diff --git a/.gitattributes b/.gitattributes
index b636f4b2a8bce63783ad05cf51f67abf223c5a3d..349f6eddbc104e454d568468f24847f626693bca 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -547,3 +547,5 @@ falcon/lib/python3.10/site-packages/sympy/core/__pycache__/numbers.cpython-310.p
falcon/lib/python3.10/site-packages/sympy/core/__pycache__/function.cpython-310.pyc filter=lfs diff=lfs merge=lfs -text
falcon/lib/python3.10/site-packages/sympy/core/__pycache__/expr.cpython-310.pyc filter=lfs diff=lfs merge=lfs -text
falcon/lib/python3.10/site-packages/sympy/printing/tests/__pycache__/test_latex.cpython-310.pyc filter=lfs diff=lfs merge=lfs -text
+falcon/lib/python3.10/site-packages/sympy/solvers/tests/__pycache__/test_solveset.cpython-310.pyc filter=lfs diff=lfs merge=lfs -text
+falcon/lib/python3.10/site-packages/sympy/solvers/tests/__pycache__/test_solvers.cpython-310.pyc filter=lfs diff=lfs merge=lfs -text
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__init__.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..9882fc2b9733554026cacebece8637f25002f985
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__init__.py
@@ -0,0 +1,142 @@
+# Copyright 2020 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from typing import TYPE_CHECKING
+
+from ...utils import (
+ OptionalDependencyNotAvailable,
+ _LazyModule,
+ is_sentencepiece_available,
+ is_tf_available,
+ is_tokenizers_available,
+ is_torch_available,
+)
+
+
+_import_structure = {
+ "configuration_camembert": ["CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "CamembertConfig", "CamembertOnnxConfig"],
+}
+
+try:
+ if not is_sentencepiece_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["tokenization_camembert"] = ["CamembertTokenizer"]
+
+try:
+ if not is_tokenizers_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["tokenization_camembert_fast"] = ["CamembertTokenizerFast"]
+
+try:
+ if not is_torch_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_camembert"] = [
+ "CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_LIST",
+ "CamembertForCausalLM",
+ "CamembertForMaskedLM",
+ "CamembertForMultipleChoice",
+ "CamembertForQuestionAnswering",
+ "CamembertForSequenceClassification",
+ "CamembertForTokenClassification",
+ "CamembertModel",
+ "CamembertPreTrainedModel",
+ ]
+
+try:
+ if not is_tf_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_tf_camembert"] = [
+ "TF_CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_LIST",
+ "TFCamembertForCausalLM",
+ "TFCamembertForMaskedLM",
+ "TFCamembertForMultipleChoice",
+ "TFCamembertForQuestionAnswering",
+ "TFCamembertForSequenceClassification",
+ "TFCamembertForTokenClassification",
+ "TFCamembertModel",
+ "TFCamembertPreTrainedModel",
+ ]
+
+
+if TYPE_CHECKING:
+ from .configuration_camembert import CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, CamembertConfig, CamembertOnnxConfig
+
+ try:
+ if not is_sentencepiece_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .tokenization_camembert import CamembertTokenizer
+
+ try:
+ if not is_tokenizers_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .tokenization_camembert_fast import CamembertTokenizerFast
+
+ try:
+ if not is_torch_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_camembert import (
+ CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
+ CamembertForCausalLM,
+ CamembertForMaskedLM,
+ CamembertForMultipleChoice,
+ CamembertForQuestionAnswering,
+ CamembertForSequenceClassification,
+ CamembertForTokenClassification,
+ CamembertModel,
+ CamembertPreTrainedModel,
+ )
+
+ try:
+ if not is_tf_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_tf_camembert import (
+ TF_CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
+ TFCamembertForCausalLM,
+ TFCamembertForMaskedLM,
+ TFCamembertForMultipleChoice,
+ TFCamembertForQuestionAnswering,
+ TFCamembertForSequenceClassification,
+ TFCamembertForTokenClassification,
+ TFCamembertModel,
+ TFCamembertPreTrainedModel,
+ )
+
+else:
+ import sys
+
+ sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__pycache__/modeling_camembert.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__pycache__/modeling_camembert.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..23fe4d0e002107b342f0ce7c520a5af93108e9d2
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__pycache__/modeling_camembert.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__pycache__/modeling_tf_camembert.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__pycache__/modeling_tf_camembert.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..5e1cd498d75a52aa0380e78bbdb5f5d43ea3b2b3
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/__pycache__/modeling_tf_camembert.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/configuration_camembert.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/configuration_camembert.py
new file mode 100644
index 0000000000000000000000000000000000000000..d712726492ae18aac88e7941ab17fbc74322e6d8
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/configuration_camembert.py
@@ -0,0 +1,162 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" CamemBERT configuration"""
+
+from collections import OrderedDict
+from typing import Mapping
+
+from ...configuration_utils import PretrainedConfig
+from ...onnx import OnnxConfig
+from ...utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+ "camembert-base": "https://huggingface.co/camembert-base/resolve/main/config.json",
+ "umberto-commoncrawl-cased-v1": (
+ "https://huggingface.co/Musixmatch/umberto-commoncrawl-cased-v1/resolve/main/config.json"
+ ),
+ "umberto-wikipedia-uncased-v1": (
+ "https://huggingface.co/Musixmatch/umberto-wikipedia-uncased-v1/resolve/main/config.json"
+ ),
+}
+
+
+class CamembertConfig(PretrainedConfig):
+ """
+ This is the configuration class to store the configuration of a [`CamembertModel`] or a [`TFCamembertModel`]. It is
+ used to instantiate a Camembert model according to the specified arguments, defining the model architecture.
+ Instantiating a configuration with the defaults will yield a similar configuration to that of the Camembert
+ [camembert-base](https://huggingface.co/camembert-base) architecture.
+
+ Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+ documentation from [`PretrainedConfig`] for more information.
+
+
+ Args:
+ vocab_size (`int`, *optional*, defaults to 30522):
+ Vocabulary size of the BERT model. Defines the number of different tokens that can be represented by the
+ `inputs_ids` passed when calling [`CamembertModel`] or [`TFCamembertModel`].
+ hidden_size (`int`, *optional*, defaults to 768):
+ Dimensionality of the encoder layers and the pooler layer.
+ num_hidden_layers (`int`, *optional*, defaults to 12):
+ Number of hidden layers in the Transformer encoder.
+ num_attention_heads (`int`, *optional*, defaults to 12):
+ Number of attention heads for each attention layer in the Transformer encoder.
+ intermediate_size (`int`, *optional*, defaults to 3072):
+ Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder.
+ hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`):
+ The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
+ `"relu"`, `"silu"` and `"gelu_new"` are supported.
+ hidden_dropout_prob (`float`, *optional*, defaults to 0.1):
+ The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
+ attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1):
+ The dropout ratio for the attention probabilities.
+ max_position_embeddings (`int`, *optional*, defaults to 512):
+ The maximum sequence length that this model might ever be used with. Typically set this to something large
+ just in case (e.g., 512 or 1024 or 2048).
+ type_vocab_size (`int`, *optional*, defaults to 2):
+ The vocabulary size of the `token_type_ids` passed when calling [`CamembertModel`] or [`TFCamembertModel`].
+ initializer_range (`float`, *optional*, defaults to 0.02):
+ The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+ layer_norm_eps (`float`, *optional*, defaults to 1e-12):
+ The epsilon used by the layer normalization layers.
+ position_embedding_type (`str`, *optional*, defaults to `"absolute"`):
+ Type of position embedding. Choose one of `"absolute"`, `"relative_key"`, `"relative_key_query"`. For
+ positional embeddings use `"absolute"`. For more information on `"relative_key"`, please refer to
+ [Self-Attention with Relative Position Representations (Shaw et al.)](https://arxiv.org/abs/1803.02155).
+ For more information on `"relative_key_query"`, please refer to *Method 4* in [Improve Transformer Models
+ with Better Relative Position Embeddings (Huang et al.)](https://arxiv.org/abs/2009.13658).
+ is_decoder (`bool`, *optional*, defaults to `False`):
+ Whether the model is used as a decoder or not. If `False`, the model is used as an encoder.
+ use_cache (`bool`, *optional*, defaults to `True`):
+ Whether or not the model should return the last key/values attentions (not used by all models). Only
+ relevant if `config.is_decoder=True`.
+ classifier_dropout (`float`, *optional*):
+ The dropout ratio for the classification head.
+
+ Example:
+
+ ```python
+ >>> from transformers import CamembertConfig, CamembertModel
+
+ >>> # Initializing a Camembert camembert-base style configuration
+ >>> configuration = CamembertConfig()
+
+ >>> # Initializing a model (with random weights) from the camembert-base style configuration
+ >>> model = CamembertModel(configuration)
+
+ >>> # Accessing the model configuration
+ >>> configuration = model.config
+ ```"""
+
+ model_type = "camembert"
+
+ def __init__(
+ self,
+ vocab_size=30522,
+ hidden_size=768,
+ num_hidden_layers=12,
+ num_attention_heads=12,
+ intermediate_size=3072,
+ hidden_act="gelu",
+ hidden_dropout_prob=0.1,
+ attention_probs_dropout_prob=0.1,
+ max_position_embeddings=512,
+ type_vocab_size=2,
+ initializer_range=0.02,
+ layer_norm_eps=1e-12,
+ pad_token_id=1,
+ bos_token_id=0,
+ eos_token_id=2,
+ position_embedding_type="absolute",
+ use_cache=True,
+ classifier_dropout=None,
+ **kwargs,
+ ):
+ super().__init__(pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs)
+
+ self.vocab_size = vocab_size
+ self.hidden_size = hidden_size
+ self.num_hidden_layers = num_hidden_layers
+ self.num_attention_heads = num_attention_heads
+ self.hidden_act = hidden_act
+ self.intermediate_size = intermediate_size
+ self.hidden_dropout_prob = hidden_dropout_prob
+ self.attention_probs_dropout_prob = attention_probs_dropout_prob
+ self.max_position_embeddings = max_position_embeddings
+ self.type_vocab_size = type_vocab_size
+ self.initializer_range = initializer_range
+ self.layer_norm_eps = layer_norm_eps
+ self.position_embedding_type = position_embedding_type
+ self.use_cache = use_cache
+ self.classifier_dropout = classifier_dropout
+
+
+class CamembertOnnxConfig(OnnxConfig):
+ @property
+ def inputs(self) -> Mapping[str, Mapping[int, str]]:
+ if self.task == "multiple-choice":
+ dynamic_axis = {0: "batch", 1: "choice", 2: "sequence"}
+ else:
+ dynamic_axis = {0: "batch", 1: "sequence"}
+ return OrderedDict(
+ [
+ ("input_ids", dynamic_axis),
+ ("attention_mask", dynamic_axis),
+ ]
+ )
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/modeling_camembert.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/modeling_camembert.py
new file mode 100644
index 0000000000000000000000000000000000000000..50fac0efd000a17ed3f97883787b3b522dcc0c68
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/modeling_camembert.py
@@ -0,0 +1,1574 @@
+# coding=utf-8
+# Copyright 2019 Inria, Facebook AI Research and the HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""PyTorch CamemBERT model."""
+
+import math
+from typing import List, Optional, Tuple, Union
+
+import torch
+import torch.utils.checkpoint
+from torch import nn
+from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
+
+from ...activations import ACT2FN, gelu
+from ...modeling_outputs import (
+ BaseModelOutputWithPastAndCrossAttentions,
+ BaseModelOutputWithPoolingAndCrossAttentions,
+ CausalLMOutputWithCrossAttentions,
+ MaskedLMOutput,
+ MultipleChoiceModelOutput,
+ QuestionAnsweringModelOutput,
+ SequenceClassifierOutput,
+ TokenClassifierOutput,
+)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
+from ...utils import (
+ add_code_sample_docstrings,
+ add_start_docstrings,
+ add_start_docstrings_to_model_forward,
+ logging,
+ replace_return_docstrings,
+)
+from .configuration_camembert import CamembertConfig
+
+
+logger = logging.get_logger(__name__)
+
+_CHECKPOINT_FOR_DOC = "camembert-base"
+_CONFIG_FOR_DOC = "CamembertConfig"
+
+CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_LIST = [
+ "camembert-base",
+ "Musixmatch/umberto-commoncrawl-cased-v1",
+ "Musixmatch/umberto-wikipedia-uncased-v1",
+ # See all CamemBERT models at https://huggingface.co/models?filter=camembert
+]
+
+CAMEMBERT_START_DOCSTRING = r"""
+
+ This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
+ library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
+ etc.)
+
+ This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
+ Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
+ and behavior.
+
+ Parameters:
+ config ([`CamembertConfig`]): Model configuration class with all the parameters of the
+ model. Initializing with a config file does not load the weights associated with the model, only the
+ configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+
+# Copied from transformers.models.roberta.modeling_roberta.RobertaEmbeddings with Roberta->Camembert
+class CamembertEmbeddings(nn.Module):
+ """
+ Same as BertEmbeddings with a tiny tweak for positional embeddings indexing.
+ """
+
+ # Copied from transformers.models.bert.modeling_bert.BertEmbeddings.__init__
+ def __init__(self, config):
+ super().__init__()
+ self.word_embeddings = nn.Embedding(config.vocab_size, config.hidden_size, padding_idx=config.pad_token_id)
+ self.position_embeddings = nn.Embedding(config.max_position_embeddings, config.hidden_size)
+ self.token_type_embeddings = nn.Embedding(config.type_vocab_size, config.hidden_size)
+
+ # self.LayerNorm is not snake-cased to stick with TensorFlow model variable name and be able to load
+ # any TensorFlow checkpoint file
+ self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+ self.dropout = nn.Dropout(config.hidden_dropout_prob)
+ # position_ids (1, len position emb) is contiguous in memory and exported when serialized
+ self.position_embedding_type = getattr(config, "position_embedding_type", "absolute")
+ self.register_buffer(
+ "position_ids", torch.arange(config.max_position_embeddings).expand((1, -1)), persistent=False
+ )
+ self.register_buffer(
+ "token_type_ids", torch.zeros(self.position_ids.size(), dtype=torch.long), persistent=False
+ )
+
+ # End copy
+ self.padding_idx = config.pad_token_id
+ self.position_embeddings = nn.Embedding(
+ config.max_position_embeddings, config.hidden_size, padding_idx=self.padding_idx
+ )
+
+ def forward(
+ self, input_ids=None, token_type_ids=None, position_ids=None, inputs_embeds=None, past_key_values_length=0
+ ):
+ if position_ids is None:
+ if input_ids is not None:
+ # Create the position ids from the input token ids. Any padded tokens remain padded.
+ position_ids = create_position_ids_from_input_ids(input_ids, self.padding_idx, past_key_values_length)
+ else:
+ position_ids = self.create_position_ids_from_inputs_embeds(inputs_embeds)
+
+ if input_ids is not None:
+ input_shape = input_ids.size()
+ else:
+ input_shape = inputs_embeds.size()[:-1]
+
+ seq_length = input_shape[1]
+
+ # Setting the token_type_ids to the registered buffer in constructor where it is all zeros, which usually occurs
+ # when its auto-generated, registered buffer helps users when tracing the model without passing token_type_ids, solves
+ # issue #5664
+ if token_type_ids is None:
+ if hasattr(self, "token_type_ids"):
+ buffered_token_type_ids = self.token_type_ids[:, :seq_length]
+ buffered_token_type_ids_expanded = buffered_token_type_ids.expand(input_shape[0], seq_length)
+ token_type_ids = buffered_token_type_ids_expanded
+ else:
+ token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=self.position_ids.device)
+
+ if inputs_embeds is None:
+ inputs_embeds = self.word_embeddings(input_ids)
+ token_type_embeddings = self.token_type_embeddings(token_type_ids)
+
+ embeddings = inputs_embeds + token_type_embeddings
+ if self.position_embedding_type == "absolute":
+ position_embeddings = self.position_embeddings(position_ids)
+ embeddings += position_embeddings
+ embeddings = self.LayerNorm(embeddings)
+ embeddings = self.dropout(embeddings)
+ return embeddings
+
+ def create_position_ids_from_inputs_embeds(self, inputs_embeds):
+ """
+ We are provided embeddings directly. We cannot infer which are padded so just generate sequential position ids.
+
+ Args:
+ inputs_embeds: torch.Tensor
+
+ Returns: torch.Tensor
+ """
+ input_shape = inputs_embeds.size()[:-1]
+ sequence_length = input_shape[1]
+
+ position_ids = torch.arange(
+ self.padding_idx + 1, sequence_length + self.padding_idx + 1, dtype=torch.long, device=inputs_embeds.device
+ )
+ return position_ids.unsqueeze(0).expand(input_shape)
+
+
+# Copied from transformers.models.roberta.modeling_roberta.RobertaSelfAttention with Roberta->Camembert
+class CamembertSelfAttention(nn.Module):
+ def __init__(self, config, position_embedding_type=None):
+ super().__init__()
+ if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
+ raise ValueError(
+ f"The hidden size ({config.hidden_size}) is not a multiple of the number of attention "
+ f"heads ({config.num_attention_heads})"
+ )
+
+ self.num_attention_heads = config.num_attention_heads
+ self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
+ self.all_head_size = self.num_attention_heads * self.attention_head_size
+
+ self.query = nn.Linear(config.hidden_size, self.all_head_size)
+ self.key = nn.Linear(config.hidden_size, self.all_head_size)
+ self.value = nn.Linear(config.hidden_size, self.all_head_size)
+
+ self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
+ self.position_embedding_type = position_embedding_type or getattr(
+ config, "position_embedding_type", "absolute"
+ )
+ if self.position_embedding_type == "relative_key" or self.position_embedding_type == "relative_key_query":
+ self.max_position_embeddings = config.max_position_embeddings
+ self.distance_embedding = nn.Embedding(2 * config.max_position_embeddings - 1, self.attention_head_size)
+
+ self.is_decoder = config.is_decoder
+
+ def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
+ new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
+ x = x.view(new_x_shape)
+ return x.permute(0, 2, 1, 3)
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ encoder_hidden_states: Optional[torch.FloatTensor] = None,
+ encoder_attention_mask: Optional[torch.FloatTensor] = None,
+ past_key_value: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
+ output_attentions: Optional[bool] = False,
+ ) -> Tuple[torch.Tensor]:
+ mixed_query_layer = self.query(hidden_states)
+
+ # If this is instantiated as a cross-attention module, the keys
+ # and values come from an encoder; the attention mask needs to be
+ # such that the encoder's padding tokens are not attended to.
+ is_cross_attention = encoder_hidden_states is not None
+
+ if is_cross_attention and past_key_value is not None:
+ # reuse k,v, cross_attentions
+ key_layer = past_key_value[0]
+ value_layer = past_key_value[1]
+ attention_mask = encoder_attention_mask
+ elif is_cross_attention:
+ key_layer = self.transpose_for_scores(self.key(encoder_hidden_states))
+ value_layer = self.transpose_for_scores(self.value(encoder_hidden_states))
+ attention_mask = encoder_attention_mask
+ elif past_key_value is not None:
+ key_layer = self.transpose_for_scores(self.key(hidden_states))
+ value_layer = self.transpose_for_scores(self.value(hidden_states))
+ key_layer = torch.cat([past_key_value[0], key_layer], dim=2)
+ value_layer = torch.cat([past_key_value[1], value_layer], dim=2)
+ else:
+ key_layer = self.transpose_for_scores(self.key(hidden_states))
+ value_layer = self.transpose_for_scores(self.value(hidden_states))
+
+ query_layer = self.transpose_for_scores(mixed_query_layer)
+
+ use_cache = past_key_value is not None
+ if self.is_decoder:
+ # if cross_attention save Tuple(torch.Tensor, torch.Tensor) of all cross attention key/value_states.
+ # Further calls to cross_attention layer can then reuse all cross-attention
+ # key/value_states (first "if" case)
+ # if uni-directional self-attention (decoder) save Tuple(torch.Tensor, torch.Tensor) of
+ # all previous decoder key/value_states. Further calls to uni-directional self-attention
+ # can concat previous decoder key/value_states to current projected key/value_states (third "elif" case)
+ # if encoder bi-directional self-attention `past_key_value` is always `None`
+ past_key_value = (key_layer, value_layer)
+
+ # Take the dot product between "query" and "key" to get the raw attention scores.
+ attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2))
+
+ if self.position_embedding_type == "relative_key" or self.position_embedding_type == "relative_key_query":
+ query_length, key_length = query_layer.shape[2], key_layer.shape[2]
+ if use_cache:
+ position_ids_l = torch.tensor(key_length - 1, dtype=torch.long, device=hidden_states.device).view(
+ -1, 1
+ )
+ else:
+ position_ids_l = torch.arange(query_length, dtype=torch.long, device=hidden_states.device).view(-1, 1)
+ position_ids_r = torch.arange(key_length, dtype=torch.long, device=hidden_states.device).view(1, -1)
+ distance = position_ids_l - position_ids_r
+
+ positional_embedding = self.distance_embedding(distance + self.max_position_embeddings - 1)
+ positional_embedding = positional_embedding.to(dtype=query_layer.dtype) # fp16 compatibility
+
+ if self.position_embedding_type == "relative_key":
+ relative_position_scores = torch.einsum("bhld,lrd->bhlr", query_layer, positional_embedding)
+ attention_scores = attention_scores + relative_position_scores
+ elif self.position_embedding_type == "relative_key_query":
+ relative_position_scores_query = torch.einsum("bhld,lrd->bhlr", query_layer, positional_embedding)
+ relative_position_scores_key = torch.einsum("bhrd,lrd->bhlr", key_layer, positional_embedding)
+ attention_scores = attention_scores + relative_position_scores_query + relative_position_scores_key
+
+ attention_scores = attention_scores / math.sqrt(self.attention_head_size)
+ if attention_mask is not None:
+ # Apply the attention mask is (precomputed for all layers in CamembertModel forward() function)
+ attention_scores = attention_scores + attention_mask
+
+ # Normalize the attention scores to probabilities.
+ attention_probs = nn.functional.softmax(attention_scores, dim=-1)
+
+ # This is actually dropping out entire tokens to attend to, which might
+ # seem a bit unusual, but is taken from the original Transformer paper.
+ attention_probs = self.dropout(attention_probs)
+
+ # Mask heads if we want to
+ if head_mask is not None:
+ attention_probs = attention_probs * head_mask
+
+ context_layer = torch.matmul(attention_probs, value_layer)
+
+ context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
+ new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
+ context_layer = context_layer.view(new_context_layer_shape)
+
+ outputs = (context_layer, attention_probs) if output_attentions else (context_layer,)
+
+ if self.is_decoder:
+ outputs = outputs + (past_key_value,)
+ return outputs
+
+
+# Copied from transformers.models.roberta.modeling_roberta.RobertaSelfOutput with Roberta->Camembert
+class CamembertSelfOutput(nn.Module):
+ def __init__(self, config):
+ super().__init__()
+ self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+ self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+ self.dropout = nn.Dropout(config.hidden_dropout_prob)
+
+ def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Tensor) -> torch.Tensor:
+ hidden_states = self.dense(hidden_states)
+ hidden_states = self.dropout(hidden_states)
+ hidden_states = self.LayerNorm(hidden_states + input_tensor)
+ return hidden_states
+
+
+# Copied from transformers.models.roberta.modeling_roberta.RobertaAttention with Roberta->Camembert
+class CamembertAttention(nn.Module):
+ def __init__(self, config, position_embedding_type=None):
+ super().__init__()
+ self.self = CamembertSelfAttention(config, position_embedding_type=position_embedding_type)
+ self.output = CamembertSelfOutput(config)
+ self.pruned_heads = set()
+
+ def prune_heads(self, heads):
+ if len(heads) == 0:
+ return
+ heads, index = find_pruneable_heads_and_indices(
+ heads, self.self.num_attention_heads, self.self.attention_head_size, self.pruned_heads
+ )
+
+ # Prune linear layers
+ self.self.query = prune_linear_layer(self.self.query, index)
+ self.self.key = prune_linear_layer(self.self.key, index)
+ self.self.value = prune_linear_layer(self.self.value, index)
+ self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
+
+ # Update hyper params and store pruned heads
+ self.self.num_attention_heads = self.self.num_attention_heads - len(heads)
+ self.self.all_head_size = self.self.attention_head_size * self.self.num_attention_heads
+ self.pruned_heads = self.pruned_heads.union(heads)
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ encoder_hidden_states: Optional[torch.FloatTensor] = None,
+ encoder_attention_mask: Optional[torch.FloatTensor] = None,
+ past_key_value: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
+ output_attentions: Optional[bool] = False,
+ ) -> Tuple[torch.Tensor]:
+ self_outputs = self.self(
+ hidden_states,
+ attention_mask,
+ head_mask,
+ encoder_hidden_states,
+ encoder_attention_mask,
+ past_key_value,
+ output_attentions,
+ )
+ attention_output = self.output(self_outputs[0], hidden_states)
+ outputs = (attention_output,) + self_outputs[1:] # add attentions if we output them
+ return outputs
+
+
+# Copied from transformers.models.bert.modeling_bert.BertIntermediate with Bert->Roberta->Camembert
+class CamembertIntermediate(nn.Module):
+ def __init__(self, config):
+ super().__init__()
+ self.dense = nn.Linear(config.hidden_size, config.intermediate_size)
+ if isinstance(config.hidden_act, str):
+ self.intermediate_act_fn = ACT2FN[config.hidden_act]
+ else:
+ self.intermediate_act_fn = config.hidden_act
+
+ def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+ hidden_states = self.dense(hidden_states)
+ hidden_states = self.intermediate_act_fn(hidden_states)
+ return hidden_states
+
+
+# Copied from transformers.models.bert.modeling_bert.BertOutput with Bert->Roberta->Camembert
+class CamembertOutput(nn.Module):
+ def __init__(self, config):
+ super().__init__()
+ self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
+ self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+ self.dropout = nn.Dropout(config.hidden_dropout_prob)
+
+ def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Tensor) -> torch.Tensor:
+ hidden_states = self.dense(hidden_states)
+ hidden_states = self.dropout(hidden_states)
+ hidden_states = self.LayerNorm(hidden_states + input_tensor)
+ return hidden_states
+
+
+# Copied from transformers.models.roberta.modeling_roberta.RobertaLayer with Roberta->Camembert
+class CamembertLayer(nn.Module):
+ def __init__(self, config):
+ super().__init__()
+ self.chunk_size_feed_forward = config.chunk_size_feed_forward
+ self.seq_len_dim = 1
+ self.attention = CamembertAttention(config)
+ self.is_decoder = config.is_decoder
+ self.add_cross_attention = config.add_cross_attention
+ if self.add_cross_attention:
+ if not self.is_decoder:
+ raise ValueError(f"{self} should be used as a decoder model if cross attention is added")
+ self.crossattention = CamembertAttention(config, position_embedding_type="absolute")
+ self.intermediate = CamembertIntermediate(config)
+ self.output = CamembertOutput(config)
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ encoder_hidden_states: Optional[torch.FloatTensor] = None,
+ encoder_attention_mask: Optional[torch.FloatTensor] = None,
+ past_key_value: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
+ output_attentions: Optional[bool] = False,
+ ) -> Tuple[torch.Tensor]:
+ # decoder uni-directional self-attention cached key/values tuple is at positions 1,2
+ self_attn_past_key_value = past_key_value[:2] if past_key_value is not None else None
+ self_attention_outputs = self.attention(
+ hidden_states,
+ attention_mask,
+ head_mask,
+ output_attentions=output_attentions,
+ past_key_value=self_attn_past_key_value,
+ )
+ attention_output = self_attention_outputs[0]
+
+ # if decoder, the last output is tuple of self-attn cache
+ if self.is_decoder:
+ outputs = self_attention_outputs[1:-1]
+ present_key_value = self_attention_outputs[-1]
+ else:
+ outputs = self_attention_outputs[1:] # add self attentions if we output attention weights
+
+ cross_attn_present_key_value = None
+ if self.is_decoder and encoder_hidden_states is not None:
+ if not hasattr(self, "crossattention"):
+ raise ValueError(
+ f"If `encoder_hidden_states` are passed, {self} has to be instantiated with cross-attention layers"
+ " by setting `config.add_cross_attention=True`"
+ )
+
+ # cross_attn cached key/values tuple is at positions 3,4 of past_key_value tuple
+ cross_attn_past_key_value = past_key_value[-2:] if past_key_value is not None else None
+ cross_attention_outputs = self.crossattention(
+ attention_output,
+ attention_mask,
+ head_mask,
+ encoder_hidden_states,
+ encoder_attention_mask,
+ cross_attn_past_key_value,
+ output_attentions,
+ )
+ attention_output = cross_attention_outputs[0]
+ outputs = outputs + cross_attention_outputs[1:-1] # add cross attentions if we output attention weights
+
+ # add cross-attn cache to positions 3,4 of present_key_value tuple
+ cross_attn_present_key_value = cross_attention_outputs[-1]
+ present_key_value = present_key_value + cross_attn_present_key_value
+
+ layer_output = apply_chunking_to_forward(
+ self.feed_forward_chunk, self.chunk_size_feed_forward, self.seq_len_dim, attention_output
+ )
+ outputs = (layer_output,) + outputs
+
+ # if decoder, return the attn key/values as the last output
+ if self.is_decoder:
+ outputs = outputs + (present_key_value,)
+
+ return outputs
+
+ def feed_forward_chunk(self, attention_output):
+ intermediate_output = self.intermediate(attention_output)
+ layer_output = self.output(intermediate_output, attention_output)
+ return layer_output
+
+
+# Copied from transformers.models.roberta.modeling_roberta.RobertaEncoder with Roberta->Camembert
+class CamembertEncoder(nn.Module):
+ def __init__(self, config):
+ super().__init__()
+ self.config = config
+ self.layer = nn.ModuleList([CamembertLayer(config) for _ in range(config.num_hidden_layers)])
+ self.gradient_checkpointing = False
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ encoder_hidden_states: Optional[torch.FloatTensor] = None,
+ encoder_attention_mask: Optional[torch.FloatTensor] = None,
+ past_key_values: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
+ use_cache: Optional[bool] = None,
+ output_attentions: Optional[bool] = False,
+ output_hidden_states: Optional[bool] = False,
+ return_dict: Optional[bool] = True,
+ ) -> Union[Tuple[torch.Tensor], BaseModelOutputWithPastAndCrossAttentions]:
+ all_hidden_states = () if output_hidden_states else None
+ all_self_attentions = () if output_attentions else None
+ all_cross_attentions = () if output_attentions and self.config.add_cross_attention else None
+
+ if self.gradient_checkpointing and self.training:
+ if use_cache:
+ logger.warning_once(
+ "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
+ )
+ use_cache = False
+
+ next_decoder_cache = () if use_cache else None
+ for i, layer_module in enumerate(self.layer):
+ if output_hidden_states:
+ all_hidden_states = all_hidden_states + (hidden_states,)
+
+ layer_head_mask = head_mask[i] if head_mask is not None else None
+ past_key_value = past_key_values[i] if past_key_values is not None else None
+
+ if self.gradient_checkpointing and self.training:
+ layer_outputs = self._gradient_checkpointing_func(
+ layer_module.__call__,
+ hidden_states,
+ attention_mask,
+ layer_head_mask,
+ encoder_hidden_states,
+ encoder_attention_mask,
+ past_key_value,
+ output_attentions,
+ )
+ else:
+ layer_outputs = layer_module(
+ hidden_states,
+ attention_mask,
+ layer_head_mask,
+ encoder_hidden_states,
+ encoder_attention_mask,
+ past_key_value,
+ output_attentions,
+ )
+
+ hidden_states = layer_outputs[0]
+ if use_cache:
+ next_decoder_cache += (layer_outputs[-1],)
+ if output_attentions:
+ all_self_attentions = all_self_attentions + (layer_outputs[1],)
+ if self.config.add_cross_attention:
+ all_cross_attentions = all_cross_attentions + (layer_outputs[2],)
+
+ if output_hidden_states:
+ all_hidden_states = all_hidden_states + (hidden_states,)
+
+ if not return_dict:
+ return tuple(
+ v
+ for v in [
+ hidden_states,
+ next_decoder_cache,
+ all_hidden_states,
+ all_self_attentions,
+ all_cross_attentions,
+ ]
+ if v is not None
+ )
+ return BaseModelOutputWithPastAndCrossAttentions(
+ last_hidden_state=hidden_states,
+ past_key_values=next_decoder_cache,
+ hidden_states=all_hidden_states,
+ attentions=all_self_attentions,
+ cross_attentions=all_cross_attentions,
+ )
+
+
+# Copied from transformers.models.bert.modeling_bert.BertPooler
+class CamembertPooler(nn.Module):
+ def __init__(self, config):
+ super().__init__()
+ self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+ self.activation = nn.Tanh()
+
+ def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+ # We "pool" the model by simply taking the hidden state corresponding
+ # to the first token.
+ first_token_tensor = hidden_states[:, 0]
+ pooled_output = self.dense(first_token_tensor)
+ pooled_output = self.activation(pooled_output)
+ return pooled_output
+
+
+class CamembertPreTrainedModel(PreTrainedModel):
+ """
+ An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
+ models.
+ """
+
+ config_class = CamembertConfig
+ base_model_prefix = "roberta"
+ supports_gradient_checkpointing = True
+
+ # Copied from transformers.models.bert.modeling_bert.BertPreTrainedModel._init_weights
+ def _init_weights(self, module):
+ """Initialize the weights"""
+ if isinstance(module, nn.Linear):
+ # Slightly different from the TF version which uses truncated_normal for initialization
+ # cf https://github.com/pytorch/pytorch/pull/5617
+ module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+ if module.bias is not None:
+ module.bias.data.zero_()
+ elif isinstance(module, nn.Embedding):
+ module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
+ if module.padding_idx is not None:
+ module.weight.data[module.padding_idx].zero_()
+ elif isinstance(module, nn.LayerNorm):
+ module.bias.data.zero_()
+ module.weight.data.fill_(1.0)
+
+
+CAMEMBERT_INPUTS_DOCSTRING = r"""
+ Args:
+ input_ids (`torch.LongTensor` of shape `({0})`):
+ Indices of input sequence tokens in the vocabulary.
+
+ Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+ [`PreTrainedTokenizer.__call__`] for details.
+
+ [What are input IDs?](../glossary#input-ids)
+ attention_mask (`torch.FloatTensor` of shape `({0})`, *optional*):
+ Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+
+ - 1 for tokens that are **not masked**,
+ - 0 for tokens that are **masked**.
+
+ [What are attention masks?](../glossary#attention-mask)
+ token_type_ids (`torch.LongTensor` of shape `({0})`, *optional*):
+ Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0,
+ 1]`:
+
+ - 0 corresponds to a *sentence A* token,
+ - 1 corresponds to a *sentence B* token.
+
+ [What are token type IDs?](../glossary#token-type-ids)
+ position_ids (`torch.LongTensor` of shape `({0})`, *optional*):
+ Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+ config.max_position_embeddings - 1]`.
+
+ [What are position IDs?](../glossary#position-ids)
+ head_mask (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
+ Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
+
+ - 1 indicates the head is **not masked**,
+ - 0 indicates the head is **masked**.
+
+ inputs_embeds (`torch.FloatTensor` of shape `({0}, hidden_size)`, *optional*):
+ Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
+ is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
+ model's internal embedding lookup matrix.
+ output_attentions (`bool`, *optional*):
+ Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+ tensors for more detail.
+ output_hidden_states (`bool`, *optional*):
+ Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+ more detail.
+ return_dict (`bool`, *optional*):
+ Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+"""
+
+
+# Copied from transformers.models.roberta.modeling_roberta.RobertaClassificationHead with Roberta->Camembert
+class CamembertClassificationHead(nn.Module):
+ """Head for sentence-level classification tasks."""
+
+ def __init__(self, config):
+ super().__init__()
+ self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+ classifier_dropout = (
+ config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+ )
+ self.dropout = nn.Dropout(classifier_dropout)
+ self.out_proj = nn.Linear(config.hidden_size, config.num_labels)
+
+ def forward(self, features, **kwargs):
+ x = features[:, 0, :] # take token (equiv. to [CLS])
+ x = self.dropout(x)
+ x = self.dense(x)
+ x = torch.tanh(x)
+ x = self.dropout(x)
+ x = self.out_proj(x)
+ return x
+
+
+# Copied from transformers.models.roberta.modeling_roberta.RobertaLMHead with Roberta->Camembert
+class CamembertLMHead(nn.Module):
+ """Camembert Head for masked language modeling."""
+
+ def __init__(self, config):
+ super().__init__()
+ self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+ self.layer_norm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+
+ self.decoder = nn.Linear(config.hidden_size, config.vocab_size)
+ self.bias = nn.Parameter(torch.zeros(config.vocab_size))
+ self.decoder.bias = self.bias
+
+ def forward(self, features, **kwargs):
+ x = self.dense(features)
+ x = gelu(x)
+ x = self.layer_norm(x)
+
+ # project back to size of vocabulary with bias
+ x = self.decoder(x)
+
+ return x
+
+ def _tie_weights(self):
+ # To tie those two weights if they get disconnected (on TPU or when the bias is resized)
+ # For accelerate compatibility and to not break backward compatibility
+ if self.decoder.bias.device.type == "meta":
+ self.decoder.bias = self.bias
+ else:
+ self.bias = self.decoder.bias
+
+
+@add_start_docstrings(
+ "The bare CamemBERT Model transformer outputting raw hidden-states without any specific head on top.",
+ CAMEMBERT_START_DOCSTRING,
+)
+class CamembertModel(CamembertPreTrainedModel):
+ """
+
+ The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer of
+ cross-attention is added between the self-attention layers, following the architecture described in *Attention is
+ all you need*_ by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz
+ Kaiser and Illia Polosukhin.
