt, self.unet.config.in_channels, height // 8, width // 8)
a__ : Optional[int] = text_embeddings.dtype
if latents is None:
if self.device.type == "mps":
# randn does not exist on mps
a__ : int = torch.randn(_snake_case , generator=_snake_case , device="""cpu""" , dtype=_snake_case ).to(
self.device )
else:
a__ : Tuple = torch.randn(_snake_case , generator=_snake_case , device=self.device , dtype=_snake_case )
else:
if latents.shape != latents_shape:
raise ValueError(F'''Unexpected latents shape, got {latents.shape}, expected {latents_shape}''' )
a__ : List[Any] = latents.to(self.device )
# set timesteps
self.scheduler.set_timesteps(_snake_case )
# Some schedulers like PNDM have timesteps as arrays
# It's more optimized to move all timesteps to correct device beforehand
a__ : Optional[int] = self.scheduler.timesteps.to(self.device )
# scale the initial noise by the standard deviation required by the scheduler
a__ : int = latents * self.scheduler.init_noise_sigma
# prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
# eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
# eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502
# and should be between [0, 1]
a__ : Tuple = """eta""" in set(inspect.signature(self.scheduler.step ).parameters.keys() )
a__ : Optional[int] = {}
if accepts_eta:
a__ : Tuple = eta
for i, t in enumerate(self.progress_bar(_snake_case ) ):
# expand the latents if we are doing classifier free guidance
a__ : int = torch.cat([latents] * 2 ) if do_classifier_free_guidance else latents
a__ : Optional[int] = self.scheduler.scale_model_input(_snake_case , _snake_case )
# predict the noise residual
a__ : List[Any] = self.unet(_snake_case , _snake_case , encoder_hidden_states=_snake_case ).sample
# perform guidance
if do_classifier_free_guidance:
a__ , a__ : Any = noise_pred.chunk(2 )
a__ : Optional[int] = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
# compute the previous noisy sample x_t -> x_t-1
a__ : str = self.scheduler.step(_snake_case , _snake_case , _snake_case , **_snake_case ).prev_sample
# call the callback, if provided
if callback is not None and i % callback_steps == 0:
callback(_snake_case , _snake_case , _snake_case )
a__ : int = 1 / 0.1_8_2_1_5 * latents
a__ : Optional[int] = self.vae.decode(_snake_case ).sample
a__ : Optional[int] = (image / 2 + 0.5).clamp(0 , 1 )
# we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
a__ : Tuple = image.cpu().permute(0 , 2 , 3 , 1 ).float().numpy()
if output_type == "pil":
a__ : Optional[int] = self.numpy_to_pil(_snake_case )
if not return_dict:
return image
return StableDiffusionPipelineOutput(images=_snake_case , nsfw_content_detected=_snake_case )
| 170
|
import hashlib
import unittest
from transformers import MODEL_FOR_DEPTH_ESTIMATION_MAPPING, is_torch_available, is_vision_available
from transformers.pipelines import DepthEstimationPipeline, pipeline
from transformers.testing_utils import (
is_pipeline_test,
nested_simplify,
require_tf,
require_timm,
require_torch,
require_vision,
slow,
)
from .test_pipelines_common import ANY
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
else:
class __snake_case :
@staticmethod
def lowerCamelCase ( *_snake_case : Optional[int] , **_snake_case : int):
"""simple docstring"""
pass
def A (__A : Image ) -> str:
"""simple docstring"""
UpperCAmelCase_ = hashlib.mda(image.tobytes() )
return m.hexdigest()
@is_pipeline_test
@require_vision
@require_timm
@require_torch
class __snake_case ( unittest.TestCase ):
UpperCAmelCase__ : Tuple = MODEL_FOR_DEPTH_ESTIMATION_MAPPING
def lowerCamelCase ( self : Tuple , _snake_case : Optional[Any] , _snake_case : Optional[Any] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = DepthEstimationPipeline(model=_snake_case , image_processor=_snake_case)
return depth_estimator, [
"./tests/fixtures/tests_samples/COCO/000000039769.png",
"./tests/fixtures/tests_samples/COCO/000000039769.png",
]
def lowerCamelCase ( self : str , _snake_case : Optional[int] , _snake_case : List[str]):
"""simple docstring"""
UpperCAmelCase_ = depth_estimator('''./tests/fixtures/tests_samples/COCO/000000039769.png''')
self.assertEqual({'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)} , _snake_case)
import datasets
UpperCAmelCase_ = datasets.load_dataset('''hf-internal-testing/fixtures_image_utils''' , '''image''' , split='''test''')
UpperCAmelCase_ = depth_estimator(
[
Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png'''),
'''http://images.cocodataset.org/val2017/000000039769.jpg''',
# RGBA
dataset[0]['''file'''],
# LA
dataset[1]['''file'''],
# L
dataset[2]['''file'''],
])
self.assertEqual(
[
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
{'''predicted_depth''': ANY(torch.Tensor), '''depth''': ANY(Image.Image)},
] , _snake_case , )
@require_tf
@unittest.skip('''Depth estimation is not implemented in TF''')
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
pass
@slow
@require_torch
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = '''Intel/dpt-large'''
UpperCAmelCase_ = pipeline('''depth-estimation''' , model=_snake_case)
UpperCAmelCase_ = depth_estimator('''http://images.cocodataset.org/val2017/000000039769.jpg''')
UpperCAmelCase_ = hashimage(outputs['''depth'''])
# This seems flaky.
# self.assertEqual(outputs["depth"], "1a39394e282e9f3b0741a90b9f108977")
self.assertEqual(nested_simplify(outputs['''predicted_depth'''].max().item()) , 2_9.3_0_4)
self.assertEqual(nested_simplify(outputs['''predicted_depth'''].min().item()) , 2.6_6_2)
@require_torch
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
self.skipTest('''There is not hf-internal-testing tiny model for either GLPN nor DPT''')
| 51
| 0
|
def A_ ( snake_case : int ) -> None:
'''simple docstring'''
__UpperCamelCase = generate_pascal_triangle(__A )
for row_idx in range(__A ):
# Print left spaces
for _ in range(num_rows - row_idx - 1 ):
print(end=''' ''' )
# Print row values
for col_idx in range(row_idx + 1 ):
if col_idx != row_idx:
print(triangle[row_idx][col_idx] , end=''' ''' )
else:
print(triangle[row_idx][col_idx] , end='''''' )
print()
def A_ ( snake_case : int ) -> list[list[int]]:
'''simple docstring'''
if not isinstance(__A , __A ):
raise TypeError('''The input value of \'num_rows\' should be \'int\'''' )
if num_rows == 0:
return []
elif num_rows < 0:
raise ValueError(
'''The input value of \'num_rows\' should be greater than or equal to 0''' )
__UpperCamelCase = []
for current_row_idx in range(__A ):
__UpperCamelCase = populate_current_row(__A , __A )
triangle.append(__A )
return triangle
def A_ ( snake_case : list[list[int]] , snake_case : int ) -> list[int]:
'''simple docstring'''
__UpperCamelCase = [-1] * (current_row_idx + 1)
# first and last elements of current row are equal to 1
__UpperCamelCase , __UpperCamelCase = 1, 1
for current_col_idx in range(1 , __A ):
calculate_current_element(
__A , __A , __A , __A )
return current_row
def A_ ( snake_case : list[list[int]] , snake_case : list[int] , snake_case : int , snake_case : int , ) -> None:
'''simple docstring'''
__UpperCamelCase = triangle[current_row_idx - 1][current_col_idx - 1]
__UpperCamelCase = triangle[current_row_idx - 1][current_col_idx]
__UpperCamelCase = above_to_left_elt + above_to_right_elt
def A_ ( snake_case : int ) -> list[list[int]]:
'''simple docstring'''
if not isinstance(__A , __A ):
raise TypeError('''The input value of \'num_rows\' should be \'int\'''' )
if num_rows == 0:
return []
elif num_rows < 0:
raise ValueError(
'''The input value of \'num_rows\' should be greater than or equal to 0''' )
__UpperCamelCase = [[1]]
for row_index in range(1 , __A ):
__UpperCamelCase = [0] + result[-1] + [0]
__UpperCamelCase = row_index + 1
# Calculate the number of distinct elements in a row
__UpperCamelCase = sum(divmod(__A , 2 ) )
__UpperCamelCase = [
temp_row[i - 1] + temp_row[i] for i in range(1 , distinct_elements + 1 )
]
__UpperCamelCase = row_first_half[: (row_index + 1) // 2]
row_second_half.reverse()
__UpperCamelCase = row_first_half + row_second_half
result.append(__A )
return result
def A_ ( ) -> None:
'''simple docstring'''
from collections.abc import Callable
from timeit import timeit
def benchmark_a_function(snake_case : Callable , snake_case : int ) -> None:
__UpperCamelCase = f"{func.__name__}({value})"
__UpperCamelCase = timeit(f"__main__.{call}" , setup='''import __main__''' )
# print(f"{call:38} = {func(value)} -- {timing:.4f} seconds")
print(f"{call:38} -- {timing:.4f} seconds" )
for value in range(15 ): # (1, 7, 14):
for func in (generate_pascal_triangle, generate_pascal_triangle_optimized):
benchmark_a_function(__A , __A )
print()
if __name__ == "__main__":
import doctest
doctest.testmod()
benchmark()
| 328
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : int = {
"configuration_deberta": ["DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP", "DebertaConfig", "DebertaOnnxConfig"],
"tokenization_deberta": ["DebertaTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : int = ["DebertaTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"DebertaForMaskedLM",
"DebertaForQuestionAnswering",
"DebertaForSequenceClassification",
"DebertaForTokenClassification",
"DebertaModel",
"DebertaPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFDebertaForMaskedLM",
"TFDebertaForQuestionAnswering",
"TFDebertaForSequenceClassification",
"TFDebertaForTokenClassification",
"TFDebertaModel",
"TFDebertaPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_deberta import DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP, DebertaConfig, DebertaOnnxConfig
from .tokenization_deberta import DebertaTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_deberta_fast import DebertaTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_deberta import (
DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
DebertaForMaskedLM,
DebertaForQuestionAnswering,
DebertaForSequenceClassification,
DebertaForTokenClassification,
DebertaModel,
DebertaPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_deberta import (
TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
TFDebertaForMaskedLM,
TFDebertaForQuestionAnswering,
TFDebertaForSequenceClassification,
TFDebertaForTokenClassification,
TFDebertaModel,
TFDebertaPreTrainedModel,
)
else:
import sys
snake_case_ : Dict = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
'''simple docstring'''
def UpperCAmelCase__ ( UpperCAmelCase__ ) -> list[int]:
if length <= 0 or not isinstance(__A, __A ):
raise ValueError("""Length must be a positive integer.""" )
return [n * (2 * n - 1) for n in range(__A )]
if __name__ == "__main__":
print(hexagonal_numbers(length=5))
print(hexagonal_numbers(length=10))
| 162
|
from typing import TYPE_CHECKING
from ...file_utils import _LazyModule, is_tokenizers_available, is_torch_available
from ...utils import OptionalDependencyNotAvailable
snake_case_ : Union[str, Any] = {"configuration_gpt_neox": ["GPT_NEOX_PRETRAINED_CONFIG_ARCHIVE_MAP", "GPTNeoXConfig"]}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Dict = ["GPTNeoXTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = [
"GPT_NEOX_PRETRAINED_MODEL_ARCHIVE_LIST",
"GPTNeoXForCausalLM",
"GPTNeoXForQuestionAnswering",
"GPTNeoXForSequenceClassification",
"GPTNeoXForTokenClassification",
"GPTNeoXLayer",
"GPTNeoXModel",
"GPTNeoXPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_gpt_neox import GPT_NEOX_PRETRAINED_CONFIG_ARCHIVE_MAP, GPTNeoXConfig
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_gpt_neox_fast import GPTNeoXTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_gpt_neox import (
GPT_NEOX_PRETRAINED_MODEL_ARCHIVE_LIST,
GPTNeoXForCausalLM,
GPTNeoXForQuestionAnswering,
GPTNeoXForSequenceClassification,
GPTNeoXForTokenClassification,
GPTNeoXLayer,
GPTNeoXModel,
GPTNeoXPreTrainedModel,
)
else:
import sys
snake_case_ : Optional[int] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
'''simple docstring'''
import argparse
import os
import re
__lowercase : Union[str, Any] = "src/transformers/models/auto"
# re pattern that matches mapping introductions:
# SUPER_MODEL_MAPPING_NAMES = OrderedDict or SUPER_MODEL_MAPPING = OrderedDict
__lowercase : Any = re.compile(r'''[A-Z_]+_MAPPING(\s+|_[A-Z_]+\s+)=\s+OrderedDict''')
# re pattern that matches identifiers in mappings
__lowercase : str = re.compile(r'''\s*\(\s*\"(\S[^\"]+)\"''')
def lowercase_ ( _lowercase , _lowercase = False ) -> Dict:
'''simple docstring'''
with open(__A , '''r''' , encoding='''utf-8''' ) as f:
lowerCamelCase_ : Dict = f.read()
lowerCamelCase_ : Dict = content.split('''\n''' )
lowerCamelCase_ : Dict = []
lowerCamelCase_ : Dict = 0
while line_idx < len(__A ):
if _re_intro_mapping.search(lines[line_idx] ) is not None:
lowerCamelCase_ : Optional[int] = len(re.search(R'''^(\s*)\S''' , lines[line_idx] ).groups()[0] ) + 8
# Start of a new mapping!
while not lines[line_idx].startswith(''' ''' * indent + '''(''' ):
new_lines.append(lines[line_idx] )
line_idx += 1
lowerCamelCase_ : List[str] = []
while lines[line_idx].strip() != "]":
# Blocks either fit in one line or not
if lines[line_idx].strip() == "(":
lowerCamelCase_ : Union[str, Any] = line_idx
while not lines[line_idx].startswith(''' ''' * indent + ''')''' ):
line_idx += 1
blocks.append('''\n'''.join(lines[start_idx : line_idx + 1] ) )
else:
blocks.append(lines[line_idx] )
line_idx += 1
# Sort blocks by their identifiers
lowerCamelCase_ : Optional[Any] = sorted(__A , key=lambda _lowercase : _re_identifier.search(__A ).groups()[0] )
new_lines += blocks
else:
new_lines.append(lines[line_idx] )
line_idx += 1
if overwrite:
with open(__A , '''w''' , encoding='''utf-8''' ) as f:
f.write('''\n'''.join(__A ) )
elif "\n".join(__A ) != content:
return True
def lowercase_ ( _lowercase = False ) -> List[Any]:
'''simple docstring'''
lowerCamelCase_ : List[str] = [os.path.join(__A , __A ) for f in os.listdir(__A ) if f.endswith('''.py''' )]
lowerCamelCase_ : List[str] = [sort_auto_mapping(__A , overwrite=__A ) for fname in fnames]
if not overwrite and any(__A ):
lowerCamelCase_ : Dict = [f for f, d in zip(__A , __A ) if d]
raise ValueError(
F"""The following files have auto mappings that need sorting: {", ".join(__A )}. Run `make style` to fix"""
''' this.''' )
if __name__ == "__main__":
__lowercase : Optional[Any] = argparse.ArgumentParser()
parser.add_argument('''--check_only''', action='''store_true''', help='''Whether to only check or fix style.''')