+
+ To behave as a decoder the model needs to be initialized with the `is_decoder` argument of the configuration set to
+ `True`. To be used in a Seq2Seq model, the model needs to initialized with both `is_decoder` argument and
+ `add_cross_attention` set to `True`; an `encoder_hidden_states` is then expected as an input to the forward pass.
+
+ .. _*Attention is all you need*: https://arxiv.org/abs/1706.03762
+
+ """
+
+ _no_split_modules = []
+
+ # Copied from transformers.models.bert.modeling_bert.BertModel.__init__ with Bert->Camembert
+ def __init__(self, config, add_pooling_layer=True):
+ super().__init__(config)
+ self.config = config
+
+ self.embeddings = CamembertEmbeddings(config)
+ self.encoder = CamembertEncoder(config)
+
+ self.pooler = CamembertPooler(config) if add_pooling_layer else None
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ def get_input_embeddings(self):
+ return self.embeddings.word_embeddings
+
+ def set_input_embeddings(self, value):
+ self.embeddings.word_embeddings = value
+
+ def _prune_heads(self, heads_to_prune):
+ """
+ Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base
+ class PreTrainedModel
+ """
+ for layer, heads in heads_to_prune.items():
+ self.encoder.layer[layer].attention.prune_heads(heads)
+
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint=_CHECKPOINT_FOR_DOC,
+ output_type=BaseModelOutputWithPoolingAndCrossAttentions,
+ config_class=_CONFIG_FOR_DOC,
+ )
+ # Copied from transformers.models.bert.modeling_bert.BertModel.forward
+ def forward(
+ self,
+ input_ids: Optional[torch.Tensor] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ token_type_ids: Optional[torch.Tensor] = None,
+ position_ids: Optional[torch.Tensor] = None,
+ head_mask: Optional[torch.Tensor] = None,
+ inputs_embeds: Optional[torch.Tensor] = None,
+ encoder_hidden_states: Optional[torch.Tensor] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ past_key_values: Optional[List[torch.FloatTensor]] = None,
+ use_cache: Optional[bool] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[Tuple[torch.Tensor], BaseModelOutputWithPoolingAndCrossAttentions]:
+ r"""
+ encoder_hidden_states (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+ Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
+ the model is configured as a decoder.
+ encoder_attention_mask (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
+ the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
+
+ - 1 for tokens that are **not masked**,
+ - 0 for tokens that are **masked**.
+ past_key_values (`tuple(tuple(torch.FloatTensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
+ Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.
+
+ If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
+ don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
+ `decoder_input_ids` of shape `(batch_size, sequence_length)`.
+ use_cache (`bool`, *optional*):
+ If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+ `past_key_values`).
+ """
+ output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+ output_hidden_states = (
+ output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+ )
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+ if self.config.is_decoder:
+ use_cache = use_cache if use_cache is not None else self.config.use_cache
+ else:
+ use_cache = False
+
+ if input_ids is not None and inputs_embeds is not None:
+ raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+ elif input_ids is not None:
+ self.warn_if_padding_and_no_attention_mask(input_ids, attention_mask)
+ input_shape = input_ids.size()
+ elif inputs_embeds is not None:
+ input_shape = inputs_embeds.size()[:-1]
+ else:
+ raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+ batch_size, seq_length = input_shape
+ device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+ # past_key_values_length
+ past_key_values_length = past_key_values[0][0].shape[2] if past_key_values is not None else 0
+
+ if attention_mask is None:
+ attention_mask = torch.ones(((batch_size, seq_length + past_key_values_length)), device=device)
+
+ if token_type_ids is None:
+ if hasattr(self.embeddings, "token_type_ids"):
+ buffered_token_type_ids = self.embeddings.token_type_ids[:, :seq_length]
+ buffered_token_type_ids_expanded = buffered_token_type_ids.expand(batch_size, seq_length)
+ token_type_ids = buffered_token_type_ids_expanded
+ else:
+ token_type_ids = torch.zeros(input_shape, dtype=torch.long, device=device)
+
+ # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
+ # ourselves in which case we just need to make it broadcastable to all heads.
+ extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(attention_mask, input_shape)
+
+ # If a 2D or 3D attention mask is provided for the cross-attention
+ # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
+ if self.config.is_decoder and encoder_hidden_states is not None:
+ encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
+ encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
+ if encoder_attention_mask is None:
+ encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
+ encoder_extended_attention_mask = self.invert_attention_mask(encoder_attention_mask)
+ else:
+ encoder_extended_attention_mask = None
+
+ # Prepare head mask if needed
+ # 1.0 in head_mask indicate we keep the head
+ # attention_probs has shape bsz x n_heads x N x N
+ # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+ # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+ head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
+
+ embedding_output = self.embeddings(
+ input_ids=input_ids,
+ position_ids=position_ids,
+ token_type_ids=token_type_ids,
+ inputs_embeds=inputs_embeds,
+ past_key_values_length=past_key_values_length,
+ )
+ encoder_outputs = self.encoder(
+ embedding_output,
+ attention_mask=extended_attention_mask,
+ head_mask=head_mask,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_extended_attention_mask,
+ past_key_values=past_key_values,
+ use_cache=use_cache,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ )
+ sequence_output = encoder_outputs[0]
+ pooled_output = self.pooler(sequence_output) if self.pooler is not None else None
+
+ if not return_dict:
+ return (sequence_output, pooled_output) + encoder_outputs[1:]
+
+ return BaseModelOutputWithPoolingAndCrossAttentions(
+ last_hidden_state=sequence_output,
+ pooler_output=pooled_output,
+ past_key_values=encoder_outputs.past_key_values,
+ hidden_states=encoder_outputs.hidden_states,
+ attentions=encoder_outputs.attentions,
+ cross_attentions=encoder_outputs.cross_attentions,
+ )
+
+
+@add_start_docstrings(
+ """CamemBERT Model with a `language modeling` head on top.""",
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_roberta.RobertaForMaskedLM with Roberta->Camembert, ROBERTA->CAMEMBERT
+class CamembertForMaskedLM(CamembertPreTrainedModel):
+ _tied_weights_keys = ["lm_head.decoder.weight", "lm_head.decoder.bias"]
+
+ def __init__(self, config):
+ super().__init__(config)
+
+ if config.is_decoder:
+ logger.warning(
+ "If you want to use `CamembertForMaskedLM` make sure `config.is_decoder=False` for "
+ "bi-directional self-attention."
+ )
+
+ self.roberta = CamembertModel(config, add_pooling_layer=False)
+ self.lm_head = CamembertLMHead(config)
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ def get_output_embeddings(self):
+ return self.lm_head.decoder
+
+ def set_output_embeddings(self, new_embeddings):
+ self.lm_head.decoder = new_embeddings
+
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint=_CHECKPOINT_FOR_DOC,
+ output_type=MaskedLMOutput,
+ config_class=_CONFIG_FOR_DOC,
+ mask="",
+ expected_output="' Paris'",
+ expected_loss=0.1,
+ )
+ def forward(
+ self,
+ input_ids: Optional[torch.LongTensor] = None,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ token_type_ids: Optional[torch.LongTensor] = None,
+ position_ids: Optional[torch.LongTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ inputs_embeds: Optional[torch.FloatTensor] = None,
+ encoder_hidden_states: Optional[torch.FloatTensor] = None,
+ encoder_attention_mask: Optional[torch.FloatTensor] = None,
+ labels: Optional[torch.LongTensor] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[Tuple[torch.Tensor], MaskedLMOutput]:
+ r"""
+ labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
+ config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
+ loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
+ kwargs (`Dict[str, any]`, optional, defaults to *{}*):
+ Used to hide legacy arguments that have been deprecated.
+ """
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ )
+ sequence_output = outputs[0]
+ prediction_scores = self.lm_head(sequence_output)
+
+ masked_lm_loss = None
+ if labels is not None:
+ # move labels to correct device to enable model parallelism
+ labels = labels.to(prediction_scores.device)
+ loss_fct = CrossEntropyLoss()
+ masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
+
+ if not return_dict:
+ output = (prediction_scores,) + outputs[2:]
+ return ((masked_lm_loss,) + output) if masked_lm_loss is not None else output
+
+ return MaskedLMOutput(
+ loss=masked_lm_loss,
+ logits=prediction_scores,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+
+@add_start_docstrings(
+ """
+ CamemBERT Model transformer with a sequence classification/regression head on top (a linear layer on top of the
+ pooled output) e.g. for GLUE tasks.
+ """,
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_roberta.RobertaForSequenceClassification with Roberta->Camembert, ROBERTA->CAMEMBERT
+class CamembertForSequenceClassification(CamembertPreTrainedModel):
+ def __init__(self, config):
+ super().__init__(config)
+ self.num_labels = config.num_labels
+ self.config = config
+
+ self.roberta = CamembertModel(config, add_pooling_layer=False)
+ self.classifier = CamembertClassificationHead(config)
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint="cardiffnlp/twitter-roberta-base-emotion",
+ output_type=SequenceClassifierOutput,
+ config_class=_CONFIG_FOR_DOC,
+ expected_output="'optimism'",
+ expected_loss=0.08,
+ )
+ def forward(
+ self,
+ input_ids: Optional[torch.LongTensor] = None,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ token_type_ids: Optional[torch.LongTensor] = None,
+ position_ids: Optional[torch.LongTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ inputs_embeds: Optional[torch.FloatTensor] = None,
+ labels: Optional[torch.LongTensor] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[Tuple[torch.Tensor], SequenceClassifierOutput]:
+ r"""
+ labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+ Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+ config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+ `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+ """
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ )
+ sequence_output = outputs[0]
+ logits = self.classifier(sequence_output)
+
+ loss = None
+ if labels is not None:
+ # move labels to correct device to enable model parallelism
+ labels = labels.to(logits.device)
+ if self.config.problem_type is None:
+ if self.num_labels == 1:
+ self.config.problem_type = "regression"
+ elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+ self.config.problem_type = "single_label_classification"
+ else:
+ self.config.problem_type = "multi_label_classification"
+
+ if self.config.problem_type == "regression":
+ loss_fct = MSELoss()
+ if self.num_labels == 1:
+ loss = loss_fct(logits.squeeze(), labels.squeeze())
+ else:
+ loss = loss_fct(logits, labels)
+ elif self.config.problem_type == "single_label_classification":
+ loss_fct = CrossEntropyLoss()
+ loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+ elif self.config.problem_type == "multi_label_classification":
+ loss_fct = BCEWithLogitsLoss()
+ loss = loss_fct(logits, labels)
+
+ if not return_dict:
+ output = (logits,) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return SequenceClassifierOutput(
+ loss=loss,
+ logits=logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+
+@add_start_docstrings(
+ """
+ CamemBERT Model with a multiple choice classification head on top (a linear layer on top of the pooled output and a
+ softmax) e.g. for RocStories/SWAG tasks.
+ """,
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_roberta.RobertaForMultipleChoice with Roberta->Camembert, ROBERTA->CAMEMBERT
+class CamembertForMultipleChoice(CamembertPreTrainedModel):
+ def __init__(self, config):
+ super().__init__(config)
+
+ self.roberta = CamembertModel(config)
+ self.dropout = nn.Dropout(config.hidden_dropout_prob)
+ self.classifier = nn.Linear(config.hidden_size, 1)
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ @add_start_docstrings_to_model_forward(
+ CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length")
+ )
+ @add_code_sample_docstrings(
+ checkpoint=_CHECKPOINT_FOR_DOC,
+ output_type=MultipleChoiceModelOutput,
+ config_class=_CONFIG_FOR_DOC,
+ )
+ def forward(
+ self,
+ input_ids: Optional[torch.LongTensor] = None,
+ token_type_ids: Optional[torch.LongTensor] = None,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ labels: Optional[torch.LongTensor] = None,
+ position_ids: Optional[torch.LongTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ inputs_embeds: Optional[torch.FloatTensor] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[Tuple[torch.Tensor], MultipleChoiceModelOutput]:
+ r"""
+ labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+ Labels for computing the multiple choice classification loss. Indices should be in `[0, ...,
+ num_choices-1]` where `num_choices` is the size of the second dimension of the input tensors. (See
+ `input_ids` above)
+ """
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+ num_choices = input_ids.shape[1] if input_ids is not None else inputs_embeds.shape[1]
+
+ flat_input_ids = input_ids.view(-1, input_ids.size(-1)) if input_ids is not None else None
+ flat_position_ids = position_ids.view(-1, position_ids.size(-1)) if position_ids is not None else None
+ flat_token_type_ids = token_type_ids.view(-1, token_type_ids.size(-1)) if token_type_ids is not None else None
+ flat_attention_mask = attention_mask.view(-1, attention_mask.size(-1)) if attention_mask is not None else None
+ flat_inputs_embeds = (
+ inputs_embeds.view(-1, inputs_embeds.size(-2), inputs_embeds.size(-1))
+ if inputs_embeds is not None
+ else None
+ )
+
+ outputs = self.roberta(
+ flat_input_ids,
+ position_ids=flat_position_ids,
+ token_type_ids=flat_token_type_ids,
+ attention_mask=flat_attention_mask,
+ head_mask=head_mask,
+ inputs_embeds=flat_inputs_embeds,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ )
+ pooled_output = outputs[1]
+
+ pooled_output = self.dropout(pooled_output)
+ logits = self.classifier(pooled_output)
+ reshaped_logits = logits.view(-1, num_choices)
+
+ loss = None
+ if labels is not None:
+ # move labels to correct device to enable model parallelism
+ labels = labels.to(reshaped_logits.device)
+ loss_fct = CrossEntropyLoss()
+ loss = loss_fct(reshaped_logits, labels)
+
+ if not return_dict:
+ output = (reshaped_logits,) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return MultipleChoiceModelOutput(
+ loss=loss,
+ logits=reshaped_logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+
+@add_start_docstrings(
+ """
+ CamemBERT Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g.
+ for Named-Entity-Recognition (NER) tasks.
+ """,
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_roberta.RobertaForTokenClassification with Roberta->Camembert, ROBERTA->CAMEMBERT
+class CamembertForTokenClassification(CamembertPreTrainedModel):
+ def __init__(self, config):
+ super().__init__(config)
+ self.num_labels = config.num_labels
+
+ self.roberta = CamembertModel(config, add_pooling_layer=False)
+ classifier_dropout = (
+ config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+ )
+ self.dropout = nn.Dropout(classifier_dropout)
+ self.classifier = nn.Linear(config.hidden_size, config.num_labels)
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint="Jean-Baptiste/roberta-large-ner-english",
+ output_type=TokenClassifierOutput,
+ config_class=_CONFIG_FOR_DOC,
+ expected_output="['O', 'ORG', 'ORG', 'O', 'O', 'O', 'O', 'O', 'LOC', 'O', 'LOC', 'LOC']",
+ expected_loss=0.01,
+ )
+ def forward(
+ self,
+ input_ids: Optional[torch.LongTensor] = None,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ token_type_ids: Optional[torch.LongTensor] = None,
+ position_ids: Optional[torch.LongTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ inputs_embeds: Optional[torch.FloatTensor] = None,
+ labels: Optional[torch.LongTensor] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[Tuple[torch.Tensor], TokenClassifierOutput]:
+ r"""
+ labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
+ """
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ )
+
+ sequence_output = outputs[0]
+
+ sequence_output = self.dropout(sequence_output)
+ logits = self.classifier(sequence_output)
+
+ loss = None
+ if labels is not None:
+ # move labels to correct device to enable model parallelism
+ labels = labels.to(logits.device)
+ loss_fct = CrossEntropyLoss()
+ loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+
+ if not return_dict:
+ output = (logits,) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return TokenClassifierOutput(
+ loss=loss,
+ logits=logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+
+@add_start_docstrings(
+ """
+ CamemBERT Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear
+ layers on top of the hidden-states output to compute `span start logits` and `span end logits`
+ """,
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_roberta.RobertaForQuestionAnswering with Roberta->Camembert, ROBERTA->CAMEMBERT
+class CamembertForQuestionAnswering(CamembertPreTrainedModel):
+ def __init__(self, config):
+ super().__init__(config)
+ self.num_labels = config.num_labels
+
+ self.roberta = CamembertModel(config, add_pooling_layer=False)
+ self.qa_outputs = nn.Linear(config.hidden_size, config.num_labels)
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint="deepset/roberta-base-squad2",
+ output_type=QuestionAnsweringModelOutput,
+ config_class=_CONFIG_FOR_DOC,
+ expected_output="' puppet'",
+ expected_loss=0.86,
+ )
+ def forward(
+ self,
+ input_ids: Optional[torch.LongTensor] = None,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ token_type_ids: Optional[torch.LongTensor] = None,
+ position_ids: Optional[torch.LongTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ inputs_embeds: Optional[torch.FloatTensor] = None,
+ start_positions: Optional[torch.LongTensor] = None,
+ end_positions: Optional[torch.LongTensor] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[Tuple[torch.Tensor], QuestionAnsweringModelOutput]:
+ r"""
+ start_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+ Labels for position (index) of the start of the labelled span for computing the token classification loss.
+ Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
+ are not taken into account for computing the loss.
+ end_positions (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+ Labels for position (index) of the end of the labelled span for computing the token classification loss.
+ Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
+ are not taken into account for computing the loss.
+ """
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ )
+
+ sequence_output = outputs[0]
+
+ logits = self.qa_outputs(sequence_output)
+ start_logits, end_logits = logits.split(1, dim=-1)
+ start_logits = start_logits.squeeze(-1).contiguous()
+ end_logits = end_logits.squeeze(-1).contiguous()
+
+ total_loss = None
+ if start_positions is not None and end_positions is not None:
+ # If we are on multi-GPU, split add a dimension
+ if len(start_positions.size()) > 1:
+ start_positions = start_positions.squeeze(-1)
+ if len(end_positions.size()) > 1:
+ end_positions = end_positions.squeeze(-1)
+ # sometimes the start/end positions are outside our model inputs, we ignore these terms
+ ignored_index = start_logits.size(1)
+ start_positions = start_positions.clamp(0, ignored_index)
+ end_positions = end_positions.clamp(0, ignored_index)
+
+ loss_fct = CrossEntropyLoss(ignore_index=ignored_index)
+ start_loss = loss_fct(start_logits, start_positions)
+ end_loss = loss_fct(end_logits, end_positions)
+ total_loss = (start_loss + end_loss) / 2
+
+ if not return_dict:
+ output = (start_logits, end_logits) + outputs[2:]
+ return ((total_loss,) + output) if total_loss is not None else output
+
+ return QuestionAnsweringModelOutput(
+ loss=total_loss,
+ start_logits=start_logits,
+ end_logits=end_logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+
+@add_start_docstrings(
+ """CamemBERT Model with a `language modeling` head on top for CLM fine-tuning.""", CAMEMBERT_START_DOCSTRING
+)
+# Copied from transformers.models.roberta.modeling_roberta.RobertaForCausalLM with Roberta->Camembert, ROBERTA->CAMEMBERT, roberta-base->camembert-base
+class CamembertForCausalLM(CamembertPreTrainedModel):
+ _tied_weights_keys = ["lm_head.decoder.weight", "lm_head.decoder.bias"]
+
+ def __init__(self, config):
+ super().__init__(config)
+
+ if not config.is_decoder:
+ logger.warning("If you want to use `CamembertLMHeadModel` as a standalone, add `is_decoder=True.`")
+
+ self.roberta = CamembertModel(config, add_pooling_layer=False)
+ self.lm_head = CamembertLMHead(config)
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ def get_output_embeddings(self):
+ return self.lm_head.decoder
+
+ def set_output_embeddings(self, new_embeddings):
+ self.lm_head.decoder = new_embeddings
+
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @replace_return_docstrings(output_type=CausalLMOutputWithCrossAttentions, config_class=_CONFIG_FOR_DOC)
+ def forward(
+ self,
+ input_ids: Optional[torch.LongTensor] = None,
+ attention_mask: Optional[torch.FloatTensor] = None,
+ token_type_ids: Optional[torch.LongTensor] = None,
+ position_ids: Optional[torch.LongTensor] = None,
+ head_mask: Optional[torch.FloatTensor] = None,
+ inputs_embeds: Optional[torch.FloatTensor] = None,
+ encoder_hidden_states: Optional[torch.FloatTensor] = None,
+ encoder_attention_mask: Optional[torch.FloatTensor] = None,
+ labels: Optional[torch.LongTensor] = None,
+ past_key_values: Tuple[Tuple[torch.FloatTensor]] = None,
+ use_cache: Optional[bool] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[Tuple[torch.Tensor], CausalLMOutputWithCrossAttentions]:
+ r"""
+ encoder_hidden_states (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+ Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
+ the model is configured as a decoder.
+ encoder_attention_mask (`torch.FloatTensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
+ the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
+
+ - 1 for tokens that are **not masked**,
+ - 0 for tokens that are **masked**.
+
+ labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Labels for computing the left-to-right language modeling loss (next word prediction). Indices should be in
+ `[-100, 0, ..., config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are
+ ignored (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
+ past_key_values (`tuple(tuple(torch.FloatTensor))` of length `config.n_layers` with each tuple having 4 tensors of shape `(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
+ Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.
+
+ If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
+ don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
+ `decoder_input_ids` of shape `(batch_size, sequence_length)`.
+ use_cache (`bool`, *optional*):
+ If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+ `past_key_values`).
+
+ Returns:
+
+ Example:
+
+ ```python
+ >>> from transformers import AutoTokenizer, CamembertForCausalLM, AutoConfig
+ >>> import torch
+
+ >>> tokenizer = AutoTokenizer.from_pretrained("camembert-base")
+ >>> config = AutoConfig.from_pretrained("camembert-base")
+ >>> config.is_decoder = True
+ >>> model = CamembertForCausalLM.from_pretrained("camembert-base", config=config)
+
+ >>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
+ >>> outputs = model(**inputs)
+
+ >>> prediction_logits = outputs.logits
+ ```"""
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+ if labels is not None:
+ use_cache = False
+
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ past_key_values=past_key_values,
+ use_cache=use_cache,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ )
+
+ sequence_output = outputs[0]
+ prediction_scores = self.lm_head(sequence_output)
+
+ lm_loss = None
+ if labels is not None:
+ # move labels to correct device to enable model parallelism
+ labels = labels.to(prediction_scores.device)
+ # we are doing next-token prediction; shift prediction scores and input ids by one
+ shifted_prediction_scores = prediction_scores[:, :-1, :].contiguous()
+ labels = labels[:, 1:].contiguous()
+ loss_fct = CrossEntropyLoss()
+ lm_loss = loss_fct(shifted_prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
+
+ if not return_dict:
+ output = (prediction_scores,) + outputs[2:]
+ return ((lm_loss,) + output) if lm_loss is not None else output
+
+ return CausalLMOutputWithCrossAttentions(
+ loss=lm_loss,
+ logits=prediction_scores,
+ past_key_values=outputs.past_key_values,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ cross_attentions=outputs.cross_attentions,
+ )
+
+ def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, **model_kwargs):
+ input_shape = input_ids.shape
+ # if model is used as a decoder in encoder-decoder model, the decoder attention mask is created on the fly
+ if attention_mask is None:
+ attention_mask = input_ids.new_ones(input_shape)
+
+ # cut decoder_input_ids if past_key_values is used
+ if past_key_values is not None:
+ past_length = past_key_values[0][0].shape[2]
+
+ # Some generation methods already pass only the last input ID
+ if input_ids.shape[1] > past_length:
+ remove_prefix_length = past_length
+ else:
+ # Default to old behavior: keep only final ID
+ remove_prefix_length = input_ids.shape[1] - 1
+
+ input_ids = input_ids[:, remove_prefix_length:]
+
+ return {"input_ids": input_ids, "attention_mask": attention_mask, "past_key_values": past_key_values}
+
+ def _reorder_cache(self, past_key_values, beam_idx):
+ reordered_past = ()
+ for layer_past in past_key_values:
+ reordered_past += (
+ tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past),
+ )
+ return reordered_past
+
+
+# Copied from transformers.models.roberta.modeling_roberta.create_position_ids_from_input_ids
+def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_values_length=0):
+ """
+ Replace non-padding symbols with their position numbers. Position numbers begin at padding_idx+1. Padding symbols
+ are ignored. This is modified from fairseq's `utils.make_positions`.
+
+ Args:
+ x: torch.Tensor x:
+
+ Returns: torch.Tensor
+ """
+ # The series of casts and type-conversions here are carefully balanced to both work with ONNX export and XLA.
+ mask = input_ids.ne(padding_idx).int()
+ incremental_indices = (torch.cumsum(mask, dim=1).type_as(mask) + past_key_values_length) * mask
+ return incremental_indices.long() + padding_idx
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/modeling_tf_camembert.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/modeling_tf_camembert.py
new file mode 100644
index 0000000000000000000000000000000000000000..850d8bccefee21be203c26c99ae533811ca25e98
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/modeling_tf_camembert.py
@@ -0,0 +1,1793 @@
+# coding=utf-8
+# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
+# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" TF 2.0 CamemBERT model."""
+
+
+from __future__ import annotations
+
+import math
+import warnings
+from typing import Optional, Tuple, Union
+
+import numpy as np
+import tensorflow as tf
+
+from ...activations_tf import get_tf_activation
+from ...modeling_tf_outputs import (
+ TFBaseModelOutputWithPastAndCrossAttentions,
+ TFBaseModelOutputWithPoolingAndCrossAttentions,
+ TFCausalLMOutputWithCrossAttentions,
+ TFMaskedLMOutput,
+ TFMultipleChoiceModelOutput,
+ TFQuestionAnsweringModelOutput,
+ TFSequenceClassifierOutput,
+ TFTokenClassifierOutput,
+)
+from ...modeling_tf_utils import (
+ TFCausalLanguageModelingLoss,
+ TFMaskedLanguageModelingLoss,
+ TFModelInputType,
+ TFMultipleChoiceLoss,
+ TFPreTrainedModel,
+ TFQuestionAnsweringLoss,
+ TFSequenceClassificationLoss,
+ TFTokenClassificationLoss,
+ get_initializer,
+ keras_serializable,
+ unpack_inputs,
+)
+from ...tf_utils import check_embeddings_within_bounds, shape_list, stable_softmax
+from ...utils import (
+ add_code_sample_docstrings,
+ add_start_docstrings,
+ add_start_docstrings_to_model_forward,
+ logging,
+)
+from .configuration_camembert import CamembertConfig
+
+
+logger = logging.get_logger(__name__)
+
+_CHECKPOINT_FOR_DOC = "camembert-base"
+_CONFIG_FOR_DOC = "CamembertConfig"
+
+TF_CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_LIST = [
+ # See all CamemBERT models at https://huggingface.co/models?filter=camembert
+]
+
+
+CAMEMBERT_START_DOCSTRING = r"""
+
+ This model inherits from [`TFPreTrainedModel`]. Check the superclass documentation for the generic methods the
+ library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
+ etc.)
+
+ This model is also a [tf.keras.Model](https://www.tensorflow.org/api_docs/python/tf/keras/Model) subclass. Use it
+ as a regular TF 2.0 Keras Model and refer to the TF 2.0 documentation for all matter related to general usage and
+ behavior.
+
+
+
+ TensorFlow models and layers in `transformers` accept two formats as input:
+
+ - having all inputs as keyword arguments (like PyTorch models), or
+ - having all inputs as a list, tuple or dict in the first positional argument.
+
+ The reason the second format is supported is that Keras methods prefer this format when passing inputs to models
+ and layers. Because of this support, when using methods like `model.fit()` things should "just work" for you - just
+ pass your inputs and labels in any format that `model.fit()` supports! If, however, you want to use the second
+ format outside of Keras methods like `fit()` and `predict()`, such as when creating your own layers or models with
+ the Keras `Functional` API, there are three possibilities you can use to gather all the input Tensors in the first
+ positional argument:
+
+ - a single Tensor with `input_ids` only and nothing else: `model(input_ids)`
+ - a list of varying length with one or several input Tensors IN THE ORDER given in the docstring:
+ `model([input_ids, attention_mask])` or `model([input_ids, attention_mask, token_type_ids])`
+ - a dictionary with one or several input Tensors associated to the input names given in the docstring:
+ `model({"input_ids": input_ids, "token_type_ids": token_type_ids})`
+
+ Note that when creating models and layers with
+ [subclassing](https://keras.io/guides/making_new_layers_and_models_via_subclassing/) then you don't need to worry
+ about any of this, as you can just pass inputs like you would to any other Python function!
+
+
+
+ Parameters:
+ config ([`CamembertConfig`]): Model configuration class with all the parameters of the
+ model. Initializing with a config file does not load the weights associated with the model, only the
+ configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+CAMEMBERT_INPUTS_DOCSTRING = r"""
+ Args:
+ input_ids (`Numpy array` or `tf.Tensor` of shape `({0})`):
+ Indices of input sequence tokens in the vocabulary.
+
+ Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.__call__`] and
+ [`PreTrainedTokenizer.encode`] for details.
+
+ [What are input IDs?](../glossary#input-ids)
+ attention_mask (`Numpy array` or `tf.Tensor` of shape `({0})`, *optional*):
+ Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+
+ - 1 for tokens that are **not masked**,
+ - 0 for tokens that are **masked**.
+
+ [What are attention masks?](../glossary#attention-mask)
+ token_type_ids (`Numpy array` or `tf.Tensor` of shape `({0})`, *optional*):
+ Segment token indices to indicate first and second portions of the inputs. Indices are selected in `[0,
+ 1]`:
+
+ - 0 corresponds to a *sentence A* token,
+ - 1 corresponds to a *sentence B* token.
+
+ [What are token type IDs?](../glossary#token-type-ids)
+ position_ids (`Numpy array` or `tf.Tensor` of shape `({0})`, *optional*):
+ Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+ config.max_position_embeddings - 1]`.
+
+ [What are position IDs?](../glossary#position-ids)
+ head_mask (`Numpy array` or `tf.Tensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
+ Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
+
+ - 1 indicates the head is **not masked**,
+ - 0 indicates the head is **masked**.
+
+ inputs_embeds (`tf.Tensor` of shape `({0}, hidden_size)`, *optional*):
+ Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
+ is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
+ model's internal embedding lookup matrix.
+ output_attentions (`bool`, *optional*):
+ Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+ tensors for more detail. This argument can be used only in eager mode, in graph mode the value in the
+ config will be used instead.
+ output_hidden_states (`bool`, *optional*):
+ Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+ more detail. This argument can be used only in eager mode, in graph mode the value in the config will be
+ used instead.
+ return_dict (`bool`, *optional*):
+ Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple. This argument can be used in
+ eager mode, in graph mode the value will always be set to True.
+ training (`bool`, *optional*, defaults to `False`):
+ Whether or not to use the model in training mode (some modules like dropout modules have different
+ behaviors between training and evaluation).
+"""
+
+
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaEmbeddings
+class TFCamembertEmbeddings(tf.keras.layers.Layer):
+ """
+ Same as BertEmbeddings with a tiny tweak for positional embeddings indexing.
+ """
+
+ def __init__(self, config, **kwargs):
+ super().__init__(**kwargs)
+
+ self.padding_idx = 1
+ self.config = config
+ self.hidden_size = config.hidden_size
+ self.max_position_embeddings = config.max_position_embeddings
+ self.initializer_range = config.initializer_range
+ self.LayerNorm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_eps, name="LayerNorm")
+ self.dropout = tf.keras.layers.Dropout(rate=config.hidden_dropout_prob)
+
+ def build(self, input_shape=None):
+ with tf.name_scope("word_embeddings"):
+ self.weight = self.add_weight(
+ name="weight",
+ shape=[self.config.vocab_size, self.hidden_size],
+ initializer=get_initializer(self.initializer_range),
+ )
+
+ with tf.name_scope("token_type_embeddings"):
+ self.token_type_embeddings = self.add_weight(
+ name="embeddings",
+ shape=[self.config.type_vocab_size, self.hidden_size],
+ initializer=get_initializer(self.initializer_range),
+ )
+
+ with tf.name_scope("position_embeddings"):
+ self.position_embeddings = self.add_weight(
+ name="embeddings",
+ shape=[self.max_position_embeddings, self.hidden_size],
+ initializer=get_initializer(self.initializer_range),
+ )
+
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "LayerNorm", None) is not None:
+ with tf.name_scope(self.LayerNorm.name):
+ self.LayerNorm.build([None, None, self.config.hidden_size])
+
+ def create_position_ids_from_input_ids(self, input_ids, past_key_values_length=0):
+ """
+ Replace non-padding symbols with their position numbers. Position numbers begin at padding_idx+1. Padding
+ symbols are ignored. This is modified from fairseq's `utils.make_positions`.
+
+ Args:
+ input_ids: tf.Tensor
+ Returns: tf.Tensor
+ """
+ mask = tf.cast(tf.math.not_equal(input_ids, self.padding_idx), dtype=input_ids.dtype)
+ incremental_indices = (tf.math.cumsum(mask, axis=1) + past_key_values_length) * mask
+
+ return incremental_indices + self.padding_idx
+
+ def call(
+ self,
+ input_ids=None,
+ position_ids=None,
+ token_type_ids=None,
+ inputs_embeds=None,
+ past_key_values_length=0,
+ training=False,
+ ):
+ """
+ Applies embedding based on inputs tensor.
+
+ Returns:
+ final_embeddings (`tf.Tensor`): output embedding tensor.
+ """
+ assert not (input_ids is None and inputs_embeds is None)
+
+ if input_ids is not None:
+ check_embeddings_within_bounds(input_ids, self.config.vocab_size)
+ inputs_embeds = tf.gather(params=self.weight, indices=input_ids)
+
+ input_shape = shape_list(inputs_embeds)[:-1]
+
+ if token_type_ids is None:
+ token_type_ids = tf.fill(dims=input_shape, value=0)
+
+ if position_ids is None:
+ if input_ids is not None:
+ # Create the position ids from the input token ids. Any padded tokens remain padded.
+ position_ids = self.create_position_ids_from_input_ids(
+ input_ids=input_ids, past_key_values_length=past_key_values_length
+ )
+ else:
+ position_ids = tf.expand_dims(
+ tf.range(start=self.padding_idx + 1, limit=input_shape[-1] + self.padding_idx + 1), axis=0
+ )
+
+ position_embeds = tf.gather(params=self.position_embeddings, indices=position_ids)
+ token_type_embeds = tf.gather(params=self.token_type_embeddings, indices=token_type_ids)
+ final_embeddings = inputs_embeds + position_embeds + token_type_embeds
+ final_embeddings = self.LayerNorm(inputs=final_embeddings)
+ final_embeddings = self.dropout(inputs=final_embeddings, training=training)
+
+ return final_embeddings
+
+
+# Copied from transformers.models.bert.modeling_tf_bert.TFBertPooler with Bert->Camembert
+class TFCamembertPooler(tf.keras.layers.Layer):
+ def __init__(self, config: CamembertConfig, **kwargs):
+ super().__init__(**kwargs)
+
+ self.dense = tf.keras.layers.Dense(
+ units=config.hidden_size,
+ kernel_initializer=get_initializer(config.initializer_range),
+ activation="tanh",
+ name="dense",
+ )
+ self.config = config
+
+ def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
+ # We "pool" the model by simply taking the hidden state corresponding
+ # to the first token.
+ first_token_tensor = hidden_states[:, 0]
+ pooled_output = self.dense(inputs=first_token_tensor)
+
+ return pooled_output
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "dense", None) is not None:
+ with tf.name_scope(self.dense.name):
+ self.dense.build([None, None, self.config.hidden_size])
+
+
+# Copied from transformers.models.bert.modeling_tf_bert.TFBertSelfAttention with Bert->Camembert
+class TFCamembertSelfAttention(tf.keras.layers.Layer):
+ def __init__(self, config: CamembertConfig, **kwargs):
+ super().__init__(**kwargs)
+
+ if config.hidden_size % config.num_attention_heads != 0:
+ raise ValueError(
+ f"The hidden size ({config.hidden_size}) is not a multiple of the number "
+ f"of attention heads ({config.num_attention_heads})"
+ )
+
+ self.num_attention_heads = config.num_attention_heads
+ self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
+ self.all_head_size = self.num_attention_heads * self.attention_head_size
+ self.sqrt_att_head_size = math.sqrt(self.attention_head_size)
+
+ self.query = tf.keras.layers.Dense(
+ units=self.all_head_size, kernel_initializer=get_initializer(config.initializer_range), name="query"
+ )
+ self.key = tf.keras.layers.Dense(
+ units=self.all_head_size, kernel_initializer=get_initializer(config.initializer_range), name="key"
+ )
+ self.value = tf.keras.layers.Dense(
+ units=self.all_head_size, kernel_initializer=get_initializer(config.initializer_range), name="value"
+ )
+ self.dropout = tf.keras.layers.Dropout(rate=config.attention_probs_dropout_prob)
+
+ self.is_decoder = config.is_decoder
+ self.config = config
+
+ def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> tf.Tensor:
+ # Reshape from [batch_size, seq_length, all_head_size] to [batch_size, seq_length, num_attention_heads, attention_head_size]
+ tensor = tf.reshape(tensor=tensor, shape=(batch_size, -1, self.num_attention_heads, self.attention_head_size))
+
+ # Transpose the tensor from [batch_size, seq_length, num_attention_heads, attention_head_size] to [batch_size, num_attention_heads, seq_length, attention_head_size]
+ return tf.transpose(tensor, perm=[0, 2, 1, 3])
+
+ def call(
+ self,
+ hidden_states: tf.Tensor,
+ attention_mask: tf.Tensor,
+ head_mask: tf.Tensor,
+ encoder_hidden_states: tf.Tensor,
+ encoder_attention_mask: tf.Tensor,
+ past_key_value: Tuple[tf.Tensor],
+ output_attentions: bool,
+ training: bool = False,
+ ) -> Tuple[tf.Tensor]:
+ batch_size = shape_list(hidden_states)[0]
+ mixed_query_layer = self.query(inputs=hidden_states)
+
+ # If this is instantiated as a cross-attention module, the keys
+ # and values come from an encoder; the attention mask needs to be
+ # such that the encoder's padding tokens are not attended to.