__lowercase : Union[str, Any] = parser.parse_args()
sort_all_auto_mappings(not args.check_only)
| 318
|
def A (__A : list , __A : int , __A : int = 0 , __A : int = 0 ) -> int:
"""simple docstring"""
UpperCAmelCase_ = right or len(__A ) - 1
if left > right:
return -1
elif list_data[left] == key:
return left
elif list_data[right] == key:
return right
else:
return search(__A , __A , left + 1 , right - 1 )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 51
| 0
|
from __future__ import absolute_import, division, print_function, unicode_literals
from torch import nn
from torch.nn import CrossEntropyLoss, MSELoss
from transformers import RobertaConfig
from transformers.file_utils import add_start_docstrings, add_start_docstrings_to_model_forward
from transformers.models.roberta.modeling_roberta import (
ROBERTA_INPUTS_DOCSTRING,
ROBERTA_START_DOCSTRING,
RobertaEmbeddings,
)
from .modeling_highway_bert import BertPreTrainedModel, DeeBertModel, HighwayException, entropy
@add_start_docstrings(
"""The RoBERTa Model transformer with early exiting (DeeRoBERTa). """ , _a , )
class __lowerCamelCase (_a ):
_lowercase = RobertaConfig
_lowercase = '''roberta'''
def __init__( self: Dict,A_: Union[str, Any] ):
'''simple docstring'''
super().__init__(_snake_case )
__UpperCamelCase = RobertaEmbeddings(_snake_case )
self.init_weights()
@add_start_docstrings(
"""RoBERTa Model (with early exiting - DeeRoBERTa) with a classifier on top,
also takes care of multi-layer training. """ , _a , )
class __lowerCamelCase (_a ):
_lowercase = RobertaConfig
_lowercase = '''roberta'''
def __init__( self: int,A_: Tuple ):
'''simple docstring'''
super().__init__(_snake_case )
__UpperCamelCase = config.num_labels
__UpperCamelCase = config.num_hidden_layers
__UpperCamelCase = DeeRobertaModel(_snake_case )
__UpperCamelCase = nn.Dropout(config.hidden_dropout_prob )
__UpperCamelCase = nn.Linear(config.hidden_size,self.config.num_labels )
@add_start_docstrings_to_model_forward(_snake_case )
def snake_case_ ( self: Union[str, Any],A_: Union[str, Any]=None,A_: Optional[int]=None,A_: Tuple=None,A_: Any=None,A_: List[str]=None,A_: int=None,A_: int=None,A_: Union[str, Any]=-1,A_: List[str]=False,):
'''simple docstring'''
__UpperCamelCase = self.num_layers
try:
__UpperCamelCase = self.roberta(
_snake_case,attention_mask=_snake_case,token_type_ids=_snake_case,position_ids=_snake_case,head_mask=_snake_case,inputs_embeds=_snake_case,)
__UpperCamelCase = outputs[1]
__UpperCamelCase = self.dropout(_snake_case )
__UpperCamelCase = self.classifier(_snake_case )
__UpperCamelCase = (logits,) + outputs[2:] # add hidden states and attention if they are here
except HighwayException as e:
__UpperCamelCase = e.message
__UpperCamelCase = e.exit_layer
__UpperCamelCase = outputs[0]
if not self.training:
__UpperCamelCase = entropy(_snake_case )
__UpperCamelCase = []
__UpperCamelCase = []
if labels is not None:
if self.num_labels == 1:
# We are doing regression
__UpperCamelCase = MSELoss()
__UpperCamelCase = loss_fct(logits.view(-1 ),labels.view(-1 ) )
else:
__UpperCamelCase = CrossEntropyLoss()
__UpperCamelCase = loss_fct(logits.view(-1,self.num_labels ),labels.view(-1 ) )
# work with highway exits
__UpperCamelCase = []
for highway_exit in outputs[-1]:
__UpperCamelCase = highway_exit[0]
if not self.training:
highway_logits_all.append(_snake_case )
highway_entropy.append(highway_exit[2] )
if self.num_labels == 1:
# We are doing regression
__UpperCamelCase = MSELoss()
__UpperCamelCase = loss_fct(highway_logits.view(-1 ),labels.view(-1 ) )
else:
__UpperCamelCase = CrossEntropyLoss()
__UpperCamelCase = loss_fct(highway_logits.view(-1,self.num_labels ),labels.view(-1 ) )
highway_losses.append(_snake_case )
if train_highway:
__UpperCamelCase = (sum(highway_losses[:-1] ),) + outputs
# exclude the final highway, of course
else:
__UpperCamelCase = (loss,) + outputs
if not self.training:
__UpperCamelCase = outputs + ((original_entropy, highway_entropy), exit_layer)
if output_layer >= 0:
__UpperCamelCase = (
(outputs[0],) + (highway_logits_all[output_layer],) + outputs[2:]
) # use the highway of the last layer
return outputs # (loss), logits, (hidden_states), (attentions), entropy
| 310
|
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case_ : int = logging.get_logger(__name__)
snake_case_ : str = {}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''llama'''
UpperCAmelCase__ : Dict = ['''past_key_values''']
def __init__( self : str , _snake_case : List[str]=32000 , _snake_case : int=4096 , _snake_case : List[str]=11008 , _snake_case : Optional[int]=32 , _snake_case : List[Any]=32 , _snake_case : Tuple=None , _snake_case : int="silu" , _snake_case : List[Any]=2048 , _snake_case : List[str]=0.0_2 , _snake_case : Any=1e-6 , _snake_case : List[str]=True , _snake_case : Optional[Any]=0 , _snake_case : Dict=1 , _snake_case : List[Any]=2 , _snake_case : str=1 , _snake_case : Union[str, Any]=False , _snake_case : str=None , **_snake_case : List[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = max_position_embeddings
UpperCAmelCase_ = hidden_size
UpperCAmelCase_ = intermediate_size
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
# for backward compatibility
if num_key_value_heads is None:
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = num_key_value_heads
UpperCAmelCase_ = hidden_act
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = rms_norm_eps
UpperCAmelCase_ = pretraining_tp
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = rope_scaling
self._rope_scaling_validation()
super().__init__(
pad_token_id=_snake_case , bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case , )
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
if self.rope_scaling is None:
return
if not isinstance(self.rope_scaling , _snake_case) or len(self.rope_scaling) != 2:
raise ValueError(
'''`rope_scaling` must be a dictionary with with two fields, `name` and `factor`, '''
F"""got {self.rope_scaling}""")
UpperCAmelCase_ = self.rope_scaling.get('''type''' , _snake_case)
UpperCAmelCase_ = self.rope_scaling.get('''factor''' , _snake_case)
if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic"]:
raise ValueError(
F"""`rope_scaling`'s name field must be one of ['linear', 'dynamic'], got {rope_scaling_type}""")
if rope_scaling_factor is None or not isinstance(_snake_case , _snake_case) or rope_scaling_factor <= 1.0:
raise ValueError(F"""`rope_scaling`'s factor field must be an float > 1, got {rope_scaling_factor}""")
| 51
| 0
|
"""simple docstring"""
import doctest
from collections import deque
import numpy as np
class __lowerCamelCase :
'''simple docstring'''
def __init__( self : int ):
lowerCAmelCase_ : Optional[Any] = [2, 1, 2, -1]
lowerCAmelCase_ : int = [1, 2, 3, 4]
def lowerCamelCase ( self : Union[str, Any] ):
lowerCAmelCase_ : int = len(self.first_signal )
lowerCAmelCase_ : Any = len(self.second_signal )
lowerCAmelCase_ : Dict = max(_snake_case , _snake_case )
# create a zero matrix of max_length x max_length
lowerCAmelCase_ : List[Any] = [[0] * max_length for i in range(_snake_case )]
# fills the smaller signal with zeros to make both signals of same length
if length_first_signal < length_second_signal:
self.first_signal += [0] * (max_length - length_first_signal)
elif length_first_signal > length_second_signal:
self.second_signal += [0] * (max_length - length_second_signal)
for i in range(_snake_case ):
lowerCAmelCase_ : Dict = deque(self.second_signal )
rotated_signal.rotate(_snake_case )
for j, item in enumerate(_snake_case ):
matrix[i][j] += item
# multiply the matrix with the first signal
lowerCAmelCase_ : Optional[int] = np.matmul(np.transpose(_snake_case ) , np.transpose(self.first_signal ) )
# rounding-off to two decimal places
return [round(_snake_case , 2 ) for i in final_signal]
if __name__ == "__main__":
doctest.testmod()
| 241
|
from collections import OrderedDict
from typing import Any, List, Mapping, Optional
from ... import PreTrainedTokenizer, TensorType, is_torch_available
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfigWithPast, PatchingSpec
from ...utils import logging
snake_case_ : List[str] = logging.get_logger(__name__)
snake_case_ : Tuple = {
"Salesforce/codegen-350M-nl": "https://huggingface.co/Salesforce/codegen-350M-nl/resolve/main/config.json",
"Salesforce/codegen-350M-multi": "https://huggingface.co/Salesforce/codegen-350M-multi/resolve/main/config.json",
"Salesforce/codegen-350M-mono": "https://huggingface.co/Salesforce/codegen-350M-mono/resolve/main/config.json",
"Salesforce/codegen-2B-nl": "https://huggingface.co/Salesforce/codegen-2B-nl/resolve/main/config.json",
"Salesforce/codegen-2B-multi": "https://huggingface.co/Salesforce/codegen-2B-multi/resolve/main/config.json",
"Salesforce/codegen-2B-mono": "https://huggingface.co/Salesforce/codegen-2B-mono/resolve/main/config.json",
"Salesforce/codegen-6B-nl": "https://huggingface.co/Salesforce/codegen-6B-nl/resolve/main/config.json",
"Salesforce/codegen-6B-multi": "https://huggingface.co/Salesforce/codegen-6B-multi/resolve/main/config.json",
"Salesforce/codegen-6B-mono": "https://huggingface.co/Salesforce/codegen-6B-mono/resolve/main/config.json",
"Salesforce/codegen-16B-nl": "https://huggingface.co/Salesforce/codegen-16B-nl/resolve/main/config.json",
"Salesforce/codegen-16B-multi": "https://huggingface.co/Salesforce/codegen-16B-multi/resolve/main/config.json",
"Salesforce/codegen-16B-mono": "https://huggingface.co/Salesforce/codegen-16B-mono/resolve/main/config.json",
}
class __snake_case ( a ):
UpperCAmelCase__ : str = '''codegen'''
UpperCAmelCase__ : int = {
'''max_position_embeddings''': '''n_positions''',
'''hidden_size''': '''n_embd''',
'''num_attention_heads''': '''n_head''',
'''num_hidden_layers''': '''n_layer''',
}
def __init__( self : Union[str, Any] , _snake_case : Union[str, Any]=50400 , _snake_case : Optional[int]=2048 , _snake_case : Union[str, Any]=2048 , _snake_case : List[str]=4096 , _snake_case : Any=28 , _snake_case : List[str]=16 , _snake_case : int=64 , _snake_case : Tuple=None , _snake_case : Dict="gelu_new" , _snake_case : Union[str, Any]=0.0 , _snake_case : Optional[Any]=0.0 , _snake_case : List[Any]=0.0 , _snake_case : List[Any]=1e-5 , _snake_case : List[str]=0.0_2 , _snake_case : Optional[Any]=True , _snake_case : int=50256 , _snake_case : Tuple=50256 , _snake_case : int=False , **_snake_case : Any , ):
"""simple docstring"""
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = n_ctx
UpperCAmelCase_ = n_positions
UpperCAmelCase_ = n_embd
UpperCAmelCase_ = n_layer
UpperCAmelCase_ = n_head
UpperCAmelCase_ = n_inner
UpperCAmelCase_ = rotary_dim
UpperCAmelCase_ = activation_function
UpperCAmelCase_ = resid_pdrop
UpperCAmelCase_ = embd_pdrop
UpperCAmelCase_ = attn_pdrop
UpperCAmelCase_ = layer_norm_epsilon
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = use_cache
UpperCAmelCase_ = bos_token_id
UpperCAmelCase_ = eos_token_id
super().__init__(
bos_token_id=_snake_case , eos_token_id=_snake_case , tie_word_embeddings=_snake_case , **_snake_case)
class __snake_case ( a ):
def __init__( self : Tuple , _snake_case : PretrainedConfig , _snake_case : str = "default" , _snake_case : List[PatchingSpec] = None , _snake_case : bool = False , ):
"""simple docstring"""
super().__init__(_snake_case , task=_snake_case , patching_specs=_snake_case , use_past=_snake_case)
if not getattr(self._config , '''pad_token_id''' , _snake_case):
# TODO: how to do that better?
UpperCAmelCase_ = 0
@property
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = OrderedDict({'''input_ids''': {0: '''batch''', 1: '''sequence'''}})
if self.use_past:
self.fill_with_past_key_values_(_snake_case , direction='''inputs''')
UpperCAmelCase_ = {0: '''batch''', 1: '''past_sequence + sequence'''}
else:
UpperCAmelCase_ = {0: '''batch''', 1: '''sequence'''}
return common_inputs
@property
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
return self._config.n_layer
@property
def lowerCamelCase ( self : int):
"""simple docstring"""
return self._config.n_head
def lowerCamelCase ( self : Optional[int] , _snake_case : PreTrainedTokenizer , _snake_case : int = -1 , _snake_case : int = -1 , _snake_case : bool = False , _snake_case : Optional[TensorType] = None , ):
"""simple docstring"""
UpperCAmelCase_ = super(_snake_case , self).generate_dummy_inputs(
_snake_case , batch_size=_snake_case , seq_length=_snake_case , is_pair=_snake_case , framework=_snake_case)
# We need to order the input in the way they appears in the forward()
UpperCAmelCase_ = OrderedDict({'''input_ids''': common_inputs['''input_ids''']})
# Need to add the past_keys
if self.use_past:
if not is_torch_available():
raise ValueError('''Cannot generate dummy past_keys inputs without PyTorch installed.''')
else:
import torch
UpperCAmelCase_ , UpperCAmelCase_ = common_inputs['''input_ids'''].shape
# Not using the same length for past_key_values
UpperCAmelCase_ = seqlen + 2
UpperCAmelCase_ = (
batch,
self.num_attention_heads,
past_key_values_length,
self._config.hidden_size // self.num_attention_heads,
)
UpperCAmelCase_ = [
(torch.zeros(_snake_case), torch.zeros(_snake_case)) for _ in range(self.num_layers)
]
UpperCAmelCase_ = common_inputs['''attention_mask''']
if self.use_past:
UpperCAmelCase_ = ordered_inputs['''attention_mask'''].dtype
UpperCAmelCase_ = torch.cat(
[ordered_inputs['''attention_mask'''], torch.ones(_snake_case , _snake_case , dtype=_snake_case)] , dim=1)
return ordered_inputs
@property
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
return 13
| 51
| 0
|
'''simple docstring'''
import unittest
import numpy as np
from transformers.testing_utils import require_torch, require_vision
from transformers.utils import is_torch_available, is_vision_available
from ...test_image_processing_common import ImageProcessingSavingTestMixin, prepare_image_inputs
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
from transformers import MobileViTImageProcessor
class __A ( unittest.TestCase ):
'''simple docstring'''
def __init__(self , A , A=7 , A=3 , A=18 , A=30 , A=400 , A=True , A=None , A=True , A=None , A=True , ) -> Optional[int]:
"""simple docstring"""
_a = size if size is not None else {'''shortest_edge''': 20}
_a = crop_size if crop_size is not None else {'''height''': 18, '''width''': 18}
_a = parent
_a = batch_size
_a = num_channels
_a = image_size
_a = min_resolution
_a = max_resolution
_a = do_resize
_a = size
_a = do_center_crop
_a = crop_size
_a = do_flip_channel_order
def a__ (self ) -> Optional[Any]:
"""simple docstring"""
return {
"do_resize": self.do_resize,
"size": self.size,
"do_center_crop": self.do_center_crop,
"crop_size": self.crop_size,
"do_flip_channel_order": self.do_flip_channel_order,
}
@require_torch
@require_vision
class __A ( A , unittest.TestCase ):
'''simple docstring'''
__lowerCamelCase : Tuple = MobileViTImageProcessor if is_vision_available() else None
def a__ (self ) -> List[Any]:
"""simple docstring"""
_a = MobileViTImageProcessingTester(self )
@property
def a__ (self ) -> str:
"""simple docstring"""
return self.image_processor_tester.prepare_image_processor_dict()
def a__ (self ) -> Tuple:
"""simple docstring"""
_a = self.image_processing_class(**self.image_processor_dict )
self.assertTrue(hasattr(_snake_case , '''do_resize''' ) )
self.assertTrue(hasattr(_snake_case , '''size''' ) )
self.assertTrue(hasattr(_snake_case , '''do_center_crop''' ) )
self.assertTrue(hasattr(_snake_case , '''center_crop''' ) )
self.assertTrue(hasattr(_snake_case , '''do_flip_channel_order''' ) )
def a__ (self ) -> List[str]:
"""simple docstring"""
_a = self.image_processing_class.from_dict(self.image_processor_dict )
self.assertEqual(image_processor.size , {'''shortest_edge''': 20} )
self.assertEqual(image_processor.crop_size , {'''height''': 18, '''width''': 18} )
_a = self.image_processing_class.from_dict(self.image_processor_dict , size=42 , crop_size=84 )
self.assertEqual(image_processor.size , {'''shortest_edge''': 42} )
self.assertEqual(image_processor.crop_size , {'''height''': 84, '''width''': 84} )
def a__ (self ) -> int:
"""simple docstring"""
pass
def a__ (self ) -> Union[str, Any]:
"""simple docstring"""
_a = self.image_processing_class(**self.image_processor_dict )
# create random PIL images
_a = prepare_image_inputs(self.image_processor_tester , equal_resolution=_snake_case )
for image in image_inputs:
self.assertIsInstance(_snake_case , Image.Image )
# Test not batched input
_a = image_processing(image_inputs[0] , return_tensors='''pt''' ).pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
# Test batched
_a = image_processing(_snake_case , return_tensors='''pt''' ).pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
def a__ (self ) -> Union[str, Any]:
"""simple docstring"""
_a = self.image_processing_class(**self.image_processor_dict )
# create random numpy tensors
_a = prepare_image_inputs(self.image_processor_tester , equal_resolution=_snake_case , numpify=_snake_case )
for image in image_inputs:
self.assertIsInstance(_snake_case , np.ndarray )
# Test not batched input
_a = image_processing(image_inputs[0] , return_tensors='''pt''' ).pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
# Test batched
_a = image_processing(_snake_case , return_tensors='''pt''' ).pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
def a__ (self ) -> Optional[Any]:
"""simple docstring"""
_a = self.image_processing_class(**self.image_processor_dict )
# create random PyTorch tensors
_a = prepare_image_inputs(self.image_processor_tester , equal_resolution=_snake_case , torchify=_snake_case )
for image in image_inputs:
self.assertIsInstance(_snake_case , torch.Tensor )
# Test not batched input
_a = image_processing(image_inputs[0] , return_tensors='''pt''' ).pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
# Test batched
_a = image_processing(_snake_case , return_tensors='''pt''' ).pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
| 211
|
import os
import unittest
from transformers.models.phobert.tokenization_phobert import VOCAB_FILES_NAMES, PhobertTokenizer
from ...test_tokenization_common import TokenizerTesterMixin
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : Any = PhobertTokenizer
UpperCAmelCase__ : List[str] = False
def lowerCamelCase ( self : str):
"""simple docstring"""
super().setUp()
# Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt
UpperCAmelCase_ = ['''T@@''', '''i''', '''I''', '''R@@''', '''r''', '''e@@''']
UpperCAmelCase_ = dict(zip(_snake_case , range(len(_snake_case))))
UpperCAmelCase_ = ['''#version: 0.2''', '''l à</w>''']
UpperCAmelCase_ = {'''unk_token''': '''<unk>'''}
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''vocab_file'''])
UpperCAmelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''merges_file'''])
with open(self.vocab_file , '''w''' , encoding='''utf-8''') as fp:
for token in vocab_tokens:
fp.write(F"""{token} {vocab_tokens[token]}\n""")
with open(self.merges_file , '''w''' , encoding='''utf-8''') as fp:
fp.write('''\n'''.join(_snake_case))
def lowerCamelCase ( self : int , **_snake_case : Any):
"""simple docstring"""
kwargs.update(self.special_tokens_map)
return PhobertTokenizer.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Optional[Any] , _snake_case : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T<unk> i <unk> <unk> <unk> <unk> <unk> <unk> I Re<unk> e<unk> <unk> <unk> <unk>'''
return input_text, output_text
def lowerCamelCase ( self : List[str]):
"""simple docstring"""
UpperCAmelCase_ = PhobertTokenizer(self.vocab_file , self.merges_file , **self.special_tokens_map)
UpperCAmelCase_ = '''Tôi là VinAI Research'''
UpperCAmelCase_ = '''T@@ ô@@ i l@@ à V@@ i@@ n@@ A@@ I R@@ e@@ s@@ e@@ a@@ r@@ c@@ h'''.split()
UpperCAmelCase_ = tokenizer.tokenize(_snake_case)
print(_snake_case)
self.assertListEqual(_snake_case , _snake_case)
UpperCAmelCase_ = tokens + [tokenizer.unk_token]
UpperCAmelCase_ = [4, 3, 5, 3, 3, 3, 3, 3, 3, 6, 7, 9, 3, 9, 3, 3, 3, 3, 3]
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case) , _snake_case)
| 51
| 0
|
from __future__ import annotations
import sys
from collections import deque
from typing import Generic, TypeVar
__lowerCamelCase : Union[str, Any] = TypeVar('''T''')
class a__ ( Generic[T] ):
A = 42 # Cache store of keys
A = 42 # References of the keys in cache
A = 10 # Maximum capacity of cache
def __init__( self : Optional[int],_A : int ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : Any = deque()
SCREAMING_SNAKE_CASE_ : Dict = set()
if not n:
SCREAMING_SNAKE_CASE_ : List[Any] = sys.maxsize
elif n < 0:
raise ValueError("n should be an integer greater than 0." )
else:
SCREAMING_SNAKE_CASE_ : Optional[Any] = n
def __UpperCamelCase ( self : int,_A : T ):
"""simple docstring"""
if x not in self.key_reference:
if len(self.dq_store ) == LRUCache._MAX_CAPACITY:
SCREAMING_SNAKE_CASE_ : List[Any] = self.dq_store.pop()
self.key_reference.remove(_snake_case )
else:
self.dq_store.remove(_snake_case )
self.dq_store.appendleft(_snake_case )
self.key_reference.add(_snake_case )
def __UpperCamelCase ( self : Any ):
"""simple docstring"""
for k in self.dq_store:
print(_snake_case )
def __repr__( self : Optional[Any] ):
"""simple docstring"""
return F'LRUCache({self._MAX_CAPACITY}) => {list(self.dq_store )}'
if __name__ == "__main__":
import doctest
doctest.testmod()
__lowerCamelCase : LRUCache[str | int] = LRUCache(4)
lru_cache.refer('''A''')
lru_cache.refer(2)
lru_cache.refer(3)
lru_cache.refer('''A''')
lru_cache.refer(4)
lru_cache.refer(5)
lru_cache.display()
print(lru_cache)
assert str(lru_cache) == "LRUCache(4) => [5, 4, 'A', 3]"
| 18
|
from typing import List, Optional, TypeVar
from .arrow_dataset import Dataset, _concatenate_map_style_datasets, _interleave_map_style_datasets
from .dataset_dict import DatasetDict, IterableDatasetDict
from .info import DatasetInfo
from .iterable_dataset import IterableDataset, _concatenate_iterable_datasets, _interleave_iterable_datasets
from .splits import NamedSplit
from .utils import logging
from .utils.py_utils import Literal
snake_case_ : Any = logging.get_logger(__name__)
snake_case_ : Optional[int] = TypeVar("DatasetType", Dataset, IterableDataset)
def A (__A : List[DatasetType] , __A : Optional[List[float]] = None , __A : Optional[int] = None , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : Literal["first_exhausted", "all_exhausted"] = "first_exhausted" , ) -> DatasetType:
"""simple docstring"""
from .arrow_dataset import Dataset
from .iterable_dataset import IterableDataset
if not datasets:
raise ValueError('''Unable to interleave an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if stopping_strategy not in ["first_exhausted", "all_exhausted"]:
raise ValueError(F"""{stopping_strategy} is not supported. Please enter a valid stopping_strategy.""" )
if dataset_type is Dataset:
return _interleave_map_style_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
else:
return _interleave_iterable_datasets(
__A , __A , __A , info=__A , split=__A , stopping_strategy=__A )
def A (__A : List[DatasetType] , __A : Optional[DatasetInfo] = None , __A : Optional[NamedSplit] = None , __A : int = 0 , ) -> DatasetType:
"""simple docstring"""
if not dsets:
raise ValueError('''Unable to concatenate an empty list of datasets.''' )
for i, dataset in enumerate(__A ):
if not isinstance(__A , (Dataset, IterableDataset) ):
if isinstance(__A , (DatasetDict, IterableDatasetDict) ):
if not dataset:
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} """
'''is an empty dataset dictionary.''' )
raise ValueError(
F"""Dataset at position {i} has at least one split: {list(__A )}\n"""
F"""Please pick one to interleave with the other datasets, for example: dataset['{next(iter(__A ) )}']""" )
raise ValueError(
F"""Expected a list of Dataset objects or a list of IterableDataset objects, but element at position {i} is a {type(__A ).__name__}.""" )
if i == 0:
UpperCAmelCase_ , UpperCAmelCase_ = (
(Dataset, IterableDataset) if isinstance(__A , __A ) else (IterableDataset, Dataset)
)
elif not isinstance(__A , __A ):
raise ValueError(
F"""Unable to interleave a {dataset_type.__name__} (at position 0) with a {other_type.__name__} (at position {i}). Expected a list of Dataset objects or a list of IterableDataset objects.""" )
if dataset_type is Dataset:
return _concatenate_map_style_datasets(__A , info=__A , split=__A , axis=__A )
else:
return _concatenate_iterable_datasets(__A , info=__A , split=__A , axis=__A )