+ is_cross_attention = encoder_hidden_states is not None
+
+ if is_cross_attention and past_key_value is not None:
+ # reuse k,v, cross_attentions
+ key_layer = past_key_value[0]
+ value_layer = past_key_value[1]
+ attention_mask = encoder_attention_mask
+ elif is_cross_attention:
+ key_layer = self.transpose_for_scores(self.key(inputs=encoder_hidden_states), batch_size)
+ value_layer = self.transpose_for_scores(self.value(inputs=encoder_hidden_states), batch_size)
+ attention_mask = encoder_attention_mask
+ elif past_key_value is not None:
+ key_layer = self.transpose_for_scores(self.key(inputs=hidden_states), batch_size)
+ value_layer = self.transpose_for_scores(self.value(inputs=hidden_states), batch_size)
+ key_layer = tf.concat([past_key_value[0], key_layer], axis=2)
+ value_layer = tf.concat([past_key_value[1], value_layer], axis=2)
+ else:
+ key_layer = self.transpose_for_scores(self.key(inputs=hidden_states), batch_size)
+ value_layer = self.transpose_for_scores(self.value(inputs=hidden_states), batch_size)
+
+ query_layer = self.transpose_for_scores(mixed_query_layer, batch_size)
+
+ if self.is_decoder:
+ # if cross_attention save Tuple(tf.Tensor, tf.Tensor) of all cross attention key/value_states.
+ # Further calls to cross_attention layer can then reuse all cross-attention
+ # key/value_states (first "if" case)
+ # if uni-directional self-attention (decoder) save Tuple(tf.Tensor, tf.Tensor) of
+ # all previous decoder key/value_states. Further calls to uni-directional self-attention
+ # can concat previous decoder key/value_states to current projected key/value_states (third "elif" case)
+ # if encoder bi-directional self-attention `past_key_value` is always `None`
+ past_key_value = (key_layer, value_layer)
+
+ # Take the dot product between "query" and "key" to get the raw attention scores.
+ # (batch size, num_heads, seq_len_q, seq_len_k)
+ attention_scores = tf.matmul(query_layer, key_layer, transpose_b=True)
+ dk = tf.cast(self.sqrt_att_head_size, dtype=attention_scores.dtype)
+ attention_scores = tf.divide(attention_scores, dk)
+
+ if attention_mask is not None:
+ # Apply the attention mask is (precomputed for all layers in TFCamembertModel call() function)
+ attention_scores = tf.add(attention_scores, attention_mask)
+
+ # Normalize the attention scores to probabilities.
+ attention_probs = stable_softmax(logits=attention_scores, axis=-1)
+
+ # This is actually dropping out entire tokens to attend to, which might
+ # seem a bit unusual, but is taken from the original Transformer paper.
+ attention_probs = self.dropout(inputs=attention_probs, training=training)
+
+ # Mask heads if we want to
+ if head_mask is not None:
+ attention_probs = tf.multiply(attention_probs, head_mask)
+
+ attention_output = tf.matmul(attention_probs, value_layer)
+ attention_output = tf.transpose(attention_output, perm=[0, 2, 1, 3])
+
+ # (batch_size, seq_len_q, all_head_size)
+ attention_output = tf.reshape(tensor=attention_output, shape=(batch_size, -1, self.all_head_size))
+ outputs = (attention_output, attention_probs) if output_attentions else (attention_output,)
+
+ if self.is_decoder:
+ outputs = outputs + (past_key_value,)
+ return outputs
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "query", None) is not None:
+ with tf.name_scope(self.query.name):
+ self.query.build([None, None, self.config.hidden_size])
+ if getattr(self, "key", None) is not None:
+ with tf.name_scope(self.key.name):
+ self.key.build([None, None, self.config.hidden_size])
+ if getattr(self, "value", None) is not None:
+ with tf.name_scope(self.value.name):
+ self.value.build([None, None, self.config.hidden_size])
+
+
+# Copied from transformers.models.bert.modeling_tf_bert.TFBertSelfOutput with Bert->Camembert
+class TFCamembertSelfOutput(tf.keras.layers.Layer):
+ def __init__(self, config: CamembertConfig, **kwargs):
+ super().__init__(**kwargs)
+
+ self.dense = tf.keras.layers.Dense(
+ units=config.hidden_size, kernel_initializer=get_initializer(config.initializer_range), name="dense"
+ )
+ self.LayerNorm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_eps, name="LayerNorm")
+ self.dropout = tf.keras.layers.Dropout(rate=config.hidden_dropout_prob)
+ self.config = config
+
+ def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, training: bool = False) -> tf.Tensor:
+ hidden_states = self.dense(inputs=hidden_states)
+ hidden_states = self.dropout(inputs=hidden_states, training=training)
+ hidden_states = self.LayerNorm(inputs=hidden_states + input_tensor)
+
+ return hidden_states
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "dense", None) is not None:
+ with tf.name_scope(self.dense.name):
+ self.dense.build([None, None, self.config.hidden_size])
+ if getattr(self, "LayerNorm", None) is not None:
+ with tf.name_scope(self.LayerNorm.name):
+ self.LayerNorm.build([None, None, self.config.hidden_size])
+
+
+# Copied from transformers.models.bert.modeling_tf_bert.TFBertAttention with Bert->Camembert
+class TFCamembertAttention(tf.keras.layers.Layer):
+ def __init__(self, config: CamembertConfig, **kwargs):
+ super().__init__(**kwargs)
+
+ self.self_attention = TFCamembertSelfAttention(config, name="self")
+ self.dense_output = TFCamembertSelfOutput(config, name="output")
+
+ def prune_heads(self, heads):
+ raise NotImplementedError
+
+ def call(
+ self,
+ input_tensor: tf.Tensor,
+ attention_mask: tf.Tensor,
+ head_mask: tf.Tensor,
+ encoder_hidden_states: tf.Tensor,
+ encoder_attention_mask: tf.Tensor,
+ past_key_value: Tuple[tf.Tensor],
+ output_attentions: bool,
+ training: bool = False,
+ ) -> Tuple[tf.Tensor]:
+ self_outputs = self.self_attention(
+ hidden_states=input_tensor,
+ attention_mask=attention_mask,
+ head_mask=head_mask,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ past_key_value=past_key_value,
+ output_attentions=output_attentions,
+ training=training,
+ )
+ attention_output = self.dense_output(
+ hidden_states=self_outputs[0], input_tensor=input_tensor, training=training
+ )
+ # add attentions (possibly with past_key_value) if we output them
+ outputs = (attention_output,) + self_outputs[1:]
+
+ return outputs
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "self_attention", None) is not None:
+ with tf.name_scope(self.self_attention.name):
+ self.self_attention.build(None)
+ if getattr(self, "dense_output", None) is not None:
+ with tf.name_scope(self.dense_output.name):
+ self.dense_output.build(None)
+
+
+# Copied from transformers.models.bert.modeling_tf_bert.TFBertIntermediate with Bert->Camembert
+class TFCamembertIntermediate(tf.keras.layers.Layer):
+ def __init__(self, config: CamembertConfig, **kwargs):
+ super().__init__(**kwargs)
+
+ self.dense = tf.keras.layers.Dense(
+ units=config.intermediate_size, kernel_initializer=get_initializer(config.initializer_range), name="dense"
+ )
+
+ if isinstance(config.hidden_act, str):
+ self.intermediate_act_fn = get_tf_activation(config.hidden_act)
+ else:
+ self.intermediate_act_fn = config.hidden_act
+ self.config = config
+
+ def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
+ hidden_states = self.dense(inputs=hidden_states)
+ hidden_states = self.intermediate_act_fn(hidden_states)
+
+ return hidden_states
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "dense", None) is not None:
+ with tf.name_scope(self.dense.name):
+ self.dense.build([None, None, self.config.hidden_size])
+
+
+# Copied from transformers.models.bert.modeling_tf_bert.TFBertOutput with Bert->Camembert
+class TFCamembertOutput(tf.keras.layers.Layer):
+ def __init__(self, config: CamembertConfig, **kwargs):
+ super().__init__(**kwargs)
+
+ self.dense = tf.keras.layers.Dense(
+ units=config.hidden_size, kernel_initializer=get_initializer(config.initializer_range), name="dense"
+ )
+ self.LayerNorm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_eps, name="LayerNorm")
+ self.dropout = tf.keras.layers.Dropout(rate=config.hidden_dropout_prob)
+ self.config = config
+
+ def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, training: bool = False) -> tf.Tensor:
+ hidden_states = self.dense(inputs=hidden_states)
+ hidden_states = self.dropout(inputs=hidden_states, training=training)
+ hidden_states = self.LayerNorm(inputs=hidden_states + input_tensor)
+
+ return hidden_states
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "dense", None) is not None:
+ with tf.name_scope(self.dense.name):
+ self.dense.build([None, None, self.config.intermediate_size])
+ if getattr(self, "LayerNorm", None) is not None:
+ with tf.name_scope(self.LayerNorm.name):
+ self.LayerNorm.build([None, None, self.config.hidden_size])
+
+
+# Copied from transformers.models.bert.modeling_tf_bert.TFBertLayer with Bert->Camembert
+class TFCamembertLayer(tf.keras.layers.Layer):
+ def __init__(self, config: CamembertConfig, **kwargs):
+ super().__init__(**kwargs)
+
+ self.attention = TFCamembertAttention(config, name="attention")
+ self.is_decoder = config.is_decoder
+ self.add_cross_attention = config.add_cross_attention
+ if self.add_cross_attention:
+ if not self.is_decoder:
+ raise ValueError(f"{self} should be used as a decoder model if cross attention is added")
+ self.crossattention = TFCamembertAttention(config, name="crossattention")
+ self.intermediate = TFCamembertIntermediate(config, name="intermediate")
+ self.bert_output = TFCamembertOutput(config, name="output")
+
+ def call(
+ self,
+ hidden_states: tf.Tensor,
+ attention_mask: tf.Tensor,
+ head_mask: tf.Tensor,
+ encoder_hidden_states: tf.Tensor | None,
+ encoder_attention_mask: tf.Tensor | None,
+ past_key_value: Tuple[tf.Tensor] | None,
+ output_attentions: bool,
+ training: bool = False,
+ ) -> Tuple[tf.Tensor]:
+ # decoder uni-directional self-attention cached key/values tuple is at positions 1,2
+ self_attn_past_key_value = past_key_value[:2] if past_key_value is not None else None
+ self_attention_outputs = self.attention(
+ input_tensor=hidden_states,
+ attention_mask=attention_mask,
+ head_mask=head_mask,
+ encoder_hidden_states=None,
+ encoder_attention_mask=None,
+ past_key_value=self_attn_past_key_value,
+ output_attentions=output_attentions,
+ training=training,
+ )
+ attention_output = self_attention_outputs[0]
+
+ # if decoder, the last output is tuple of self-attn cache
+ if self.is_decoder:
+ outputs = self_attention_outputs[1:-1]
+ present_key_value = self_attention_outputs[-1]
+ else:
+ outputs = self_attention_outputs[1:] # add self attentions if we output attention weights
+
+ cross_attn_present_key_value = None
+ if self.is_decoder and encoder_hidden_states is not None:
+ if not hasattr(self, "crossattention"):
+ raise ValueError(
+ f"If `encoder_hidden_states` are passed, {self} has to be instantiated with cross-attention layers"
+ " by setting `config.add_cross_attention=True`"
+ )
+
+ # cross_attn cached key/values tuple is at positions 3,4 of past_key_value tuple
+ cross_attn_past_key_value = past_key_value[-2:] if past_key_value is not None else None
+ cross_attention_outputs = self.crossattention(
+ input_tensor=attention_output,
+ attention_mask=attention_mask,
+ head_mask=head_mask,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ past_key_value=cross_attn_past_key_value,
+ output_attentions=output_attentions,
+ training=training,
+ )
+ attention_output = cross_attention_outputs[0]
+ outputs = outputs + cross_attention_outputs[1:-1] # add cross attentions if we output attention weights
+
+ # add cross-attn cache to positions 3,4 of present_key_value tuple
+ cross_attn_present_key_value = cross_attention_outputs[-1]
+ present_key_value = present_key_value + cross_attn_present_key_value
+
+ intermediate_output = self.intermediate(hidden_states=attention_output)
+ layer_output = self.bert_output(
+ hidden_states=intermediate_output, input_tensor=attention_output, training=training
+ )
+ outputs = (layer_output,) + outputs # add attentions if we output them
+
+ # if decoder, return the attn key/values as the last output
+ if self.is_decoder:
+ outputs = outputs + (present_key_value,)
+
+ return outputs
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "attention", None) is not None:
+ with tf.name_scope(self.attention.name):
+ self.attention.build(None)
+ if getattr(self, "intermediate", None) is not None:
+ with tf.name_scope(self.intermediate.name):
+ self.intermediate.build(None)
+ if getattr(self, "bert_output", None) is not None:
+ with tf.name_scope(self.bert_output.name):
+ self.bert_output.build(None)
+ if getattr(self, "crossattention", None) is not None:
+ with tf.name_scope(self.crossattention.name):
+ self.crossattention.build(None)
+
+
+# Copied from transformers.models.bert.modeling_tf_bert.TFBertEncoder with Bert->Camembert
+class TFCamembertEncoder(tf.keras.layers.Layer):
+ def __init__(self, config: CamembertConfig, **kwargs):
+ super().__init__(**kwargs)
+ self.config = config
+ self.layer = [TFCamembertLayer(config, name=f"layer_._{i}") for i in range(config.num_hidden_layers)]
+
+ def call(
+ self,
+ hidden_states: tf.Tensor,
+ attention_mask: tf.Tensor,
+ head_mask: tf.Tensor,
+ encoder_hidden_states: tf.Tensor | None,
+ encoder_attention_mask: tf.Tensor | None,
+ past_key_values: Tuple[Tuple[tf.Tensor]] | None,
+ use_cache: Optional[bool],
+ output_attentions: bool,
+ output_hidden_states: bool,
+ return_dict: bool,
+ training: bool = False,
+ ) -> Union[TFBaseModelOutputWithPastAndCrossAttentions, Tuple[tf.Tensor]]:
+ all_hidden_states = () if output_hidden_states else None
+ all_attentions = () if output_attentions else None
+ all_cross_attentions = () if output_attentions and self.config.add_cross_attention else None
+
+ next_decoder_cache = () if use_cache else None
+ for i, layer_module in enumerate(self.layer):
+ if output_hidden_states:
+ all_hidden_states = all_hidden_states + (hidden_states,)
+
+ past_key_value = past_key_values[i] if past_key_values is not None else None
+
+ layer_outputs = layer_module(
+ hidden_states=hidden_states,
+ attention_mask=attention_mask,
+ head_mask=head_mask[i],
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ past_key_value=past_key_value,
+ output_attentions=output_attentions,
+ training=training,
+ )
+ hidden_states = layer_outputs[0]
+
+ if use_cache:
+ next_decoder_cache += (layer_outputs[-1],)
+
+ if output_attentions:
+ all_attentions = all_attentions + (layer_outputs[1],)
+ if self.config.add_cross_attention and encoder_hidden_states is not None:
+ all_cross_attentions = all_cross_attentions + (layer_outputs[2],)
+
+ # Add last layer
+ if output_hidden_states:
+ all_hidden_states = all_hidden_states + (hidden_states,)
+
+ if not return_dict:
+ return tuple(
+ v for v in [hidden_states, all_hidden_states, all_attentions, all_cross_attentions] if v is not None
+ )
+
+ return TFBaseModelOutputWithPastAndCrossAttentions(
+ last_hidden_state=hidden_states,
+ past_key_values=next_decoder_cache,
+ hidden_states=all_hidden_states,
+ attentions=all_attentions,
+ cross_attentions=all_cross_attentions,
+ )
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "layer", None) is not None:
+ for layer in self.layer:
+ with tf.name_scope(layer.name):
+ layer.build(None)
+
+
+@keras_serializable
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaMainLayer with Roberta->Camembert
+class TFCamembertMainLayer(tf.keras.layers.Layer):
+ config_class = CamembertConfig
+
+ def __init__(self, config, add_pooling_layer=True, **kwargs):
+ super().__init__(**kwargs)
+
+ self.config = config
+ self.is_decoder = config.is_decoder
+
+ self.num_hidden_layers = config.num_hidden_layers
+ self.initializer_range = config.initializer_range
+ self.output_attentions = config.output_attentions
+ self.output_hidden_states = config.output_hidden_states
+ self.return_dict = config.use_return_dict
+ self.encoder = TFCamembertEncoder(config, name="encoder")
+ self.pooler = TFCamembertPooler(config, name="pooler") if add_pooling_layer else None
+ # The embeddings must be the last declaration in order to follow the weights order
+ self.embeddings = TFCamembertEmbeddings(config, name="embeddings")
+
+ # Copied from transformers.models.bert.modeling_tf_bert.TFBertMainLayer.get_input_embeddings
+ def get_input_embeddings(self) -> tf.keras.layers.Layer:
+ return self.embeddings
+
+ # Copied from transformers.models.bert.modeling_tf_bert.TFBertMainLayer.set_input_embeddings
+ def set_input_embeddings(self, value: tf.Variable):
+ self.embeddings.weight = value
+ self.embeddings.vocab_size = shape_list(value)[0]
+
+ # Copied from transformers.models.bert.modeling_tf_bert.TFBertMainLayer._prune_heads
+ def _prune_heads(self, heads_to_prune):
+ """
+ Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base
+ class PreTrainedModel
+ """
+ raise NotImplementedError
+
+ @unpack_inputs
+ # Copied from transformers.models.bert.modeling_tf_bert.TFBertMainLayer.call
+ def call(
+ self,
+ input_ids: TFModelInputType | None = None,
+ attention_mask: np.ndarray | tf.Tensor | None = None,
+ token_type_ids: np.ndarray | tf.Tensor | None = None,
+ position_ids: np.ndarray | tf.Tensor | None = None,
+ head_mask: np.ndarray | tf.Tensor | None = None,
+ inputs_embeds: np.ndarray | tf.Tensor | None = None,
+ encoder_hidden_states: np.ndarray | tf.Tensor | None = None,
+ encoder_attention_mask: np.ndarray | tf.Tensor | None = None,
+ past_key_values: Optional[Tuple[Tuple[Union[np.ndarray, tf.Tensor]]]] = None,
+ use_cache: Optional[bool] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ training: bool = False,
+ ) -> Union[TFBaseModelOutputWithPoolingAndCrossAttentions, Tuple[tf.Tensor]]:
+ if not self.config.is_decoder:
+ use_cache = False
+
+ if input_ids is not None and inputs_embeds is not None:
+ raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+ elif input_ids is not None:
+ input_shape = shape_list(input_ids)
+ elif inputs_embeds is not None:
+ input_shape = shape_list(inputs_embeds)[:-1]
+ else:
+ raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+ batch_size, seq_length = input_shape
+
+ if past_key_values is None:
+ past_key_values_length = 0
+ past_key_values = [None] * len(self.encoder.layer)
+ else:
+ past_key_values_length = shape_list(past_key_values[0][0])[-2]
+
+ if attention_mask is None:
+ attention_mask = tf.fill(dims=(batch_size, seq_length + past_key_values_length), value=1)
+
+ if token_type_ids is None:
+ token_type_ids = tf.fill(dims=input_shape, value=0)
+
+ embedding_output = self.embeddings(
+ input_ids=input_ids,
+ position_ids=position_ids,
+ token_type_ids=token_type_ids,
+ inputs_embeds=inputs_embeds,
+ past_key_values_length=past_key_values_length,
+ training=training,
+ )
+
+ # We create a 3D attention mask from a 2D tensor mask.
+ # Sizes are [batch_size, 1, 1, to_seq_length]
+ # So we can broadcast to [batch_size, num_heads, from_seq_length, to_seq_length]
+ # this attention mask is more simple than the triangular masking of causal attention
+ # used in OpenAI GPT, we just need to prepare the broadcast dimension here.
+ attention_mask_shape = shape_list(attention_mask)
+
+ mask_seq_length = seq_length + past_key_values_length
+ # Copied from `modeling_tf_t5.py`
+ # Provided a padding mask of dimensions [batch_size, mask_seq_length]
+ # - if the model is a decoder, apply a causal mask in addition to the padding mask
+ # - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, mask_seq_length, mask_seq_length]
+ if self.is_decoder:
+ seq_ids = tf.range(mask_seq_length)
+ causal_mask = tf.less_equal(
+ tf.tile(seq_ids[None, None, :], (batch_size, mask_seq_length, 1)),
+ seq_ids[None, :, None],
+ )
+ causal_mask = tf.cast(causal_mask, dtype=attention_mask.dtype)
+ extended_attention_mask = causal_mask * attention_mask[:, None, :]
+ attention_mask_shape = shape_list(extended_attention_mask)
+ extended_attention_mask = tf.reshape(
+ extended_attention_mask, (attention_mask_shape[0], 1, attention_mask_shape[1], attention_mask_shape[2])
+ )
+ if past_key_values[0] is not None:
+ # attention_mask needs to be sliced to the shape `[batch_size, 1, from_seq_length - cached_seq_length, to_seq_length]
+ extended_attention_mask = extended_attention_mask[:, :, -seq_length:, :]
+ else:
+ extended_attention_mask = tf.reshape(
+ attention_mask, (attention_mask_shape[0], 1, 1, attention_mask_shape[1])
+ )
+
+ # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
+ # masked positions, this operation will create a tensor which is 0.0 for
+ # positions we want to attend and -10000.0 for masked positions.
+ # Since we are adding it to the raw scores before the softmax, this is
+ # effectively the same as removing these entirely.
+ extended_attention_mask = tf.cast(extended_attention_mask, dtype=embedding_output.dtype)
+ one_cst = tf.constant(1.0, dtype=embedding_output.dtype)
+ ten_thousand_cst = tf.constant(-10000.0, dtype=embedding_output.dtype)
+ extended_attention_mask = tf.multiply(tf.subtract(one_cst, extended_attention_mask), ten_thousand_cst)
+
+ # Copied from `modeling_tf_t5.py` with -1e9 -> -10000
+ if self.is_decoder and encoder_attention_mask is not None:
+ # If a 2D ou 3D attention mask is provided for the cross-attention
+ # we need to make broadcastable to [batch_size, num_heads, mask_seq_length, mask_seq_length]
+ # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
+ encoder_attention_mask = tf.cast(encoder_attention_mask, dtype=extended_attention_mask.dtype)
+ num_dims_encoder_attention_mask = len(shape_list(encoder_attention_mask))
+ if num_dims_encoder_attention_mask == 3:
+ encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
+ if num_dims_encoder_attention_mask == 2:
+ encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
+
+ # T5 has a mask that can compare sequence ids, we can simulate this here with this transposition
+ # Cf. https://github.com/tensorflow/mesh/blob/8d2465e9bc93129b913b5ccc6a59aa97abd96ec6/mesh_tensorflow/transformer/transformer_layers.py#L270
+ # encoder_extended_attention_mask = tf.math.equal(encoder_extended_attention_mask,
+ # tf.transpose(encoder_extended_attention_mask, perm=(-1, -2)))
+
+ encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
+ else:
+ encoder_extended_attention_mask = None
+
+ # Prepare head mask if needed
+ # 1.0 in head_mask indicate we keep the head
+ # attention_probs has shape bsz x n_heads x N x N
+ # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+ # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+ if head_mask is not None:
+ raise NotImplementedError
+ else:
+ head_mask = [None] * self.config.num_hidden_layers
+
+ encoder_outputs = self.encoder(
+ hidden_states=embedding_output,
+ attention_mask=extended_attention_mask,
+ head_mask=head_mask,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_extended_attention_mask,
+ past_key_values=past_key_values,
+ use_cache=use_cache,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ training=training,
+ )
+
+ sequence_output = encoder_outputs[0]
+ pooled_output = self.pooler(hidden_states=sequence_output) if self.pooler is not None else None
+
+ if not return_dict:
+ return (
+ sequence_output,
+ pooled_output,
+ ) + encoder_outputs[1:]
+
+ return TFBaseModelOutputWithPoolingAndCrossAttentions(
+ last_hidden_state=sequence_output,
+ pooler_output=pooled_output,
+ past_key_values=encoder_outputs.past_key_values,
+ hidden_states=encoder_outputs.hidden_states,
+ attentions=encoder_outputs.attentions,
+ cross_attentions=encoder_outputs.cross_attentions,
+ )
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "encoder", None) is not None:
+ with tf.name_scope(self.encoder.name):
+ self.encoder.build(None)
+ if getattr(self, "pooler", None) is not None:
+ with tf.name_scope(self.pooler.name):
+ self.pooler.build(None)
+ if getattr(self, "embeddings", None) is not None:
+ with tf.name_scope(self.embeddings.name):
+ self.embeddings.build(None)
+
+
+class TFCamembertPreTrainedModel(TFPreTrainedModel):
+ """
+ An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
+ models.
+ """
+
+ config_class = CamembertConfig
+ base_model_prefix = "roberta"
+
+
+@add_start_docstrings(
+ "The bare CamemBERT Model transformer outputting raw hidden-states without any specific head on top.",
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaModel with Roberta->Camembert, ROBERTA->CAMEMBERT
+class TFCamembertModel(TFCamembertPreTrainedModel):
+ def __init__(self, config, *inputs, **kwargs):
+ super().__init__(config, *inputs, **kwargs)
+ self.roberta = TFCamembertMainLayer(config, name="roberta")
+
+ @unpack_inputs
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint=_CHECKPOINT_FOR_DOC,
+ output_type=TFBaseModelOutputWithPoolingAndCrossAttentions,
+ config_class=_CONFIG_FOR_DOC,
+ )
+ def call(
+ self,
+ input_ids: TFModelInputType | None = None,
+ attention_mask: np.ndarray | tf.Tensor | None = None,
+ token_type_ids: np.ndarray | tf.Tensor | None = None,
+ position_ids: np.ndarray | tf.Tensor | None = None,
+ head_mask: np.ndarray | tf.Tensor | None = None,
+ inputs_embeds: np.ndarray | tf.Tensor | None = None,
+ encoder_hidden_states: np.ndarray | tf.Tensor | None = None,
+ encoder_attention_mask: np.ndarray | tf.Tensor | None = None,
+ past_key_values: Optional[Tuple[Tuple[Union[np.ndarray, tf.Tensor]]]] = None,
+ use_cache: Optional[bool] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ training: Optional[bool] = False,
+ ) -> Union[Tuple, TFBaseModelOutputWithPoolingAndCrossAttentions]:
+ r"""
+ encoder_hidden_states (`tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+ Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
+ the model is configured as a decoder.
+ encoder_attention_mask (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
+ the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
+
+ - 1 for tokens that are **not masked**,
+ - 0 for tokens that are **masked**.
+
+ past_key_values (`Tuple[Tuple[tf.Tensor]]` of length `config.n_layers`)
+ contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.
+ If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
+ don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
+ `decoder_input_ids` of shape `(batch_size, sequence_length)`.
+ use_cache (`bool`, *optional*, defaults to `True`):
+ If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+ `past_key_values`). Set to `False` during training, `True` during generation
+ """
+ outputs = self.roberta(
+ input_ids=input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ past_key_values=past_key_values,
+ use_cache=use_cache,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ training=training,
+ )
+
+ return outputs
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "roberta", None) is not None:
+ with tf.name_scope(self.roberta.name):
+ self.roberta.build(None)
+
+
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaLMHead with Roberta->Camembert
+class TFCamembertLMHead(tf.keras.layers.Layer):
+ """Camembert Head for masked language modeling."""
+
+ def __init__(self, config, input_embeddings, **kwargs):
+ super().__init__(**kwargs)
+
+ self.config = config
+ self.hidden_size = config.hidden_size
+ self.dense = tf.keras.layers.Dense(
+ config.hidden_size, kernel_initializer=get_initializer(config.initializer_range), name="dense"
+ )
+ self.layer_norm = tf.keras.layers.LayerNormalization(epsilon=config.layer_norm_eps, name="layer_norm")
+ self.act = get_tf_activation("gelu")
+
+ # The output weights are the same as the input embeddings, but there is
+ # an output-only bias for each token.
+ self.decoder = input_embeddings
+
+ def build(self, input_shape=None):
+ self.bias = self.add_weight(shape=(self.config.vocab_size,), initializer="zeros", trainable=True, name="bias")
+
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "dense", None) is not None:
+ with tf.name_scope(self.dense.name):
+ self.dense.build([None, None, self.config.hidden_size])
+ if getattr(self, "layer_norm", None) is not None:
+ with tf.name_scope(self.layer_norm.name):
+ self.layer_norm.build([None, None, self.config.hidden_size])
+
+ def get_output_embeddings(self):
+ return self.decoder
+
+ def set_output_embeddings(self, value):
+ self.decoder.weight = value
+ self.decoder.vocab_size = shape_list(value)[0]
+
+ def get_bias(self):
+ return {"bias": self.bias}
+
+ def set_bias(self, value):
+ self.bias = value["bias"]
+ self.config.vocab_size = shape_list(value["bias"])[0]
+
+ def call(self, hidden_states):
+ hidden_states = self.dense(hidden_states)
+ hidden_states = self.act(hidden_states)
+ hidden_states = self.layer_norm(hidden_states)
+
+ # project back to size of vocabulary with bias
+ seq_length = shape_list(tensor=hidden_states)[1]
+ hidden_states = tf.reshape(tensor=hidden_states, shape=[-1, self.hidden_size])
+ hidden_states = tf.matmul(a=hidden_states, b=self.decoder.weight, transpose_b=True)
+ hidden_states = tf.reshape(tensor=hidden_states, shape=[-1, seq_length, self.config.vocab_size])
+ hidden_states = tf.nn.bias_add(value=hidden_states, bias=self.bias)
+
+ return hidden_states
+
+
+@add_start_docstrings(
+ """CamemBERT Model with a `language modeling` head on top.""",
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaForMaskedLM with Roberta->Camembert, ROBERTA->CAMEMBERT
+class TFCamembertForMaskedLM(TFCamembertPreTrainedModel, TFMaskedLanguageModelingLoss):
+ # names with a '.' represents the authorized unexpected/missing layers when a TF model is loaded from a PT model
+ _keys_to_ignore_on_load_unexpected = [r"pooler", r"lm_head.decoder.weight"]
+
+ def __init__(self, config, *inputs, **kwargs):
+ super().__init__(config, *inputs, **kwargs)
+
+ self.roberta = TFCamembertMainLayer(config, add_pooling_layer=False, name="roberta")
+ self.lm_head = TFCamembertLMHead(config, self.roberta.embeddings, name="lm_head")
+
+ def get_lm_head(self):
+ return self.lm_head
+
+ def get_prefix_bias_name(self):
+ warnings.warn("The method get_prefix_bias_name is deprecated. Please use `get_bias` instead.", FutureWarning)
+ return self.name + "/" + self.lm_head.name
+
+ @unpack_inputs
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint=_CHECKPOINT_FOR_DOC,
+ output_type=TFMaskedLMOutput,
+ config_class=_CONFIG_FOR_DOC,
+ mask="",
+ expected_output="' Paris'",
+ expected_loss=0.1,
+ )
+ def call(
+ self,
+ input_ids: TFModelInputType | None = None,
+ attention_mask: np.ndarray | tf.Tensor | None = None,
+ token_type_ids: np.ndarray | tf.Tensor | None = None,
+ position_ids: np.ndarray | tf.Tensor | None = None,
+ head_mask: np.ndarray | tf.Tensor | None = None,
+ inputs_embeds: np.ndarray | tf.Tensor | None = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ labels: np.ndarray | tf.Tensor | None = None,
+ training: Optional[bool] = False,
+ ) -> Union[TFMaskedLMOutput, Tuple[tf.Tensor]]:
+ r"""
+ labels (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
+ config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked), the
+ loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
+ """
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ training=training,
+ )
+
+ sequence_output = outputs[0]
+ prediction_scores = self.lm_head(sequence_output)
+
+ loss = None if labels is None else self.hf_compute_loss(labels, prediction_scores)
+
+ if not return_dict:
+ output = (prediction_scores,) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return TFMaskedLMOutput(
+ loss=loss,
+ logits=prediction_scores,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "roberta", None) is not None:
+ with tf.name_scope(self.roberta.name):
+ self.roberta.build(None)
+ if getattr(self, "lm_head", None) is not None:
+ with tf.name_scope(self.lm_head.name):
+ self.lm_head.build(None)
+
+
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaClassificationHead
+class TFCamembertClassificationHead(tf.keras.layers.Layer):
+ """Head for sentence-level classification tasks."""
+
+ def __init__(self, config, **kwargs):
+ super().__init__(**kwargs)
+ self.dense = tf.keras.layers.Dense(
+ config.hidden_size,
+ kernel_initializer=get_initializer(config.initializer_range),
+ activation="tanh",
+ name="dense",
+ )
+ classifier_dropout = (
+ config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+ )
+ self.dropout = tf.keras.layers.Dropout(classifier_dropout)
+ self.out_proj = tf.keras.layers.Dense(
+ config.num_labels, kernel_initializer=get_initializer(config.initializer_range), name="out_proj"
+ )
+ self.config = config
+
+ def call(self, features, training=False):
+ x = features[:, 0, :] # take token (equiv. to [CLS])
+ x = self.dropout(x, training=training)
+ x = self.dense(x)
+ x = self.dropout(x, training=training)
+ x = self.out_proj(x)
+ return x
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "dense", None) is not None:
+ with tf.name_scope(self.dense.name):
+ self.dense.build([None, None, self.config.hidden_size])
+ if getattr(self, "out_proj", None) is not None:
+ with tf.name_scope(self.out_proj.name):
+ self.out_proj.build([None, None, self.config.hidden_size])
+
+
+@add_start_docstrings(
+ """
+ CamemBERT Model transformer with a sequence classification/regression head on top (a linear layer on top of the
+ pooled output) e.g. for GLUE tasks.
+ """,
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaForSequenceClassification with Roberta->Camembert, ROBERTA->CAMEMBERT
+class TFCamembertForSequenceClassification(TFCamembertPreTrainedModel, TFSequenceClassificationLoss):
+ # names with a '.' represents the authorized unexpected/missing layers when a TF model is loaded from a PT model
+ _keys_to_ignore_on_load_unexpected = [r"pooler", r"lm_head"]
+
+ def __init__(self, config, *inputs, **kwargs):
+ super().__init__(config, *inputs, **kwargs)
+ self.num_labels = config.num_labels
+
+ self.roberta = TFCamembertMainLayer(config, add_pooling_layer=False, name="roberta")
+ self.classifier = TFCamembertClassificationHead(config, name="classifier")
+
+ @unpack_inputs
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint="cardiffnlp/twitter-roberta-base-emotion",
+ output_type=TFSequenceClassifierOutput,
+ config_class=_CONFIG_FOR_DOC,
+ expected_output="'optimism'",
+ expected_loss=0.08,
+ )
+ def call(
+ self,
+ input_ids: TFModelInputType | None = None,
+ attention_mask: np.ndarray | tf.Tensor | None = None,
+ token_type_ids: np.ndarray | tf.Tensor | None = None,
+ position_ids: np.ndarray | tf.Tensor | None = None,
+ head_mask: np.ndarray | tf.Tensor | None = None,
+ inputs_embeds: np.ndarray | tf.Tensor | None = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ labels: np.ndarray | tf.Tensor | None = None,
+ training: Optional[bool] = False,
+ ) -> Union[TFSequenceClassifierOutput, Tuple[tf.Tensor]]:
+ r"""
+ labels (`tf.Tensor` of shape `(batch_size,)`, *optional*):
+ Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+ config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+ `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+ """
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ training=training,
+ )
+ sequence_output = outputs[0]
+ logits = self.classifier(sequence_output, training=training)
+
+ loss = None if labels is None else self.hf_compute_loss(labels, logits)
+
+ if not return_dict:
+ output = (logits,) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return TFSequenceClassifierOutput(
+ loss=loss,
+ logits=logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "roberta", None) is not None:
+ with tf.name_scope(self.roberta.name):
+ self.roberta.build(None)
+ if getattr(self, "classifier", None) is not None:
+ with tf.name_scope(self.classifier.name):
+ self.classifier.build(None)
+
+
+@add_start_docstrings(
+ """
+ CamemBERT Model with a token classification head on top (a linear layer on top of the hidden-states output) e.g.
+ for Named-Entity-Recognition (NER) tasks.