| 51
| 0
|
"""simple docstring"""
import json
import os
import unittest
from transformers.models.gptsan_japanese.tokenization_gptsan_japanese import (
VOCAB_FILES_NAMES,
GPTSanJapaneseTokenizer,
)
from transformers.testing_utils import require_tokenizers, slow
from ...test_tokenization_common import TokenizerTesterMixin
@require_tokenizers
class _SCREAMING_SNAKE_CASE ( A__ , unittest.TestCase ):
UpperCAmelCase_ :int = GPTSanJapaneseTokenizer
UpperCAmelCase_ :Optional[int] = False
UpperCAmelCase_ :List[Any] = {'''do_clean_text''': False, '''add_prefix_space''': False}
def __lowerCAmelCase ( self ) -> int:
super().setUp()
# fmt: off
lowerCAmelCase_ :Tuple = ["""こん""", """こんに""", """にちは""", """ばんは""", """世界,㔺界""", """、""", """。""", """<BR>""", """<SP>""", """<TAB>""", """<URL>""", """<EMAIL>""", """<TEL>""", """<DATE>""", """<PRICE>""", """<BLOCK>""", """<KIGOU>""", """<U2000U2BFF>""", """<|emoji1|>""", """<unk>""", """<|bagoftoken|>""", """<|endoftext|>"""]
# fmt: on
lowerCAmelCase_ :Dict = {"""emoji""": {"""\ud83d\ude00""": """<|emoji1|>"""}, """emoji_inv""": {"""<|emoji1|>""": """\ud83d\ude00"""}} # 😀
lowerCAmelCase_ :Union[str, Any] = {"""unk_token""": """<unk>"""}
lowerCAmelCase_ :Any = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""vocab_file"""] )
lowerCAmelCase_ :Optional[Any] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""emoji_file"""] )
with open(self.vocab_file , """w""" , encoding="""utf-8""" ) as vocab_writer:
vocab_writer.write("""""".join([x + """\n""" for x in vocab_tokens] ) )
with open(self.emoji_file , """w""" ) as emoji_writer:
emoji_writer.write(json.dumps(_snake_case ) )
def __lowerCAmelCase ( self , **__A ) -> Dict:
kwargs.update(self.special_tokens_map )
return GPTSanJapaneseTokenizer.from_pretrained(self.tmpdirname , **_snake_case )
def __lowerCAmelCase ( self , __A ) -> int:
lowerCAmelCase_ :str = """こんにちは、世界。 \nこんばんは、㔺界。😀"""
lowerCAmelCase_ :Union[str, Any] = """こんにちは、世界。 \nこんばんは、世界。😀"""
return input_text, output_text
def __lowerCAmelCase ( self , __A ) -> Tuple:
lowerCAmelCase_ , lowerCAmelCase_ :Optional[int] = self.get_input_output_texts(_snake_case )
lowerCAmelCase_ :List[str] = tokenizer.encode(_snake_case , add_special_tokens=_snake_case )
lowerCAmelCase_ :List[str] = tokenizer.decode(_snake_case , clean_up_tokenization_spaces=_snake_case )
return text, ids
def __lowerCAmelCase ( self ) -> int:
pass # TODO add if relevant
def __lowerCAmelCase ( self ) -> List[str]:
pass # TODO add if relevant
def __lowerCAmelCase ( self ) -> List[str]:
pass # TODO add if relevant
def __lowerCAmelCase ( self ) -> Optional[Any]:
lowerCAmelCase_ :Any = self.get_tokenizer()
# Testing tokenization
lowerCAmelCase_ :str = """こんにちは、世界。 こんばんは、㔺界。"""
lowerCAmelCase_ :Union[str, Any] = ["""こん""", """にちは""", """、""", """世界""", """。""", """<SP>""", """こん""", """ばんは""", """、""", """㔺界""", """。"""]
lowerCAmelCase_ :Tuple = tokenizer.tokenize(_snake_case )
self.assertListEqual(_snake_case , _snake_case )
# Testing conversion to ids without special tokens
lowerCAmelCase_ :str = [0, 2, 5, 4, 6, 8, 0, 3, 5, 4, 6]
lowerCAmelCase_ :Any = tokenizer.convert_tokens_to_ids(_snake_case )
self.assertListEqual(_snake_case , _snake_case )
# Testing conversion to ids with special tokens
lowerCAmelCase_ :Optional[Any] = tokens + [tokenizer.unk_token]
lowerCAmelCase_ :List[Any] = [0, 2, 5, 4, 6, 8, 0, 3, 5, 4, 6, 19]
lowerCAmelCase_ :List[str] = tokenizer.convert_tokens_to_ids(_snake_case )
self.assertListEqual(_snake_case , _snake_case )
def __lowerCAmelCase ( self ) -> int:
lowerCAmelCase_ :Optional[Any] = self.get_tokenizer()
# Testing tokenization
lowerCAmelCase_ :List[str] = """こんにちは、<|bagoftoken|>世界。こんばんは、<|bagoftoken|>㔺界。"""
lowerCAmelCase_ :str = """こんにちは、、、、世界。こんばんは、、、、世界。"""
lowerCAmelCase_ :Union[str, Any] = tokenizer.encode(_snake_case )
lowerCAmelCase_ :Any = tokenizer.decode(_snake_case )
self.assertEqual(_snake_case , _snake_case )
@slow
def __lowerCAmelCase ( self ) -> Optional[Any]:
lowerCAmelCase_ :Tuple = self.tokenizer_class.from_pretrained("""Tanrei/GPTSAN-japanese""" )
# Testing tokenization
lowerCAmelCase_ :Optional[Any] = """こんにちは、世界。"""
lowerCAmelCase_ :Tuple = """こんばんは、㔺界。😀"""
lowerCAmelCase_ :Optional[int] = """こんにちは、世界。こんばんは、世界。😀"""
lowerCAmelCase_ :Any = tokenizer.encode(prefix_text + input_text )
lowerCAmelCase_ :int = tokenizer.encode("""""" , prefix_text=prefix_text + input_text )
lowerCAmelCase_ :Tuple = tokenizer.encode(_snake_case , prefix_text=_snake_case )
lowerCAmelCase_ :List[Any] = tokenizer.decode(_snake_case )
lowerCAmelCase_ :Union[str, Any] = tokenizer.decode(_snake_case )
lowerCAmelCase_ :Tuple = tokenizer.decode(_snake_case )
self.assertEqual(_snake_case , _snake_case )
self.assertEqual(_snake_case , _snake_case )
self.assertEqual(_snake_case , _snake_case )
@slow
def __lowerCAmelCase ( self ) -> Tuple:
lowerCAmelCase_ :Tuple = self.tokenizer_class.from_pretrained("""Tanrei/GPTSAN-japanese""" )
# Testing tokenization
lowerCAmelCase_ :str = """こんにちは、世界。"""
lowerCAmelCase_ :Union[str, Any] = """こんばんは、㔺界。😀"""
lowerCAmelCase_ :Dict = len(tokenizer.encode(_snake_case ) ) - 2
lowerCAmelCase_ :str = len(tokenizer.encode(_snake_case ) ) - 2
lowerCAmelCase_ :Optional[Any] = [1] + [0] * (len_prefix + len_text + 1)
lowerCAmelCase_ :Tuple = [1] * (len_prefix + len_text + 1) + [0]
lowerCAmelCase_ :List[str] = [1] + [1] * (len_prefix) + [0] * (len_text + 1)
lowerCAmelCase_ :List[Any] = tokenizer(prefix_text + input_text ).token_type_ids
lowerCAmelCase_ :Any = tokenizer("""""" , prefix_text=prefix_text + input_text ).token_type_ids
lowerCAmelCase_ :str = tokenizer(_snake_case , prefix_text=_snake_case ).token_type_ids
self.assertListEqual(_snake_case , _snake_case )
self.assertListEqual(_snake_case , _snake_case )
self.assertListEqual(_snake_case , _snake_case )
@slow
def __lowerCAmelCase ( self ) -> Tuple:
lowerCAmelCase_ :int = self.tokenizer_class.from_pretrained("""Tanrei/GPTSAN-japanese""" )
lowerCAmelCase_ :Any = tokenizer.encode("""あンいワ""" )
lowerCAmelCase_ :List[Any] = tokenizer.encode("""""" , prefix_text="""あンいワ""" )
lowerCAmelCase_ :Any = tokenizer.encode("""いワ""" , prefix_text="""あン""" )
self.assertEqual(tokenizer.decode(_snake_case ) , tokenizer.decode(_snake_case ) )
self.assertEqual(tokenizer.decode(_snake_case ) , tokenizer.decode(_snake_case ) )
self.assertNotEqual(_snake_case , _snake_case )
self.assertNotEqual(_snake_case , _snake_case )
self.assertEqual(x_token_a[1] , x_token_a[-1] ) # SEG token
self.assertEqual(x_token_a[1] , x_token_a[3] ) # SEG token
@slow
def __lowerCAmelCase ( self ) -> Any:
lowerCAmelCase_ :List[Any] = self.tokenizer_class.from_pretrained("""Tanrei/GPTSAN-japanese""" )
lowerCAmelCase_ :Union[str, Any] = [["""武田信玄""", """は、"""], ["""織田信長""", """の配下の、"""]]
lowerCAmelCase_ :Optional[int] = tokenizer(_snake_case , padding=_snake_case )
lowerCAmelCase_ :Dict = tokenizer.batch_encode_plus(_snake_case , padding=_snake_case )
# fmt: off
lowerCAmelCase_ :List[str] = [[3_5993, 8640, 2_5948, 3_5998, 3_0647, 3_5675, 3_5999, 3_5999], [3_5993, 1_0382, 9868, 3_5998, 3_0646, 9459, 3_0646, 3_5675]]
lowerCAmelCase_ :int = [[1, 1, 1, 0, 0, 0, 0, 0], [1, 1, 1, 0, 0, 0, 0, 0]]
lowerCAmelCase_ :Optional[Any] = [[1, 1, 1, 1, 1, 1, 0, 0], [1, 1, 1, 1, 1, 1, 1, 1]]
# fmt: on
self.assertListEqual(x_token.input_ids , _snake_case )
self.assertListEqual(x_token.token_type_ids , _snake_case )
self.assertListEqual(x_token.attention_mask , _snake_case )
self.assertListEqual(x_token_a.input_ids , _snake_case )
self.assertListEqual(x_token_a.token_type_ids , _snake_case )
self.assertListEqual(x_token_a.attention_mask , _snake_case )
def __lowerCAmelCase ( self ) -> Union[str, Any]:
pass
def __lowerCAmelCase ( self ) -> List[str]:
pass
| 84
|
import json
import os
import re
import shutil
import tempfile
import unittest
from typing import Tuple
from transformers import AddedToken, BatchEncoding, ByTaTokenizer
from transformers.utils import cached_property, is_tf_available, is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
if is_torch_available():
snake_case_ : Optional[Any] = "pt"
elif is_tf_available():
snake_case_ : Union[str, Any] = "tf"
else:
snake_case_ : str = "jax"
class __snake_case ( a , unittest.TestCase ):
UpperCAmelCase__ : List[Any] = ByTaTokenizer
UpperCAmelCase__ : int = False
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
super().setUp()
UpperCAmelCase_ = ByTaTokenizer()
tokenizer.save_pretrained(self.tmpdirname)
@cached_property
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
return ByTaTokenizer.from_pretrained('''google/byt5-small''')
def lowerCamelCase ( self : List[str] , **_snake_case : Union[str, Any]):
"""simple docstring"""
return self.tokenizer_class.from_pretrained(self.tmpdirname , **_snake_case)
def lowerCamelCase ( self : Dict , _snake_case : int , _snake_case : Tuple=False , _snake_case : Dict=20 , _snake_case : Optional[Any]=5):
"""simple docstring"""
UpperCAmelCase_ = []
for i in range(len(_snake_case)):
try:
UpperCAmelCase_ = tokenizer.decode([i] , clean_up_tokenization_spaces=_snake_case)
except UnicodeDecodeError:
pass
toks.append((i, tok))
UpperCAmelCase_ = list(filter(lambda _snake_case: re.match(r'''^[ a-zA-Z]+$''' , t[1]) , _snake_case))
UpperCAmelCase_ = list(filter(lambda _snake_case: [t[0]] == tokenizer.encode(t[1] , add_special_tokens=_snake_case) , _snake_case))
if max_length is not None and len(_snake_case) > max_length:
UpperCAmelCase_ = toks[:max_length]
if min_length is not None and len(_snake_case) < min_length and len(_snake_case) > 0:
while len(_snake_case) < min_length:
UpperCAmelCase_ = toks + toks
# toks_str = [t[1] for t in toks]
UpperCAmelCase_ = [t[0] for t in toks]
# Ensure consistency
UpperCAmelCase_ = tokenizer.decode(_snake_case , clean_up_tokenization_spaces=_snake_case)
if " " not in output_txt and len(_snake_case) > 1:
UpperCAmelCase_ = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=_snake_case)
+ ''' '''
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=_snake_case)
)
if with_prefix_space:
UpperCAmelCase_ = ''' ''' + output_txt
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
return output_txt, output_ids
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = tokenizer(['''hi</s>''', '''I went to the gym</s>''', '''</s>'''])
UpperCAmelCase_ = tokenizer(['''hi''', '''I went to the gym''', ''''''])
self.assertListEqual(batch_with_eos_added['''input_ids'''] , batch_without_eos_added['''input_ids'''])
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = '''Unicode €.'''
UpperCAmelCase_ = tokenizer(_snake_case)
UpperCAmelCase_ = [88, 113, 108, 102, 114, 103, 104, 35, 229, 133, 175, 49, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''Unicode €.</s>''')
UpperCAmelCase_ = tokenizer('''e è é ê ë''')
UpperCAmelCase_ = [104, 35, 198, 171, 35, 198, 172, 35, 198, 173, 35, 198, 174, 1]
self.assertEqual(encoded['''input_ids'''] , _snake_case)
# decoding
UpperCAmelCase_ = tokenizer.decode(_snake_case)
self.assertEqual(_snake_case , '''e è é ê ë</s>''')
# encode/decode, but with `encode` instead of `__call__`
self.assertEqual(tokenizer.decode(tokenizer.encode('''e è é ê ë''')) , '''e è é ê ë</s>''')
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 1, 0]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
if FRAMEWORK != "jax":
UpperCAmelCase_ = list(batch.input_ids.numpy()[0])
else:
UpperCAmelCase_ = list(batch.input_ids.tolist()[0])
self.assertListEqual(_snake_case , _snake_case)
self.assertEqual((2, 37) , batch.input_ids.shape)
self.assertEqual((2, 37) , batch.attention_mask.shape)
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization.''', '''Another paragraph for summarization.''']