+ """,
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaForTokenClassification with Roberta->Camembert, ROBERTA->CAMEMBERT
+class TFCamembertForTokenClassification(TFCamembertPreTrainedModel, TFTokenClassificationLoss):
+ # names with a '.' represents the authorized unexpected/missing layers when a TF model is loaded from a PT model
+ _keys_to_ignore_on_load_unexpected = [r"pooler", r"lm_head"]
+ _keys_to_ignore_on_load_missing = [r"dropout"]
+
+ def __init__(self, config, *inputs, **kwargs):
+ super().__init__(config, *inputs, **kwargs)
+ self.num_labels = config.num_labels
+
+ self.roberta = TFCamembertMainLayer(config, add_pooling_layer=False, name="roberta")
+ classifier_dropout = (
+ config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
+ )
+ self.dropout = tf.keras.layers.Dropout(classifier_dropout)
+ self.classifier = tf.keras.layers.Dense(
+ config.num_labels, kernel_initializer=get_initializer(config.initializer_range), name="classifier"
+ )
+ self.config = config
+
+ @unpack_inputs
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint="ydshieh/roberta-large-ner-english",
+ output_type=TFTokenClassifierOutput,
+ config_class=_CONFIG_FOR_DOC,
+ expected_output="['O', 'ORG', 'ORG', 'O', 'O', 'O', 'O', 'O', 'LOC', 'O', 'LOC', 'LOC']",
+ expected_loss=0.01,
+ )
+ def call(
+ self,
+ input_ids: TFModelInputType | None = None,
+ attention_mask: np.ndarray | tf.Tensor | None = None,
+ token_type_ids: np.ndarray | tf.Tensor | None = None,
+ position_ids: np.ndarray | tf.Tensor | None = None,
+ head_mask: np.ndarray | tf.Tensor | None = None,
+ inputs_embeds: np.ndarray | tf.Tensor | None = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ labels: np.ndarray | tf.Tensor | None = None,
+ training: Optional[bool] = False,
+ ) -> Union[TFTokenClassifierOutput, Tuple[tf.Tensor]]:
+ r"""
+ labels (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
+ """
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ training=training,
+ )
+ sequence_output = outputs[0]
+
+ sequence_output = self.dropout(sequence_output, training=training)
+ logits = self.classifier(sequence_output)
+
+ loss = None if labels is None else self.hf_compute_loss(labels, logits)
+
+ if not return_dict:
+ output = (logits,) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return TFTokenClassifierOutput(
+ loss=loss,
+ logits=logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "roberta", None) is not None:
+ with tf.name_scope(self.roberta.name):
+ self.roberta.build(None)
+ if getattr(self, "classifier", None) is not None:
+ with tf.name_scope(self.classifier.name):
+ self.classifier.build([None, None, self.config.hidden_size])
+
+
+@add_start_docstrings(
+ """
+ CamemBERT Model with a multiple choice classification head on top (a linear layer on top of the pooled output and a
+ softmax) e.g. for RocStories/SWAG tasks.
+ """,
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaForMultipleChoice with Roberta->Camembert, ROBERTA->CAMEMBERT
+class TFCamembertForMultipleChoice(TFCamembertPreTrainedModel, TFMultipleChoiceLoss):
+ # names with a '.' represents the authorized unexpected/missing layers when a TF model is loaded from a PT model
+ _keys_to_ignore_on_load_unexpected = [r"lm_head"]
+ _keys_to_ignore_on_load_missing = [r"dropout"]
+
+ def __init__(self, config, *inputs, **kwargs):
+ super().__init__(config, *inputs, **kwargs)
+
+ self.roberta = TFCamembertMainLayer(config, name="roberta")
+ self.dropout = tf.keras.layers.Dropout(config.hidden_dropout_prob)
+ self.classifier = tf.keras.layers.Dense(
+ 1, kernel_initializer=get_initializer(config.initializer_range), name="classifier"
+ )
+ self.config = config
+
+ @unpack_inputs
+ @add_start_docstrings_to_model_forward(
+ CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, num_choices, sequence_length")
+ )
+ @add_code_sample_docstrings(
+ checkpoint=_CHECKPOINT_FOR_DOC,
+ output_type=TFMultipleChoiceModelOutput,
+ config_class=_CONFIG_FOR_DOC,
+ )
+ def call(
+ self,
+ input_ids: TFModelInputType | None = None,
+ attention_mask: np.ndarray | tf.Tensor | None = None,
+ token_type_ids: np.ndarray | tf.Tensor | None = None,
+ position_ids: np.ndarray | tf.Tensor | None = None,
+ head_mask: np.ndarray | tf.Tensor | None = None,
+ inputs_embeds: np.ndarray | tf.Tensor | None = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ labels: np.ndarray | tf.Tensor | None = None,
+ training: Optional[bool] = False,
+ ) -> Union[TFMultipleChoiceModelOutput, Tuple[tf.Tensor]]:
+ r"""
+ labels (`tf.Tensor` of shape `(batch_size,)`, *optional*):
+ Labels for computing the multiple choice classification loss. Indices should be in `[0, ..., num_choices]`
+ where `num_choices` is the size of the second dimension of the input tensors. (See `input_ids` above)
+ """
+
+ if input_ids is not None:
+ num_choices = shape_list(input_ids)[1]
+ seq_length = shape_list(input_ids)[2]
+ else:
+ num_choices = shape_list(inputs_embeds)[1]
+ seq_length = shape_list(inputs_embeds)[2]
+
+ flat_input_ids = tf.reshape(input_ids, (-1, seq_length)) if input_ids is not None else None
+ flat_attention_mask = tf.reshape(attention_mask, (-1, seq_length)) if attention_mask is not None else None
+ flat_token_type_ids = tf.reshape(token_type_ids, (-1, seq_length)) if token_type_ids is not None else None
+ flat_position_ids = tf.reshape(position_ids, (-1, seq_length)) if position_ids is not None else None
+ outputs = self.roberta(
+ flat_input_ids,
+ flat_attention_mask,
+ flat_token_type_ids,
+ flat_position_ids,
+ head_mask,
+ inputs_embeds,
+ output_attentions,
+ output_hidden_states,
+ return_dict=return_dict,
+ training=training,
+ )
+ pooled_output = outputs[1]
+ pooled_output = self.dropout(pooled_output, training=training)
+ logits = self.classifier(pooled_output)
+ reshaped_logits = tf.reshape(logits, (-1, num_choices))
+
+ loss = None if labels is None else self.hf_compute_loss(labels, reshaped_logits)
+
+ if not return_dict:
+ output = (reshaped_logits,) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return TFMultipleChoiceModelOutput(
+ loss=loss,
+ logits=reshaped_logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "roberta", None) is not None:
+ with tf.name_scope(self.roberta.name):
+ self.roberta.build(None)
+ if getattr(self, "classifier", None) is not None:
+ with tf.name_scope(self.classifier.name):
+ self.classifier.build([None, None, self.config.hidden_size])
+
+
+@add_start_docstrings(
+ """
+ CamemBERT Model with a span classification head on top for extractive question-answering tasks like SQuAD (a linear
+ layers on top of the hidden-states output to compute `span start logits` and `span end logits`).
+ """,
+ CAMEMBERT_START_DOCSTRING,
+)
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaForQuestionAnswering with Roberta->Camembert, ROBERTA->CAMEMBERT
+class TFCamembertForQuestionAnswering(TFCamembertPreTrainedModel, TFQuestionAnsweringLoss):
+ # names with a '.' represents the authorized unexpected/missing layers when a TF model is loaded from a PT model
+ _keys_to_ignore_on_load_unexpected = [r"pooler", r"lm_head"]
+
+ def __init__(self, config, *inputs, **kwargs):
+ super().__init__(config, *inputs, **kwargs)
+ self.num_labels = config.num_labels
+
+ self.roberta = TFCamembertMainLayer(config, add_pooling_layer=False, name="roberta")
+ self.qa_outputs = tf.keras.layers.Dense(
+ config.num_labels, kernel_initializer=get_initializer(config.initializer_range), name="qa_outputs"
+ )
+ self.config = config
+
+ @unpack_inputs
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint="ydshieh/roberta-base-squad2",
+ output_type=TFQuestionAnsweringModelOutput,
+ config_class=_CONFIG_FOR_DOC,
+ expected_output="' puppet'",
+ expected_loss=0.86,
+ )
+ def call(
+ self,
+ input_ids: TFModelInputType | None = None,
+ attention_mask: np.ndarray | tf.Tensor | None = None,
+ token_type_ids: np.ndarray | tf.Tensor | None = None,
+ position_ids: np.ndarray | tf.Tensor | None = None,
+ head_mask: np.ndarray | tf.Tensor | None = None,
+ inputs_embeds: np.ndarray | tf.Tensor | None = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ start_positions: np.ndarray | tf.Tensor | None = None,
+ end_positions: np.ndarray | tf.Tensor | None = None,
+ training: Optional[bool] = False,
+ ) -> Union[TFQuestionAnsweringModelOutput, Tuple[tf.Tensor]]:
+ r"""
+ start_positions (`tf.Tensor` of shape `(batch_size,)`, *optional*):
+ Labels for position (index) of the start of the labelled span for computing the token classification loss.
+ Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
+ are not taken into account for computing the loss.
+ end_positions (`tf.Tensor` of shape `(batch_size,)`, *optional*):
+ Labels for position (index) of the end of the labelled span for computing the token classification loss.
+ Positions are clamped to the length of the sequence (`sequence_length`). Position outside of the sequence
+ are not taken into account for computing the loss.
+ """
+ outputs = self.roberta(
+ input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ training=training,
+ )
+ sequence_output = outputs[0]
+
+ logits = self.qa_outputs(sequence_output)
+ start_logits, end_logits = tf.split(logits, 2, axis=-1)
+ start_logits = tf.squeeze(start_logits, axis=-1)
+ end_logits = tf.squeeze(end_logits, axis=-1)
+
+ loss = None
+ if start_positions is not None and end_positions is not None:
+ labels = {"start_position": start_positions}
+ labels["end_position"] = end_positions
+ loss = self.hf_compute_loss(labels, (start_logits, end_logits))
+
+ if not return_dict:
+ output = (start_logits, end_logits) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return TFQuestionAnsweringModelOutput(
+ loss=loss,
+ start_logits=start_logits,
+ end_logits=end_logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "roberta", None) is not None:
+ with tf.name_scope(self.roberta.name):
+ self.roberta.build(None)
+ if getattr(self, "qa_outputs", None) is not None:
+ with tf.name_scope(self.qa_outputs.name):
+ self.qa_outputs.build([None, None, self.config.hidden_size])
+
+
+@add_start_docstrings(
+ """CamemBERT Model with a `language modeling` head on top for CLM fine-tuning.""", CAMEMBERT_START_DOCSTRING
+)
+# Copied from transformers.models.roberta.modeling_tf_roberta.TFRobertaForCausalLM with Roberta->Camembert, ROBERTA->CAMEMBERT
+class TFCamembertForCausalLM(TFCamembertPreTrainedModel, TFCausalLanguageModelingLoss):
+ # names with a '.' represents the authorized unexpected/missing layers when a TF model is loaded from a PT model
+ _keys_to_ignore_on_load_unexpected = [r"pooler", r"lm_head.decoder.weight"]
+
+ def __init__(self, config: CamembertConfig, *inputs, **kwargs):
+ super().__init__(config, *inputs, **kwargs)
+
+ if not config.is_decoder:
+ logger.warning("If you want to use `TFCamembertLMHeadModel` as a standalone, add `is_decoder=True.`")
+
+ self.roberta = TFCamembertMainLayer(config, add_pooling_layer=False, name="roberta")
+ self.lm_head = TFCamembertLMHead(config, input_embeddings=self.roberta.embeddings, name="lm_head")
+
+ def get_lm_head(self):
+ return self.lm_head
+
+ def get_prefix_bias_name(self):
+ warnings.warn("The method get_prefix_bias_name is deprecated. Please use `get_bias` instead.", FutureWarning)
+ return self.name + "/" + self.lm_head.name
+
+ # Copied from transformers.models.bert.modeling_tf_bert.TFBertLMHeadModel.prepare_inputs_for_generation
+ def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, **model_kwargs):
+ input_shape = input_ids.shape
+ # if model is used as a decoder in encoder-decoder model, the decoder attention mask is created on the fly
+ if attention_mask is None:
+ attention_mask = tf.ones(input_shape)
+
+ # cut decoder_input_ids if past is used
+ if past_key_values is not None:
+ input_ids = input_ids[:, -1:]
+
+ return {"input_ids": input_ids, "attention_mask": attention_mask, "past_key_values": past_key_values}
+
+ @unpack_inputs
+ @add_start_docstrings_to_model_forward(CAMEMBERT_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
+ @add_code_sample_docstrings(
+ checkpoint=_CHECKPOINT_FOR_DOC,
+ output_type=TFCausalLMOutputWithCrossAttentions,
+ config_class=_CONFIG_FOR_DOC,
+ )
+ def call(
+ self,
+ input_ids: TFModelInputType | None = None,
+ attention_mask: np.ndarray | tf.Tensor | None = None,
+ token_type_ids: np.ndarray | tf.Tensor | None = None,
+ position_ids: np.ndarray | tf.Tensor | None = None,
+ head_mask: np.ndarray | tf.Tensor | None = None,
+ inputs_embeds: np.ndarray | tf.Tensor | None = None,
+ encoder_hidden_states: np.ndarray | tf.Tensor | None = None,
+ encoder_attention_mask: np.ndarray | tf.Tensor | None = None,
+ past_key_values: Optional[Tuple[Tuple[Union[np.ndarray, tf.Tensor]]]] = None,
+ use_cache: Optional[bool] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ labels: np.ndarray | tf.Tensor | None = None,
+ training: Optional[bool] = False,
+ ) -> Union[TFCausalLMOutputWithCrossAttentions, Tuple[tf.Tensor]]:
+ r"""
+ encoder_hidden_states (`tf.Tensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+ Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if
+ the model is configured as a decoder.
+ encoder_attention_mask (`tf.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+ Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in
+ the cross-attention if the model is configured as a decoder. Mask values selected in `[0, 1]`:
+
+ - 1 for tokens that are **not masked**,
+ - 0 for tokens that are **masked**.
+
+ past_key_values (`Tuple[Tuple[tf.Tensor]]` of length `config.n_layers`)
+ contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding.
+ If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
+ don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
+ `decoder_input_ids` of shape `(batch_size, sequence_length)`.
+ use_cache (`bool`, *optional*, defaults to `True`):
+ If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+ `past_key_values`). Set to `False` during training, `True` during generation
+ labels (`tf.Tensor` or `np.ndarray` of shape `(batch_size, sequence_length)`, *optional*):
+ Labels for computing the cross entropy classification loss. Indices should be in `[0, ...,
+ config.vocab_size - 1]`.
+ """
+ outputs = self.roberta(
+ input_ids=input_ids,
+ attention_mask=attention_mask,
+ token_type_ids=token_type_ids,
+ position_ids=position_ids,
+ head_mask=head_mask,
+ inputs_embeds=inputs_embeds,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ past_key_values=past_key_values,
+ use_cache=use_cache,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ training=training,
+ )
+
+ sequence_output = outputs[0]
+ logits = self.lm_head(hidden_states=sequence_output, training=training)
+ loss = None
+
+ if labels is not None:
+ # shift labels to the left and cut last logit token
+ shifted_logits = logits[:, :-1]
+ labels = labels[:, 1:]
+ loss = self.hf_compute_loss(labels=labels, logits=shifted_logits)
+
+ if not return_dict:
+ output = (logits,) + outputs[2:]
+ return ((loss,) + output) if loss is not None else output
+
+ return TFCausalLMOutputWithCrossAttentions(
+ loss=loss,
+ logits=logits,
+ past_key_values=outputs.past_key_values,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ cross_attentions=outputs.cross_attentions,
+ )
+
+ def build(self, input_shape=None):
+ if self.built:
+ return
+ self.built = True
+ if getattr(self, "roberta", None) is not None:
+ with tf.name_scope(self.roberta.name):
+ self.roberta.build(None)
+ if getattr(self, "lm_head", None) is not None:
+ with tf.name_scope(self.lm_head.name):
+ self.lm_head.build(None)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/tokenization_camembert.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/tokenization_camembert.py
new file mode 100644
index 0000000000000000000000000000000000000000..40755494901791370f7d072f8eb0fb398c85805d
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/camembert/tokenization_camembert.py
@@ -0,0 +1,330 @@
+# coding=utf-8
+# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License
+""" Tokenization classes for Camembert model."""
+
+
+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
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "sentencepiece.bpe.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+ "vocab_file": {
+ "camembert-base": "https://huggingface.co/camembert-base/resolve/main/sentencepiece.bpe.model",
+ }
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+ "camembert-base": 512,
+}
+
+SPIECE_UNDERLINE = "▁"
+
+
+class CamembertTokenizer(PreTrainedTokenizer):
+ """
+ Adapted from [`RobertaTokenizer`] and [`XLNetTokenizer`]. Construct a CamemBERT tokenizer. Based on
+ [SentencePiece](https://github.com/google/sentencepiece).
+
+ This tokenizer inherits from [`PreTrainedTokenizer`] which contains most of the main methods. Users should refer to
+ this superclass for more information regarding those methods.
+
+ Args:
+ vocab_file (`str`):
+ [SentencePiece](https://github.com/google/sentencepiece) file (generally has a *.spm* extension) that
+ contains the vocabulary necessary to instantiate a tokenizer.
+ bos_token (`str`, *optional*, defaults to `""`):
+ The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.
+
+
+
+ When building a sequence using special tokens, this is not the token that is used for the beginning of
+ sequence. The token used is the `cls_token`.
+
+
+
+ eos_token (`str`, *optional*, defaults to `""`):
+ The end of sequence token.
+
+
+
+ When building a sequence using special tokens, this is not the token that is used for the end of sequence.
+ The token used is the `sep_token`.
+
+
+
+ sep_token (`str`, *optional*, defaults to `""`):
+ The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for
+ sequence classification or for a text and a question for question answering. It is also used as the last
+ token of a sequence built with special tokens.
+ cls_token (`str`, *optional*, defaults to `""`):
+ The classifier token which is used when doing sequence classification (classification of the whole sequence
+ instead of per-token classification). It is the first token of the sequence when built with special tokens.
+ unk_token (`str`, *optional*, defaults to `""`):
+ The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
+ token instead.
+ pad_token (`str`, *optional*, defaults to `""`):
+ The token used for padding, for example when batching sequences of different lengths.
+ mask_token (`str`, *optional*, defaults to `""`):
+ The token used for masking values. This is the token used when training this model with masked language
+ modeling. This is the token which the model will try to predict.
+ additional_special_tokens (`List[str]`, *optional*, defaults to `['NOTUSED', 'NOTUSED', 'NOTUSED']`):
+ Additional special tokens used by the tokenizer.
+ sp_model_kwargs (`dict`, *optional*):
+ Will be passed to the `SentencePieceProcessor.__init__()` method. The [Python wrapper for
+ SentencePiece](https://github.com/google/sentencepiece/tree/master/python) can be used, among other things,
+ to set:
+
+ - `enable_sampling`: Enable subword regularization.
+ - `nbest_size`: Sampling parameters for unigram. Invalid for BPE-Dropout.
+
+ - `nbest_size = {0,1}`: No sampling is performed.
+ - `nbest_size > 1`: samples from the nbest_size results.
+ - `nbest_size < 0`: assuming that nbest_size is infinite and samples from the all hypothesis (lattice)
+ using forward-filtering-and-backward-sampling algorithm.
+
+ - `alpha`: Smoothing parameter for unigram sampling, and dropout probability of merge operations for
+ BPE-dropout.
+
+ Attributes:
+ sp_model (`SentencePieceProcessor`):
+ The *SentencePiece* processor that is used for every conversion (string, tokens and IDs).
+ """
+
+ vocab_files_names = VOCAB_FILES_NAMES
+ pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+ max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+ model_input_names = ["input_ids", "attention_mask"]
+
+ def __init__(
+ self,
+ vocab_file,
+ bos_token="",
+ eos_token="",
+ sep_token="",
+ cls_token="",
+ unk_token="",
+ pad_token="",
+ mask_token="",
+ additional_special_tokens=["NOTUSED", "NOTUSED", "NOTUSED"],
+ sp_model_kwargs: Optional[Dict[str, Any]] = None,
+ **kwargs,
+ ) -> None:
+ # Mask token behave like a normal word, i.e. include the space before it
+ mask_token = (
+ AddedToken(mask_token, lstrip=True, rstrip=False, normalized=False, special=True)
+ if isinstance(mask_token, str)
+ else mask_token
+ )
+
+ self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+
+ self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+ self.sp_model.Load(str(vocab_file))
+ self.vocab_file = vocab_file
+
+ # HACK: These tokens were added by the author for an obscure reason as they were already part of the
+ # sentencepiece vocabulary (this is the case for and and ).
+ # In this case it is recommended to properly set the tokens by hand.
+ self._added_tokens_decoder = {
+ 0: AddedToken("NOTUSED", special=True),
+ 1: AddedToken(pad_token, special=True) if isinstance(pad_token, str) else pad_token,
+ 2: AddedToken("NOTUSED", special=True),
+ 3: AddedToken(unk_token, special=True) if isinstance(unk_token, str) else unk_token,
+ 4: AddedToken("NOTUSED", special=True),
+ }
+
+ self.fairseq_offset = 4 # 3 tokens are newly added, but the offset starts from 4
+
+ # legacy: camemebert is a particular case were we have to make sure `"NOTUSED"` is here
+ if "added_tokens_decoder" in kwargs:
+ # this is the only class that requires this unfortunately.....
+ # the reason is that the fast version has a whole.
+ kwargs["added_tokens_decoder"].update(self._added_tokens_decoder)
+
+ super().__init__(
+ bos_token=bos_token,
+ eos_token=eos_token,
+ unk_token=unk_token,
+ sep_token=sep_token,
+ cls_token=cls_token,
+ pad_token=pad_token,
+ mask_token=mask_token,
+ additional_special_tokens=additional_special_tokens,
+ sp_model_kwargs=self.sp_model_kwargs,
+ **kwargs,
+ )
+
+ @property
+ def vocab_size(self):
+ # The length of the vocabulary without added tokens is len(self.sp_model) but the added tokens are added at the beginning.
+ return len(self.sp_model)
+
+ def get_vocab(self):
+ vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size + self.fairseq_offset)}
+ vocab.update(self.added_tokens_encoder)
+ return vocab
+
+ def _tokenize(self, text: str) -> List[str]:
+ return self.sp_model.encode(text, out_type=str)
+
+ def _convert_token_to_id(self, token):
+ """Converts a token (str) in an id using the vocab."""
+ # specifi to camembert, both 3 and 4 point to the unk token.
+ if self.sp_model.PieceToId(token) == 0:
+ # Convert sentence piece unk token to fairseq unk token index
+ return self.unk_token_id
+ return self.fairseq_offset + self.sp_model.PieceToId(token)
+
+ def _convert_id_to_token(self, index):
+ """Converts an index (integer) in a token (str) using the vocab."""
+ return self.sp_model.IdToPiece(index - self.fairseq_offset)
+
+ def convert_tokens_to_string(self, tokens):
+ """Converts a sequence of tokens (string) in a single string."""
+ # TODO decode outputs do not match between fast and slow
+ current_sub_tokens = []
+ out_string = ""
+ prev_is_special = 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(current_sub_tokens) + token
+ prev_is_special = True
+ current_sub_tokens = []
+ else:
+ current_sub_tokens.append(token)
+ prev_is_special = False
+ out_string += self.sp_model.decode(current_sub_tokens)
+ return out_string.strip()
+
+ def __getstate__(self):
+ state = self.__dict__.copy()
+ state["sp_model"] = None
+ return state
+
+ def __setstate__(self, d):
+ self.__dict__ = d
+
+ # for backward compatibility
+ if not hasattr(self, "sp_model_kwargs"):
+ self.sp_model_kwargs = {}
+
+ self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+ self.sp_model.Load(self.vocab_file)
+
+ def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+ if not os.path.isdir(save_directory):
+ logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+ return
+ out_vocab_file = os.path.join(
+ save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+ )
+
+ if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+ copyfile(self.vocab_file, out_vocab_file)
+ elif not os.path.isfile(self.vocab_file):
+ with open(out_vocab_file, "wb") as fi:
+ content_spiece_model = self.sp_model.serialized_model_proto()
+ fi.write(content_spiece_model)
+
+ return (out_vocab_file,)
+
+ def build_inputs_with_special_tokens(
+ self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+ ) -> List[int]:
+ """
+ Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
+ adding special tokens. An CamemBERT sequence has the following format:
+
+ - single sequence: ` X `
+ - pair of sequences: ` A B `
+
+ Args:
+ token_ids_0 (`List[int]`):
+ List of IDs to which the special tokens will be added.
+ token_ids_1 (`List[int]`, *optional*):
+ Optional second list of IDs for sequence pairs.
+
+ Returns:
+ `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
+ """
+
+ if token_ids_1 is None:
+ return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
+ cls = [self.cls_token_id]
+ sep = [self.sep_token_id]
+ return cls + token_ids_0 + sep + sep + token_ids_1 + sep
+
+ def get_special_tokens_mask(
+ self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+ ) -> List[int]:
+ """
+ Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+ special tokens using the tokenizer `prepare_for_model` method.
+
+ Args:
+ token_ids_0 (`List[int]`):
+ List of IDs.
+ token_ids_1 (`List[int]`, *optional*):
+ Optional second list of IDs for sequence pairs.
+ already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+ Whether or not the token list is already formatted with special tokens for the model.
+
+ Returns:
+ `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+ """
+ if already_has_special_tokens:
+ return super().get_special_tokens_mask(
+ token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+ )
+
+ if token_ids_1 is None:
+ return [1] + ([0] * len(token_ids_0)) + [1]
+ return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+ def create_token_type_ids_from_sequences(
+ self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+ ) -> List[int]:
+ """
+ Create a mask from the two sequences passed to be used in a sequence-pair classification task. CamemBERT, like
+ RoBERTa, does not make use of token type ids, therefore a list of zeros is returned.
+
+ Args:
+ token_ids_0 (`List[int]`):
+ List of IDs.
+ token_ids_1 (`List[int]`, *optional*):
+ Optional second list of IDs for sequence pairs.
+
+ Returns:
+ `List[int]`: List of zeros.
+ """
+ sep = [self.sep_token_id]
+ cls = [self.cls_token_id]
+
+ if token_ids_1 is None:
+ return len(cls + token_ids_0 + sep) * [0]
+ return len(cls + token_ids_0 + sep + sep + token_ids_1 + sep) * [0]
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__init__.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..e172dd1dc791010141fb4555c663558a0498612d
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__init__.py
@@ -0,0 +1,120 @@
+# Copyright 2020 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from typing import TYPE_CHECKING
+
+from ...utils import (
+ OptionalDependencyNotAvailable,
+ _LazyModule,
+ is_tf_available,
+ is_tokenizers_available,
+ is_torch_available,
+)
+
+
+_import_structure = {
+ "configuration_layoutlm": ["LAYOUTLM_PRETRAINED_CONFIG_ARCHIVE_MAP", "LayoutLMConfig", "LayoutLMOnnxConfig"],
+ "tokenization_layoutlm": ["LayoutLMTokenizer"],
+}
+
+try:
+ if not is_tokenizers_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["tokenization_layoutlm_fast"] = ["LayoutLMTokenizerFast"]
+
+try:
+ if not is_torch_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_layoutlm"] = [
+ "LAYOUTLM_PRETRAINED_MODEL_ARCHIVE_LIST",
+ "LayoutLMForMaskedLM",
+ "LayoutLMForSequenceClassification",
+ "LayoutLMForTokenClassification",
+ "LayoutLMForQuestionAnswering",
+ "LayoutLMModel",
+ "LayoutLMPreTrainedModel",
+ ]
+
+try:
+ if not is_tf_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_tf_layoutlm"] = [
+ "TF_LAYOUTLM_PRETRAINED_MODEL_ARCHIVE_LIST",
+ "TFLayoutLMForMaskedLM",
+ "TFLayoutLMForSequenceClassification",
+ "TFLayoutLMForTokenClassification",
+ "TFLayoutLMForQuestionAnswering",
+ "TFLayoutLMMainLayer",
+ "TFLayoutLMModel",
+ "TFLayoutLMPreTrainedModel",
+ ]
+
+
+if TYPE_CHECKING:
+ from .configuration_layoutlm import LAYOUTLM_PRETRAINED_CONFIG_ARCHIVE_MAP, LayoutLMConfig, LayoutLMOnnxConfig
+ from .tokenization_layoutlm import LayoutLMTokenizer
+
+ try:
+ if not is_tokenizers_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .tokenization_layoutlm_fast import LayoutLMTokenizerFast
+
+ try:
+ if not is_torch_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_layoutlm import (
+ LAYOUTLM_PRETRAINED_MODEL_ARCHIVE_LIST,
+ LayoutLMForMaskedLM,
+ LayoutLMForQuestionAnswering,
+ LayoutLMForSequenceClassification,
+ LayoutLMForTokenClassification,
+ LayoutLMModel,
+ LayoutLMPreTrainedModel,
+ )
+ try:
+ if not is_tf_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_tf_layoutlm import (
+ TF_LAYOUTLM_PRETRAINED_MODEL_ARCHIVE_LIST,
+ TFLayoutLMForMaskedLM,
+ TFLayoutLMForQuestionAnswering,
+ TFLayoutLMForSequenceClassification,
+ TFLayoutLMForTokenClassification,
+ TFLayoutLMMainLayer,
+ TFLayoutLMModel,
+ TFLayoutLMPreTrainedModel,
+ )
+
+else:
+ import sys
+
+ sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/configuration_layoutlm.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/configuration_layoutlm.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..eae98a43c714ee98b32cd27b0b0fe8842a9dccd2
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/configuration_layoutlm.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/modeling_tf_layoutlm.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/modeling_tf_layoutlm.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..2ae17450c736c1eb41d3d8487c09b5ed9f254942
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/modeling_tf_layoutlm.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/tokenization_layoutlm_fast.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/tokenization_layoutlm_fast.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..2c88dcb14fce04f927a9c808582cce84b2ec1674
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/__pycache__/tokenization_layoutlm_fast.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/configuration_layoutlm.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/configuration_layoutlm.py
new file mode 100644
index 0000000000000000000000000000000000000000..77d62ded403b92b6cafcd762044d81367cd1bffa
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/configuration_layoutlm.py
@@ -0,0 +1,204 @@
+# coding=utf-8
+# Copyright 2010, The Microsoft Research Asia LayoutLM Team authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" LayoutLM model configuration"""
+from collections import OrderedDict
+from typing import Any, List, Mapping, Optional
+
+from ... import PretrainedConfig, PreTrainedTokenizer
+from ...onnx import OnnxConfig, PatchingSpec
+from ...utils import TensorType, is_torch_available, logging
+
+
+logger = logging.get_logger(__name__)
+
+LAYOUTLM_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+ "microsoft/layoutlm-base-uncased": (
+ "https://huggingface.co/microsoft/layoutlm-base-uncased/resolve/main/config.json"
+ ),
+ "microsoft/layoutlm-large-uncased": (
+ "https://huggingface.co/microsoft/layoutlm-large-uncased/resolve/main/config.json"
+ ),
+}
+
+
+class LayoutLMConfig(PretrainedConfig):
+ r"""
+ This is the configuration class to store the configuration of a [`LayoutLMModel`]. It is used to instantiate a
+ LayoutLM model according to the specified arguments, defining the model architecture. Instantiating a configuration
+ with the defaults will yield a similar configuration to that of the LayoutLM
+ [microsoft/layoutlm-base-uncased](https://huggingface.co/microsoft/layoutlm-base-uncased) architecture.
+
+ Configuration objects inherit from [`BertConfig`] and can be used to control the model outputs. Read the
+ documentation from [`BertConfig`] for more information.
+
+
+ Args:
+ vocab_size (`int`, *optional*, defaults to 30522):
+ Vocabulary size of the LayoutLM model. Defines the different tokens that can be represented by the
+ *inputs_ids* passed to the forward method of [`LayoutLMModel`].
+ hidden_size (`int`, *optional*, defaults to 768):
+ Dimensionality of the encoder layers and the pooler layer.
+ num_hidden_layers (`int`, *optional*, defaults to 12):
+ Number of hidden layers in the Transformer encoder.
+ num_attention_heads (`int`, *optional*, defaults to 12):
+ Number of attention heads for each attention layer in the Transformer encoder.
+ intermediate_size (`int`, *optional*, defaults to 3072):
+ Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
+ hidden_act (`str` or `function`, *optional*, defaults to `"gelu"`):
+ The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
+ `"relu"`, `"silu"` and `"gelu_new"` are supported.
+ hidden_dropout_prob (`float`, *optional*, defaults to 0.1):
+ The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
+ attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1):
+ The dropout ratio for the attention probabilities.
+ max_position_embeddings (`int`, *optional*, defaults to 512):
+ The maximum sequence length that this model might ever be used with. Typically set this to something large
+ just in case (e.g., 512 or 1024 or 2048).
+ type_vocab_size (`int`, *optional*, defaults to 2):
+ The vocabulary size of the `token_type_ids` passed into [`LayoutLMModel`].
+ initializer_range (`float`, *optional*, defaults to 0.02):
+ The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+ layer_norm_eps (`float`, *optional*, defaults to 1e-12):
+ The epsilon used by the layer normalization layers.
+ pad_token_id (`int`, *optional*, defaults to 0):
+ The value used to pad input_ids.
+ position_embedding_type (`str`, *optional*, defaults to `"absolute"`):
+ Type of position embedding. Choose one of `"absolute"`, `"relative_key"`, `"relative_key_query"`. For
+ positional embeddings use `"absolute"`. For more information on `"relative_key"`, please refer to
+ [Self-Attention with Relative Position Representations (Shaw et al.)](https://arxiv.org/abs/1803.02155).
+ For more information on `"relative_key_query"`, please refer to *Method 4* in [Improve Transformer Models
+ with Better Relative Position Embeddings (Huang et al.)](https://arxiv.org/abs/2009.13658).
+ use_cache (`bool`, *optional*, defaults to `True`):
+ Whether or not the model should return the last key/values attentions (not used by all models). Only
+ relevant if `config.is_decoder=True`.
+ max_2d_position_embeddings (`int`, *optional*, defaults to 1024):
+ The maximum value that the 2D position embedding might ever used. Typically set this to something large
+ just in case (e.g., 1024).
+
+ Examples:
+
+ ```python
+ >>> from transformers import LayoutLMConfig, LayoutLMModel
+
+ >>> # Initializing a LayoutLM configuration
+ >>> configuration = LayoutLMConfig()
+
+ >>> # Initializing a model (with random weights) from the configuration
+ >>> model = LayoutLMModel(configuration)
+
+ >>> # Accessing the model configuration
+ >>> configuration = model.config
+ ```"""
+
+ model_type = "layoutlm"
+
+ def __init__(
+ self,
+ vocab_size=30522,
+ hidden_size=768,
+ num_hidden_layers=12,
+ num_attention_heads=12,
+ intermediate_size=3072,
+ hidden_act="gelu",
+ hidden_dropout_prob=0.1,
+ attention_probs_dropout_prob=0.1,
+ max_position_embeddings=512,
+ type_vocab_size=2,
+ initializer_range=0.02,
+ layer_norm_eps=1e-12,
+ pad_token_id=0,
+ position_embedding_type="absolute",
+ use_cache=True,
+ max_2d_position_embeddings=1024,
+ **kwargs,
+ ):
+ super().__init__(pad_token_id=pad_token_id, **kwargs)
+ self.vocab_size = vocab_size
+ self.hidden_size = hidden_size
+ self.num_hidden_layers = num_hidden_layers
+ self.num_attention_heads = num_attention_heads
+ self.hidden_act = hidden_act
+ self.intermediate_size = intermediate_size
+ self.hidden_dropout_prob = hidden_dropout_prob
+ self.attention_probs_dropout_prob = attention_probs_dropout_prob
+ self.max_position_embeddings = max_position_embeddings
+ self.type_vocab_size = type_vocab_size
+ self.initializer_range = initializer_range
+ self.layer_norm_eps = layer_norm_eps
+ self.position_embedding_type = position_embedding_type
+ self.use_cache = use_cache
+ self.max_2d_position_embeddings = max_2d_position_embeddings
+
+
+class LayoutLMOnnxConfig(OnnxConfig):
+ def __init__(
+ self,
+ config: PretrainedConfig,
+ task: str = "default",
+ patching_specs: List[PatchingSpec] = None,
+ ):
+ super().__init__(config, task=task, patching_specs=patching_specs)
+ self.max_2d_positions = config.max_2d_position_embeddings - 1
+
+ @property
+ def inputs(self) -> Mapping[str, Mapping[int, str]]:
+ return OrderedDict(
+ [
+ ("input_ids", {0: "batch", 1: "sequence"}),
+ ("bbox", {0: "batch", 1: "sequence"}),
+ ("attention_mask", {0: "batch", 1: "sequence"}),
+ ("token_type_ids", {0: "batch", 1: "sequence"}),
+ ]
+ )
+
+ def generate_dummy_inputs(
+ self,
+ tokenizer: PreTrainedTokenizer,
+ batch_size: int = -1,
+ seq_length: int = -1,
+ is_pair: bool = False,
+ framework: Optional[TensorType] = None,
+ ) -> Mapping[str, Any]:
+ """
+ Generate inputs to provide to the ONNX exporter for the specific framework
+
+ Args:
+ tokenizer: The tokenizer associated with this model configuration
+ batch_size: The batch size (int) to export the model for (-1 means dynamic axis)
+ seq_length: The sequence length (int) to export the model for (-1 means dynamic axis)
+ is_pair: Indicate if the input is a pair (sentence 1, sentence 2)
+ framework: The framework (optional) the tokenizer will generate tensor for
+
+ Returns:
+ Mapping[str, Tensor] holding the kwargs to provide to the model's forward function
+ """
+
+ input_dict = super().generate_dummy_inputs(
+ tokenizer, batch_size=batch_size, seq_length=seq_length, is_pair=is_pair, framework=framework
+ )
+
+ # Generate a dummy bbox
+ box = [48, 84, 73, 128]
+
+ if not framework == TensorType.PYTORCH:
+ raise NotImplementedError("Exporting LayoutLM to ONNX is currently only supported for PyTorch.")