UpperCAmelCase_ = tokenizer(_snake_case , padding=_snake_case , return_tensors=_snake_case)
# check if input_ids are returned and no decoder_input_ids
self.assertIn('''input_ids''' , _snake_case)
self.assertIn('''attention_mask''' , _snake_case)
self.assertNotIn('''decoder_input_ids''' , _snake_case)
self.assertNotIn('''decoder_attention_mask''' , _snake_case)
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = [
'''Summary of the text.''',
'''Another summary.''',
]
UpperCAmelCase_ = tokenizer(
text_target=_snake_case , max_length=32 , padding='''max_length''' , truncation=_snake_case , return_tensors=_snake_case)
self.assertEqual(32 , targets['''input_ids'''].shape[1])
def lowerCamelCase ( self : int):
"""simple docstring"""
UpperCAmelCase_ = self.ta_base_tokenizer
UpperCAmelCase_ = ['''A long paragraph for summarization. </s>''']
UpperCAmelCase_ = ['''Summary of the text. </s>''']
# fmt: off
UpperCAmelCase_ = [68, 35, 111, 114, 113, 106, 35, 115, 100, 117, 100, 106, 117, 100, 115, 107, 35, 105, 114, 117, 35, 118, 120, 112, 112, 100, 117, 108, 125, 100, 119, 108, 114, 113, 49, 35, 1]
UpperCAmelCase_ = [86, 120, 112, 112, 100, 117, 124, 35, 114, 105, 35, 119, 107, 104, 35, 119, 104, 123, 119, 49, 35, 1]
# fmt: on
UpperCAmelCase_ = tokenizer(_snake_case , text_target=_snake_case)
self.assertEqual(_snake_case , batch['''input_ids'''][0])
self.assertEqual(_snake_case , batch['''labels'''][0])
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
self.assertNotEqual(tokenizer.model_max_length , 42)
# Now let's start the test
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
shutil.rmtree(_snake_case)
UpperCAmelCase_ = self.get_tokenizers(model_max_length=42)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
# Isolate this from the other tests because we save additional tokens/etc
UpperCAmelCase_ = tempfile.mkdtemp()
UpperCAmelCase_ = ''' He is very happy, UNwant\u00E9d,running'''
tokenizer.add_tokens(['''bim''', '''bambam'''])
UpperCAmelCase_ = tokenizer.additional_special_tokens
additional_special_tokens.append('''new_additional_special_token''')
tokenizer.add_special_tokens({'''additional_special_tokens''': additional_special_tokens})
UpperCAmelCase_ = tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
tokenizer.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case)
UpperCAmelCase_ = after_tokenizer.encode(_snake_case , add_special_tokens=_snake_case)
self.assertListEqual(_snake_case , _snake_case)
self.assertIn('''new_additional_special_token''' , after_tokenizer.additional_special_tokens)
self.assertEqual(after_tokenizer.model_max_length , 42)
UpperCAmelCase_ = tokenizer.__class__.from_pretrained(_snake_case , model_max_length=43)
self.assertEqual(tokenizer.model_max_length , 43)
shutil.rmtree(_snake_case)
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , encoding='''utf-8''') as json_file:
UpperCAmelCase_ = json.load(_snake_case)
UpperCAmelCase_ = [F"""<extra_id_{i}>""" for i in range(125)]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
UpperCAmelCase_ = added_tokens_extra_ids + [
'''an_additional_special_token'''
]
with open(os.path.join(_snake_case , '''special_tokens_map.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
with open(os.path.join(_snake_case , '''tokenizer_config.json''') , '''w''' , encoding='''utf-8''') as outfile:
json.dump(_snake_case , _snake_case)
# the following checks allow us to verify that our test works as expected, i.e. that the tokenizer takes
# into account the new value of additional_special_tokens given in the "tokenizer_config.json" and
# "special_tokens_map.json" files
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , )
self.assertIn(
'''an_additional_special_token''' , tokenizer_without_change_in_init.additional_special_tokens)
# self.assertIn("an_additional_special_token",tokenizer_without_change_in_init.get_vocab()) # ByT5Tokenization no vocab
self.assertEqual(
['''an_additional_special_token'''] , tokenizer_without_change_in_init.convert_ids_to_tokens(
tokenizer_without_change_in_init.convert_tokens_to_ids(['''an_additional_special_token'''])) , )
# Now we test that we can change the value of additional_special_tokens in the from_pretrained
UpperCAmelCase_ = added_tokens_extra_ids + [AddedToken('''a_new_additional_special_token''' , lstrip=_snake_case)]
UpperCAmelCase_ = tokenizer_class.from_pretrained(
_snake_case , additional_special_tokens=_snake_case , )
self.assertIn('''a_new_additional_special_token''' , tokenizer.additional_special_tokens)
self.assertEqual(
['''a_new_additional_special_token'''] , tokenizer.convert_ids_to_tokens(
tokenizer.convert_tokens_to_ids(['''a_new_additional_special_token'''])) , )
def lowerCamelCase ( self : Any):
"""simple docstring"""
UpperCAmelCase_ = []
if self.test_slow_tokenizer:
tokenizer_list.append((self.tokenizer_class, self.get_tokenizer()))
if self.test_rust_tokenizer:
tokenizer_list.append((self.rust_tokenizer_class, self.get_rust_tokenizer()))
for tokenizer_class, tokenizer_utils in tokenizer_list:
with tempfile.TemporaryDirectory() as tmp_dir:
tokenizer_utils.save_pretrained(_snake_case)
UpperCAmelCase_ = tokenizer_class.from_pretrained(_snake_case)
self.assertTrue(tokenizer.decode([255]) == '''''')
def lowerCamelCase ( self : int):
"""simple docstring"""
pass
def lowerCamelCase ( self : Optional[int]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Dict):
"""simple docstring"""
pass
def lowerCamelCase ( self : List[Any]):
"""simple docstring"""
pass
def lowerCamelCase ( self : Tuple):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers(fast=_snake_case , do_lower_case=_snake_case)
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = ['''t''', '''h''', '''i''', '''s''', ''' ''', '''i''', '''s''', ''' ''', '''a''', ''' ''', '''t''', '''e''', '''x''', '''t''', '''</s>''']
UpperCAmelCase_ = tokenizer.convert_tokens_to_string(_snake_case)
self.assertIsInstance(_snake_case , _snake_case)
def lowerCamelCase ( self : Union[str, Any]):
"""simple docstring"""
UpperCAmelCase_ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}"""):
UpperCAmelCase_ = [
'''bos_token''',
'''eos_token''',
'''unk_token''',
'''sep_token''',
'''pad_token''',
'''cls_token''',
'''mask_token''',
]
UpperCAmelCase_ = 0
UpperCAmelCase_ = tokenizer.convert_ids_to_tokens(
_snake_case , skip_special_tokens=_snake_case)
for attr in attributes_list:
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , attr + '''_id''' , _snake_case)
self.assertEqual(getattr(_snake_case , _snake_case) , _snake_case)
self.assertEqual(getattr(_snake_case , attr + '''_id''') , _snake_case)
setattr(_snake_case , '''additional_special_tokens_ids''' , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [])
setattr(_snake_case , '''additional_special_tokens_ids''' , [token_id_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens''') , [token_to_test_setters])
self.assertListEqual(getattr(_snake_case , '''additional_special_tokens_ids''') , [token_id_to_test_setters])
| 51
| 0
|
'''simple docstring'''
import timeit
import numpy as np
import datasets
from datasets.arrow_writer import ArrowWriter
from datasets.features.features import _ArrayXD
def snake_case_ ( _lowerCAmelCase : List[str] ) -> Tuple:
def wrapper(*_lowerCAmelCase : Any , **_lowerCAmelCase : Union[str, Any] ):
UpperCAmelCase : Any = timeit.default_timer()
UpperCAmelCase : int = func(*__A , **__A )
UpperCAmelCase : Union[str, Any] = timeit.default_timer() - starttime
return delta
UpperCAmelCase : Optional[Any] = func.__name__
return wrapper
def snake_case_ ( _lowerCAmelCase : dict , _lowerCAmelCase : str=100 , _lowerCAmelCase : Dict=None ) -> str:
UpperCAmelCase : List[Any] = []
UpperCAmelCase : Tuple = seq_shapes or {}
for i in range(__A ):
UpperCAmelCase : List[Any] = {}
for col_id, (k, v) in enumerate(features.items() ):
if isinstance(__A , _ArrayXD ):
UpperCAmelCase : List[str] = np.random.rand(*v.shape ).astype(v.dtype )
elif isinstance(__A , datasets.Value ):
if v.dtype == "string":
UpperCAmelCase : List[Any] = '''The small grey turtle was surprisingly fast when challenged.'''
else:
UpperCAmelCase : Optional[Any] = np.random.randint(10 , size=1 ).astype(v.dtype ).item()
elif isinstance(__A , datasets.Sequence ):
while isinstance(__A , datasets.Sequence ):
UpperCAmelCase : Optional[Any] = v.feature
UpperCAmelCase : Optional[int] = seq_shapes[k]
UpperCAmelCase : Union[str, Any] = np.random.rand(*__A ).astype(v.dtype )
UpperCAmelCase : List[str] = data
dummy_data.append((i, example) )
return dummy_data
def snake_case_ ( _lowerCAmelCase : Optional[Any] , _lowerCAmelCase : List[Any] , _lowerCAmelCase : Dict=100 , _lowerCAmelCase : str=None ) -> Optional[Any]:
UpperCAmelCase : List[Any] = generate_examples(__A , num_examples=__A , seq_shapes=__A )
with ArrowWriter(features=__A , path=__A ) as writer:
for key, record in dummy_data:
UpperCAmelCase : Union[str, Any] = features.encode_example(__A )
writer.write(__A )
UpperCAmelCase , UpperCAmelCase : str = writer.finalize()
if not num_final_examples == num_examples:
raise ValueError(
f"""Error writing the dataset, wrote {num_final_examples} examples but should have written {num_examples}.""" )
UpperCAmelCase : Dict = datasets.Dataset.from_file(filename=__A , info=datasets.DatasetInfo(features=__A ) )
return dataset
| 23
|
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_sentencepiece_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case_ : Dict = {"configuration_mbart": ["MBART_PRETRAINED_CONFIG_ARCHIVE_MAP", "MBartConfig", "MBartOnnxConfig"]}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Tuple = ["MBartTokenizer"]
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : str = ["MBartTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[Any] = [
"MBART_PRETRAINED_MODEL_ARCHIVE_LIST",
"MBartForCausalLM",
"MBartForConditionalGeneration",
"MBartForQuestionAnswering",
"MBartForSequenceClassification",
"MBartModel",
"MBartPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : Any = [
"TFMBartForConditionalGeneration",
"TFMBartModel",
"TFMBartPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ : List[str] = [
"FlaxMBartForConditionalGeneration",
"FlaxMBartForQuestionAnswering",
"FlaxMBartForSequenceClassification",
"FlaxMBartModel",
"FlaxMBartPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_mbart import MBART_PRETRAINED_CONFIG_ARCHIVE_MAP, MBartConfig, MBartOnnxConfig
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart import MBartTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_mbart_fast import MBartTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_mbart import (
MBART_PRETRAINED_MODEL_ARCHIVE_LIST,
MBartForCausalLM,
MBartForConditionalGeneration,
MBartForQuestionAnswering,
MBartForSequenceClassification,
MBartModel,
MBartPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_mbart import TFMBartForConditionalGeneration, TFMBartModel, TFMBartPreTrainedModel
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_mbart import (
FlaxMBartForConditionalGeneration,
FlaxMBartForQuestionAnswering,
FlaxMBartForSequenceClassification,
FlaxMBartModel,
FlaxMBartPreTrainedModel,
)
else:
import sys
snake_case_ : Any = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 51
| 0
|
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase ) -> int:
lowerCamelCase__ : Any = hex_num.strip()
if not hex_num:
raise ValueError('No value was passed to the function' )
lowerCamelCase__ : Tuple = hex_num[0] == '-'
if is_negative:
lowerCamelCase__ : List[str] = hex_num[1:]
try:
lowerCamelCase__ : Union[str, Any] = int(__A , 16 )
except ValueError:
raise ValueError('Invalid value was passed to the function' )
lowerCamelCase__ : Optional[Any] = ''
while int_num > 0:
lowerCamelCase__ : int = str(int_num % 2 ) + bin_str
int_num >>= 1
return int(('-' + bin_str) if is_negative else bin_str )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 50
|
import warnings
from typing import List, Optional, Union
from ...image_utils import ImageInput
from ...processing_utils import ProcessorMixin
from ...tokenization_utils_base import BatchEncoding, PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy
from ...utils import TensorType
class __snake_case ( a ):
UpperCAmelCase__ : Dict = ['''image_processor''', '''tokenizer''']
UpperCAmelCase__ : Dict = '''FlavaImageProcessor'''
UpperCAmelCase__ : Dict = ('''BertTokenizer''', '''BertTokenizerFast''')
def __init__( self : Union[str, Any] , _snake_case : List[str]=None , _snake_case : str=None , **_snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = None
if "feature_extractor" in kwargs:
warnings.warn(
'''The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor`'''
''' instead.''' , _snake_case , )
UpperCAmelCase_ = kwargs.pop('''feature_extractor''')
UpperCAmelCase_ = image_processor if image_processor is not None else feature_extractor
if image_processor is None:
raise ValueError('''You need to specify an `image_processor`.''')
if tokenizer is None:
raise ValueError('''You need to specify a `tokenizer`.''')
super().__init__(_snake_case , _snake_case)
UpperCAmelCase_ = self.image_processor
def __call__( self : List[Any] , _snake_case : Optional[ImageInput] = None , _snake_case : Optional[Union[TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]]] = None , _snake_case : bool = True , _snake_case : Union[bool, str, PaddingStrategy] = False , _snake_case : Union[bool, str, TruncationStrategy] = False , _snake_case : Optional[int] = None , _snake_case : int = 0 , _snake_case : Optional[int] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : Optional[bool] = None , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = False , _snake_case : bool = True , _snake_case : Optional[Union[str, TensorType]] = None , **_snake_case : Any , ):
"""simple docstring"""
if text is None and images is None:
raise ValueError('''You have to specify either text or images. Both cannot be none.''')