+
+ if not is_torch_available():
+ raise ValueError("Cannot generate dummy inputs without PyTorch installed.")
+ import torch
+
+ batch_size, seq_length = input_dict["input_ids"].shape
+ input_dict["bbox"] = torch.tensor([*[box] * seq_length]).tile(batch_size, 1, 1)
+ return input_dict
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/tokenization_layoutlm_fast.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/tokenization_layoutlm_fast.py
new file mode 100644
index 0000000000000000000000000000000000000000..c0bc1072f7f5f1da9d88dcbbab0c4d38653edb98
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/layoutlm/tokenization_layoutlm_fast.py
@@ -0,0 +1,205 @@
+# coding=utf-8
+# Copyright 2018 The Microsoft Research Asia LayoutLM Team Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Tokenization class for model LayoutLM."""
+
+import json
+from typing import List, Optional, Tuple
+
+from tokenizers import normalizers
+
+from ...tokenization_utils_fast import PreTrainedTokenizerFast
+from ...utils import logging
+from .tokenization_layoutlm import LayoutLMTokenizer
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "vocab.txt", "tokenizer_file": "tokenizer.json"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+ "vocab_file": {
+ "microsoft/layoutlm-base-uncased": (
+ "https://huggingface.co/microsoft/layoutlm-base-uncased/resolve/main/vocab.txt"
+ ),
+ "microsoft/layoutlm-large-uncased": (
+ "https://huggingface.co/microsoft/layoutlm-large-uncased/resolve/main/vocab.txt"
+ ),
+ },
+ "tokenizer_file": {
+ "microsoft/layoutlm-base-uncased": (
+ "https://huggingface.co/microsoft/layoutlm-base-uncased/resolve/main/tokenizer.json"
+ ),
+ "microsoft/layoutlm-large-uncased": (
+ "https://huggingface.co/microsoft/layoutlm-large-uncased/resolve/main/tokenizer.json"
+ ),
+ },
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+ "microsoft/layoutlm-base-uncased": 512,
+ "microsoft/layoutlm-large-uncased": 512,
+}
+
+PRETRAINED_INIT_CONFIGURATION = {
+ "microsoft/layoutlm-base-uncased": {"do_lower_case": True},
+ "microsoft/layoutlm-large-uncased": {"do_lower_case": True},
+}
+
+
+# Copied from transformers.models.bert.tokenization_bert_fast.BertTokenizerFast with Bert->LayoutLM,BERT->LayoutLM
+class LayoutLMTokenizerFast(PreTrainedTokenizerFast):
+ r"""
+ Construct a "fast" LayoutLM tokenizer (backed by HuggingFace's *tokenizers* library). Based on WordPiece.
+
+ This tokenizer inherits from [`PreTrainedTokenizerFast`] which contains most of the main methods. Users should
+ refer to this superclass for more information regarding those methods.
+
+ Args:
+ vocab_file (`str`):
+ File containing the vocabulary.
+ do_lower_case (`bool`, *optional*, defaults to `True`):
+ Whether or not to lowercase the input when tokenizing.
+ unk_token (`str`, *optional*, defaults to `"[UNK]"`):
+ The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
+ token instead.
+ sep_token (`str`, *optional*, defaults to `"[SEP]"`):
+ The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for
+ sequence classification or for a text and a question for question answering. It is also used as the last
+ token of a sequence built with special tokens.
+ pad_token (`str`, *optional*, defaults to `"[PAD]"`):
+ The token used for padding, for example when batching sequences of different lengths.
+ cls_token (`str`, *optional*, defaults to `"[CLS]"`):
+ The classifier token which is used when doing sequence classification (classification of the whole sequence
+ instead of per-token classification). It is the first token of the sequence when built with special tokens.
+ mask_token (`str`, *optional*, defaults to `"[MASK]"`):
+ The token used for masking values. This is the token used when training this model with masked language
+ modeling. This is the token which the model will try to predict.
+ clean_text (`bool`, *optional*, defaults to `True`):
+ Whether or not to clean the text before tokenization by removing any control characters and replacing all
+ whitespaces by the classic one.
+ tokenize_chinese_chars (`bool`, *optional*, defaults to `True`):
+ Whether or not to tokenize Chinese characters. This should likely be deactivated for Japanese (see [this
+ issue](https://github.com/huggingface/transformers/issues/328)).
+ strip_accents (`bool`, *optional*):
+ Whether or not to strip all accents. If this option is not specified, then it will be determined by the
+ value for `lowercase` (as in the original LayoutLM).
+ wordpieces_prefix (`str`, *optional*, defaults to `"##"`):
+ The prefix for subwords.
+ """
+
+ vocab_files_names = VOCAB_FILES_NAMES
+ pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+ pretrained_init_configuration = PRETRAINED_INIT_CONFIGURATION
+ max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+ slow_tokenizer_class = LayoutLMTokenizer
+
+ def __init__(
+ self,
+ vocab_file=None,
+ tokenizer_file=None,
+ do_lower_case=True,
+ unk_token="[UNK]",
+ sep_token="[SEP]",
+ pad_token="[PAD]",
+ cls_token="[CLS]",
+ mask_token="[MASK]",
+ tokenize_chinese_chars=True,
+ strip_accents=None,
+ **kwargs,
+ ):
+ super().__init__(
+ vocab_file,
+ tokenizer_file=tokenizer_file,
+ do_lower_case=do_lower_case,
+ unk_token=unk_token,
+ sep_token=sep_token,
+ pad_token=pad_token,
+ cls_token=cls_token,
+ mask_token=mask_token,
+ tokenize_chinese_chars=tokenize_chinese_chars,
+ strip_accents=strip_accents,
+ **kwargs,
+ )
+
+ normalizer_state = json.loads(self.backend_tokenizer.normalizer.__getstate__())
+ if (
+ normalizer_state.get("lowercase", do_lower_case) != do_lower_case
+ or normalizer_state.get("strip_accents", strip_accents) != strip_accents
+ or normalizer_state.get("handle_chinese_chars", tokenize_chinese_chars) != tokenize_chinese_chars
+ ):
+ normalizer_class = getattr(normalizers, normalizer_state.pop("type"))
+ normalizer_state["lowercase"] = do_lower_case
+ normalizer_state["strip_accents"] = strip_accents
+ normalizer_state["handle_chinese_chars"] = tokenize_chinese_chars
+ self.backend_tokenizer.normalizer = normalizer_class(**normalizer_state)
+
+ self.do_lower_case = do_lower_case
+
+ def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+ """
+ Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
+ adding special tokens. A LayoutLM sequence has the following format:
+
+ - single sequence: `[CLS] X [SEP]`
+ - pair of sequences: `[CLS] A [SEP] B [SEP]`
+
+ Args:
+ token_ids_0 (`List[int]`):
+ List of IDs to which the special tokens will be added.
+ token_ids_1 (`List[int]`, *optional*):
+ Optional second list of IDs for sequence pairs.
+
+ Returns:
+ `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
+ """
+ output = [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
+
+ if token_ids_1 is not None:
+ output += token_ids_1 + [self.sep_token_id]
+
+ return output
+
+ def create_token_type_ids_from_sequences(
+ self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+ ) -> List[int]:
+ """
+ Create a mask from the two sequences passed to be used in a sequence-pair classification task. A LayoutLM sequence
+ pair mask has the following format:
+
+ ```
+ 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
+ | first sequence | second sequence |
+ ```
+
+ If `token_ids_1` is `None`, this method only returns the first portion of the mask (0s).
+
+ Args:
+ token_ids_0 (`List[int]`):
+ List of IDs.
+ token_ids_1 (`List[int]`, *optional*):
+ Optional second list of IDs for sequence pairs.
+
+ Returns:
+ `List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s).
+ """
+ sep = [self.sep_token_id]
+ cls = [self.cls_token_id]
+ if token_ids_1 is None:
+ return len(cls + token_ids_0 + sep) * [0]
+ return len(cls + token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1]
+
+ def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+ files = self._tokenizer.model.save(save_directory, name=filename_prefix)
+ return tuple(files)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__init__.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..66ef7c953cff4385424b208313445962d4facf28
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__init__.py
@@ -0,0 +1,135 @@
+# Copyright 2020 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from typing import TYPE_CHECKING
+
+from ...utils import (
+ OptionalDependencyNotAvailable,
+ _LazyModule,
+ is_tf_available,
+ is_tokenizers_available,
+ is_torch_available,
+)
+
+
+_import_structure = {
+ "configuration_longformer": [
+ "LONGFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP",
+ "LongformerConfig",
+ "LongformerOnnxConfig",
+ ],
+ "tokenization_longformer": ["LongformerTokenizer"],
+}
+
+try:
+ if not is_tokenizers_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["tokenization_longformer_fast"] = ["LongformerTokenizerFast"]
+
+try:
+ if not is_torch_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_longformer"] = [
+ "LONGFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
+ "LongformerForMaskedLM",
+ "LongformerForMultipleChoice",
+ "LongformerForQuestionAnswering",
+ "LongformerForSequenceClassification",
+ "LongformerForTokenClassification",
+ "LongformerModel",
+ "LongformerPreTrainedModel",
+ "LongformerSelfAttention",
+ ]
+
+try:
+ if not is_tf_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_tf_longformer"] = [
+ "TF_LONGFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
+ "TFLongformerForMaskedLM",
+ "TFLongformerForMultipleChoice",
+ "TFLongformerForQuestionAnswering",
+ "TFLongformerForSequenceClassification",
+ "TFLongformerForTokenClassification",
+ "TFLongformerModel",
+ "TFLongformerPreTrainedModel",
+ "TFLongformerSelfAttention",
+ ]
+
+
+if TYPE_CHECKING:
+ from .configuration_longformer import (
+ LONGFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP,
+ LongformerConfig,
+ LongformerOnnxConfig,
+ )
+ from .tokenization_longformer import LongformerTokenizer
+
+ try:
+ if not is_tokenizers_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .tokenization_longformer_fast import LongformerTokenizerFast
+
+ try:
+ if not is_torch_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_longformer import (
+ LONGFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
+ LongformerForMaskedLM,
+ LongformerForMultipleChoice,
+ LongformerForQuestionAnswering,
+ LongformerForSequenceClassification,
+ LongformerForTokenClassification,
+ LongformerModel,
+ LongformerPreTrainedModel,
+ LongformerSelfAttention,
+ )
+
+ try:
+ if not is_tf_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_tf_longformer import (
+ TF_LONGFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
+ TFLongformerForMaskedLM,
+ TFLongformerForMultipleChoice,
+ TFLongformerForQuestionAnswering,
+ TFLongformerForSequenceClassification,
+ TFLongformerForTokenClassification,
+ TFLongformerModel,
+ TFLongformerPreTrainedModel,
+ TFLongformerSelfAttention,
+ )
+
+else:
+ import sys
+
+ sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/__init__.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/__init__.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..ff02e770c2ac14a24263adc84c4aa92c3323e126
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/__init__.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/configuration_longformer.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/configuration_longformer.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..9741891c04428f1c8a8f72e9136dcd6528da4588
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/configuration_longformer.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/convert_longformer_original_pytorch_lightning_to_pytorch.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/convert_longformer_original_pytorch_lightning_to_pytorch.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..c5522cdab239d5cf5a4312be47018dafe6675b01
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/convert_longformer_original_pytorch_lightning_to_pytorch.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/modeling_longformer.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/modeling_longformer.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..5c0d9031354332a2276eabc7399270d408fbed3c
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/modeling_longformer.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/modeling_tf_longformer.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/modeling_tf_longformer.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..042fbf5644f7c745bb0f5b797448c1de9b66b943
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/modeling_tf_longformer.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/tokenization_longformer.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/tokenization_longformer.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..88e376bf980a886d6816760e54bc8dfdc40948d3
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/__pycache__/tokenization_longformer.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/configuration_longformer.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/configuration_longformer.py
new file mode 100644
index 0000000000000000000000000000000000000000..2935dd4aaaae2550af22eb9e504fa80f21f29965
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/configuration_longformer.py
@@ -0,0 +1,214 @@
+# coding=utf-8
+# Copyright 2020 The Allen Institute for AI team and The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" Longformer configuration"""
+from collections import OrderedDict
+from typing import TYPE_CHECKING, Any, List, Mapping, Optional, Union
+
+from ...configuration_utils import PretrainedConfig
+from ...onnx import OnnxConfig
+from ...utils import TensorType, logging
+
+
+if TYPE_CHECKING:
+ from ...onnx.config import PatchingSpec
+ from ...tokenization_utils_base import PreTrainedTokenizerBase
+
+
+logger = logging.get_logger(__name__)
+
+LONGFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+ "allenai/longformer-base-4096": "https://huggingface.co/allenai/longformer-base-4096/resolve/main/config.json",
+ "allenai/longformer-large-4096": "https://huggingface.co/allenai/longformer-large-4096/resolve/main/config.json",
+ "allenai/longformer-large-4096-finetuned-triviaqa": (
+ "https://huggingface.co/allenai/longformer-large-4096-finetuned-triviaqa/resolve/main/config.json"
+ ),
+ "allenai/longformer-base-4096-extra.pos.embd.only": (
+ "https://huggingface.co/allenai/longformer-base-4096-extra.pos.embd.only/resolve/main/config.json"
+ ),
+ "allenai/longformer-large-4096-extra.pos.embd.only": (
+ "https://huggingface.co/allenai/longformer-large-4096-extra.pos.embd.only/resolve/main/config.json"
+ ),
+}
+
+
+class LongformerConfig(PretrainedConfig):
+ r"""
+ This is the configuration class to store the configuration of a [`LongformerModel`] or a [`TFLongformerModel`]. It
+ is used to instantiate a Longformer model according to the specified arguments, defining the model architecture.
+
+ This is the configuration class to store the configuration of a [`LongformerModel`]. It is used to instantiate an
+ Longformer model according to the specified arguments, defining the model architecture. Instantiating a
+ configuration with the defaults will yield a similar configuration to that of the LongFormer
+ [allenai/longformer-base-4096](https://huggingface.co/allenai/longformer-base-4096) architecture with a sequence
+ length 4,096.
+
+ Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+ documentation from [`PretrainedConfig`] for more information.
+
+ Args:
+ vocab_size (`int`, *optional*, defaults to 30522):
+ Vocabulary size of the Longformer model. Defines the number of different tokens that can be represented by
+ the `inputs_ids` passed when calling [`LongformerModel`] or [`TFLongformerModel`].
+ hidden_size (`int`, *optional*, defaults to 768):
+ Dimensionality of the encoder layers and the pooler layer.
+ num_hidden_layers (`int`, *optional*, defaults to 12):
+ Number of hidden layers in the Transformer encoder.
+ num_attention_heads (`int`, *optional*, defaults to 12):
+ Number of attention heads for each attention layer in the Transformer encoder.
+ intermediate_size (`int`, *optional*, defaults to 3072):
+ Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder.
+ hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`):
+ The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
+ `"relu"`, `"silu"` and `"gelu_new"` are supported.
+ hidden_dropout_prob (`float`, *optional*, defaults to 0.1):
+ The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
+ attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1):
+ The dropout ratio for the attention probabilities.
+ max_position_embeddings (`int`, *optional*, defaults to 512):
+ The maximum sequence length that this model might ever be used with. Typically set this to something large
+ just in case (e.g., 512 or 1024 or 2048).
+ type_vocab_size (`int`, *optional*, defaults to 2):
+ The vocabulary size of the `token_type_ids` passed when calling [`LongformerModel`] or
+ [`TFLongformerModel`].
+ initializer_range (`float`, *optional*, defaults to 0.02):
+ The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+ layer_norm_eps (`float`, *optional*, defaults to 1e-12):
+ The epsilon used by the layer normalization layers.
+ attention_window (`int` or `List[int]`, *optional*, defaults to 512):
+ Size of an attention window around each token. If an `int`, use the same size for all layers. To specify a
+ different window size for each layer, use a `List[int]` where `len(attention_window) == num_hidden_layers`.
+
+ Example:
+
+ ```python
+ >>> from transformers import LongformerConfig, LongformerModel
+
+ >>> # Initializing a Longformer configuration
+ >>> configuration = LongformerConfig()
+
+ >>> # Initializing a model from the configuration
+ >>> model = LongformerModel(configuration)
+
+ >>> # Accessing the model configuration
+ >>> configuration = model.config
+ ```"""
+
+ model_type = "longformer"
+
+ def __init__(
+ self,
+ attention_window: Union[List[int], int] = 512,
+ sep_token_id: int = 2,
+ pad_token_id: int = 1,
+ bos_token_id: int = 0,
+ eos_token_id: int = 2,
+ vocab_size: int = 30522,
+ hidden_size: int = 768,
+ num_hidden_layers: int = 12,
+ num_attention_heads: int = 12,
+ intermediate_size: int = 3072,
+ hidden_act: str = "gelu",
+ hidden_dropout_prob: float = 0.1,
+ attention_probs_dropout_prob: float = 0.1,
+ max_position_embeddings: int = 512,
+ type_vocab_size: int = 2,
+ initializer_range: float = 0.02,
+ layer_norm_eps: float = 1e-12,
+ onnx_export: bool = False,
+ **kwargs,
+ ):
+ """Constructs LongformerConfig."""
+ super().__init__(pad_token_id=pad_token_id, **kwargs)
+
+ self.attention_window = attention_window
+ self.sep_token_id = sep_token_id
+ self.bos_token_id = bos_token_id
+ self.eos_token_id = eos_token_id
+ self.vocab_size = vocab_size
+ self.hidden_size = hidden_size
+ self.num_hidden_layers = num_hidden_layers
+ self.num_attention_heads = num_attention_heads
+ self.hidden_act = hidden_act
+ self.intermediate_size = intermediate_size
+ self.hidden_dropout_prob = hidden_dropout_prob
+ self.attention_probs_dropout_prob = attention_probs_dropout_prob
+ self.max_position_embeddings = max_position_embeddings
+ self.type_vocab_size = type_vocab_size
+ self.initializer_range = initializer_range
+ self.layer_norm_eps = layer_norm_eps
+ self.onnx_export = onnx_export
+
+
+class LongformerOnnxConfig(OnnxConfig):
+ def __init__(self, config: "PretrainedConfig", task: str = "default", patching_specs: "List[PatchingSpec]" = None):
+ super().__init__(config, task, patching_specs)
+ config.onnx_export = True
+
+ @property
+ def inputs(self) -> Mapping[str, Mapping[int, str]]:
+ if self.task == "multiple-choice":
+ dynamic_axis = {0: "batch", 1: "choice", 2: "sequence"}
+ else:
+ dynamic_axis = {0: "batch", 1: "sequence"}
+ return OrderedDict(
+ [
+ ("input_ids", dynamic_axis),
+ ("attention_mask", dynamic_axis),
+ ("global_attention_mask", dynamic_axis),
+ ]
+ )
+
+ @property
+ def outputs(self) -> Mapping[str, Mapping[int, str]]:
+ outputs = super().outputs
+ if self.task == "default":
+ outputs["pooler_output"] = {0: "batch"}
+ return outputs
+
+ @property
+ def atol_for_validation(self) -> float:
+ """
+ What absolute tolerance value to use during model conversion validation.
+
+ Returns:
+ Float absolute tolerance value.
+ """
+ return 1e-4
+
+ @property
+ def default_onnx_opset(self) -> int:
+ # needs to be >= 14 to support tril operator
+ return max(super().default_onnx_opset, 14)
+
+ def generate_dummy_inputs(
+ self,
+ tokenizer: "PreTrainedTokenizerBase",
+ batch_size: int = -1,
+ seq_length: int = -1,
+ is_pair: bool = False,
+ framework: Optional[TensorType] = None,
+ ) -> Mapping[str, Any]:
+ inputs = super().generate_dummy_inputs(
+ preprocessor=tokenizer, batch_size=batch_size, seq_length=seq_length, is_pair=is_pair, framework=framework
+ )
+ import torch
+
+ # for some reason, replacing this code by inputs["global_attention_mask"] = torch.randint(2, inputs["input_ids"].shape, dtype=torch.int64)
+ # makes the export fail randomly
+ inputs["global_attention_mask"] = torch.zeros_like(inputs["input_ids"])
+ # make every second token global
+ inputs["global_attention_mask"][:, ::2] = 1
+
+ return inputs
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/tokenization_longformer_fast.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/tokenization_longformer_fast.py
new file mode 100644
index 0000000000000000000000000000000000000000..fb35a8b67bba7a11a90c0927d9c41f3467ec489e
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/longformer/tokenization_longformer_fast.py
@@ -0,0 +1,329 @@
+# coding=utf-8
+# Copyright 2020 The Allen Institute for AI team and The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Fast Tokenization classes for Longformer."""
+import json
+from typing import List, Optional, Tuple
+
+from tokenizers import pre_tokenizers, processors
+
+from ...tokenization_utils_base import AddedToken, BatchEncoding
+from ...tokenization_utils_fast import PreTrainedTokenizerFast
+from ...utils import logging
+from .tokenization_longformer import LongformerTokenizer
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "vocab.json", "merges_file": "merges.txt", "tokenizer_file": "tokenizer.json"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+ "vocab_file": {
+ "allenai/longformer-base-4096": "https://huggingface.co/allenai/longformer-base-4096/resolve/main/vocab.json",
+ "allenai/longformer-large-4096": (
+ "https://huggingface.co/allenai/longformer-large-4096/resolve/main/vocab.json"
+ ),
+ "allenai/longformer-large-4096-finetuned-triviaqa": (
+ "https://huggingface.co/allenai/longformer-large-4096-finetuned-triviaqa/resolve/main/vocab.json"
+ ),
+ "allenai/longformer-base-4096-extra.pos.embd.only": (
+ "https://huggingface.co/allenai/longformer-base-4096-extra.pos.embd.only/resolve/main/vocab.json"
+ ),
+ "allenai/longformer-large-4096-extra.pos.embd.only": (
+ "https://huggingface.co/allenai/longformer-large-4096-extra.pos.embd.only/resolve/main/vocab.json"
+ ),
+ },
+ "merges_file": {
+ "allenai/longformer-base-4096": "https://huggingface.co/allenai/longformer-base-4096/resolve/main/merges.txt",
+ "allenai/longformer-large-4096": (
+ "https://huggingface.co/allenai/longformer-large-4096/resolve/main/merges.txt"
+ ),
+ "allenai/longformer-large-4096-finetuned-triviaqa": (
+ "https://huggingface.co/allenai/longformer-large-4096-finetuned-triviaqa/resolve/main/merges.txt"
+ ),
+ "allenai/longformer-base-4096-extra.pos.embd.only": (
+ "https://huggingface.co/allenai/longformer-base-4096-extra.pos.embd.only/resolve/main/merges.txt"
+ ),
+ "allenai/longformer-large-4096-extra.pos.embd.only": (
+ "https://huggingface.co/allenai/longformer-large-4096-extra.pos.embd.only/resolve/main/merges.txt"
+ ),
+ },
+ "tokenizer_file": {
+ "allenai/longformer-base-4096": (
+ "https://huggingface.co/allenai/longformer-base-4096/resolve/main/tokenizer.json"
+ ),
+ "allenai/longformer-large-4096": (
+ "https://huggingface.co/allenai/longformer-large-4096/resolve/main/tokenizer.json"
+ ),
+ "allenai/longformer-large-4096-finetuned-triviaqa": (
+ "https://huggingface.co/allenai/longformer-large-4096-finetuned-triviaqa/resolve/main/tokenizer.json"
+ ),
+ "allenai/longformer-base-4096-extra.pos.embd.only": (
+ "https://huggingface.co/allenai/longformer-base-4096-extra.pos.embd.only/resolve/main/tokenizer.json"
+ ),
+ "allenai/longformer-large-4096-extra.pos.embd.only": (
+ "https://huggingface.co/allenai/longformer-large-4096-extra.pos.embd.only/resolve/main/tokenizer.json"
+ ),
+ },
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
+ "allenai/longformer-base-4096": 4096,
+ "allenai/longformer-large-4096": 4096,
+ "allenai/longformer-large-4096-finetuned-triviaqa": 4096,
+ "allenai/longformer-base-4096-extra.pos.embd.only": 4096,
+ "allenai/longformer-large-4096-extra.pos.embd.only": 4096,
+}
+
+
+# Copied from transformers.models.roberta.tokenization_roberta_fast.RobertaTokenizerFast with roberta-base->allenai/longformer-base-4096, RoBERTa->Longformer all-casing, Roberta->Longformer
+class LongformerTokenizerFast(PreTrainedTokenizerFast):
+ """
+ Construct a "fast" Longformer tokenizer (backed by HuggingFace's *tokenizers* library), derived from the GPT-2
+ tokenizer, using byte-level Byte-Pair-Encoding.
+
+ This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will
+ be encoded differently whether it is at the beginning of the sentence (without space) or not:
+
+ ```python
+ >>> from transformers import LongformerTokenizerFast
+
+ >>> tokenizer = LongformerTokenizerFast.from_pretrained("allenai/longformer-base-4096")
+ >>> tokenizer("Hello world")["input_ids"]
+ [0, 31414, 232, 2]
+
+ >>> tokenizer(" Hello world")["input_ids"]
+ [0, 20920, 232, 2]
+ ```
+
+ You can get around that behavior by passing `add_prefix_space=True` when instantiating this tokenizer or when you
+ call it on some text, but since the model was not pretrained this way, it might yield a decrease in performance.
+
+
+
+ When used with `is_split_into_words=True`, this tokenizer needs to be instantiated with `add_prefix_space=True`.
+
+
+
+ This tokenizer inherits from [`PreTrainedTokenizerFast`] which contains most of the main methods. Users should
+ refer to this superclass for more information regarding those methods.
+
+ Args:
+ vocab_file (`str`):
+ Path to the vocabulary file.
+ merges_file (`str`):
+ Path to the merges file.
+ errors (`str`, *optional*, defaults to `"replace"`):
+ Paradigm to follow when decoding bytes to UTF-8. See
+ [bytes.decode](https://docs.python.org/3/library/stdtypes.html#bytes.decode) for more information.
+ bos_token (`str`, *optional*, defaults to `""`):
+ The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.
+
+
+
+ When building a sequence using special tokens, this is not the token that is used for the beginning of
+ sequence. The token used is the `cls_token`.
+
+
+
+ eos_token (`str`, *optional*, defaults to `""`):
+ The end of sequence token.
+
+
+
+ When building a sequence using special tokens, this is not the token that is used for the end of sequence.
+ The token used is the `sep_token`.
+
+
+
+ sep_token (`str`, *optional*, defaults to `""`):
+ The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for
+ sequence classification or for a text and a question for question answering. It is also used as the last
+ token of a sequence built with special tokens.
+ cls_token (`str`, *optional*, defaults to `""`):
+ The classifier token which is used when doing sequence classification (classification of the whole sequence
+ instead of per-token classification). It is the first token of the sequence when built with special tokens.
+ unk_token (`str`, *optional*, defaults to `""`):
+ The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
+ token instead.
+ pad_token (`str`, *optional*, defaults to `""`):
+ The token used for padding, for example when batching sequences of different lengths.
+ mask_token (`str`, *optional*, defaults to `""`):
+ The token used for masking values. This is the token used when training this model with masked language
+ modeling. This is the token which the model will try to predict.
+ add_prefix_space (`bool`, *optional*, defaults to `False`):
+ Whether or not to add an initial space to the input. This allows to treat the leading word just as any
+ other word. (Longformer tokenizer detect beginning of words by the preceding space).
+ trim_offsets (`bool`, *optional*, defaults to `True`):
+ Whether the post processing step should trim offsets to avoid including whitespaces.
+ """
+
+ vocab_files_names = VOCAB_FILES_NAMES
+ pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+ max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+ model_input_names = ["input_ids", "attention_mask"]
+ slow_tokenizer_class = LongformerTokenizer
+
+ def __init__(
+ self,
+ vocab_file=None,
+ merges_file=None,
+ tokenizer_file=None,
+ errors="replace",
+ bos_token="",
+ eos_token="",
+ sep_token="",
+ cls_token="",
+ unk_token="",
+ pad_token="",
+ mask_token="",
+ add_prefix_space=False,
+ trim_offsets=True,
+ **kwargs,
+ ):
+ mask_token = (
+ AddedToken(mask_token, lstrip=True, rstrip=False, normalized=False)
+ if isinstance(mask_token, str)
+ else mask_token
+ )
+ super().__init__(
+ vocab_file,
+ merges_file,
+ tokenizer_file=tokenizer_file,
+ errors=errors,
+ bos_token=bos_token,
+ eos_token=eos_token,
+ sep_token=sep_token,
+ cls_token=cls_token,
+ unk_token=unk_token,
+ pad_token=pad_token,
+ mask_token=mask_token,
+ add_prefix_space=add_prefix_space,
+ trim_offsets=trim_offsets,
+ **kwargs,
+ )
+
+ pre_tok_state = json.loads(self.backend_tokenizer.pre_tokenizer.__getstate__())
+ if pre_tok_state.get("add_prefix_space", add_prefix_space) != add_prefix_space:
+ pre_tok_class = getattr(pre_tokenizers, pre_tok_state.pop("type"))
+ pre_tok_state["add_prefix_space"] = add_prefix_space
+ self.backend_tokenizer.pre_tokenizer = pre_tok_class(**pre_tok_state)
+
+ self.add_prefix_space = add_prefix_space
+
+ tokenizer_component = "post_processor"
+ tokenizer_component_instance = getattr(self.backend_tokenizer, tokenizer_component, None)
+ if tokenizer_component_instance:
+ state = json.loads(tokenizer_component_instance.__getstate__())
+
+ # The lists 'sep' and 'cls' must be cased in tuples for the object `post_processor_class`
+ if "sep" in state:
+ state["sep"] = tuple(state["sep"])
+ if "cls" in state:
+ state["cls"] = tuple(state["cls"])
+
+ changes_to_apply = False
+
+ if state.get("add_prefix_space", add_prefix_space) != add_prefix_space:
+ state["add_prefix_space"] = add_prefix_space
+ changes_to_apply = True
+
+ if state.get("trim_offsets", trim_offsets) != trim_offsets:
+ state["trim_offsets"] = trim_offsets
+ changes_to_apply = True
+
+ if changes_to_apply:
+ component_class = getattr(processors, state.pop("type"))
+ new_value = component_class(**state)
+ setattr(self.backend_tokenizer, tokenizer_component, new_value)
+
+ @property
+ def mask_token(self) -> str:
+ """
+ `str`: Mask token, to use when training a model with masked-language modeling. Log an error if used while not
+ having been set.
+
+ Longformer tokenizer has a special mask token to be usable in the fill-mask pipeline. The mask token will greedily
+ comprise the space before the **.
+ """
+ if self._mask_token is None:
+ if self.verbose:
+ logger.error("Using mask_token, but it is not set yet.")
+ return None
+ return str(self._mask_token)
+
+ @mask_token.setter
+ def mask_token(self, value):
+ """
+ Overriding the default behavior of the mask token to have it eat the space before it.
+
+ This is needed to preserve backward compatibility with all the previously used models based on Longformer.
+ """
+ # Mask token behave like a normal word, i.e. include the space before it
+ # So we set lstrip to True
+ value = AddedToken(value, lstrip=True, rstrip=False) if isinstance(value, str) else value
+ self._mask_token = value
+
+ def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
+ is_split_into_words = kwargs.get("is_split_into_words", False)
+ assert self.add_prefix_space or not is_split_into_words, (
+ f"You need to instantiate {self.__class__.__name__} with add_prefix_space=True "
+ "to use it with pretokenized inputs."
+ )
+
+ return super()._batch_encode_plus(*args, **kwargs)
+
+ def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
+ is_split_into_words = kwargs.get("is_split_into_words", False)
+
+ assert self.add_prefix_space or not is_split_into_words, (
+ f"You need to instantiate {self.__class__.__name__} with add_prefix_space=True "
+ "to use it with pretokenized inputs."
+ )
+
+ return super()._encode_plus(*args, **kwargs)
+
+ def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+ files = self._tokenizer.model.save(save_directory, name=filename_prefix)
+ return tuple(files)
+
+ def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+ output = [self.bos_token_id] + token_ids_0 + [self.eos_token_id]
+ if token_ids_1 is None:
+ return output
+
+ return output + [self.eos_token_id] + token_ids_1 + [self.eos_token_id]
+
+ def create_token_type_ids_from_sequences(
+ self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+ ) -> List[int]:
+ """
+ Create a mask from the two sequences passed to be used in a sequence-pair classification task. Longformer does not
+ make use of token type ids, therefore a list of zeros is returned.
+
+ Args:
+ token_ids_0 (`List[int]`):
+ List of IDs.
+ token_ids_1 (`List[int]`, *optional*):
+ Optional second list of IDs for sequence pairs.
+
+ Returns:
+ `List[int]`: List of zeros.
+ """
+ sep = [self.sep_token_id]
+ cls = [self.cls_token_id]
+
+ if token_ids_1 is None:
+ return len(cls + token_ids_0 + sep) * [0]
+ return len(cls + token_ids_0 + sep + sep + token_ids_1 + sep) * [0]
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart/convert_mbart_original_checkpoint_to_pytorch.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart/convert_mbart_original_checkpoint_to_pytorch.py
new file mode 100644
index 0000000000000000000000000000000000000000..eb7f00bf77107ff858a6131305f2e8bf6a17654b
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart/convert_mbart_original_checkpoint_to_pytorch.py
@@ -0,0 +1,83 @@
+# Copyright 2020 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+
+import torch
+from torch import nn
+
+from transformers import MBartConfig, MBartForConditionalGeneration
+
+
+def remove_ignore_keys_(state_dict):
+ ignore_keys = [
+ "encoder.version",
+ "decoder.version",
+ "model.encoder.version",
+ "model.decoder.version",
+ "_float_tensor",
+ "decoder.output_projection.weight",
+ ]
+ for k in ignore_keys:
+ state_dict.pop(k, None)
+
+
+def make_linear_from_emb(emb):
+ vocab_size, emb_size = emb.weight.shape
+ lin_layer = nn.Linear(vocab_size, emb_size, bias=False)
+ lin_layer.weight.data = emb.weight.data
+ return lin_layer
+
+
+def convert_fairseq_mbart_checkpoint_from_disk(
+ checkpoint_path, hf_config_path="facebook/mbart-large-en-ro", finetuned=False, mbart_50=False
+):
+ state_dict = torch.load(checkpoint_path, map_location="cpu")["model"]
+ remove_ignore_keys_(state_dict)
+ vocab_size = state_dict["encoder.embed_tokens.weight"].shape[0]
+
+ mbart_config = MBartConfig.from_pretrained(hf_config_path, vocab_size=vocab_size)
+ if mbart_50 and finetuned:
+ mbart_config.activation_function = "relu"
+
+ state_dict["shared.weight"] = state_dict["decoder.embed_tokens.weight"]
+ model = MBartForConditionalGeneration(mbart_config)
+ model.model.load_state_dict(state_dict)
+
+ if finetuned:
+ model.lm_head = make_linear_from_emb(model.model.shared)
+
+ return model
+
+
+if __name__ == "__main__":
+ parser = argparse.ArgumentParser()
+ # Required parameters
+ parser.add_argument(
+ "fairseq_path", type=str, help="bart.large, bart.large.cnn or a path to a model.pt on local filesystem."