if text is not None:
UpperCAmelCase_ = self.tokenizer(
text=_snake_case , add_special_tokens=_snake_case , padding=_snake_case , truncation=_snake_case , max_length=_snake_case , stride=_snake_case , pad_to_multiple_of=_snake_case , return_token_type_ids=_snake_case , return_attention_mask=_snake_case , return_overflowing_tokens=_snake_case , return_special_tokens_mask=_snake_case , return_offsets_mapping=_snake_case , return_length=_snake_case , verbose=_snake_case , return_tensors=_snake_case , **_snake_case , )
if images is not None:
UpperCAmelCase_ = self.image_processor(
_snake_case , return_image_mask=_snake_case , return_codebook_pixels=_snake_case , return_tensors=_snake_case , **_snake_case , )
if text is not None and images is not None:
encoding.update(_snake_case)
return encoding
elif text is not None:
return encoding
else:
return BatchEncoding(data=dict(**_snake_case) , tensor_type=_snake_case)
def lowerCamelCase ( self : Any , *_snake_case : Optional[Any] , **_snake_case : int):
"""simple docstring"""
return self.tokenizer.batch_decode(*_snake_case , **_snake_case)
def lowerCamelCase ( self : Optional[int] , *_snake_case : int , **_snake_case : Dict):
"""simple docstring"""
return self.tokenizer.decode(*_snake_case , **_snake_case)
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
UpperCAmelCase_ = self.tokenizer.model_input_names
UpperCAmelCase_ = self.image_processor.model_input_names
return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names))
@property
def lowerCamelCase ( self : str):
"""simple docstring"""
warnings.warn(
'''`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.''' , _snake_case , )
return self.image_processor_class
@property
def lowerCamelCase ( self : Any):
"""simple docstring"""
warnings.warn(
'''`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.''' , _snake_case , )
return self.image_processor
| 51
| 0
|
from __future__ import annotations
def lowerCAmelCase_ ( _lowercase : str) -> list[int]:
"""simple docstring"""
return [ord(__A) - 96 for elem in plain]
def lowerCAmelCase_ ( _lowercase : list[int]) -> str:
"""simple docstring"""
return "".join(chr(elem + 96) for elem in encoded)
def lowerCAmelCase_ ( ) -> None:
"""simple docstring"""
a__ : Any = encode(input("""-> """).strip().lower())
print("""Encoded: """ , __A)
print("""Decoded:""" , decode(__A))
if __name__ == "__main__":
main()
| 170
|
from diffusers.utils.testing_utils import require_onnxruntime
@require_onnxruntime
class __snake_case :
pass
| 51
| 0
|
import os
import pickle
import unittest
from transformers import AutoTokenizer
from transformers.models.bert.tokenization_bert import BertTokenizer
from transformers.models.bert_japanese.tokenization_bert_japanese import (
VOCAB_FILES_NAMES,
BertJapaneseTokenizer,
CharacterTokenizer,
JumanppTokenizer,
MecabTokenizer,
SudachiTokenizer,
WordpieceTokenizer,
)
from transformers.testing_utils import custom_tokenizers, require_jumanpp, require_sudachi
from ...test_tokenization_common import TokenizerTesterMixin
@custom_tokenizers
class SCREAMING_SNAKE_CASE__ ( SCREAMING_SNAKE_CASE_ , unittest.TestCase ):
"""simple docstring"""
_snake_case = BertJapaneseTokenizer
_snake_case = False
_snake_case = True
def A__ ( self )-> Any:
'''simple docstring'''
super().setUp()
__UpperCamelCase = [
'''[UNK]''',
'''[CLS]''',
'''[SEP]''',
'''こんにちは''',
'''こん''',
'''にちは''',
'''ばんは''',
'''##こん''',
'''##にちは''',
'''##ばんは''',
'''世界''',
'''##世界''',
'''、''',
'''##、''',
'''。''',
'''##。''',
]
__UpperCamelCase = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''vocab_file'''] )
with open(self.vocab_file , '''w''' , encoding='''utf-8''' ) as vocab_writer:
vocab_writer.write(''''''.join([x + '''\n''' for x in vocab_tokens] ) )
def A__ ( self , SCREAMING_SNAKE_CASE_ )-> List[str]:
'''simple docstring'''
__UpperCamelCase = '''こんにちは、世界。 \nこんばんは、世界。'''
__UpperCamelCase = '''こんにちは 、 世界 。 こんばんは 、 世界 。'''
return input_text, output_text
def A__ ( self , SCREAMING_SNAKE_CASE_ )-> Optional[Any]:
'''simple docstring'''
__UpperCamelCase , __UpperCamelCase = self.get_input_output_texts(_snake_case )
__UpperCamelCase = tokenizer.encode(_snake_case , add_special_tokens=_snake_case )
__UpperCamelCase = tokenizer.decode(_snake_case , clean_up_tokenization_spaces=_snake_case )
return text, ids
def A__ ( self )-> Any:
'''simple docstring'''
pass # TODO add if relevant
def A__ ( self )-> Optional[Any]:
'''simple docstring'''
pass # TODO add if relevant
def A__ ( self )-> Tuple:
'''simple docstring'''
pass # TODO add if relevant
def A__ ( self )-> List[Any]:
'''simple docstring'''
__UpperCamelCase = self.tokenizer_class(self.vocab_file )
__UpperCamelCase = tokenizer.tokenize('''こんにちは、世界。\nこんばんは、世界。''' )
self.assertListEqual(_snake_case , ['''こんにちは''', '''、''', '''世界''', '''。''', '''こん''', '''##ばんは''', '''、''', '''世界''', '''。'''] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case ) , [3, 12, 10, 14, 4, 9, 12, 10, 14] )
def A__ ( self )-> List[Any]:
'''simple docstring'''
__UpperCamelCase = self.tokenizer_class(self.vocab_file , word_tokenizer_type='''mecab''' )
self.assertIsNotNone(_snake_case )
__UpperCamelCase = '''こんにちは、世界。\nこんばんは、世界。'''
__UpperCamelCase = tokenizer.tokenize(_snake_case )
self.assertListEqual(_snake_case , ['''こんにちは''', '''、''', '''世界''', '''。''', '''こん''', '''##ばんは''', '''、''', '''世界''', '''。'''] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case ) , [3, 12, 10, 14, 4, 9, 12, 10, 14] )
__UpperCamelCase = os.path.join(self.tmpdirname , '''tokenizer.bin''' )
with open(_snake_case , '''wb''' ) as handle:
pickle.dump(_snake_case , _snake_case )
with open(_snake_case , '''rb''' ) as handle:
__UpperCamelCase = pickle.load(_snake_case )
__UpperCamelCase = tokenizer_new.tokenize(_snake_case )
self.assertListEqual(_snake_case , _snake_case )
def A__ ( self )-> List[str]:
'''simple docstring'''
__UpperCamelCase = MecabTokenizer(mecab_dic='''ipadic''' )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''アップルストア''', '''で''', '''iPhone''', '''8''', '''が''', '''発売''', '''さ''', '''れ''', '''た''', '''。'''] , )
def A__ ( self )-> Any:
'''simple docstring'''
try:
__UpperCamelCase = MecabTokenizer(mecab_dic='''unidic_lite''' )
except ModuleNotFoundError:
return
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''アップル''', '''ストア''', '''で''', '''iPhone''', '''8''', '''が''', '''発売''', '''さ''', '''れ''', '''た''', '''。'''] , )
def A__ ( self )-> Optional[Any]:
'''simple docstring'''
try:
__UpperCamelCase = MecabTokenizer(mecab_dic='''unidic''' )
except ModuleNotFoundError:
return
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''アップル''', '''ストア''', '''で''', '''iPhone''', '''8''', '''が''', '''発売''', '''さ''', '''れ''', '''た''', '''。'''] , )
def A__ ( self )-> Any:
'''simple docstring'''
__UpperCamelCase = MecabTokenizer(do_lower_case=_snake_case , mecab_dic='''ipadic''' )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''アップルストア''', '''で''', '''iphone''', '''8''', '''が''', '''発売''', '''さ''', '''れ''', '''た''', '''。'''] , )
def A__ ( self )-> Any:
'''simple docstring'''
try:
__UpperCamelCase = MecabTokenizer(
do_lower_case=_snake_case , normalize_text=_snake_case , mecab_option='''-d /usr/local/lib/mecab/dic/jumandic''' )
except RuntimeError:
# if dict doesn't exist in the system, previous code raises this error.
return
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''ｱｯﾌﾟﾙストア''', '''で''', '''iPhone''', '''８''', '''が''', '''発売''', '''さ''', '''れた''', '''\u3000''', '''。'''] , )
def A__ ( self )-> Any:
'''simple docstring'''
__UpperCamelCase = MecabTokenizer(normalize_text=_snake_case , mecab_dic='''ipadic''' )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''ｱｯﾌﾟﾙストア''', '''で''', '''iPhone''', '''８''', '''が''', '''発売''', '''さ''', '''れ''', '''た''', ''' ''', '''。'''] , )
@require_sudachi
def A__ ( self )-> Optional[Any]:
'''simple docstring'''
__UpperCamelCase = self.tokenizer_class(self.vocab_file , word_tokenizer_type='''sudachi''' )
self.assertIsNotNone(_snake_case )
__UpperCamelCase = '''こんにちは、世界。\nこんばんは、世界。'''
__UpperCamelCase = tokenizer.tokenize(_snake_case )
self.assertListEqual(_snake_case , ['''こんにちは''', '''、''', '''世界''', '''。''', '''こん''', '''##ばんは''', '''、''', '''世界''', '''。'''] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case ) , [3, 12, 10, 14, 4, 9, 12, 10, 14] )
__UpperCamelCase = os.path.join(self.tmpdirname , '''tokenizer.bin''' )
with open(_snake_case , '''wb''' ) as handle:
pickle.dump(_snake_case , _snake_case )
with open(_snake_case , '''rb''' ) as handle:
__UpperCamelCase = pickle.load(_snake_case )
__UpperCamelCase = tokenizer_new.tokenize(_snake_case )
self.assertListEqual(_snake_case , _snake_case )
@require_sudachi
def A__ ( self )-> List[Any]:
'''simple docstring'''
__UpperCamelCase = SudachiTokenizer(sudachi_dict_type='''core''' )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , [''' ''', '''\t''', '''アップル''', '''ストア''', '''で''', '''iPhone''', '''8''', ''' ''', '''が''', ''' ''', ''' ''', '''\n ''', '''発売''', '''さ''', '''れ''', '''た''', ''' ''', '''。''', ''' ''', ''' '''] , )
@require_sudachi
def A__ ( self )-> List[str]:
'''simple docstring'''
__UpperCamelCase = SudachiTokenizer(sudachi_dict_type='''core''' , sudachi_split_mode='''A''' )
self.assertListEqual(tokenizer.tokenize('''外国人参政権''' ) , ['''外国''', '''人''', '''参政''', '''権'''] )
@require_sudachi
def A__ ( self )-> Any:
'''simple docstring'''
__UpperCamelCase = SudachiTokenizer(sudachi_dict_type='''core''' , sudachi_split_mode='''B''' )
self.assertListEqual(tokenizer.tokenize('''外国人参政権''' ) , ['''外国人''', '''参政権'''] )
@require_sudachi
def A__ ( self )-> str:
'''simple docstring'''
__UpperCamelCase = SudachiTokenizer(sudachi_dict_type='''core''' , sudachi_split_mode='''C''' )
self.assertListEqual(tokenizer.tokenize('''外国人参政権''' ) , ['''外国人参政権'''] )
@require_sudachi
def A__ ( self )-> Tuple:
'''simple docstring'''
__UpperCamelCase = SudachiTokenizer(do_lower_case=_snake_case , sudachi_dict_type='''core''' )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , [''' ''', '''\t''', '''アップル''', '''ストア''', '''で''', '''iphone''', '''8''', ''' ''', '''が''', ''' ''', ''' ''', '''\n ''', '''発売''', '''さ''', '''れ''', '''た''', ''' ''', '''。''', ''' ''', ''' '''] , )
@require_sudachi
def A__ ( self )-> str:
'''simple docstring'''
__UpperCamelCase = SudachiTokenizer(normalize_text=_snake_case , sudachi_dict_type='''core''' )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , [''' ''', '''\t''', '''ｱｯﾌﾟﾙ''', '''ストア''', '''で''', '''iPhone''', '''８''', ''' ''', '''が''', ''' ''', ''' ''', '''\n ''', '''発売''', '''さ''', '''れ''', '''た''', '''\u3000''', '''。''', ''' ''', ''' '''] , )
@require_sudachi
def A__ ( self )-> Dict:
'''simple docstring'''
__UpperCamelCase = SudachiTokenizer(trim_whitespace=_snake_case , sudachi_dict_type='''core''' )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''アップル''', '''ストア''', '''で''', '''iPhone''', '''8''', '''が''', '''発売''', '''さ''', '''れ''', '''た''', '''。'''] , )
@require_jumanpp
def A__ ( self )-> List[str]:
'''simple docstring'''
__UpperCamelCase = self.tokenizer_class(self.vocab_file , word_tokenizer_type='''jumanpp''' )
self.assertIsNotNone(_snake_case )
__UpperCamelCase = '''こんにちは、世界。\nこんばんは、世界。'''
__UpperCamelCase = tokenizer.tokenize(_snake_case )
self.assertListEqual(_snake_case , ['''こんにちは''', '''、''', '''世界''', '''。''', '''こん''', '''##ばんは''', '''、''', '''世界''', '''。'''] )
self.assertListEqual(tokenizer.convert_tokens_to_ids(_snake_case ) , [3, 12, 10, 14, 4, 9, 12, 10, 14] )
__UpperCamelCase = os.path.join(self.tmpdirname , '''tokenizer.bin''' )
with open(_snake_case , '''wb''' ) as handle:
pickle.dump(_snake_case , _snake_case )
with open(_snake_case , '''rb''' ) as handle:
__UpperCamelCase = pickle.load(_snake_case )
__UpperCamelCase = tokenizer_new.tokenize(_snake_case )
self.assertListEqual(_snake_case , _snake_case )
@require_jumanpp
def A__ ( self )-> Dict:
'''simple docstring'''
__UpperCamelCase = JumanppTokenizer()
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''アップル''', '''ストア''', '''で''', '''iPhone''', '''8''', '''\u3000''', '''が''', '''\u3000''', '''\u3000''', '''\u3000''', '''発売''', '''さ''', '''れた''', '''\u3000''', '''。'''] , )
@require_jumanpp
def A__ ( self )-> str:
'''simple docstring'''
__UpperCamelCase = JumanppTokenizer(do_lower_case=_snake_case )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''アップル''', '''ストア''', '''で''', '''iphone''', '''8''', '''\u3000''', '''が''', '''\u3000''', '''\u3000''', '''\u3000''', '''発売''', '''さ''', '''れた''', '''\u3000''', '''。'''] , )
@require_jumanpp
def A__ ( self )-> List[Any]:
'''simple docstring'''
__UpperCamelCase = JumanppTokenizer(normalize_text=_snake_case )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''ｱ''', '''ｯ''', '''ﾌ''', '''ﾟ''', '''ﾙ''', '''ストア''', '''で''', '''iPhone''', '''８''', '''\u3000''', '''が''', '''\u3000''', '''\u3000''', '''\u3000''', '''発売''', '''さ''', '''れた''', '''\u3000''', '''。'''] , )
@require_jumanpp
def A__ ( self )-> str:
'''simple docstring'''
__UpperCamelCase = JumanppTokenizer(trim_whitespace=_snake_case )
self.assertListEqual(
tokenizer.tokenize(''' \tｱｯﾌﾟﾙストアでiPhone８ が \n 発売された 。 ''' ) , ['''アップル''', '''ストア''', '''で''', '''iPhone''', '''8''', '''が''', '''発売''', '''さ''', '''れた''', '''。'''] , )
@require_jumanpp
def A__ ( self )-> Dict:
'''simple docstring'''
__UpperCamelCase = JumanppTokenizer()
self.assertListEqual(
tokenizer.tokenize('''ありがとうございますm(_ _)ｍ見つけるのが大変です。''' ) , ['''ありがとう''', '''ございます''', '''m(_ _)m''', '''見つける''', '''の''', '''が''', '''大変です''', '''。'''] , )
def A__ ( self )-> str:
'''simple docstring'''
__UpperCamelCase = ['''[UNK]''', '''[CLS]''', '''[SEP]''', '''こんにちは''', '''こん''', '''にちは''', '''ばんは''', '''##こん''', '''##にちは''', '''##ばんは''']
__UpperCamelCase = {}
for i, token in enumerate(_snake_case ):
__UpperCamelCase = i
__UpperCamelCase = WordpieceTokenizer(vocab=_snake_case , unk_token='''[UNK]''' )
self.assertListEqual(tokenizer.tokenize('''''' ) , [] )
self.assertListEqual(tokenizer.tokenize('''こんにちは''' ) , ['''こんにちは'''] )
self.assertListEqual(tokenizer.tokenize('''こんばんは''' ) , ['''こん''', '''##ばんは'''] )
self.assertListEqual(tokenizer.tokenize('''こんばんは こんばんにちは こんにちは''' ) , ['''こん''', '''##ばんは''', '''[UNK]''', '''こんにちは'''] )
def A__ ( self )-> Tuple:
'''simple docstring'''
__UpperCamelCase = BertJapaneseTokenizer.from_pretrained('''nlp-waseda/roberta-base-japanese-with-auto-jumanpp''' )
__UpperCamelCase = tokenizer.subword_tokenizer
__UpperCamelCase = subword_tokenizer.tokenize('''国境 の 長い トンネル を 抜ける と 雪国 であった 。''' )
self.assertListEqual(_snake_case , ['''▁国境''', '''▁の''', '''▁長い''', '''▁トンネル''', '''▁を''', '''▁抜ける''', '''▁と''', '''▁雪''', '''国''', '''▁であった''', '''▁。'''] )
__UpperCamelCase = subword_tokenizer.tokenize('''こんばんは こんばん にち は こんにちは''' )
self.assertListEqual(_snake_case , ['''▁こん''', '''ばん''', '''は''', '''▁こん''', '''ばん''', '''▁に''', '''ち''', '''▁は''', '''▁こんにちは'''] )
def A__ ( self )-> Optional[Any]:
'''simple docstring'''
__UpperCamelCase = self.tokenizer_class.from_pretrained('''cl-tohoku/bert-base-japanese''' )
__UpperCamelCase = tokenizer.encode('''ありがとう。''' , add_special_tokens=_snake_case )
__UpperCamelCase = tokenizer.encode('''どういたしまして。''' , add_special_tokens=_snake_case )
__UpperCamelCase = tokenizer.build_inputs_with_special_tokens(_snake_case )
__UpperCamelCase = tokenizer.build_inputs_with_special_tokens(_snake_case , _snake_case )
# 2 is for "[CLS]", 3 is for "[SEP]"
assert encoded_sentence == [2] + text + [3]
assert encoded_pair == [2] + text + [3] + text_a + [3]
@custom_tokenizers
class SCREAMING_SNAKE_CASE__ ( SCREAMING_SNAKE_CASE_ , unittest.TestCase ):
"""simple docstring"""
_snake_case = BertJapaneseTokenizer
_snake_case = False
def A__ ( self )-> str:
'''simple docstring'''
super().setUp()
__UpperCamelCase = ['''[UNK]''', '''[CLS]''', '''[SEP]''', '''こ''', '''ん''', '''に''', '''ち''', '''は''', '''ば''', '''世''', '''界''', '''、''', '''。''']
__UpperCamelCase = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''vocab_file'''] )
with open(self.vocab_file , '''w''' , encoding='''utf-8''' ) as vocab_writer:
vocab_writer.write(''''''.join([x + '''\n''' for x in vocab_tokens] ) )
def A__ ( self , **SCREAMING_SNAKE_CASE_ )-> Dict:
'''simple docstring'''
return BertJapaneseTokenizer.from_pretrained(self.tmpdirname , subword_tokenizer_type='''character''' , **_snake_case )
def A__ ( self , SCREAMING_SNAKE_CASE_ )-> Dict:
'''simple docstring'''
__UpperCamelCase = '''こんにちは、世界。 \nこんばんは、世界。'''
__UpperCamelCase = '''こ ん に ち は 、 世 界 。 こ ん ば ん は 、 世 界 。'''
return input_text, output_text
def A__ ( self )-> Optional[Any]:
'''simple docstring'''
pass # TODO add if relevant
def A__ ( self )-> Optional[int]:
'''simple docstring'''
pass # TODO add if relevant
def A__ ( self )-> int:
'''simple docstring'''
pass # TODO add if relevant
def A__ ( self )-> Optional[int]:
'''simple docstring'''
__UpperCamelCase = self.tokenizer_class(self.vocab_file , subword_tokenizer_type='''character''' )
__UpperCamelCase = tokenizer.tokenize('''こんにちは、世界。 \nこんばんは、世界。''' )
self.assertListEqual(
_snake_case , ['''こ''', '''ん''', '''に''', '''ち''', '''は''', '''、''', '''世''', '''界''', '''。''', '''こ''', '''ん''', '''ば''', '''ん''', '''は''', '''、''', '''世''', '''界''', '''。'''] )
self.assertListEqual(
tokenizer.convert_tokens_to_ids(_snake_case ) , [3, 4, 5, 6, 7, 11, 9, 10, 12, 3, 4, 8, 4, 7, 11, 9, 10, 12] )
def A__ ( self )-> List[str]:
'''simple docstring'''
__UpperCamelCase = ['''[UNK]''', '''[CLS]''', '''[SEP]''', '''こ''', '''ん''', '''に''', '''ち''', '''は''', '''ば''', '''世''', '''界''', '''、''', '''。''']
__UpperCamelCase = {}
for i, token in enumerate(_snake_case ):
__UpperCamelCase = i
__UpperCamelCase = CharacterTokenizer(vocab=_snake_case , unk_token='''[UNK]''' )
self.assertListEqual(tokenizer.tokenize('''''' ) , [] )
self.assertListEqual(tokenizer.tokenize('''こんにちは''' ) , ['''こ''', '''ん''', '''に''', '''ち''', '''は'''] )
self.assertListEqual(tokenizer.tokenize('''こんにちほ''' ) , ['''こ''', '''ん''', '''に''', '''ち''', '''[UNK]'''] )
def A__ ( self )-> str:
'''simple docstring'''
__UpperCamelCase = self.tokenizer_class.from_pretrained('''cl-tohoku/bert-base-japanese-char''' )
__UpperCamelCase = tokenizer.encode('''ありがとう。''' , add_special_tokens=_snake_case )
__UpperCamelCase = tokenizer.encode('''どういたしまして。''' , add_special_tokens=_snake_case )
__UpperCamelCase = tokenizer.build_inputs_with_special_tokens(_snake_case )
__UpperCamelCase = tokenizer.build_inputs_with_special_tokens(_snake_case , _snake_case )
# 2 is for "[CLS]", 3 is for "[SEP]"
assert encoded_sentence == [2] + text + [3]
assert encoded_pair == [2] + text + [3] + text_a + [3]
@custom_tokenizers
class SCREAMING_SNAKE_CASE__ ( unittest.TestCase ):
"""simple docstring"""
def A__ ( self )-> Union[str, Any]:
'''simple docstring'''
__UpperCamelCase = '''cl-tohoku/bert-base-japanese'''
__UpperCamelCase = AutoTokenizer.from_pretrained(_snake_case )
self.assertIsInstance(_snake_case , _snake_case )
class SCREAMING_SNAKE_CASE__ ( unittest.TestCase ):
"""simple docstring"""
def A__ ( self )-> int:
'''simple docstring'''
__UpperCamelCase = '''cl-tohoku/bert-base-japanese'''
with self.assertLogs('''transformers''' , level='''WARNING''' ) as cm:
BertTokenizer.from_pretrained(_snake_case )
self.assertTrue(
cm.records[0].message.startswith(
'''The tokenizer class you load from this checkpoint is not the same type as the class this function'''
''' is called from.''' ) )
__UpperCamelCase = '''bert-base-cased'''
with self.assertLogs('''transformers''' , level='''WARNING''' ) as cm:
BertJapaneseTokenizer.from_pretrained(_snake_case )
self.assertTrue(
cm.records[0].message.startswith(
'''The tokenizer class you load from this checkpoint is not the same type as the class this function'''
''' is called from.''' ) )
| 328
|
import argparse
import os
import pickle
import sys
import torch
from transformers import TransfoXLConfig, TransfoXLLMHeadModel, load_tf_weights_in_transfo_xl
from transformers.models.transfo_xl import tokenization_transfo_xl as data_utils
from transformers.models.transfo_xl.tokenization_transfo_xl import CORPUS_NAME, VOCAB_FILES_NAMES
from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging
logging.set_verbosity_info()
# We do this to be able to load python 2 datasets pickles
# See e.g. https://stackoverflow.com/questions/2121874/python-pickling-after-changing-a-modules-directory/2121918#2121918
snake_case_ : List[Any] = data_utils.TransfoXLTokenizer
snake_case_ : int = data_utils.TransfoXLCorpus
snake_case_ : List[Any] = data_utils
snake_case_ : int = data_utils
def A (__A : Dict , __A : List[Any] , __A : Union[str, Any] , __A : Tuple ) -> Union[str, Any]:
"""simple docstring"""
if transfo_xl_dataset_file:
# Convert a pre-processed corpus (see original TensorFlow repo)
with open(__A , '''rb''' ) as fp:
UpperCAmelCase_ = pickle.load(__A , encoding='''latin1''' )
# Save vocabulary and dataset cache as Dictionaries (should be better than pickles for the long-term)
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + VOCAB_FILES_NAMES['''pretrained_vocab_file''']
print(F"""Save vocabulary to {pytorch_vocab_dump_path}""" )
UpperCAmelCase_ = corpus.vocab.__dict__
torch.save(__A , __A )
UpperCAmelCase_ = corpus.__dict__
corpus_dict_no_vocab.pop('''vocab''' , __A )
UpperCAmelCase_ = pytorch_dump_folder_path + '''/''' + CORPUS_NAME
print(F"""Save dataset to {pytorch_dataset_dump_path}""" )
torch.save(__A , __A )
if tf_checkpoint_path:
# Convert a pre-trained TensorFlow model
UpperCAmelCase_ = os.path.abspath(__A )
UpperCAmelCase_ = os.path.abspath(__A )
print(F"""Converting Transformer XL checkpoint from {tf_path} with config at {config_path}.""" )
# Initialise PyTorch model
if transfo_xl_config_file == "":
UpperCAmelCase_ = TransfoXLConfig()
else:
UpperCAmelCase_ = TransfoXLConfig.from_json_file(__A )
print(F"""Building PyTorch model from configuration: {config}""" )
UpperCAmelCase_ = TransfoXLLMHeadModel(__A )
UpperCAmelCase_ = load_tf_weights_in_transfo_xl(__A , __A , __A )
# Save pytorch-model
UpperCAmelCase_ = os.path.join(__A , __A )
UpperCAmelCase_ = os.path.join(__A , __A )
print(F"""Save PyTorch model to {os.path.abspath(__A )}""" )
torch.save(model.state_dict() , __A )
print(F"""Save configuration file to {os.path.abspath(__A )}""" )
with open(__A , '''w''' , encoding='''utf-8''' ) as f:
f.write(config.to_json_string() )
if __name__ == "__main__":
snake_case_ : List[str] = argparse.ArgumentParser()
parser.add_argument(
"--pytorch_dump_folder_path",
default=None,
type=str,
required=True,
help="Path to the folder to store the PyTorch model or dataset/vocab.",
)
parser.add_argument(
"--tf_checkpoint_path",
default="",
type=str,
help="An optional path to a TensorFlow checkpoint path to be converted.",
)
parser.add_argument(
"--transfo_xl_config_file",
default="",
type=str,
help=(
"An optional config json file corresponding to the pre-trained BERT model. \n"
"This specifies the model architecture."