+ )
+ parser.add_argument("pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
+ parser.add_argument(
+ "--hf_config",
+ default="facebook/mbart-large-cc25",
+ type=str,
+ help="Which huggingface architecture to use: mbart-large",
+ )
+ parser.add_argument("--mbart_50", action="store_true", help="whether the model is mMART-50 checkpoint")
+ parser.add_argument("--finetuned", action="store_true", help="whether the model is a fine-tuned checkpoint")
+ args = parser.parse_args()
+ model = convert_fairseq_mbart_checkpoint_from_disk(
+ args.fairseq_path, hf_config_path=args.hf_config, finetuned=args.finetuned, mbart_50=args.mbart_50
+ )
+ model.save_pretrained(args.pytorch_dump_folder_path)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart50/__init__.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart50/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..b889e374bb6d1e3afbf0b5f40cd34cbdc2ed468a
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart50/__init__.py
@@ -0,0 +1,58 @@
+# Copyright 2020 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import TYPE_CHECKING
+
+from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available, is_tokenizers_available
+
+
+_import_structure = {}
+
+try:
+ if not is_sentencepiece_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["tokenization_mbart50"] = ["MBart50Tokenizer"]
+
+try:
+ if not is_tokenizers_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["tokenization_mbart50_fast"] = ["MBart50TokenizerFast"]
+
+
+if TYPE_CHECKING:
+ try:
+ if not is_sentencepiece_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .tokenization_mbart50 import MBart50Tokenizer
+
+ try:
+ if not is_tokenizers_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .tokenization_mbart50_fast import MBart50TokenizerFast
+
+else:
+ import sys
+
+ sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart50/__pycache__/__init__.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart50/__pycache__/__init__.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..a715942d843615190f050975010b771bd1587e3c
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/mbart50/__pycache__/__init__.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/musicgen/__pycache__/__init__.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/musicgen/__pycache__/__init__.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..c6070896bd0981640632ab5b740ffbc65bbca2a4
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/musicgen/__pycache__/__init__.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__init__.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..d426ec93bf5859bc3ba040421c54ae4eefbbb32e
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__init__.py
@@ -0,0 +1,121 @@
+# Copyright 2021 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import TYPE_CHECKING
+
+from ...utils import (
+ OptionalDependencyNotAvailable,
+ _LazyModule,
+ is_flax_available,
+ is_tf_available,
+ is_torch_available,
+ is_vision_available,
+)
+
+
+_import_structure = {"configuration_vit": ["VIT_PRETRAINED_CONFIG_ARCHIVE_MAP", "ViTConfig", "ViTOnnxConfig"]}
+
+try:
+ if not is_vision_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["feature_extraction_vit"] = ["ViTFeatureExtractor"]
+ _import_structure["image_processing_vit"] = ["ViTImageProcessor"]
+
+try:
+ if not is_torch_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_vit"] = [
+ "VIT_PRETRAINED_MODEL_ARCHIVE_LIST",
+ "ViTForImageClassification",
+ "ViTForMaskedImageModeling",
+ "ViTModel",
+ "ViTPreTrainedModel",
+ ]
+
+try:
+ if not is_tf_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_tf_vit"] = [
+ "TFViTForImageClassification",
+ "TFViTModel",
+ "TFViTPreTrainedModel",
+ ]
+
+try:
+ if not is_flax_available():
+ raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+ pass
+else:
+ _import_structure["modeling_flax_vit"] = [
+ "FlaxViTForImageClassification",
+ "FlaxViTModel",
+ "FlaxViTPreTrainedModel",
+ ]
+
+if TYPE_CHECKING:
+ from .configuration_vit import VIT_PRETRAINED_CONFIG_ARCHIVE_MAP, ViTConfig, ViTOnnxConfig
+
+ try:
+ if not is_vision_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .feature_extraction_vit import ViTFeatureExtractor
+ from .image_processing_vit import ViTImageProcessor
+
+ try:
+ if not is_torch_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_vit import (
+ VIT_PRETRAINED_MODEL_ARCHIVE_LIST,
+ ViTForImageClassification,
+ ViTForMaskedImageModeling,
+ ViTModel,
+ ViTPreTrainedModel,
+ )
+
+ try:
+ if not is_tf_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_tf_vit import TFViTForImageClassification, TFViTModel, TFViTPreTrainedModel
+
+ try:
+ if not is_flax_available():
+ raise OptionalDependencyNotAvailable()
+ except OptionalDependencyNotAvailable:
+ pass
+ else:
+ from .modeling_flax_vit import FlaxViTForImageClassification, FlaxViTModel, FlaxViTPreTrainedModel
+
+
+else:
+ import sys
+
+ sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/__init__.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/__init__.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..caa90661aa36d39ac6a8321279f69f2317bac2cf
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/__init__.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/feature_extraction_vit.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/feature_extraction_vit.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..e46d9ba260161f841988029af85e5edae6a16ade
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/feature_extraction_vit.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/modeling_flax_vit.cpython-310.pyc b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/modeling_flax_vit.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..9fba775f71ca8a4352bf5d54ad263240b614d897
Binary files /dev/null and b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/__pycache__/modeling_flax_vit.cpython-310.pyc differ
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/configuration_vit.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/configuration_vit.py
new file mode 100644
index 0000000000000000000000000000000000000000..5eda0385c30c1ddd86933dc89bbb01f987757f9d
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/configuration_vit.py
@@ -0,0 +1,143 @@
+# coding=utf-8
+# Copyright 2021 Google AI and The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" ViT model configuration"""
+
+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
+
+
+logger = logging.get_logger(__name__)
+
+VIT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+ "google/vit-base-patch16-224": "https://huggingface.co/vit-base-patch16-224/resolve/main/config.json",
+ # See all ViT models at https://huggingface.co/models?filter=vit
+}
+
+
+class ViTConfig(PretrainedConfig):
+ r"""
+ This is the configuration class to store the configuration of a [`ViTModel`]. It is used to instantiate an ViT
+ model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
+ defaults will yield a similar configuration to that of the ViT
+ [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) architecture.
+
+ Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+ documentation from [`PretrainedConfig`] for more information.
+
+
+ Args:
+ hidden_size (`int`, *optional*, defaults to 768):
+ Dimensionality of the encoder layers and the pooler layer.
+ num_hidden_layers (`int`, *optional*, defaults to 12):
+ Number of hidden layers in the Transformer encoder.
+ num_attention_heads (`int`, *optional*, defaults to 12):
+ Number of attention heads for each attention layer in the Transformer encoder.
+ intermediate_size (`int`, *optional*, defaults to 3072):
+ Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
+ hidden_act (`str` or `function`, *optional*, defaults to `"gelu"`):
+ The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
+ `"relu"`, `"selu"` and `"gelu_new"` are supported.
+ hidden_dropout_prob (`float`, *optional*, defaults to 0.0):
+ The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
+ attention_probs_dropout_prob (`float`, *optional*, defaults to 0.0):
+ The dropout ratio for the attention probabilities.
+ initializer_range (`float`, *optional*, defaults to 0.02):
+ The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+ layer_norm_eps (`float`, *optional*, defaults to 1e-12):
+ The epsilon used by the layer normalization layers.
+ image_size (`int`, *optional*, defaults to 224):
+ The size (resolution) of each image.
+ patch_size (`int`, *optional*, defaults to 16):
+ The size (resolution) of each patch.
+ num_channels (`int`, *optional*, defaults to 3):
+ The number of input channels.
+ qkv_bias (`bool`, *optional*, defaults to `True`):
+ Whether to add a bias to the queries, keys and values.
+ encoder_stride (`int`, *optional*, defaults to 16):
+ Factor to increase the spatial resolution by in the decoder head for masked image modeling.
+
+ Example:
+
+ ```python
+ >>> from transformers import ViTConfig, ViTModel
+
+ >>> # Initializing a ViT vit-base-patch16-224 style configuration
+ >>> configuration = ViTConfig()
+
+ >>> # Initializing a model (with random weights) from the vit-base-patch16-224 style configuration
+ >>> model = ViTModel(configuration)
+
+ >>> # Accessing the model configuration
+ >>> configuration = model.config
+ ```"""
+
+ model_type = "vit"
+
+ def __init__(
+ self,
+ hidden_size=768,
+ num_hidden_layers=12,
+ num_attention_heads=12,
+ intermediate_size=3072,
+ hidden_act="gelu",
+ hidden_dropout_prob=0.0,
+ attention_probs_dropout_prob=0.0,
+ initializer_range=0.02,
+ layer_norm_eps=1e-12,
+ image_size=224,
+ patch_size=16,
+ num_channels=3,
+ qkv_bias=True,
+ encoder_stride=16,
+ **kwargs,
+ ):
+ super().__init__(**kwargs)
+
+ self.hidden_size = hidden_size
+ self.num_hidden_layers = num_hidden_layers
+ self.num_attention_heads = num_attention_heads
+ self.intermediate_size = intermediate_size
+ self.hidden_act = hidden_act
+ self.hidden_dropout_prob = hidden_dropout_prob
+ self.attention_probs_dropout_prob = attention_probs_dropout_prob
+ self.initializer_range = initializer_range
+ self.layer_norm_eps = layer_norm_eps
+ self.image_size = image_size
+ self.patch_size = patch_size
+ self.num_channels = num_channels
+ self.qkv_bias = qkv_bias
+ self.encoder_stride = encoder_stride
+
+
+class ViTOnnxConfig(OnnxConfig):
+ torch_onnx_minimum_version = version.parse("1.11")
+
+ @property
+ def inputs(self) -> Mapping[str, Mapping[int, str]]:
+ return OrderedDict(
+ [
+ ("pixel_values", {0: "batch", 1: "num_channels", 2: "height", 3: "width"}),
+ ]
+ )
+
+ @property
+ def atol_for_validation(self) -> float:
+ return 1e-4
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/convert_dino_to_pytorch.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/convert_dino_to_pytorch.py
new file mode 100644
index 0000000000000000000000000000000000000000..7eec823ad5d1d80a5a438693dbaee49189d7731f
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/convert_dino_to_pytorch.py
@@ -0,0 +1,219 @@
+# coding=utf-8
+# Copyright 2021 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Convert ViT checkpoints trained with the DINO method."""
+
+
+import argparse
+import json
+from pathlib import Path
+
+import requests
+import torch
+from huggingface_hub import hf_hub_download
+from PIL import Image
+
+from transformers import ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
+from transformers.utils import logging
+
+
+logging.set_verbosity_info()
+logger = logging.get_logger(__name__)
+
+
+# here we list all keys to be renamed (original name on the left, our name on the right)
+def create_rename_keys(config, base_model=False):
+ rename_keys = []
+ 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"),
+ ]
+ )
+
+ # if just the base model, we should remove "vit" from all keys that start with "vit"
+ rename_keys = [(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
+
+
+# we split up the matrix of each encoder layer into queries, keys and values
+def read_in_q_k_v(state_dict, config, base_model=False):
+ for i in range(config.num_hidden_layers):
+ if base_model:
+ prefix = ""
+ else:
+ prefix = "vit."
+ # read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
+ in_proj_weight = state_dict.pop(f"blocks.{i}.attn.qkv.weight")
+ in_proj_bias = state_dict.pop(f"blocks.{i}.attn.qkv.bias")
+ # next, add query, keys and values (in that order) to the state dict
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.query.weight"] = in_proj_weight[
+ : config.hidden_size, :
+ ]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.query.bias"] = in_proj_bias[: config.hidden_size]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.key.weight"] = in_proj_weight[
+ config.hidden_size : config.hidden_size * 2, :
+ ]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.key.bias"] = in_proj_bias[
+ config.hidden_size : config.hidden_size * 2
+ ]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.value.weight"] = in_proj_weight[
+ -config.hidden_size :, :
+ ]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.value.bias"] = in_proj_bias[-config.hidden_size :]
+
+
+def remove_classification_head_(state_dict):
+ ignore_keys = ["head.weight", "head.bias"]
+ for k in ignore_keys:
+ state_dict.pop(k, None)
+
+
+def rename_key(dct, old, new):
+ val = dct.pop(old)
+ dct[new] = val
+
+
+# We will verify our results on an image of cute cats
+def prepare_img():
+ url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+ im = Image.open(requests.get(url, stream=True).raw)
+ return im
+
+
+@torch.no_grad()
+def convert_vit_checkpoint(model_name, pytorch_dump_folder_path, base_model=True):
+ """
+ Copy/paste/tweak model's weights to our ViT structure.
+ """
+
+ # define default ViT configuration
+ config = ViTConfig()
+ # patch_size
+ if model_name[-1] == "8":
+ config.patch_size = 8
+ # set labels if required
+ if not base_model:
+ config.num_labels = 1000
+ repo_id = "huggingface/label-files"
+ filename = "imagenet-1k-id2label.json"
+ id2label = json.load(open(hf_hub_download(repo_id, filename, repo_type="dataset"), "r"))
+ id2label = {int(k): v for k, v in id2label.items()}
+ config.id2label = id2label
+ config.label2id = {v: k for k, v in id2label.items()}
+ # size of the architecture
+ if model_name in ["dino_vits8", "dino_vits16"]:
+ config.hidden_size = 384
+ config.intermediate_size = 1536
+ config.num_hidden_layers = 12
+ config.num_attention_heads = 6
+
+ # load original model from torch hub
+ original_model = torch.hub.load("facebookresearch/dino:main", model_name)
+ original_model.eval()
+
+ # load state_dict of original model, remove and rename some keys
+ state_dict = original_model.state_dict()
+ if base_model:
+ remove_classification_head_(state_dict)
+ rename_keys = create_rename_keys(config, base_model=base_model)
+ for src, dest in rename_keys:
+ rename_key(state_dict, src, dest)
+ read_in_q_k_v(state_dict, config, base_model)
+
+ # load HuggingFace model
+ if base_model:
+ model = ViTModel(config, add_pooling_layer=False).eval()
+ else:
+ model = ViTForImageClassification(config).eval()
+ model.load_state_dict(state_dict)
+
+ # Check outputs on an image, prepared by ViTImageProcessor
+ image_processor = ViTImageProcessor()
+ encoding = image_processor(images=prepare_img(), return_tensors="pt")
+ pixel_values = encoding["pixel_values"]
+ outputs = model(pixel_values)
+
+ if base_model:
+ final_hidden_state_cls_token = original_model(pixel_values)
+ assert torch.allclose(final_hidden_state_cls_token, outputs.last_hidden_state[:, 0, :], atol=1e-1)
+ else:
+ logits = original_model(pixel_values)
+ assert logits.shape == outputs.logits.shape
+ assert torch.allclose(logits, outputs.logits, atol=1e-3)
+
+ Path(pytorch_dump_folder_path).mkdir(exist_ok=True)
+ print(f"Saving model {model_name} to {pytorch_dump_folder_path}")
+ model.save_pretrained(pytorch_dump_folder_path)
+ print(f"Saving image processor to {pytorch_dump_folder_path}")
+ image_processor.save_pretrained(pytorch_dump_folder_path)
+
+
+if __name__ == "__main__":
+ parser = argparse.ArgumentParser()
+ # Required parameters
+ parser.add_argument(
+ "--model_name",
+ default="dino_vitb16",
+ type=str,
+ help="Name of the model trained with DINO 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(
+ "--base_model",
+ action="store_true",
+ help="Whether to only convert the base model (no projection head weights).",
+ )
+
+ parser.set_defaults(base_model=True)
+ args = parser.parse_args()
+ convert_vit_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.base_model)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/convert_vit_timm_to_pytorch.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/convert_vit_timm_to_pytorch.py
new file mode 100644
index 0000000000000000000000000000000000000000..0ccd9b9f6685fe375955fdee7298c17cf308de86
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/convert_vit_timm_to_pytorch.py
@@ -0,0 +1,255 @@
+# coding=utf-8
+# Copyright 2021 The HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Convert ViT and non-distilled DeiT checkpoints from the timm library."""
+
+
+import argparse
+from pathlib import Path
+
+import requests
+import timm
+import torch
+from PIL import Image
+from timm.data import ImageNetInfo, infer_imagenet_subset
+
+from transformers import DeiTImageProcessor, ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
+from transformers.utils import logging
+
+
+logging.set_verbosity_info()
+logger = logging.get_logger(__name__)
+
+
+# here we list all keys to be renamed (original name on the left, our name on the right)
+def create_rename_keys(config, base_model=False):
+ rename_keys = []
+ 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
+ rename_keys.extend(
+ [
+ ("norm.weight", "layernorm.weight"),
+ ("norm.bias", "layernorm.bias"),
+ ]
+ )
+
+ # if just the base model, we should remove "vit" from all keys that start with "vit"
+ rename_keys = [(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
+
+
+# we split up the matrix of each encoder layer into queries, keys and values
+def read_in_q_k_v(state_dict, config, base_model=False):
+ for i in range(config.num_hidden_layers):
+ if base_model:
+ prefix = ""
+ else:
+ prefix = "vit."
+ # read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
+ in_proj_weight = state_dict.pop(f"blocks.{i}.attn.qkv.weight")
+ in_proj_bias = state_dict.pop(f"blocks.{i}.attn.qkv.bias")
+ # next, add query, keys and values (in that order) to the state dict
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.query.weight"] = in_proj_weight[
+ : config.hidden_size, :
+ ]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.query.bias"] = in_proj_bias[: config.hidden_size]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.key.weight"] = in_proj_weight[
+ config.hidden_size : config.hidden_size * 2, :
+ ]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.key.bias"] = in_proj_bias[
+ config.hidden_size : config.hidden_size * 2
+ ]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.value.weight"] = in_proj_weight[
+ -config.hidden_size :, :
+ ]
+ state_dict[f"{prefix}encoder.layer.{i}.attention.attention.value.bias"] = in_proj_bias[-config.hidden_size :]
+
+
+def remove_classification_head_(state_dict):
+ ignore_keys = ["head.weight", "head.bias"]
+ for k in ignore_keys:
+ state_dict.pop(k, None)
+
+
+def rename_key(dct, old, new):
+ val = dct.pop(old)
+ dct[new] = val
+
+
+# We will verify our results on an image of cute cats
+def prepare_img():
+ url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+ im = Image.open(requests.get(url, stream=True).raw)
+ return im
+
+
+@torch.no_grad()
+def convert_vit_checkpoint(vit_name, pytorch_dump_folder_path):
+ """
+ Copy/paste/tweak model's weights to our ViT structure.
+ """
+
+ # define default ViT configuration
+ config = ViTConfig()
+ base_model = False
+
+ # load original model from timm
+ timm_model = timm.create_model(vit_name, pretrained=True)
+ timm_model.eval()
+
+ # detect unsupported ViT models in transformers
+ # fc_norm is present
+ if not isinstance(getattr(timm_model, "fc_norm", None), torch.nn.Identity):
+ raise ValueError(f"{vit_name} is not supported in transformers because of the presence of fc_norm.")
+
+ # use of global average pooling in combination (or without) class token
+ if getattr(timm_model, "global_pool", None) == "avg":
+ raise ValueError(f"{vit_name} is not supported in transformers because of use of global average pooling.")
+
+ # CLIP style vit with norm_pre layer present
+ if "clip" in vit_name and not isinstance(getattr(timm_model, "norm_pre", None), torch.nn.Identity):
+ raise ValueError(
+ f"{vit_name} is not supported in transformers because it's a CLIP style ViT with norm_pre layer."
+ )
+
+ # SigLIP style vit with attn_pool layer present
+ if "siglip" in vit_name and getattr(timm_model, "global_pool", None) == "map":
+ raise ValueError(
+ f"{vit_name} is not supported in transformers because it's a SigLIP style ViT with attn_pool."
+ )
+
+ # use of layer scale in ViT model blocks
+ if not isinstance(getattr(timm_model.blocks[0], "ls1", None), torch.nn.Identity) or not isinstance(
+ getattr(timm_model.blocks[0], "ls2", None), torch.nn.Identity
+ ):
+ raise ValueError(f"{vit_name} is not supported in transformers because it uses a layer scale in its blocks.")
+
+ # Hybrid ResNet-ViTs
+ if not isinstance(timm_model.patch_embed, timm.layers.PatchEmbed):
+ raise ValueError(f"{vit_name} is not supported in transformers because it is a hybrid ResNet-ViT.")
+
+ # get patch size and image size from the patch embedding submodule
+ config.patch_size = timm_model.patch_embed.patch_size[0]
+ config.image_size = timm_model.patch_embed.img_size[0]
+
+ # retrieve architecture-specific parameters from the timm model
+ config.hidden_size = timm_model.embed_dim
+ config.intermediate_size = timm_model.blocks[0].mlp.fc1.out_features
+ config.num_hidden_layers = len(timm_model.blocks)
+ config.num_attention_heads = timm_model.blocks[0].attn.num_heads
+
+ # check whether the model has a classification head or not
+ if timm_model.num_classes != 0:
+ config.num_labels = timm_model.num_classes
+ # infer ImageNet subset from timm model
+ imagenet_subset = infer_imagenet_subset(timm_model)
+ dataset_info = ImageNetInfo(imagenet_subset)
+ config.id2label = {i: dataset_info.index_to_label_name(i) for i in range(dataset_info.num_classes())}
+ config.label2id = {v: k for k, v in config.id2label.items()}
+ else:
+ print(f"{vit_name} is going to be converted as a feature extractor only.")
+ base_model = True
+
+ # load state_dict of original model
+ state_dict = timm_model.state_dict()
+
+ # remove and rename some keys in the state dict
+ if base_model:
+ remove_classification_head_(state_dict)
+ rename_keys = create_rename_keys(config, base_model)
+ for src, dest in rename_keys:
+ rename_key(state_dict, src, dest)
+ read_in_q_k_v(state_dict, config, base_model)
+
+ # load HuggingFace model
+ if base_model:
+ model = ViTModel(config, add_pooling_layer=False).eval()
+ else:
+ model = ViTForImageClassification(config).eval()
+ model.load_state_dict(state_dict)
+
+ # Check outputs on an image, prepared by ViTImageProcessor/DeiTImageProcessor
+ if "deit" in vit_name:
+ image_processor = DeiTImageProcessor(size=config.image_size)
+ else:
+ image_processor = ViTImageProcessor(size=config.image_size)
+ encoding = image_processor(images=prepare_img(), return_tensors="pt")
+ pixel_values = encoding["pixel_values"]
+ outputs = model(pixel_values)
+
+ if base_model:
+ timm_pooled_output = timm_model.forward_features(pixel_values)
+ assert timm_pooled_output.shape == outputs.last_hidden_state.shape
+ assert torch.allclose(timm_pooled_output, outputs.last_hidden_state, atol=1e-1)
+ else:
+ timm_logits = timm_model(pixel_values)
+ assert timm_logits.shape == outputs.logits.shape
+ assert torch.allclose(timm_logits, outputs.logits, atol=1e-3)
+
+ Path(pytorch_dump_folder_path).mkdir(exist_ok=True)
+ print(f"Saving model {vit_name} to {pytorch_dump_folder_path}")
+ model.save_pretrained(pytorch_dump_folder_path)
+ print(f"Saving image processor to {pytorch_dump_folder_path}")
+ image_processor.save_pretrained(pytorch_dump_folder_path)
+
+
+if __name__ == "__main__":
+ parser = 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."
+ )
+
+ args = parser.parse_args()
+ convert_vit_checkpoint(args.vit_name, args.pytorch_dump_folder_path)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/image_processing_vit.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/image_processing_vit.py
new file mode 100644
index 0000000000000000000000000000000000000000..be806d94c4d2f296d1b8c300caac8c5fd337673a
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/image_processing_vit.py
@@ -0,0 +1,267 @@
+# coding=utf-8
+# Copyright 2022 The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Image processor class for ViT."""
+
+from typing import Dict, List, Optional, Union
+
+import numpy as np
+
+from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
+from ...image_transforms import resize, to_channel_dimension_format
+from ...image_utils import (
+ IMAGENET_STANDARD_MEAN,
+ IMAGENET_STANDARD_STD,
+ ChannelDimension,
+ ImageInput,
+ PILImageResampling,
+ infer_channel_dimension_format,
+ is_scaled_image,
+ make_list_of_images,
+ to_numpy_array,
+ valid_images,
+)
+from ...utils import TensorType, logging
+
+
+logger = logging.get_logger(__name__)
+
+
+class ViTImageProcessor(BaseImageProcessor):
+ r"""
+ Constructs a ViT image processor.
+
+ Args:
+ do_resize (`bool`, *optional*, defaults to `True`):
+ Whether to resize the image's (height, width) dimensions to the specified `(size["height"],
+ size["width"])`. Can be overridden by the `do_resize` parameter in the `preprocess` method.
+ size (`dict`, *optional*, defaults to `{"height": 224, "width": 224}`):
+ Size of the output image after resizing. Can be overridden by the `size` parameter in the `preprocess`
+ method.
+ resample (`PILImageResampling`, *optional*, defaults to `Resampling.BILINEAR`):
+ Resampling filter to use if resizing the image. Can be overridden by the `resample` parameter in the
+ `preprocess` method.
+ do_rescale (`bool`, *optional*, defaults to `True`):
+ Whether to rescale the image by the specified scale `rescale_factor`. Can be overridden by the `do_rescale`
+ parameter in the `preprocess` method.
+ rescale_factor (`int` or `float`, *optional*, defaults to `1/255`):
+ Scale factor to use if rescaling the image. Can be overridden by the `rescale_factor` parameter in the
+ `preprocess` method.
+ do_normalize (`bool`, *optional*, defaults to `True`):
+ Whether to normalize the image. Can be overridden by the `do_normalize` parameter in the `preprocess`
+ method.
+ image_mean (`float` or `List[float]`, *optional*, defaults to `IMAGENET_STANDARD_MEAN`):
+ Mean to use if normalizing the image. This is a float or list of floats the length of the number of
+ channels in the image. Can be overridden by the `image_mean` parameter in the `preprocess` method.
+ image_std (`float` or `List[float]`, *optional*, defaults to `IMAGENET_STANDARD_STD`):
+ Standard deviation to use if normalizing the image. This is a float or list of floats the length of the
+ number of channels in the image. Can be overridden by the `image_std` parameter in the `preprocess` method.
+ """
+
+ model_input_names = ["pixel_values"]
+
+ def __init__(
+ self,
+ do_resize: bool = True,
+ size: Optional[Dict[str, int]] = None,
+ resample: PILImageResampling = PILImageResampling.BILINEAR,
+ do_rescale: bool = True,
+ rescale_factor: Union[int, float] = 1 / 255,
+ do_normalize: bool = True,
+ image_mean: Optional[Union[float, List[float]]] = None,
+ image_std: Optional[Union[float, List[float]]] = None,
+ **kwargs,
+ ) -> None:
+ super().__init__(**kwargs)
+ size = size if size is not None else {"height": 224, "width": 224}
+ size = get_size_dict(size)
+ self.do_resize = do_resize
+ self.do_rescale = do_rescale
+ self.do_normalize = do_normalize
+ self.size = size
+ self.resample = resample
+ self.rescale_factor = rescale_factor
+ self.image_mean = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN
+ self.image_std = image_std if image_std is not None else IMAGENET_STANDARD_STD
+
+ def resize(
+ self,
+ image: np.ndarray,
+ size: Dict[str, int],
+ resample: PILImageResampling = PILImageResampling.BILINEAR,
+ data_format: Optional[Union[str, ChannelDimension]] = None,
+ input_data_format: Optional[Union[str, ChannelDimension]] = None,
+ **kwargs,
+ ) -> np.ndarray:
+ """
+ Resize an image to `(size["height"], size["width"])`.
+
+ Args:
+ image (`np.ndarray`):
+ Image to resize.
+ size (`Dict[str, int]`):
+ Dictionary in the format `{"height": int, "width": int}` specifying the size of the output image.
+ resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`):
+ `PILImageResampling` filter to use when resizing the image e.g. `PILImageResampling.BILINEAR`.
+ data_format (`ChannelDimension` or `str`, *optional*):
+ The channel dimension format for the output image. If unset, the channel dimension format of the input
+ image is used. Can be one of:
+ - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+ - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+ - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+ input_data_format (`ChannelDimension` or `str`, *optional*):
+ The channel dimension format for the input image. If unset, the channel dimension format is inferred
+ from the input image. Can be one of:
+ - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+ - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+ - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+
+ Returns:
+ `np.ndarray`: The resized image.
+ """
+ size = get_size_dict(size)
+ if "height" not in size or "width" not in size:
+ raise ValueError(f"The `size` dictionary must contain the keys `height` and `width`. Got {size.keys()}")
+ output_size = (size["height"], size["width"])
+ return resize(
+ image,
+ size=output_size,
+ resample=resample,
+ data_format=data_format,
+ input_data_format=input_data_format,
+ **kwargs,
+ )
+
+ def preprocess(
+ self,
+ images: ImageInput,
+ do_resize: Optional[bool] = None,
+ size: Dict[str, int] = None,
+ resample: PILImageResampling = None,
+ do_rescale: Optional[bool] = None,
+ rescale_factor: Optional[float] = None,
+ do_normalize: Optional[bool] = None,
+ image_mean: Optional[Union[float, List[float]]] = None,
+ image_std: Optional[Union[float, List[float]]] = None,
+ return_tensors: Optional[Union[str, TensorType]] = None,
+ data_format: Union[str, ChannelDimension] = ChannelDimension.FIRST,
+ input_data_format: Optional[Union[str, ChannelDimension]] = None,
+ **kwargs,
+ ):
+ """
+ Preprocess an image or batch of images.
+
+ Args:
+ images (`ImageInput`):
+ Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If
+ passing in images with pixel values between 0 and 1, set `do_rescale=False`.
+ do_resize (`bool`, *optional*, defaults to `self.do_resize`):
+ Whether to resize the image.
+ size (`Dict[str, int]`, *optional*, defaults to `self.size`):
+ Dictionary in the format `{"height": h, "width": w}` specifying the size of the output image after
+ resizing.
+ resample (`PILImageResampling` filter, *optional*, defaults to `self.resample`):
+ `PILImageResampling` filter to use if resizing the image e.g. `PILImageResampling.BILINEAR`. Only has
+ an effect if `do_resize` is set to `True`.
+ do_rescale (`bool`, *optional*, defaults to `self.do_rescale`):
+ Whether to rescale the image values between [0 - 1].
+ rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`):
+ Rescale factor to rescale the image by if `do_rescale` is set to `True`.
+ do_normalize (`bool`, *optional*, defaults to `self.do_normalize`):
+ Whether to normalize the image.
+ image_mean (`float` or `List[float]`, *optional*, defaults to `self.image_mean`):
+ Image mean to use if `do_normalize` is set to `True`.
+ image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`):
+ Image standard deviation to use if `do_normalize` is set to `True`.
+ return_tensors (`str` or `TensorType`, *optional*):
+ The type of tensors to return. Can be one of:
+ - Unset: Return a list of `np.ndarray`.
+ - `TensorType.TENSORFLOW` or `'tf'`: Return a batch of type `tf.Tensor`.
+ - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`.
+ - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.ndarray`.
+ - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
+ data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
+ The channel dimension format for the output image. Can be one of:
+ - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+ - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+ - Unset: Use the channel dimension format of the input image.
+ input_data_format (`ChannelDimension` or `str`, *optional*):
+ The channel dimension format for the input image. If unset, the channel dimension format is inferred
+ from the input image. Can be one of:
+ - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+ - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+ - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+ """
+ do_resize = do_resize if do_resize is not None else self.do_resize
+ do_rescale = do_rescale if do_rescale is not None else self.do_rescale
+ do_normalize = do_normalize if do_normalize is not None else self.do_normalize
+ resample = resample if resample is not None else self.resample
+ rescale_factor = rescale_factor if rescale_factor is not None else self.rescale_factor
+ image_mean = image_mean if image_mean is not None else self.image_mean
+ image_std = image_std if image_std is not None else self.image_std
+
+ size = size if size is not None else self.size
+ size_dict = get_size_dict(size)
+
+ images = make_list_of_images(images)
+
+ if not valid_images(images):
+ raise ValueError(
+ "Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, "
+ "torch.Tensor, tf.Tensor or jax.ndarray."
+ )
+
+ if do_resize and size is None:
+ raise ValueError("Size must be specified if do_resize is True.")
+
+ if do_rescale and rescale_factor is None:
+ raise ValueError("Rescale factor must be specified if do_rescale is True.")
+
+ # All transformations expect numpy arrays.
+ images = [to_numpy_array(image) for image in images]
+
+ if is_scaled_image(images[0]) and do_rescale:
+ logger.warning_once(
+ "It looks like you are trying to rescale already rescaled images. If the input"
+ " images have pixel values between 0 and 1, set `do_rescale=False` to avoid rescaling them again."
+ )
+
+ if input_data_format is None:
+ # We assume that all images have the same channel dimension format.
+ input_data_format = infer_channel_dimension_format(images[0])
+
+ if do_resize:
+ images = [
+ self.resize(image=image, size=size_dict, resample=resample, input_data_format=input_data_format)
+ for image in images
+ ]
+
+ if do_rescale:
+ images = [
+ self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+ for image in images
+ ]
+
+ if do_normalize:
+ images = [
+ self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format)
+ for image in images
+ ]
+
+ images = [
+ to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) for image in images
+ ]
+
+ data = {"pixel_values": images}
+ return BatchFeature(data=data, tensor_type=return_tensors)
diff --git a/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/modeling_vit.py b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/modeling_vit.py
new file mode 100644
index 0000000000000000000000000000000000000000..734ccf6a9e80f4a5f231b92a00bc0c3a1037c5a8
--- /dev/null
+++ b/evalkit_tf437/lib/python3.10/site-packages/transformers/models/vit/modeling_vit.py
@@ -0,0 +1,841 @@
+# coding=utf-8
+# Copyright 2021 Google AI, Ross Wightman, The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" PyTorch ViT model."""
+
+
+import collections.abc
+import math
+from typing import Dict, List, Optional, Set, Tuple, Union
+
+import torch
+import torch.utils.checkpoint
+from torch import nn
+from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
+
+from ...activations import ACT2FN
+from ...modeling_outputs import (
+ BaseModelOutput,
+ BaseModelOutputWithPooling,
+ ImageClassifierOutput,
+ MaskedImageModelingOutput,
+)
+from ...modeling_utils import PreTrainedModel
+from ...pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer
+from ...utils import (
+ add_code_sample_docstrings,
+ add_start_docstrings,
+ add_start_docstrings_to_model_forward,
+ logging,
+ replace_return_docstrings,
+)
+from .configuration_vit import ViTConfig
+
+
+logger = logging.get_logger(__name__)
+
+# General docstring
+_CONFIG_FOR_DOC = "ViTConfig"
+
+# Base docstring
+_CHECKPOINT_FOR_DOC = "google/vit-base-patch16-224-in21k"
+_EXPECTED_OUTPUT_SHAPE = [1, 197, 768]
+
+# Image classification docstring
+_IMAGE_CLASS_CHECKPOINT = "google/vit-base-patch16-224"
+_IMAGE_CLASS_EXPECTED_OUTPUT = "Egyptian cat"
+
+
+VIT_PRETRAINED_MODEL_ARCHIVE_LIST = [
+ "google/vit-base-patch16-224",
+ # See all ViT models at https://huggingface.co/models?filter=vit
+]
+
+
+class ViTEmbeddings(nn.Module):
+ """
+ Construct the CLS token, position and patch embeddings. Optionally, also the mask token.
+ """
+
+ def __init__(self, config: ViTConfig, use_mask_token: bool = False) -> None:
+ super().__init__()
+
+ self.cls_token = nn.Parameter(torch.randn(1, 1, config.hidden_size))
+ self.mask_token = nn.Parameter(torch.zeros(1, 1, config.hidden_size)) if use_mask_token else None
+ self.patch_embeddings = ViTPatchEmbeddings(config)
+ num_patches = self.patch_embeddings.num_patches
+ self.position_embeddings = nn.Parameter(torch.randn(1, num_patches + 1, config.hidden_size))
+ self.dropout = nn.Dropout(config.hidden_dropout_prob)
+ self.config = config
+
+ def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: int, width: int) -> torch.Tensor:
+ """
+ This method allows to interpolate the pre-trained position encodings, to be able to use the model on higher
+ resolution images.
+
+ Source:
+ https://github.com/facebookresearch/dino/blob/de9ee3df6cf39fac952ab558447af1fa1365362a/vision_transformer.py#L174
+ """
+
+ num_patches = embeddings.shape[1] - 1
+ num_positions = self.position_embeddings.shape[1] - 1
+ if num_patches == num_positions and height == width:
+ return self.position_embeddings
+ class_pos_embed = self.position_embeddings[:, 0]
+ patch_pos_embed = self.position_embeddings[:, 1:]
+ dim = embeddings.shape[-1]
+ h0 = height // self.config.patch_size
+ w0 = width // self.config.patch_size
+ # we add a small number to avoid floating point error in the interpolation
+ # see discussion at https://github.com/facebookresearch/dino/issues/8
+ h0, w0 = h0 + 0.1, w0 + 0.1
+ patch_pos_embed = patch_pos_embed.reshape(1, int(math.sqrt(num_positions)), int(math.sqrt(num_positions)), dim)
+ patch_pos_embed = patch_pos_embed.permute(0, 3, 1, 2)
+ patch_pos_embed = nn.functional.interpolate(
+ patch_pos_embed,
+ scale_factor=(h0 / math.sqrt(num_positions), w0 / math.sqrt(num_positions)),
+ mode="bicubic",
+ align_corners=False,
+ )
+ assert int(h0) == patch_pos_embed.shape[-2] and int(w0) == patch_pos_embed.shape[-1]
+ patch_pos_embed = patch_pos_embed.permute(0, 2, 3, 1).view(1, -1, dim)
+ return torch.cat((class_pos_embed.unsqueeze(0), patch_pos_embed), dim=1)
+
+ def forward(
+ self,
+ pixel_values: torch.Tensor,
+ bool_masked_pos: Optional[torch.BoolTensor] = None,
+ interpolate_pos_encoding: bool = False,
+ ) -> torch.Tensor:
+ batch_size, num_channels, height, width = pixel_values.shape
+ embeddings = self.patch_embeddings(pixel_values, interpolate_pos_encoding=interpolate_pos_encoding)
+
+ if bool_masked_pos is not None:
+ seq_length = embeddings.shape[1]
+ mask_tokens = self.mask_token.expand(batch_size, seq_length, -1)
+ # replace the masked visual tokens by mask_tokens
+ mask = bool_masked_pos.unsqueeze(-1).type_as(mask_tokens)
+ embeddings = embeddings * (1.0 - mask) + mask_tokens * mask
+
+ # add the [CLS] token to the embedded patch tokens
+ cls_tokens = self.cls_token.expand(batch_size, -1, -1)
+ embeddings = torch.cat((cls_tokens, embeddings), dim=1)
+
+ # add positional encoding to each token
+ if interpolate_pos_encoding:
+ embeddings = embeddings + self.interpolate_pos_encoding(embeddings, height, width)
+ else:
+ embeddings = embeddings + self.position_embeddings
+
+ embeddings = self.dropout(embeddings)
+
+ return embeddings
+
+
+class ViTPatchEmbeddings(nn.Module):
+ """
+ This class turns `pixel_values` of shape `(batch_size, num_channels, height, width)` into the initial
+ `hidden_states` (patch embeddings) of shape `(batch_size, seq_length, hidden_size)` to be consumed by a
+ Transformer.
+ """
+
+ def __init__(self, config):
+ super().__init__()
+ image_size, patch_size = config.image_size, config.patch_size
+ num_channels, hidden_size = config.num_channels, config.hidden_size
+
+ image_size = image_size if isinstance(image_size, collections.abc.Iterable) else (image_size, image_size)
+ patch_size = patch_size if isinstance(patch_size, collections.abc.Iterable) else (patch_size, patch_size)
+ num_patches = (image_size[1] // patch_size[1]) * (image_size[0] // patch_size[0])
+ self.image_size = image_size
+ self.patch_size = patch_size
+ self.num_channels = num_channels
+ self.num_patches = num_patches
+
+ self.projection = nn.Conv2d(num_channels, hidden_size, kernel_size=patch_size, stride=patch_size)
+
+ def forward(self, pixel_values: torch.Tensor, interpolate_pos_encoding: bool = False) -> torch.Tensor:
+ batch_size, num_channels, height, width = pixel_values.shape
+ if num_channels != self.num_channels:
+ raise ValueError(
+ "Make sure that the channel dimension of the pixel values match with the one set in the configuration."