),
)
parser.add_argument(
"--transfo_xl_dataset_file",
default="",
type=str,
help="An optional dataset file to be converted in a vocabulary.",
)
snake_case_ : int = parser.parse_args()
convert_transfo_xl_checkpoint_to_pytorch(
args.tf_checkpoint_path,
args.transfo_xl_config_file,
args.pytorch_dump_folder_path,
args.transfo_xl_dataset_file,
)
| 51
| 0
|
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
__lowerCamelCase = {
"configuration_deberta": ["DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP", "DebertaConfig", "DebertaOnnxConfig"],
"tokenization_deberta": ["DebertaTokenizer"],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__lowerCamelCase = ["DebertaTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__lowerCamelCase = [
"DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"DebertaForMaskedLM",
"DebertaForQuestionAnswering",
"DebertaForSequenceClassification",
"DebertaForTokenClassification",
"DebertaModel",
"DebertaPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__lowerCamelCase = [
"TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST",
"TFDebertaForMaskedLM",
"TFDebertaForQuestionAnswering",
"TFDebertaForSequenceClassification",
"TFDebertaForTokenClassification",
"TFDebertaModel",
"TFDebertaPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_deberta import DEBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP, DebertaConfig, DebertaOnnxConfig
from .tokenization_deberta import DebertaTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_deberta_fast import DebertaTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_deberta import (
DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
DebertaForMaskedLM,
DebertaForQuestionAnswering,
DebertaForSequenceClassification,
DebertaForTokenClassification,
DebertaModel,
DebertaPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_deberta import (
TF_DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
TFDebertaForMaskedLM,
TFDebertaForQuestionAnswering,
TFDebertaForSequenceClassification,
TFDebertaForTokenClassification,
TFDebertaModel,
TFDebertaPreTrainedModel,
)
else:
import sys
__lowerCamelCase = _LazyModule(__name__, globals()['''__file__'''], _import_structure, module_spec=__spec__)
| 162
|
from typing import Optional, Tuple, Union
import torch
from einops import rearrange, reduce
from diffusers import DDIMScheduler, DDPMScheduler, DiffusionPipeline, ImagePipelineOutput, UNetaDConditionModel
from diffusers.schedulers.scheduling_ddim import DDIMSchedulerOutput
from diffusers.schedulers.scheduling_ddpm import DDPMSchedulerOutput
snake_case_ : List[str] = 8
def A (__A : Union[str, Any] , __A : List[Any]=BITS ) -> Tuple:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x * 255).int().clamp(0 , 255 )
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b c h w -> b c 1 h w''' )
UpperCAmelCase_ = ((x & mask) != 0).float()
UpperCAmelCase_ = rearrange(__A , '''b c d h w -> b (c d) h w''' )
UpperCAmelCase_ = bits * 2 - 1
return bits
def A (__A : Dict , __A : Tuple=BITS ) -> List[str]:
"""simple docstring"""
UpperCAmelCase_ = x.device
UpperCAmelCase_ = (x > 0).int()
UpperCAmelCase_ = 2 ** torch.arange(bits - 1 , -1 , -1 , device=__A , dtype=torch.intaa )
UpperCAmelCase_ = rearrange(__A , '''d -> d 1 1''' )
UpperCAmelCase_ = rearrange(__A , '''b (c d) h w -> b c d h w''' , d=8 )
UpperCAmelCase_ = reduce(x * mask , '''b c d h w -> b c h w''' , '''sum''' )
return (dec / 255).clamp(0.0 , 1.0 )
def A (self : List[Any] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : float = 0.0 , __A : bool = True , __A : Tuple=None , __A : bool = True , ) -> Union[DDIMSchedulerOutput, Tuple]:
"""simple docstring"""
if self.num_inference_steps is None:
raise ValueError(
'''Number of inference steps is \'None\', you need to run \'set_timesteps\' after creating the scheduler''' )
# See formulas (12) and (16) of DDIM paper https://arxiv.org/pdf/2010.02502.pdf
# Ideally, read DDIM paper in-detail understanding
# Notation (<variable name> -> <name in paper>
# - pred_noise_t -> e_theta(x_t, t)
# - pred_original_sample -> f_theta(x_t, t) or x_0
# - std_dev_t -> sigma_t
# - eta -> η
# - pred_sample_direction -> "direction pointing to x_t"
# - pred_prev_sample -> "x_t-1"
# 1. get previous step value (=t-1)
UpperCAmelCase_ = timestep - self.config.num_train_timesteps // self.num_inference_steps
# 2. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[timestep]
UpperCAmelCase_ = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
UpperCAmelCase_ = 1 - alpha_prod_t
# 3. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
# 4. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 5. compute variance: "sigma_t(η)" -> see formula (16)
# σ_t = sqrt((1 − α_t−1)/(1 − α_t)) * sqrt(1 − α_t/α_t−1)
UpperCAmelCase_ = self._get_variance(__A , __A )
UpperCAmelCase_ = eta * variance ** 0.5
if use_clipped_model_output:
# the model_output is always re-derived from the clipped x_0 in Glide
UpperCAmelCase_ = (sample - alpha_prod_t ** 0.5 * pred_original_sample) / beta_prod_t ** 0.5
# 6. compute "direction pointing to x_t" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = (1 - alpha_prod_t_prev - std_dev_t**2) ** 0.5 * model_output
# 7. compute x_t without "random noise" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
UpperCAmelCase_ = alpha_prod_t_prev ** 0.5 * pred_original_sample + pred_sample_direction
if eta > 0:
# randn_like does not support generator https://github.com/pytorch/pytorch/issues/27072
UpperCAmelCase_ = model_output.device if torch.is_tensor(__A ) else '''cpu'''
UpperCAmelCase_ = torch.randn(model_output.shape , dtype=model_output.dtype , generator=__A ).to(__A )
UpperCAmelCase_ = self._get_variance(__A , __A ) ** 0.5 * eta * noise
UpperCAmelCase_ = prev_sample + variance
if not return_dict:
return (prev_sample,)
return DDIMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
def A (self : Optional[int] , __A : torch.FloatTensor , __A : int , __A : torch.FloatTensor , __A : int="epsilon" , __A : Optional[Any]=None , __A : bool = True , ) -> Union[DDPMSchedulerOutput, Tuple]:
"""simple docstring"""
UpperCAmelCase_ = timestep
if model_output.shape[1] == sample.shape[1] * 2 and self.variance_type in ["learned", "learned_range"]:
UpperCAmelCase_ , UpperCAmelCase_ = torch.split(__A , sample.shape[1] , dim=1 )
else:
UpperCAmelCase_ = None
# 1. compute alphas, betas
UpperCAmelCase_ = self.alphas_cumprod[t]
UpperCAmelCase_ = self.alphas_cumprod[t - 1] if t > 0 else self.one
UpperCAmelCase_ = 1 - alpha_prod_t
UpperCAmelCase_ = 1 - alpha_prod_t_prev
# 2. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
if prediction_type == "epsilon":
UpperCAmelCase_ = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
elif prediction_type == "sample":
UpperCAmelCase_ = model_output
else:
raise ValueError(F"""Unsupported prediction_type {prediction_type}.""" )
# 3. Clip "predicted x_0"
UpperCAmelCase_ = self.bit_scale
if self.config.clip_sample:
UpperCAmelCase_ = torch.clamp(__A , -scale , __A )
# 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = (alpha_prod_t_prev ** 0.5 * self.betas[t]) / beta_prod_t
UpperCAmelCase_ = self.alphas[t] ** 0.5 * beta_prod_t_prev / beta_prod_t
# 5. Compute predicted previous sample µ_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
UpperCAmelCase_ = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * sample
# 6. Add noise
UpperCAmelCase_ = 0
if t > 0:
UpperCAmelCase_ = torch.randn(
model_output.size() , dtype=model_output.dtype , layout=model_output.layout , generator=__A ).to(model_output.device )
UpperCAmelCase_ = (self._get_variance(__A , predicted_variance=__A ) ** 0.5) * noise
UpperCAmelCase_ = pred_prev_sample + variance
if not return_dict:
return (pred_prev_sample,)
return DDPMSchedulerOutput(prev_sample=__A , pred_original_sample=__A )
class __snake_case ( a ):
def __init__( self : Union[str, Any] , _snake_case : UNetaDConditionModel , _snake_case : Union[DDIMScheduler, DDPMScheduler] , _snake_case : Optional[float] = 1.0 , ):
"""simple docstring"""
super().__init__()
UpperCAmelCase_ = bit_scale
UpperCAmelCase_ = (
ddim_bit_scheduler_step if isinstance(_snake_case , _snake_case) else ddpm_bit_scheduler_step
)
self.register_modules(unet=_snake_case , scheduler=_snake_case)
@torch.no_grad()
def __call__( self : Union[str, Any] , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 256 , _snake_case : Optional[int] = 50 , _snake_case : Optional[torch.Generator] = None , _snake_case : Optional[int] = 1 , _snake_case : Optional[str] = "pil" , _snake_case : bool = True , **_snake_case : Optional[Any] , ):
"""simple docstring"""
UpperCAmelCase_ = torch.randn(
(batch_size, self.unet.config.in_channels, height, width) , generator=_snake_case , )
UpperCAmelCase_ = decimal_to_bits(_snake_case) * self.bit_scale
UpperCAmelCase_ = latents.to(self.device)
self.scheduler.set_timesteps(_snake_case)
for t in self.progress_bar(self.scheduler.timesteps):
# predict the noise residual
UpperCAmelCase_ = self.unet(_snake_case , _snake_case).sample
# compute the previous noisy sample x_t -> x_t-1
UpperCAmelCase_ = self.scheduler.step(_snake_case , _snake_case , _snake_case).prev_sample
UpperCAmelCase_ = bits_to_decimal(_snake_case)
if output_type == "pil":
UpperCAmelCase_ = self.numpy_to_pil(_snake_case)
if not return_dict:
return (image,)
return ImagePipelineOutput(images=_snake_case)
| 51
| 0
|
'''simple docstring'''
import pytest
import datasets
# Import fixture modules as plugins
__lowercase : Tuple = ["tests.fixtures.files", "tests.fixtures.hub", "tests.fixtures.fsspec"]
def lowercase_ ( _lowercase , _lowercase ) -> Optional[int]:
'''simple docstring'''
for item in items:
if any(marker in item.keywords for marker in ['''integration''', '''unit'''] ):
continue
item.add_marker(pytest.mark.unit )
def lowercase_ ( _lowercase ) -> Optional[int]:
'''simple docstring'''
config.addinivalue_line('''markers''' , '''torchaudio_latest: mark test to run with torchaudio>=0.12''' )
@pytest.fixture(autouse=__A )
def lowercase_ ( _lowercase , _lowercase ) -> List[Any]:
'''simple docstring'''
lowerCamelCase_ : str = tmp_path_factory.getbasetemp() / '''cache'''
lowerCamelCase_ : Dict = test_hf_cache_home / '''datasets'''
lowerCamelCase_ : Tuple = test_hf_cache_home / '''metrics'''
lowerCamelCase_ : Union[str, Any] = test_hf_cache_home / '''modules'''
monkeypatch.setattr('''datasets.config.HF_DATASETS_CACHE''' , str(__A ) )
monkeypatch.setattr('''datasets.config.HF_METRICS_CACHE''' , str(__A ) )
monkeypatch.setattr('''datasets.config.HF_MODULES_CACHE''' , str(__A ) )
lowerCamelCase_ : str = test_hf_datasets_cache / '''downloads'''
monkeypatch.setattr('''datasets.config.DOWNLOADED_DATASETS_PATH''' , str(__A ) )
lowerCamelCase_ : str = test_hf_datasets_cache / '''downloads''' / '''extracted'''
monkeypatch.setattr('''datasets.config.EXTRACTED_DATASETS_PATH''' , str(__A ) )
@pytest.fixture(autouse=__A , scope='''session''' )
def lowercase_ ( ) -> Dict:
'''simple docstring'''
datasets.disable_progress_bar()
@pytest.fixture(autouse=__A )
def lowercase_ ( _lowercase ) -> str:
'''simple docstring'''
monkeypatch.setattr('''datasets.config.HF_UPDATE_DOWNLOAD_COUNTS''' , __A )
@pytest.fixture
def lowercase_ ( _lowercase ) -> int:
'''simple docstring'''
monkeypatch.setattr('''sqlalchemy.util.deprecations.SILENCE_UBER_WARNING''' , __A )
| 318
|
snake_case_ : Dict = {
"Pillow": "Pillow",
"accelerate": "accelerate>=0.11.0",
"compel": "compel==0.1.8",
"black": "black~=23.1",
"datasets": "datasets",
"filelock": "filelock",
"flax": "flax>=0.4.1",
"hf-doc-builder": "hf-doc-builder>=0.3.0",
"huggingface-hub": "huggingface-hub>=0.13.2",
"requests-mock": "requests-mock==1.10.0",
"importlib_metadata": "importlib_metadata",
"invisible-watermark": "invisible-watermark",
"isort": "isort>=5.5.4",
"jax": "jax>=0.2.8,!=0.3.2",
"jaxlib": "jaxlib>=0.1.65",
"Jinja2": "Jinja2",
"k-diffusion": "k-diffusion>=0.0.12",
"torchsde": "torchsde",
"note_seq": "note_seq",
"librosa": "librosa",
"numpy": "numpy",
"omegaconf": "omegaconf",
"parameterized": "parameterized",
"protobuf": "protobuf>=3.20.3,<4",
"pytest": "pytest",
"pytest-timeout": "pytest-timeout",
"pytest-xdist": "pytest-xdist",
"ruff": "ruff>=0.0.241",
"safetensors": "safetensors",
"sentencepiece": "sentencepiece>=0.1.91,!=0.1.92",
"scipy": "scipy",
"onnx": "onnx",
"regex": "regex!=2019.12.17",
"requests": "requests",
"tensorboard": "tensorboard",
"torch": "torch>=1.4",
"torchvision": "torchvision",
"transformers": "transformers>=4.25.1",
"urllib3": "urllib3<=2.0.0",
}
| 51
| 0
|
import warnings
from ..trainer import Trainer
from ..utils import logging
__snake_case = logging.get_logger(__name__)
class __lowerCamelCase (_a ):
def __init__( self: Optional[Any],A_: List[str]=None,**A_: Any ):
'''simple docstring'''
warnings.warn(
'`SageMakerTrainer` is deprecated and will be removed in v5 of Transformers. You can use `Trainer` '
'instead.',_snake_case,)
super().__init__(args=_snake_case,**_snake_case )
| 310
|
from datetime import datetime
import requests
def A (__A : str ) -> bytes:
"""simple docstring"""
UpperCAmelCase_ = '''https://downloadgram.net/wp-json/wppress/video-downloader/video?url='''
UpperCAmelCase_ = requests.get(base_url + url ).json()[0]['''urls'''][0]['''src''']
return requests.get(__A ).content
if __name__ == "__main__":
snake_case_ : Optional[Any] = input("Enter Video/IGTV url: ").strip()
snake_case_ : Any = f"{datetime.now():%Y-%m-%d_%H:%M:%S}.mp4"
with open(file_name, "wb") as fp:
fp.write(download_video(url))
print(f"Done. Video saved to disk as {file_name}.")
| 51
| 0
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.