+ f" Expected {self.num_channels} but got {num_channels}."
+ )
+ if not interpolate_pos_encoding:
+ if height != self.image_size[0] or width != self.image_size[1]:
+ raise ValueError(
+ f"Input image size ({height}*{width}) doesn't match model"
+ f" ({self.image_size[0]}*{self.image_size[1]})."
+ )
+ embeddings = self.projection(pixel_values).flatten(2).transpose(1, 2)
+ return embeddings
+
+
+class ViTSelfAttention(nn.Module):
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__()
+ if config.hidden_size % config.num_attention_heads != 0 and not hasattr(config, "embedding_size"):
+ raise ValueError(
+ f"The hidden size {config.hidden_size,} is not a multiple of the number of attention "
+ f"heads {config.num_attention_heads}."
+ )
+
+ self.num_attention_heads = config.num_attention_heads
+ self.attention_head_size = int(config.hidden_size / config.num_attention_heads)
+ self.all_head_size = self.num_attention_heads * self.attention_head_size
+
+ self.query = nn.Linear(config.hidden_size, self.all_head_size, bias=config.qkv_bias)
+ self.key = nn.Linear(config.hidden_size, self.all_head_size, bias=config.qkv_bias)
+ self.value = nn.Linear(config.hidden_size, self.all_head_size, bias=config.qkv_bias)
+
+ self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
+
+ def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
+ new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.attention_head_size)
+ x = x.view(new_x_shape)
+ return x.permute(0, 2, 1, 3)
+
+ def forward(
+ self, hidden_states, head_mask: Optional[torch.Tensor] = None, output_attentions: bool = False
+ ) -> Union[Tuple[torch.Tensor, torch.Tensor], Tuple[torch.Tensor]]:
+ mixed_query_layer = self.query(hidden_states)
+
+ key_layer = self.transpose_for_scores(self.key(hidden_states))
+ value_layer = self.transpose_for_scores(self.value(hidden_states))
+ query_layer = self.transpose_for_scores(mixed_query_layer)
+
+ # Take the dot product between "query" and "key" to get the raw attention scores.
+ attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2))
+
+ attention_scores = attention_scores / math.sqrt(self.attention_head_size)
+
+ # Normalize the attention scores to probabilities.
+ attention_probs = nn.functional.softmax(attention_scores, dim=-1)
+
+ # This is actually dropping out entire tokens to attend to, which might
+ # seem a bit unusual, but is taken from the original Transformer paper.
+ attention_probs = self.dropout(attention_probs)
+
+ # Mask heads if we want to
+ if head_mask is not None:
+ attention_probs = attention_probs * head_mask
+
+ context_layer = torch.matmul(attention_probs, value_layer)
+
+ context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
+ new_context_layer_shape = context_layer.size()[:-2] + (self.all_head_size,)
+ context_layer = context_layer.view(new_context_layer_shape)
+
+ outputs = (context_layer, attention_probs) if output_attentions else (context_layer,)
+
+ return outputs
+
+
+class ViTSelfOutput(nn.Module):
+ """
+ The residual connection is defined in ViTLayer instead of here (as is the case with other models), due to the
+ layernorm applied before each block.
+ """
+
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__()
+ self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+ self.dropout = nn.Dropout(config.hidden_dropout_prob)
+
+ def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Tensor) -> torch.Tensor:
+ hidden_states = self.dense(hidden_states)
+ hidden_states = self.dropout(hidden_states)
+
+ return hidden_states
+
+
+class ViTAttention(nn.Module):
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__()
+ self.attention = ViTSelfAttention(config)
+ self.output = ViTSelfOutput(config)
+ self.pruned_heads = set()
+
+ def prune_heads(self, heads: Set[int]) -> None:
+ if len(heads) == 0:
+ return
+ heads, index = find_pruneable_heads_and_indices(
+ heads, self.attention.num_attention_heads, self.attention.attention_head_size, self.pruned_heads
+ )
+
+ # Prune linear layers
+ self.attention.query = prune_linear_layer(self.attention.query, index)
+ self.attention.key = prune_linear_layer(self.attention.key, index)
+ self.attention.value = prune_linear_layer(self.attention.value, index)
+ self.output.dense = prune_linear_layer(self.output.dense, index, dim=1)
+
+ # Update hyper params and store pruned heads
+ self.attention.num_attention_heads = self.attention.num_attention_heads - len(heads)
+ self.attention.all_head_size = self.attention.attention_head_size * self.attention.num_attention_heads
+ self.pruned_heads = self.pruned_heads.union(heads)
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ head_mask: Optional[torch.Tensor] = None,
+ output_attentions: bool = False,
+ ) -> Union[Tuple[torch.Tensor, torch.Tensor], Tuple[torch.Tensor]]:
+ self_outputs = self.attention(hidden_states, head_mask, output_attentions)
+
+ attention_output = self.output(self_outputs[0], hidden_states)
+
+ outputs = (attention_output,) + self_outputs[1:] # add attentions if we output them
+ return outputs
+
+
+class ViTIntermediate(nn.Module):
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__()
+ self.dense = nn.Linear(config.hidden_size, config.intermediate_size)
+ if isinstance(config.hidden_act, str):
+ self.intermediate_act_fn = ACT2FN[config.hidden_act]
+ else:
+ self.intermediate_act_fn = config.hidden_act
+
+ def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+ hidden_states = self.dense(hidden_states)
+ hidden_states = self.intermediate_act_fn(hidden_states)
+
+ return hidden_states
+
+
+class ViTOutput(nn.Module):
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__()
+ self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
+ self.dropout = nn.Dropout(config.hidden_dropout_prob)
+
+ def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Tensor) -> torch.Tensor:
+ hidden_states = self.dense(hidden_states)
+ hidden_states = self.dropout(hidden_states)
+
+ hidden_states = hidden_states + input_tensor
+
+ return hidden_states
+
+
+class ViTLayer(nn.Module):
+ """This corresponds to the Block class in the timm implementation."""
+
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__()
+ self.chunk_size_feed_forward = config.chunk_size_feed_forward
+ self.seq_len_dim = 1
+ self.attention = ViTAttention(config)
+ self.intermediate = ViTIntermediate(config)
+ self.output = ViTOutput(config)
+ self.layernorm_before = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+ self.layernorm_after = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ head_mask: Optional[torch.Tensor] = None,
+ output_attentions: bool = False,
+ ) -> Union[Tuple[torch.Tensor, torch.Tensor], Tuple[torch.Tensor]]:
+ self_attention_outputs = self.attention(
+ self.layernorm_before(hidden_states), # in ViT, layernorm is applied before self-attention
+ head_mask,
+ output_attentions=output_attentions,
+ )
+ attention_output = self_attention_outputs[0]
+ outputs = self_attention_outputs[1:] # add self attentions if we output attention weights
+
+ # first residual connection
+ hidden_states = attention_output + hidden_states
+
+ # in ViT, layernorm is also applied after self-attention
+ layer_output = self.layernorm_after(hidden_states)
+ layer_output = self.intermediate(layer_output)
+
+ # second residual connection is done here
+ layer_output = self.output(layer_output, hidden_states)
+
+ outputs = (layer_output,) + outputs
+
+ return outputs
+
+
+class ViTEncoder(nn.Module):
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__()
+ self.config = config
+ self.layer = nn.ModuleList([ViTLayer(config) for _ in range(config.num_hidden_layers)])
+ self.gradient_checkpointing = False
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ head_mask: Optional[torch.Tensor] = None,
+ output_attentions: bool = False,
+ output_hidden_states: bool = False,
+ return_dict: bool = True,
+ ) -> Union[tuple, BaseModelOutput]:
+ all_hidden_states = () if output_hidden_states else None
+ all_self_attentions = () if output_attentions else None
+
+ for i, layer_module in enumerate(self.layer):
+ if output_hidden_states:
+ all_hidden_states = all_hidden_states + (hidden_states,)
+
+ layer_head_mask = head_mask[i] if head_mask is not None else None
+
+ if self.gradient_checkpointing and self.training:
+ layer_outputs = self._gradient_checkpointing_func(
+ layer_module.__call__,
+ hidden_states,
+ layer_head_mask,
+ output_attentions,
+ )
+ else:
+ layer_outputs = layer_module(hidden_states, layer_head_mask, output_attentions)
+
+ hidden_states = layer_outputs[0]
+
+ if output_attentions:
+ all_self_attentions = all_self_attentions + (layer_outputs[1],)
+
+ if output_hidden_states:
+ all_hidden_states = all_hidden_states + (hidden_states,)
+
+ if not return_dict:
+ return tuple(v for v in [hidden_states, all_hidden_states, all_self_attentions] if v is not None)
+ return BaseModelOutput(
+ last_hidden_state=hidden_states,
+ hidden_states=all_hidden_states,
+ attentions=all_self_attentions,
+ )
+
+
+class ViTPreTrainedModel(PreTrainedModel):
+ """
+ An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
+ models.
+ """
+
+ config_class = ViTConfig
+ base_model_prefix = "vit"
+ main_input_name = "pixel_values"
+ supports_gradient_checkpointing = True
+ _no_split_modules = ["ViTEmbeddings", "ViTLayer"]
+
+ def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNorm]) -> None:
+ """Initialize the weights"""
+ if isinstance(module, (nn.Linear, nn.Conv2d)):
+ # Upcast the input in `fp32` and cast it back to desired `dtype` to avoid
+ # `trunc_normal_cpu` not implemented in `half` issues
+ module.weight.data = nn.init.trunc_normal_(
+ module.weight.data.to(torch.float32), mean=0.0, std=self.config.initializer_range
+ ).to(module.weight.dtype)
+ if module.bias is not None:
+ module.bias.data.zero_()
+ elif isinstance(module, nn.LayerNorm):
+ module.bias.data.zero_()
+ module.weight.data.fill_(1.0)
+ elif isinstance(module, ViTEmbeddings):
+ module.position_embeddings.data = nn.init.trunc_normal_(
+ module.position_embeddings.data.to(torch.float32),
+ mean=0.0,
+ std=self.config.initializer_range,
+ ).to(module.position_embeddings.dtype)
+
+ module.cls_token.data = nn.init.trunc_normal_(
+ module.cls_token.data.to(torch.float32),
+ mean=0.0,
+ std=self.config.initializer_range,
+ ).to(module.cls_token.dtype)
+
+
+VIT_START_DOCSTRING = r"""
+ This model is a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass. Use it
+ as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and
+ behavior.
+
+ Parameters:
+ config ([`ViTConfig`]): Model configuration class with all the parameters of the model.
+ Initializing with a config file does not load the weights associated with the model, only the
+ configuration. Check out the [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+VIT_INPUTS_DOCSTRING = r"""
+ Args:
+ pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):
+ Pixel values. Pixel values can be obtained using [`AutoImageProcessor`]. See [`ViTImageProcessor.__call__`]
+ for details.
+
+ head_mask (`torch.FloatTensor` of shape `(num_heads,)` or `(num_layers, num_heads)`, *optional*):
+ Mask to nullify selected heads of the self-attention modules. Mask values selected in `[0, 1]`:
+
+ - 1 indicates the head is **not masked**,
+ - 0 indicates the head is **masked**.
+
+ output_attentions (`bool`, *optional*):
+ Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+ tensors for more detail.
+ output_hidden_states (`bool`, *optional*):
+ Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+ more detail.
+ interpolate_pos_encoding (`bool`, *optional*):
+ Whether to interpolate the pre-trained position encodings.
+ return_dict (`bool`, *optional*):
+ Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+"""
+
+
+@add_start_docstrings(
+ "The bare ViT Model transformer outputting raw hidden-states without any specific head on top.",
+ VIT_START_DOCSTRING,
+)
+class ViTModel(ViTPreTrainedModel):
+ def __init__(self, config: ViTConfig, add_pooling_layer: bool = True, use_mask_token: bool = False):
+ super().__init__(config)
+ self.config = config
+
+ self.embeddings = ViTEmbeddings(config, use_mask_token=use_mask_token)
+ self.encoder = ViTEncoder(config)
+
+ self.layernorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)
+ self.pooler = ViTPooler(config) if add_pooling_layer else None
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ def get_input_embeddings(self) -> ViTPatchEmbeddings:
+ return self.embeddings.patch_embeddings
+
+ def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
+ """
+ Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base
+ class PreTrainedModel
+ """
+ for layer, heads in heads_to_prune.items():
+ self.encoder.layer[layer].attention.prune_heads(heads)
+
+ @add_start_docstrings_to_model_forward(VIT_INPUTS_DOCSTRING)
+ @add_code_sample_docstrings(
+ checkpoint=_CHECKPOINT_FOR_DOC,
+ output_type=BaseModelOutputWithPooling,
+ config_class=_CONFIG_FOR_DOC,
+ modality="vision",
+ expected_output=_EXPECTED_OUTPUT_SHAPE,
+ )
+ def forward(
+ self,
+ pixel_values: Optional[torch.Tensor] = None,
+ bool_masked_pos: Optional[torch.BoolTensor] = None,
+ head_mask: Optional[torch.Tensor] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ interpolate_pos_encoding: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[Tuple, BaseModelOutputWithPooling]:
+ r"""
+ bool_masked_pos (`torch.BoolTensor` of shape `(batch_size, num_patches)`, *optional*):
+ Boolean masked positions. Indicates which patches are masked (1) and which aren't (0).
+ """
+ output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+ output_hidden_states = (
+ output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+ )
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+ if pixel_values is None:
+ raise ValueError("You have to specify pixel_values")
+
+ # Prepare head mask if needed
+ # 1.0 in head_mask indicate we keep the head
+ # attention_probs has shape bsz x n_heads x N x N
+ # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
+ # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
+ head_mask = self.get_head_mask(head_mask, self.config.num_hidden_layers)
+
+ # TODO: maybe have a cleaner way to cast the input (from `ImageProcessor` side?)
+ expected_dtype = self.embeddings.patch_embeddings.projection.weight.dtype
+ if pixel_values.dtype != expected_dtype:
+ pixel_values = pixel_values.to(expected_dtype)
+
+ embedding_output = self.embeddings(
+ pixel_values, bool_masked_pos=bool_masked_pos, interpolate_pos_encoding=interpolate_pos_encoding
+ )
+
+ encoder_outputs = self.encoder(
+ embedding_output,
+ head_mask=head_mask,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ return_dict=return_dict,
+ )
+ sequence_output = encoder_outputs[0]
+ sequence_output = self.layernorm(sequence_output)
+ pooled_output = self.pooler(sequence_output) if self.pooler is not None else None
+
+ if not return_dict:
+ head_outputs = (sequence_output, pooled_output) if pooled_output is not None else (sequence_output,)
+ return head_outputs + encoder_outputs[1:]
+
+ return BaseModelOutputWithPooling(
+ last_hidden_state=sequence_output,
+ pooler_output=pooled_output,
+ hidden_states=encoder_outputs.hidden_states,
+ attentions=encoder_outputs.attentions,
+ )
+
+
+class ViTPooler(nn.Module):
+ def __init__(self, config: ViTConfig):
+ super().__init__()
+ self.dense = nn.Linear(config.hidden_size, config.hidden_size)
+ self.activation = nn.Tanh()
+
+ def forward(self, hidden_states):
+ # We "pool" the model by simply taking the hidden state corresponding
+ # to the first token.
+ first_token_tensor = hidden_states[:, 0]
+ pooled_output = self.dense(first_token_tensor)
+ pooled_output = self.activation(pooled_output)
+ return pooled_output
+
+
+@add_start_docstrings(
+ """ViT Model with a decoder on top for masked image modeling, as proposed in [SimMIM](https://arxiv.org/abs/2111.09886).
+
+
+
+ Note that we provide a script to pre-train this model on custom data in our [examples
+ directory](https://github.com/huggingface/transformers/tree/main/examples/pytorch/image-pretraining).
+
+
+ """,
+ VIT_START_DOCSTRING,
+)
+class ViTForMaskedImageModeling(ViTPreTrainedModel):
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__(config)
+
+ self.vit = ViTModel(config, add_pooling_layer=False, use_mask_token=True)
+
+ self.decoder = nn.Sequential(
+ nn.Conv2d(
+ in_channels=config.hidden_size,
+ out_channels=config.encoder_stride**2 * config.num_channels,
+ kernel_size=1,
+ ),
+ nn.PixelShuffle(config.encoder_stride),
+ )
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ @add_start_docstrings_to_model_forward(VIT_INPUTS_DOCSTRING)
+ @replace_return_docstrings(output_type=MaskedImageModelingOutput, config_class=_CONFIG_FOR_DOC)
+ def forward(
+ self,
+ pixel_values: Optional[torch.Tensor] = None,
+ bool_masked_pos: Optional[torch.BoolTensor] = None,
+ head_mask: Optional[torch.Tensor] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ interpolate_pos_encoding: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[tuple, MaskedImageModelingOutput]:
+ r"""
+ bool_masked_pos (`torch.BoolTensor` of shape `(batch_size, num_patches)`):
+ Boolean masked positions. Indicates which patches are masked (1) and which aren't (0).
+
+ Returns:
+
+ Examples:
+ ```python
+ >>> from transformers import AutoImageProcessor, ViTForMaskedImageModeling
+ >>> import torch
+ >>> from PIL import Image
+ >>> import requests
+
+ >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+ >>> image = Image.open(requests.get(url, stream=True).raw)
+
+ >>> image_processor = AutoImageProcessor.from_pretrained("google/vit-base-patch16-224-in21k")
+ >>> model = ViTForMaskedImageModeling.from_pretrained("google/vit-base-patch16-224-in21k")
+
+ >>> num_patches = (model.config.image_size // model.config.patch_size) ** 2
+ >>> pixel_values = image_processor(images=image, return_tensors="pt").pixel_values
+ >>> # create random boolean mask of shape (batch_size, num_patches)
+ >>> bool_masked_pos = torch.randint(low=0, high=2, size=(1, num_patches)).bool()
+
+ >>> outputs = model(pixel_values, bool_masked_pos=bool_masked_pos)
+ >>> loss, reconstructed_pixel_values = outputs.loss, outputs.reconstruction
+ >>> list(reconstructed_pixel_values.shape)
+ [1, 3, 224, 224]
+ ```"""
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+ if bool_masked_pos is not None and (self.config.patch_size != self.config.encoder_stride):
+ raise ValueError(
+ "When `bool_masked_pos` is provided, `patch_size` must be equal to `encoder_stride` to ensure that "
+ "the reconstructed image has the same dimensions as the input. "
+ f"Got `patch_size` = {self.config.patch_size} and `encoder_stride` = {self.config.encoder_stride}."
+ )
+
+ outputs = self.vit(
+ pixel_values,
+ bool_masked_pos=bool_masked_pos,
+ head_mask=head_mask,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ interpolate_pos_encoding=interpolate_pos_encoding,
+ return_dict=return_dict,
+ )
+
+ sequence_output = outputs[0]
+
+ # Reshape to (batch_size, num_channels, height, width)
+ sequence_output = sequence_output[:, 1:]
+ batch_size, sequence_length, num_channels = sequence_output.shape
+ height = width = math.floor(sequence_length**0.5)
+ sequence_output = sequence_output.permute(0, 2, 1).reshape(batch_size, num_channels, height, width)
+
+ # Reconstruct pixel values
+ reconstructed_pixel_values = self.decoder(sequence_output)
+
+ masked_im_loss = None
+ if bool_masked_pos is not None:
+ size = self.config.image_size // self.config.patch_size
+ bool_masked_pos = bool_masked_pos.reshape(-1, size, size)
+ mask = (
+ bool_masked_pos.repeat_interleave(self.config.patch_size, 1)
+ .repeat_interleave(self.config.patch_size, 2)
+ .unsqueeze(1)
+ .contiguous()
+ )
+ reconstruction_loss = nn.functional.l1_loss(pixel_values, reconstructed_pixel_values, reduction="none")
+ masked_im_loss = (reconstruction_loss * mask).sum() / (mask.sum() + 1e-5) / self.config.num_channels
+
+ if not return_dict:
+ output = (reconstructed_pixel_values,) + outputs[1:]
+ return ((masked_im_loss,) + output) if masked_im_loss is not None else output
+
+ return MaskedImageModelingOutput(
+ loss=masked_im_loss,
+ reconstruction=reconstructed_pixel_values,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
+
+
+@add_start_docstrings(
+ """
+ ViT Model transformer with an image classification head on top (a linear layer on top of the final hidden state of
+ the [CLS] token) e.g. for ImageNet.
+
+
+
+ Note that it's possible to fine-tune ViT on higher resolution images than the ones it has been trained on, by
+ setting `interpolate_pos_encoding` to `True` in the forward of the model. This will interpolate the pre-trained
+ position embeddings to the higher resolution.
+
+
+ """,
+ VIT_START_DOCSTRING,
+)
+class ViTForImageClassification(ViTPreTrainedModel):
+ def __init__(self, config: ViTConfig) -> None:
+ super().__init__(config)
+
+ self.num_labels = config.num_labels
+ self.vit = ViTModel(config, add_pooling_layer=False)
+
+ # Classifier head
+ self.classifier = nn.Linear(config.hidden_size, config.num_labels) if config.num_labels > 0 else nn.Identity()
+
+ # Initialize weights and apply final processing
+ self.post_init()
+
+ @add_start_docstrings_to_model_forward(VIT_INPUTS_DOCSTRING)
+ @add_code_sample_docstrings(
+ checkpoint=_IMAGE_CLASS_CHECKPOINT,
+ output_type=ImageClassifierOutput,
+ config_class=_CONFIG_FOR_DOC,
+ expected_output=_IMAGE_CLASS_EXPECTED_OUTPUT,
+ )
+ def forward(
+ self,
+ pixel_values: Optional[torch.Tensor] = None,
+ head_mask: Optional[torch.Tensor] = None,
+ labels: Optional[torch.Tensor] = None,
+ output_attentions: Optional[bool] = None,
+ output_hidden_states: Optional[bool] = None,
+ interpolate_pos_encoding: Optional[bool] = None,
+ return_dict: Optional[bool] = None,
+ ) -> Union[tuple, ImageClassifierOutput]:
+ r"""
+ labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+ Labels for computing the image classification/regression loss. Indices should be in `[0, ...,
+ config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+ `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+ """
+ return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+ outputs = self.vit(
+ pixel_values,
+ head_mask=head_mask,
+ output_attentions=output_attentions,
+ output_hidden_states=output_hidden_states,
+ interpolate_pos_encoding=interpolate_pos_encoding,
+ return_dict=return_dict,
+ )
+
+ sequence_output = outputs[0]
+
+ logits = self.classifier(sequence_output[:, 0, :])
+
+ loss = None
+ if labels is not None:
+ # move labels to correct device to enable model parallelism
+ labels = labels.to(logits.device)
+ if self.config.problem_type is None:
+ if self.num_labels == 1:
+ self.config.problem_type = "regression"
+ elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+ self.config.problem_type = "single_label_classification"
+ else:
+ self.config.problem_type = "multi_label_classification"
+
+ if self.config.problem_type == "regression":
+ loss_fct = MSELoss()
+ if self.num_labels == 1:
+ loss = loss_fct(logits.squeeze(), labels.squeeze())
+ else:
+ loss = loss_fct(logits, labels)
+ elif self.config.problem_type == "single_label_classification":
+ loss_fct = CrossEntropyLoss()
+ loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
+ elif self.config.problem_type == "multi_label_classification":
+ loss_fct = BCEWithLogitsLoss()
+ loss = loss_fct(logits, labels)
+
+ if not return_dict:
+ output = (logits,) + outputs[1:]
+ return ((loss,) + output) if loss is not None else output
+
+ return ImageClassifierOutput(
+ loss=loss,
+ logits=logits,
+ hidden_states=outputs.hidden_states,
+ attentions=outputs.attentions,
+ )
diff --git a/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/INSTALLER b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/INSTALLER
new file mode 100644
index 0000000000000000000000000000000000000000..a1b589e38a32041e49332e5e81c2d363dc418d68
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/INSTALLER
@@ -0,0 +1 @@
+pip
diff --git a/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/RECORD b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/RECORD
new file mode 100644
index 0000000000000000000000000000000000000000..a516050a5392527c9a4b5e413bd7a6183a3541b8
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/RECORD
@@ -0,0 +1,32 @@
+colorama-0.4.6.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4
+colorama-0.4.6.dist-info/METADATA,sha256=e67SnrUMOym9sz_4TjF3vxvAV4T3aF7NyqRHHH3YEMw,17158
+colorama-0.4.6.dist-info/RECORD,,
+colorama-0.4.6.dist-info/REQUESTED,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+colorama-0.4.6.dist-info/WHEEL,sha256=cdcF4Fbd0FPtw2EMIOwH-3rSOTUdTCeOSXRMD1iLUb8,105
+colorama-0.4.6.dist-info/licenses/LICENSE.txt,sha256=ysNcAmhuXQSlpxQL-zs25zrtSWZW6JEQLkKIhteTAxg,1491
+colorama/__init__.py,sha256=wePQA4U20tKgYARySLEC047ucNX-g8pRLpYBuiHlLb8,266
+colorama/__pycache__/__init__.cpython-310.pyc,,
+colorama/__pycache__/ansi.cpython-310.pyc,,
+colorama/__pycache__/ansitowin32.cpython-310.pyc,,
+colorama/__pycache__/initialise.cpython-310.pyc,,
+colorama/__pycache__/win32.cpython-310.pyc,,
+colorama/__pycache__/winterm.cpython-310.pyc,,
+colorama/ansi.py,sha256=Top4EeEuaQdBWdteKMEcGOTeKeF19Q-Wo_6_Cj5kOzQ,2522
+colorama/ansitowin32.py,sha256=vPNYa3OZbxjbuFyaVo0Tmhmy1FZ1lKMWCnT7odXpItk,11128
+colorama/initialise.py,sha256=-hIny86ClXo39ixh5iSCfUIa2f_h_bgKRDW7gqs-KLU,3325
+colorama/tests/__init__.py,sha256=MkgPAEzGQd-Rq0w0PZXSX2LadRWhUECcisJY8lSrm4Q,75
+colorama/tests/__pycache__/__init__.cpython-310.pyc,,
+colorama/tests/__pycache__/ansi_test.cpython-310.pyc,,
+colorama/tests/__pycache__/ansitowin32_test.cpython-310.pyc,,
+colorama/tests/__pycache__/initialise_test.cpython-310.pyc,,
+colorama/tests/__pycache__/isatty_test.cpython-310.pyc,,
+colorama/tests/__pycache__/utils.cpython-310.pyc,,
+colorama/tests/__pycache__/winterm_test.cpython-310.pyc,,
+colorama/tests/ansi_test.py,sha256=FeViDrUINIZcr505PAxvU4AjXz1asEiALs9GXMhwRaE,2839
+colorama/tests/ansitowin32_test.py,sha256=RN7AIhMJ5EqDsYaCjVo-o4u8JzDD4ukJbmevWKS70rY,10678
+colorama/tests/initialise_test.py,sha256=BbPy-XfyHwJ6zKozuQOvNvQZzsx9vdb_0bYXn7hsBTc,6741
+colorama/tests/isatty_test.py,sha256=Pg26LRpv0yQDB5Ac-sxgVXG7hsA1NYvapFgApZfYzZg,1866
+colorama/tests/utils.py,sha256=1IIRylG39z5-dzq09R_ngufxyPZxgldNbrxKxUGwGKE,1079
+colorama/tests/winterm_test.py,sha256=qoWFPEjym5gm2RuMwpf3pOis3a5r_PJZFCzK254JL8A,3709
+colorama/win32.py,sha256=YQOKwMTwtGBbsY4dL5HYTvwTeP9wIQra5MvPNddpxZs,6181
+colorama/winterm.py,sha256=XCQFDHjPi6AHYNdZwy0tA02H-Jh48Jp-HvCjeLeLp3U,7134
diff --git a/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/REQUESTED b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/REQUESTED
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/WHEEL b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/WHEEL
new file mode 100644
index 0000000000000000000000000000000000000000..d79189fda3251187de18c3998f23ae6fec11b20f
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/WHEEL
@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.11.1
+Root-Is-Purelib: true
+Tag: py2-none-any
+Tag: py3-none-any
diff --git a/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/licenses/LICENSE.txt b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/licenses/LICENSE.txt
new file mode 100644
index 0000000000000000000000000000000000000000..3105888ec149d10cad51c11d332779e94b548661
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/colorama-0.4.6.dist-info/licenses/LICENSE.txt
@@ -0,0 +1,27 @@
+Copyright (c) 2010 Jonathan Hartley
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+ list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+ this list of conditions and the following disclaimer in the documentation
+ and/or other materials provided with the distribution.
+
+* Neither the name of the copyright holders, nor those of its contributors
+ may be used to endorse or promote products derived from this software without
+ specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/falcon/lib/python3.10/site-packages/dateutil/__init__.py b/falcon/lib/python3.10/site-packages/dateutil/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..a2c19c06fe14476a9bfa4f1f60de7a997a41191c
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/__init__.py
@@ -0,0 +1,24 @@
+# -*- coding: utf-8 -*-
+import sys
+
+try:
+ from ._version import version as __version__
+except ImportError:
+ __version__ = 'unknown'
+
+__all__ = ['easter', 'parser', 'relativedelta', 'rrule', 'tz',
+ 'utils', 'zoneinfo']
+
+def __getattr__(name):
+ import importlib
+
+ if name in __all__:
+ return importlib.import_module("." + name, __name__)
+ raise AttributeError(
+ "module {!r} has not attribute {!r}".format(__name__, name)
+ )
+
+
+def __dir__():
+ # __dir__ should include all the lazy-importable modules as well.
+ return [x for x in globals() if x not in sys.modules] + __all__
diff --git a/falcon/lib/python3.10/site-packages/dateutil/__pycache__/relativedelta.cpython-310.pyc b/falcon/lib/python3.10/site-packages/dateutil/__pycache__/relativedelta.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..a8d3d87fff7ad6d2ba925654fe3c14c020d371fe
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/dateutil/__pycache__/relativedelta.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/dateutil/__pycache__/tzwin.cpython-310.pyc b/falcon/lib/python3.10/site-packages/dateutil/__pycache__/tzwin.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..5ac40b4e98b7faa40cae56588c74091070231fe6
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/dateutil/__pycache__/tzwin.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/dateutil/__pycache__/utils.cpython-310.pyc b/falcon/lib/python3.10/site-packages/dateutil/__pycache__/utils.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..0c2ecc4c337ed26559e8394a185111586b0342db
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/dateutil/__pycache__/utils.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/dateutil/_common.py b/falcon/lib/python3.10/site-packages/dateutil/_common.py
new file mode 100644
index 0000000000000000000000000000000000000000..4eb2659bd2986125fcfb4afea5bae9efc2dcd1a0
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/_common.py
@@ -0,0 +1,43 @@
+"""
+Common code used in multiple modules.
+"""
+
+
+class weekday(object):
+ __slots__ = ["weekday", "n"]
+
+ def __init__(self, weekday, n=None):
+ self.weekday = weekday
+ self.n = n
+
+ def __call__(self, n):
+ if n == self.n:
+ return self
+ else:
+ return self.__class__(self.weekday, n)
+
+ def __eq__(self, other):
+ try:
+ if self.weekday != other.weekday or self.n != other.n:
+ return False
+ except AttributeError:
+ return False
+ return True
+
+ def __hash__(self):
+ return hash((
+ self.weekday,
+ self.n,
+ ))
+
+ def __ne__(self, other):
+ return not (self == other)
+
+ def __repr__(self):
+ s = ("MO", "TU", "WE", "TH", "FR", "SA", "SU")[self.weekday]
+ if not self.n:
+ return s
+ else:
+ return "%s(%+d)" % (s, self.n)
+
+# vim:ts=4:sw=4:et
diff --git a/falcon/lib/python3.10/site-packages/dateutil/_version.py b/falcon/lib/python3.10/site-packages/dateutil/_version.py
new file mode 100644
index 0000000000000000000000000000000000000000..ddda98098527a73348e694c2edb691fd625475fc
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/_version.py
@@ -0,0 +1,4 @@
+# file generated by setuptools_scm
+# don't change, don't track in version control
+__version__ = version = '2.9.0.post0'
+__version_tuple__ = version_tuple = (2, 9, 0)
diff --git a/falcon/lib/python3.10/site-packages/dateutil/easter.py b/falcon/lib/python3.10/site-packages/dateutil/easter.py
new file mode 100644
index 0000000000000000000000000000000000000000..f74d1f7442473997245ac683b8a269a3574d1ba4
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/easter.py
@@ -0,0 +1,89 @@
+# -*- coding: utf-8 -*-
+"""
+This module offers a generic Easter computing method for any given year, using
+Western, Orthodox or Julian algorithms.
+"""
+
+import datetime
+
+__all__ = ["easter", "EASTER_JULIAN", "EASTER_ORTHODOX", "EASTER_WESTERN"]
+
+EASTER_JULIAN = 1
+EASTER_ORTHODOX = 2
+EASTER_WESTERN = 3
+
+
+def easter(year, method=EASTER_WESTERN):
+ """
+ This method was ported from the work done by GM Arts,
+ on top of the algorithm by Claus Tondering, which was
+ based in part on the algorithm of Ouding (1940), as
+ quoted in "Explanatory Supplement to the Astronomical
+ Almanac", P. Kenneth Seidelmann, editor.
+
+ This algorithm implements three different Easter
+ calculation methods:
+
+ 1. Original calculation in Julian calendar, valid in
+ dates after 326 AD
+ 2. Original method, with date converted to Gregorian
+ calendar, valid in years 1583 to 4099
+ 3. Revised method, in Gregorian calendar, valid in
+ years 1583 to 4099 as well
+
+ These methods are represented by the constants:
+
+ * ``EASTER_JULIAN = 1``
+ * ``EASTER_ORTHODOX = 2``
+ * ``EASTER_WESTERN = 3``
+
+ The default method is method 3.
+
+ More about the algorithm may be found at:
+
+ `GM Arts: Easter Algorithms `_
+
+ and
+
+ `The Calendar FAQ: Easter `_
+
+ """
+
+ if not (1 <= method <= 3):
+ raise ValueError("invalid method")
+
+ # g - Golden year - 1
+ # c - Century
+ # h - (23 - Epact) mod 30
+ # i - Number of days from March 21 to Paschal Full Moon
+ # j - Weekday for PFM (0=Sunday, etc)
+ # p - Number of days from March 21 to Sunday on or before PFM
+ # (-6 to 28 methods 1 & 3, to 56 for method 2)
+ # e - Extra days to add for method 2 (converting Julian
+ # date to Gregorian date)
+
+ y = year
+ g = y % 19
+ e = 0
+ if method < 3:
+ # Old method
+ i = (19*g + 15) % 30
+ j = (y + y//4 + i) % 7
+ if method == 2:
+ # Extra dates to convert Julian to Gregorian date
+ e = 10
+ if y > 1600:
+ e = e + y//100 - 16 - (y//100 - 16)//4
+ else:
+ # New method
+ c = y//100
+ h = (c - c//4 - (8*c + 13)//25 + 19*g + 15) % 30
+ i = h - (h//28)*(1 - (h//28)*(29//(h + 1))*((21 - g)//11))
+ j = (y + y//4 + i + 2 - c + c//4) % 7
+
+ # p can be from -6 to 56 corresponding to dates 22 March to 23 May
+ # (later dates apply to method 2, although 23 May never actually occurs)
+ p = i - j + e
+ d = 1 + (p + 27 + (p + 6)//40) % 31
+ m = 3 + (p + 26)//30
+ return datetime.date(int(y), int(m), int(d))
diff --git a/falcon/lib/python3.10/site-packages/dateutil/parser/__init__.py b/falcon/lib/python3.10/site-packages/dateutil/parser/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..d174b0e4dcc472999b75e55ebb88af320ae38081
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/parser/__init__.py
@@ -0,0 +1,61 @@
+# -*- coding: utf-8 -*-
+from ._parser import parse, parser, parserinfo, ParserError
+from ._parser import DEFAULTPARSER, DEFAULTTZPARSER
+from ._parser import UnknownTimezoneWarning
+
+from ._parser import __doc__
+
+from .isoparser import isoparser, isoparse
+
+__all__ = ['parse', 'parser', 'parserinfo',
+ 'isoparse', 'isoparser',
+ 'ParserError',
+ 'UnknownTimezoneWarning']
+
+
+###
+# Deprecate portions of the private interface so that downstream code that
+# is improperly relying on it is given *some* notice.
+
+
+def __deprecated_private_func(f):
+ from functools import wraps
+ import warnings
+
+ msg = ('{name} is a private function and may break without warning, '
+ 'it will be moved and or renamed in future versions.')
+ msg = msg.format(name=f.__name__)
+
+ @wraps(f)
+ def deprecated_func(*args, **kwargs):
+ warnings.warn(msg, DeprecationWarning)
+ return f(*args, **kwargs)
+
+ return deprecated_func
+
+def __deprecate_private_class(c):
+ import warnings
+
+ msg = ('{name} is a private class and may break without warning, '
+ 'it will be moved and or renamed in future versions.')
+ msg = msg.format(name=c.__name__)
+
+ class private_class(c):
+ __doc__ = c.__doc__
+
+ def __init__(self, *args, **kwargs):
+ warnings.warn(msg, DeprecationWarning)
+ super(private_class, self).__init__(*args, **kwargs)
+
+ private_class.__name__ = c.__name__
+
+ return private_class
+
+
+from ._parser import _timelex, _resultbase
+from ._parser import _tzparser, _parsetz
+
+_timelex = __deprecate_private_class(_timelex)
+_tzparser = __deprecate_private_class(_tzparser)
+_resultbase = __deprecate_private_class(_resultbase)
+_parsetz = __deprecated_private_func(_parsetz)
diff --git a/falcon/lib/python3.10/site-packages/dateutil/parser/__pycache__/_parser.cpython-310.pyc b/falcon/lib/python3.10/site-packages/dateutil/parser/__pycache__/_parser.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..41f656bce72f43efe5ef15910dba9e756186513c
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/dateutil/parser/__pycache__/_parser.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/dateutil/parser/__pycache__/isoparser.cpython-310.pyc b/falcon/lib/python3.10/site-packages/dateutil/parser/__pycache__/isoparser.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..27a9c78859e84d8b388c841c0f7eb11d91936888
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/dateutil/parser/__pycache__/isoparser.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/dateutil/parser/_parser.py b/falcon/lib/python3.10/site-packages/dateutil/parser/_parser.py
new file mode 100644
index 0000000000000000000000000000000000000000..37d1663b2f72447800d9a553929e3de932244289
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/parser/_parser.py
@@ -0,0 +1,1613 @@
+# -*- coding: utf-8 -*-
+"""
+This module offers a generic date/time string parser which is able to parse
+most known formats to represent a date and/or time.
+
+This module attempts to be forgiving with regards to unlikely input formats,
+returning a datetime object even for dates which are ambiguous. If an element
+of a date/time stamp is omitted, the following rules are applied:
+
+- If AM or PM is left unspecified, a 24-hour clock is assumed, however, an hour
+ on a 12-hour clock (``0 <= hour <= 12``) *must* be specified if AM or PM is
+ specified.
+- If a time zone is omitted, a timezone-naive datetime is returned.
+
+If any other elements are missing, they are taken from the
+:class:`datetime.datetime` object passed to the parameter ``default``. If this
+results in a day number exceeding the valid number of days per month, the
+value falls back to the end of the month.
+
+Additional resources about date/time string formats can be found below:
+
+- `A summary of the international standard date and time notation
+ `_
+- `W3C Date and Time Formats `_
+- `Time Formats (Planetary Rings Node) `_
+- `CPAN ParseDate module
+ `_
+- `Java SimpleDateFormat Class
+ `_
+"""
+from __future__ import unicode_literals
+
+import datetime
+import re
+import string
+import time
+import warnings
+
+from calendar import monthrange
+from io import StringIO
+
+import six
+from six import integer_types, text_type
+
+from decimal import Decimal
+
+from warnings import warn
+
+from .. import relativedelta
+from .. import tz
+
+__all__ = ["parse", "parserinfo", "ParserError"]
+
+
+# TODO: pandas.core.tools.datetimes imports this explicitly. Might be worth
+# making public and/or figuring out if there is something we can
+# take off their plate.
+class _timelex(object):
+ # Fractional seconds are sometimes split by a comma
+ _split_decimal = re.compile("([.,])")
+
+ def __init__(self, instream):
+ if isinstance(instream, (bytes, bytearray)):
+ instream = instream.decode()
+
+ if isinstance(instream, text_type):
+ instream = StringIO(instream)
+ elif getattr(instream, 'read', None) is None:
+ raise TypeError('Parser must be a string or character stream, not '
+ '{itype}'.format(itype=instream.__class__.__name__))
+
+ self.instream = instream
+ self.charstack = []
+ self.tokenstack = []
+ self.eof = False
+
+ def get_token(self):
+ """
+ This function breaks the time string into lexical units (tokens), which
+ can be parsed by the parser. Lexical units are demarcated by changes in
+ the character set, so any continuous string of letters is considered
+ one unit, any continuous string of numbers is considered one unit.
+
+ The main complication arises from the fact that dots ('.') can be used
+ both as separators (e.g. "Sep.20.2009") or decimal points (e.g.
+ "4:30:21.447"). As such, it is necessary to read the full context of
+ any dot-separated strings before breaking it into tokens; as such, this
+ function maintains a "token stack", for when the ambiguous context
+ demands that multiple tokens be parsed at once.
+ """
+ if self.tokenstack:
+ return self.tokenstack.pop(0)
+
+ seenletters = False
+ token = None
+ state = None
+
+ while not self.eof:
+ # We only realize that we've reached the end of a token when we
+ # find a character that's not part of the current token - since
+ # that character may be part of the next token, it's stored in the
+ # charstack.
+ if self.charstack:
+ nextchar = self.charstack.pop(0)
+ else:
+ nextchar = self.instream.read(1)
+ while nextchar == '\x00':
+ nextchar = self.instream.read(1)
+
+ if not nextchar:
+ self.eof = True
+ break
+ elif not state:
+ # First character of the token - determines if we're starting
+ # to parse a word, a number or something else.
+ token = nextchar
+ if self.isword(nextchar):
+ state = 'a'
+ elif self.isnum(nextchar):
+ state = '0'
+ elif self.isspace(nextchar):
+ token = ' '
+ break # emit token
+ else:
+ break # emit token
+ elif state == 'a':
+ # If we've already started reading a word, we keep reading
+ # letters until we find something that's not part of a word.
+ seenletters = True
+ if self.isword(nextchar):
+ token += nextchar
+ elif nextchar == '.':
+ token += nextchar
+ state = 'a.'
+ else:
+ self.charstack.append(nextchar)
+ break # emit token
+ elif state == '0':
+ # If we've already started reading a number, we keep reading
+ # numbers until we find something that doesn't fit.
+ if self.isnum(nextchar):
+ token += nextchar
+ elif nextchar == '.' or (nextchar == ',' and len(token) >= 2):
+ token += nextchar
+ state = '0.'
+ else:
+ self.charstack.append(nextchar)
+ break # emit token
+ elif state == 'a.':
+ # If we've seen some letters and a dot separator, continue
+ # parsing, and the tokens will be broken up later.
+ seenletters = True
+ if nextchar == '.' or self.isword(nextchar):
+ token += nextchar
+ elif self.isnum(nextchar) and token[-1] == '.':
+ token += nextchar
+ state = '0.'
+ else:
+ self.charstack.append(nextchar)
+ break # emit token
+ elif state == '0.':
+ # If we've seen at least one dot separator, keep going, we'll
+ # break up the tokens later.
+ if nextchar == '.' or self.isnum(nextchar):
+ token += nextchar
+ elif self.isword(nextchar) and token[-1] == '.':
+ token += nextchar
+ state = 'a.'
+ else:
+ self.charstack.append(nextchar)
+ break # emit token
+
+ if (state in ('a.', '0.') and (seenletters or token.count('.') > 1 or
+ token[-1] in '.,')):
+ l = self._split_decimal.split(token)
+ token = l[0]
+ for tok in l[1:]:
+ if tok:
+ self.tokenstack.append(tok)
+
+ if state == '0.' and token.count('.') == 0:
+ token = token.replace(',', '.')
+
+ return token
+
+ def __iter__(self):
+ return self
+
+ def __next__(self):
+ token = self.get_token()
+ if token is None:
+ raise StopIteration
+
+ return token
+
+ def next(self):
+ return self.__next__() # Python 2.x support
+
+ @classmethod
+ def split(cls, s):
+ return list(cls(s))
+
+ @classmethod
+ def isword(cls, nextchar):
+ """ Whether or not the next character is part of a word """
+ return nextchar.isalpha()
+
+ @classmethod
+ def isnum(cls, nextchar):
+ """ Whether the next character is part of a number """
+ return nextchar.isdigit()
+
+ @classmethod
+ def isspace(cls, nextchar):
+ """ Whether the next character is whitespace """
+ return nextchar.isspace()
+
+
+class _resultbase(object):
+
+ def __init__(self):
+ for attr in self.__slots__:
+ setattr(self, attr, None)
+
+ def _repr(self, classname):
+ l = []
+ for attr in self.__slots__:
+ value = getattr(self, attr)
+ if value is not None:
+ l.append("%s=%s" % (attr, repr(value)))
+ return "%s(%s)" % (classname, ", ".join(l))
+
+ def __len__(self):
+ return (sum(getattr(self, attr) is not None
+ for attr in self.__slots__))
+
+ def __repr__(self):
+ return self._repr(self.__class__.__name__)
+
+
+class parserinfo(object):
+ """
+ Class which handles what inputs are accepted. Subclass this to customize
+ the language and acceptable values for each parameter.
+
+ :param dayfirst:
+ Whether to interpret the first value in an ambiguous 3-integer date
+ (e.g. 01/05/09) as the day (``True``) or month (``False``). If
+ ``yearfirst`` is set to ``True``, this distinguishes between YDM
+ and YMD. Default is ``False``.
+
+ :param yearfirst:
+ Whether to interpret the first value in an ambiguous 3-integer date
+ (e.g. 01/05/09) as the year. If ``True``, the first number is taken
+ to be the year, otherwise the last number is taken to be the year.
+ Default is ``False``.
+ """
+
+ # m from a.m/p.m, t from ISO T separator
+ JUMP = [" ", ".", ",", ";", "-", "/", "'",
+ "at", "on", "and", "ad", "m", "t", "of",
+ "st", "nd", "rd", "th"]
+
+ WEEKDAYS = [("Mon", "Monday"),
+ ("Tue", "Tuesday"), # TODO: "Tues"
+ ("Wed", "Wednesday"),
+ ("Thu", "Thursday"), # TODO: "Thurs"
+ ("Fri", "Friday"),
+ ("Sat", "Saturday"),
+ ("Sun", "Sunday")]
+ MONTHS = [("Jan", "January"),
+ ("Feb", "February"), # TODO: "Febr"
+ ("Mar", "March"),
+ ("Apr", "April"),
+ ("May", "May"),
+ ("Jun", "June"),
+ ("Jul", "July"),
+ ("Aug", "August"),
+ ("Sep", "Sept", "September"),
+ ("Oct", "October"),
+ ("Nov", "November"),
+ ("Dec", "December")]
+ HMS = [("h", "hour", "hours"),
+ ("m", "minute", "minutes"),
+ ("s", "second", "seconds")]
+ AMPM = [("am", "a"),
+ ("pm", "p")]
+ UTCZONE = ["UTC", "GMT", "Z", "z"]
+ PERTAIN = ["of"]
+ TZOFFSET = {}
+ # TODO: ERA = ["AD", "BC", "CE", "BCE", "Stardate",
+ # "Anno Domini", "Year of Our Lord"]
+
+ def __init__(self, dayfirst=False, yearfirst=False):
+ self._jump = self._convert(self.JUMP)
+ self._weekdays = self._convert(self.WEEKDAYS)
+ self._months = self._convert(self.MONTHS)
+ self._hms = self._convert(self.HMS)
+ self._ampm = self._convert(self.AMPM)
+ self._utczone = self._convert(self.UTCZONE)
+ self._pertain = self._convert(self.PERTAIN)
+
+ self.dayfirst = dayfirst
+ self.yearfirst = yearfirst
+
+ self._year = time.localtime().tm_year
+ self._century = self._year // 100 * 100
+
+ def _convert(self, lst):
+ dct = {}
+ for i, v in enumerate(lst):
+ if isinstance(v, tuple):
+ for v in v:
+ dct[v.lower()] = i
+ else:
+ dct[v.lower()] = i
+ return dct
+
+ def jump(self, name):
+ return name.lower() in self._jump
+
+ def weekday(self, name):
+ try:
+ return self._weekdays[name.lower()]
+ except KeyError:
+ pass
+ return None
+
+ def month(self, name):
+ try:
+ return self._months[name.lower()] + 1
+ except KeyError:
+ pass
+ return None
+
+ def hms(self, name):
+ try:
+ return self._hms[name.lower()]
+ except KeyError:
+ return None
+
+ def ampm(self, name):
+ try:
+ return self._ampm[name.lower()]
+ except KeyError:
+ return None
+
+ def pertain(self, name):
+ return name.lower() in self._pertain
+
+ def utczone(self, name):
+ return name.lower() in self._utczone
+
+ def tzoffset(self, name):
+ if name in self._utczone:
+ return 0
+
+ return self.TZOFFSET.get(name)
+
+ def convertyear(self, year, century_specified=False):
+ """
+ Converts two-digit years to year within [-50, 49]
+ range of self._year (current local time)
+ """
+
+ # Function contract is that the year is always positive
+ assert year >= 0
+
+ if year < 100 and not century_specified:
+ # assume current century to start
+ year += self._century
+
+ if year >= self._year + 50: # if too far in future
+ year -= 100
+ elif year < self._year - 50: # if too far in past
+ year += 100
+
+ return year
+
+ def validate(self, res):
+ # move to info
+ if res.year is not None:
+ res.year = self.convertyear(res.year, res.century_specified)
+
+ if ((res.tzoffset == 0 and not res.tzname) or
+ (res.tzname == 'Z' or res.tzname == 'z')):
+ res.tzname = "UTC"
+ res.tzoffset = 0
+ elif res.tzoffset != 0 and res.tzname and self.utczone(res.tzname):
+ res.tzoffset = 0
+ return True
+
+
+class _ymd(list):
+ def __init__(self, *args, **kwargs):
+ super(self.__class__, self).__init__(*args, **kwargs)
+ self.century_specified = False
+ self.dstridx = None
+ self.mstridx = None
+ self.ystridx = None
+
+ @property
+ def has_year(self):
+ return self.ystridx is not None
+
+ @property
+ def has_month(self):
+ return self.mstridx is not None
+
+ @property
+ def has_day(self):
+ return self.dstridx is not None
+
+ def could_be_day(self, value):
+ if self.has_day:
+ return False
+ elif not self.has_month:
+ return 1 <= value <= 31
+ elif not self.has_year:
+ # Be permissive, assume leap year
+ month = self[self.mstridx]
+ return 1 <= value <= monthrange(2000, month)[1]
+ else:
+ month = self[self.mstridx]
+ year = self[self.ystridx]
+ return 1 <= value <= monthrange(year, month)[1]
+
+ def append(self, val, label=None):
+ if hasattr(val, '__len__'):
+ if val.isdigit() and len(val) > 2:
+ self.century_specified = True
+ if label not in [None, 'Y']: # pragma: no cover
+ raise ValueError(label)
+ label = 'Y'
+ elif val > 100:
+ self.century_specified = True
+ if label not in [None, 'Y']: # pragma: no cover
+ raise ValueError(label)
+ label = 'Y'
+
+ super(self.__class__, self).append(int(val))
+
+ if label == 'M':
+ if self.has_month:
+ raise ValueError('Month is already set')
+ self.mstridx = len(self) - 1
+ elif label == 'D':
+ if self.has_day:
+ raise ValueError('Day is already set')
+ self.dstridx = len(self) - 1
+ elif label == 'Y':
+ if self.has_year:
+ raise ValueError('Year is already set')
+ self.ystridx = len(self) - 1
+
+ def _resolve_from_stridxs(self, strids):
+ """
+ Try to resolve the identities of year/month/day elements using
+ ystridx, mstridx, and dstridx, if enough of these are specified.
+ """
+ if len(self) == 3 and len(strids) == 2:
+ # we can back out the remaining stridx value
+ missing = [x for x in range(3) if x not in strids.values()]
+ key = [x for x in ['y', 'm', 'd'] if x not in strids]
+ assert len(missing) == len(key) == 1
+ key = key[0]
+ val = missing[0]
+ strids[key] = val
+
+ assert len(self) == len(strids) # otherwise this should not be called
+ out = {key: self[strids[key]] for key in strids}
+ return (out.get('y'), out.get('m'), out.get('d'))
+
+ def resolve_ymd(self, yearfirst, dayfirst):
+ len_ymd = len(self)
+ year, month, day = (None, None, None)
+
+ strids = (('y', self.ystridx),
+ ('m', self.mstridx),
+ ('d', self.dstridx))
+
+ strids = {key: val for key, val in strids if val is not None}
+ if (len(self) == len(strids) > 0 or
+ (len(self) == 3 and len(strids) == 2)):
+ return self._resolve_from_stridxs(strids)
+
+ mstridx = self.mstridx
+
+ if len_ymd > 3:
+ raise ValueError("More than three YMD values")
+ elif len_ymd == 1 or (mstridx is not None and len_ymd == 2):
+ # One member, or two members with a month string
+ if mstridx is not None:
+ month = self[mstridx]
+ # since mstridx is 0 or 1, self[mstridx-1] always
+ # looks up the other element
+ other = self[mstridx - 1]
+ else:
+ other = self[0]
+
+ if len_ymd > 1 or mstridx is None:
+ if other > 31:
+ year = other
+ else:
+ day = other
+
+ elif len_ymd == 2:
+ # Two members with numbers
+ if self[0] > 31:
+ # 99-01
+ year, month = self
+ elif self[1] > 31:
+ # 01-99
+ month, year = self
+ elif dayfirst and self[1] <= 12:
+ # 13-01
+ day, month = self
+ else:
+ # 01-13
+ month, day = self
+
+ elif len_ymd == 3:
+ # Three members
+ if mstridx == 0:
+ if self[1] > 31:
+ # Apr-2003-25
+ month, year, day = self
+ else:
+ month, day, year = self
+ elif mstridx == 1:
+ if self[0] > 31 or (yearfirst and self[2] <= 31):
+ # 99-Jan-01
+ year, month, day = self
+ else:
+ # 01-Jan-01
+ # Give precedence to day-first, since
+ # two-digit years is usually hand-written.
+ day, month, year = self
+
+ elif mstridx == 2:
+ # WTF!?
+ if self[1] > 31:
+ # 01-99-Jan
+ day, year, month = self
+ else:
+ # 99-01-Jan
+ year, day, month = self
+
+ else:
+ if (self[0] > 31 or
+ self.ystridx == 0 or
+ (yearfirst and self[1] <= 12 and self[2] <= 31)):
+ # 99-01-01
+ if dayfirst and self[2] <= 12:
+ year, day, month = self
+ else:
+ year, month, day = self
+ elif self[0] > 12 or (dayfirst and self[1] <= 12):
+ # 13-01-01
+ day, month, year = self
+ else:
+ # 01-13-01
+ month, day, year = self
+
+ return year, month, day
+
+
+class parser(object):
+ def __init__(self, info=None):
+ self.info = info or parserinfo()
+
+ def parse(self, timestr, default=None,
+ ignoretz=False, tzinfos=None, **kwargs):
+ """
+ Parse the date/time string into a :class:`datetime.datetime` object.
+
+ :param timestr:
+ Any date/time string using the supported formats.
+
+ :param default:
+ The default datetime object, if this is a datetime object and not
+ ``None``, elements specified in ``timestr`` replace elements in the
+ default object.
+
+ :param ignoretz:
+ If set ``True``, time zones in parsed strings are ignored and a
+ naive :class:`datetime.datetime` object is returned.
+
+ :param tzinfos:
+ Additional time zone names / aliases which may be present in the
+ string. This argument maps time zone names (and optionally offsets
+ from those time zones) to time zones. This parameter can be a
+ dictionary with timezone aliases mapping time zone names to time
+ zones or a function taking two parameters (``tzname`` and
+ ``tzoffset``) and returning a time zone.
+
+ The timezones to which the names are mapped can be an integer
+ offset from UTC in seconds or a :class:`tzinfo` object.
+
+ .. doctest::
+ :options: +NORMALIZE_WHITESPACE
+
+ >>> from dateutil.parser import parse
+ >>> from dateutil.tz import gettz
+ >>> tzinfos = {"BRST": -7200, "CST": gettz("America/Chicago")}
+ >>> parse("2012-01-19 17:21:00 BRST", tzinfos=tzinfos)
+ datetime.datetime(2012, 1, 19, 17, 21, tzinfo=tzoffset(u'BRST', -7200))
+ >>> parse("2012-01-19 17:21:00 CST", tzinfos=tzinfos)
+ datetime.datetime(2012, 1, 19, 17, 21,
+ tzinfo=tzfile('/usr/share/zoneinfo/America/Chicago'))
+
+ This parameter is ignored if ``ignoretz`` is set.
+
+ :param \\*\\*kwargs:
+ Keyword arguments as passed to ``_parse()``.
+
+ :return:
+ Returns a :class:`datetime.datetime` object or, if the
+ ``fuzzy_with_tokens`` option is ``True``, returns a tuple, the
+ first element being a :class:`datetime.datetime` object, the second
+ a tuple containing the fuzzy tokens.
+
+ :raises ParserError:
+ Raised for invalid or unknown string format, if the provided
+ :class:`tzinfo` is not in a valid format, or if an invalid date
+ would be created.
+
+ :raises TypeError:
+ Raised for non-string or character stream input.
+
+ :raises OverflowError:
+ Raised if the parsed date exceeds the largest valid C integer on
+ your system.
+ """
+
+ if default is None:
+ default = datetime.datetime.now().replace(hour=0, minute=0,
+ second=0, microsecond=0)
+
+ res, skipped_tokens = self._parse(timestr, **kwargs)
+
+ if res is None:
+ raise ParserError("Unknown string format: %s", timestr)
+
+ if len(res) == 0:
+ raise ParserError("String does not contain a date: %s", timestr)
+
+ try:
+ ret = self._build_naive(res, default)
+ except ValueError as e:
+ six.raise_from(ParserError(str(e) + ": %s", timestr), e)
+
+ if not ignoretz:
+ ret = self._build_tzaware(ret, res, tzinfos)
+
+ if kwargs.get('fuzzy_with_tokens', False):
+ return ret, skipped_tokens
+ else:
+ return ret
+
+ class _result(_resultbase):
+ __slots__ = ["year", "month", "day", "weekday",
+ "hour", "minute", "second", "microsecond",
+ "tzname", "tzoffset", "ampm","any_unused_tokens"]
+
+ def _parse(self, timestr, dayfirst=None, yearfirst=None, fuzzy=False,
+ fuzzy_with_tokens=False):
+ """
+ Private method which performs the heavy lifting of parsing, called from
+ ``parse()``, which passes on its ``kwargs`` to this function.
+
+ :param timestr:
+ The string to parse.
+
+ :param dayfirst:
+ Whether to interpret the first value in an ambiguous 3-integer date
+ (e.g. 01/05/09) as the day (``True``) or month (``False``). If
+ ``yearfirst`` is set to ``True``, this distinguishes between YDM
+ and YMD. If set to ``None``, this value is retrieved from the
+ current :class:`parserinfo` object (which itself defaults to
+ ``False``).
+
+ :param yearfirst:
+ Whether to interpret the first value in an ambiguous 3-integer date
+ (e.g. 01/05/09) as the year. If ``True``, the first number is taken
+ to be the year, otherwise the last number is taken to be the year.
+ If this is set to ``None``, the value is retrieved from the current
+ :class:`parserinfo` object (which itself defaults to ``False``).
+
+ :param fuzzy:
+ Whether to allow fuzzy parsing, allowing for string like "Today is
+ January 1, 2047 at 8:21:00AM".
+
+ :param fuzzy_with_tokens:
+ If ``True``, ``fuzzy`` is automatically set to True, and the parser
+ will return a tuple where the first element is the parsed
+ :class:`datetime.datetime` datetimestamp and the second element is
+ a tuple containing the portions of the string which were ignored:
+
+ .. doctest::
+
+ >>> from dateutil.parser import parse
+ >>> parse("Today is January 1, 2047 at 8:21:00AM", fuzzy_with_tokens=True)
+ (datetime.datetime(2047, 1, 1, 8, 21), (u'Today is ', u' ', u'at '))
+
+ """
+ if fuzzy_with_tokens:
+ fuzzy = True
+
+ info = self.info
+
+ if dayfirst is None:
+ dayfirst = info.dayfirst
+
+ if yearfirst is None:
+ yearfirst = info.yearfirst
+
+ res = self._result()
+ l = _timelex.split(timestr) # Splits the timestr into tokens
+
+ skipped_idxs = []
+
+ # year/month/day list
+ ymd = _ymd()
+
+ len_l = len(l)
+ i = 0
+ try:
+ while i < len_l:
+
+ # Check if it's a number
+ value_repr = l[i]
+ try:
+ value = float(value_repr)
+ except ValueError:
+ value = None
+
+ if value is not None:
+ # Numeric token
+ i = self._parse_numeric_token(l, i, info, ymd, res, fuzzy)
+
+ # Check weekday
+ elif info.weekday(l[i]) is not None:
+ value = info.weekday(l[i])
+ res.weekday = value
+
+ # Check month name
+ elif info.month(l[i]) is not None:
+ value = info.month(l[i])
+ ymd.append(value, 'M')
+
+ if i + 1 < len_l:
+ if l[i + 1] in ('-', '/'):
+ # Jan-01[-99]
+ sep = l[i + 1]
+ ymd.append(l[i + 2])
+
+ if i + 3 < len_l and l[i + 3] == sep:
+ # Jan-01-99
+ ymd.append(l[i + 4])
+ i += 2
+
+ i += 2
+
+ elif (i + 4 < len_l and l[i + 1] == l[i + 3] == ' ' and
+ info.pertain(l[i + 2])):
+ # Jan of 01
+ # In this case, 01 is clearly year
+ if l[i + 4].isdigit():
+ # Convert it here to become unambiguous
+ value = int(l[i + 4])
+ year = str(info.convertyear(value))
+ ymd.append(year, 'Y')
+ else:
+ # Wrong guess
+ pass
+ # TODO: not hit in tests
+ i += 4
+
+ # Check am/pm
+ elif info.ampm(l[i]) is not None:
+ value = info.ampm(l[i])
+ val_is_ampm = self._ampm_valid(res.hour, res.ampm, fuzzy)
+
+ if val_is_ampm:
+ res.hour = self._adjust_ampm(res.hour, value)
+ res.ampm = value
+
+ elif fuzzy:
+ skipped_idxs.append(i)
+
+ # Check for a timezone name
+ elif self._could_be_tzname(res.hour, res.tzname, res.tzoffset, l[i]):
+ res.tzname = l[i]
+ res.tzoffset = info.tzoffset(res.tzname)
+
+ # Check for something like GMT+3, or BRST+3. Notice
+ # that it doesn't mean "I am 3 hours after GMT", but
+ # "my time +3 is GMT". If found, we reverse the
+ # logic so that timezone parsing code will get it
+ # right.
+ if i + 1 < len_l and l[i + 1] in ('+', '-'):
+ l[i + 1] = ('+', '-')[l[i + 1] == '+']
+ res.tzoffset = None
+ if info.utczone(res.tzname):
+ # With something like GMT+3, the timezone
+ # is *not* GMT.
+ res.tzname = None
+
+ # Check for a numbered timezone
+ elif res.hour is not None and l[i] in ('+', '-'):
+ signal = (-1, 1)[l[i] == '+']
+ len_li = len(l[i + 1])
+
+ # TODO: check that l[i + 1] is integer?
+ if len_li == 4:
+ # -0300
+ hour_offset = int(l[i + 1][:2])
+ min_offset = int(l[i + 1][2:])
+ elif i + 2 < len_l and l[i + 2] == ':':
+ # -03:00
+ hour_offset = int(l[i + 1])
+ min_offset = int(l[i + 3]) # TODO: Check that l[i+3] is minute-like?
+ i += 2
+ elif len_li <= 2:
+ # -[0]3
+ hour_offset = int(l[i + 1][:2])
+ min_offset = 0
+ else:
+ raise ValueError(timestr)
+
+ res.tzoffset = signal * (hour_offset * 3600 + min_offset * 60)
+
+ # Look for a timezone name between parenthesis
+ if (i + 5 < len_l and
+ info.jump(l[i + 2]) and l[i + 3] == '(' and
+ l[i + 5] == ')' and
+ 3 <= len(l[i + 4]) and
+ self._could_be_tzname(res.hour, res.tzname,
+ None, l[i + 4])):
+ # -0300 (BRST)
+ res.tzname = l[i + 4]
+ i += 4
+
+ i += 1
+
+ # Check jumps
+ elif not (info.jump(l[i]) or fuzzy):
+ raise ValueError(timestr)
+
+ else:
+ skipped_idxs.append(i)
+ i += 1
+
+ # Process year/month/day
+ year, month, day = ymd.resolve_ymd(yearfirst, dayfirst)
+
+ res.century_specified = ymd.century_specified
+ res.year = year
+ res.month = month
+ res.day = day
+
+ except (IndexError, ValueError):
+ return None, None
+
+ if not info.validate(res):
+ return None, None
+
+ if fuzzy_with_tokens:
+ skipped_tokens = self._recombine_skipped(l, skipped_idxs)
+ return res, tuple(skipped_tokens)
+ else:
+ return res, None
+
+ def _parse_numeric_token(self, tokens, idx, info, ymd, res, fuzzy):
+ # Token is a number
+ value_repr = tokens[idx]
+ try:
+ value = self._to_decimal(value_repr)
+ except Exception as e:
+ six.raise_from(ValueError('Unknown numeric token'), e)
+
+ len_li = len(value_repr)
+
+ len_l = len(tokens)
+
+ if (len(ymd) == 3 and len_li in (2, 4) and
+ res.hour is None and
+ (idx + 1 >= len_l or
+ (tokens[idx + 1] != ':' and
+ info.hms(tokens[idx + 1]) is None))):
+ # 19990101T23[59]
+ s = tokens[idx]
+ res.hour = int(s[:2])
+
+ if len_li == 4:
+ res.minute = int(s[2:])
+
+ elif len_li == 6 or (len_li > 6 and tokens[idx].find('.') == 6):
+ # YYMMDD or HHMMSS[.ss]
+ s = tokens[idx]
+
+ if not ymd and '.' not in tokens[idx]:
+ ymd.append(s[:2])
+ ymd.append(s[2:4])
+ ymd.append(s[4:])
+ else:
+ # 19990101T235959[.59]
+
+ # TODO: Check if res attributes already set.
+ res.hour = int(s[:2])
+ res.minute = int(s[2:4])
+ res.second, res.microsecond = self._parsems(s[4:])
+
+ elif len_li in (8, 12, 14):
+ # YYYYMMDD
+ s = tokens[idx]
+ ymd.append(s[:4], 'Y')
+ ymd.append(s[4:6])
+ ymd.append(s[6:8])
+
+ if len_li > 8:
+ res.hour = int(s[8:10])
+ res.minute = int(s[10:12])
+
+ if len_li > 12:
+ res.second = int(s[12:])
+
+ elif self._find_hms_idx(idx, tokens, info, allow_jump=True) is not None:
+ # HH[ ]h or MM[ ]m or SS[.ss][ ]s
+ hms_idx = self._find_hms_idx(idx, tokens, info, allow_jump=True)
+ (idx, hms) = self._parse_hms(idx, tokens, info, hms_idx)
+ if hms is not None:
+ # TODO: checking that hour/minute/second are not
+ # already set?
+ self._assign_hms(res, value_repr, hms)
+
+ elif idx + 2 < len_l and tokens[idx + 1] == ':':
+ # HH:MM[:SS[.ss]]
+ res.hour = int(value)
+ value = self._to_decimal(tokens[idx + 2]) # TODO: try/except for this?
+ (res.minute, res.second) = self._parse_min_sec(value)
+
+ if idx + 4 < len_l and tokens[idx + 3] == ':':
+ res.second, res.microsecond = self._parsems(tokens[idx + 4])
+
+ idx += 2
+
+ idx += 2
+
+ elif idx + 1 < len_l and tokens[idx + 1] in ('-', '/', '.'):
+ sep = tokens[idx + 1]
+ ymd.append(value_repr)
+
+ if idx + 2 < len_l and not info.jump(tokens[idx + 2]):
+ if tokens[idx + 2].isdigit():
+ # 01-01[-01]
+ ymd.append(tokens[idx + 2])
+ else:
+ # 01-Jan[-01]
+ value = info.month(tokens[idx + 2])
+
+ if value is not None:
+ ymd.append(value, 'M')
+ else:
+ raise ValueError()
+
+ if idx + 3 < len_l and tokens[idx + 3] == sep:
+ # We have three members
+ value = info.month(tokens[idx + 4])
+
+ if value is not None:
+ ymd.append(value, 'M')
+ else:
+ ymd.append(tokens[idx + 4])
+ idx += 2
+
+ idx += 1
+ idx += 1
+
+ elif idx + 1 >= len_l or info.jump(tokens[idx + 1]):
+ if idx + 2 < len_l and info.ampm(tokens[idx + 2]) is not None:
+ # 12 am
+ hour = int(value)
+ res.hour = self._adjust_ampm(hour, info.ampm(tokens[idx + 2]))
+ idx += 1
+ else:
+ # Year, month or day
+ ymd.append(value)
+ idx += 1
+
+ elif info.ampm(tokens[idx + 1]) is not None and (0 <= value < 24):
+ # 12am
+ hour = int(value)
+ res.hour = self._adjust_ampm(hour, info.ampm(tokens[idx + 1]))
+ idx += 1
+
+ elif ymd.could_be_day(value):
+ ymd.append(value)
+
+ elif not fuzzy:
+ raise ValueError()
+
+ return idx
+
+ def _find_hms_idx(self, idx, tokens, info, allow_jump):
+ len_l = len(tokens)
+
+ if idx+1 < len_l and info.hms(tokens[idx+1]) is not None:
+ # There is an "h", "m", or "s" label following this token. We take
+ # assign the upcoming label to the current token.
+ # e.g. the "12" in 12h"
+ hms_idx = idx + 1
+
+ elif (allow_jump and idx+2 < len_l and tokens[idx+1] == ' ' and
+ info.hms(tokens[idx+2]) is not None):
+ # There is a space and then an "h", "m", or "s" label.
+ # e.g. the "12" in "12 h"
+ hms_idx = idx + 2
+
+ elif idx > 0 and info.hms(tokens[idx-1]) is not None:
+ # There is a "h", "m", or "s" preceding this token. Since neither
+ # of the previous cases was hit, there is no label following this
+ # token, so we use the previous label.
+ # e.g. the "04" in "12h04"
+ hms_idx = idx-1
+
+ elif (1 < idx == len_l-1 and tokens[idx-1] == ' ' and
+ info.hms(tokens[idx-2]) is not None):
+ # If we are looking at the final token, we allow for a
+ # backward-looking check to skip over a space.
+ # TODO: Are we sure this is the right condition here?
+ hms_idx = idx - 2
+
+ else:
+ hms_idx = None
+
+ return hms_idx
+
+ def _assign_hms(self, res, value_repr, hms):
+ # See GH issue #427, fixing float rounding
+ value = self._to_decimal(value_repr)
+
+ if hms == 0:
+ # Hour
+ res.hour = int(value)
+ if value % 1:
+ res.minute = int(60*(value % 1))
+
+ elif hms == 1:
+ (res.minute, res.second) = self._parse_min_sec(value)
+
+ elif hms == 2:
+ (res.second, res.microsecond) = self._parsems(value_repr)
+
+ def _could_be_tzname(self, hour, tzname, tzoffset, token):
+ return (hour is not None and
+ tzname is None and
+ tzoffset is None and
+ len(token) <= 5 and
+ (all(x in string.ascii_uppercase for x in token)
+ or token in self.info.UTCZONE))
+
+ def _ampm_valid(self, hour, ampm, fuzzy):
+ """
+ For fuzzy parsing, 'a' or 'am' (both valid English words)
+ may erroneously trigger the AM/PM flag. Deal with that
+ here.
+ """
+ val_is_ampm = True
+
+ # If there's already an AM/PM flag, this one isn't one.
+ if fuzzy and ampm is not None:
+ val_is_ampm = False
+
+ # If AM/PM is found and hour is not, raise a ValueError
+ if hour is None:
+ if fuzzy:
+ val_is_ampm = False
+ else:
+ raise ValueError('No hour specified with AM or PM flag.')
+ elif not 0 <= hour <= 12:
+ # If AM/PM is found, it's a 12 hour clock, so raise
+ # an error for invalid range
+ if fuzzy:
+ val_is_ampm = False
+ else:
+ raise ValueError('Invalid hour specified for 12-hour clock.')
+
+ return val_is_ampm
+
+ def _adjust_ampm(self, hour, ampm):
+ if hour < 12 and ampm == 1:
+ hour += 12
+ elif hour == 12 and ampm == 0:
+ hour = 0
+ return hour
+
+ def _parse_min_sec(self, value):
+ # TODO: Every usage of this function sets res.second to the return
+ # value. Are there any cases where second will be returned as None and
+ # we *don't* want to set res.second = None?
+ minute = int(value)
+ second = None
+
+ sec_remainder = value % 1
+ if sec_remainder:
+ second = int(60 * sec_remainder)
+ return (minute, second)
+
+ def _parse_hms(self, idx, tokens, info, hms_idx):
+ # TODO: Is this going to admit a lot of false-positives for when we
+ # just happen to have digits and "h", "m" or "s" characters in non-date
+ # text? I guess hex hashes won't have that problem, but there's plenty
+ # of random junk out there.
+ if hms_idx is None:
+ hms = None
+ new_idx = idx
+ elif hms_idx > idx:
+ hms = info.hms(tokens[hms_idx])
+ new_idx = hms_idx
+ else:
+ # Looking backwards, increment one.
+ hms = info.hms(tokens[hms_idx]) + 1
+ new_idx = idx
+
+ return (new_idx, hms)
+
+ # ------------------------------------------------------------------
+ # Handling for individual tokens. These are kept as methods instead
+ # of functions for the sake of customizability via subclassing.
+
+ def _parsems(self, value):
+ """Parse a I[.F] seconds value into (seconds, microseconds)."""
+ if "." not in value:
+ return int(value), 0
+ else:
+ i, f = value.split(".")
+ return int(i), int(f.ljust(6, "0")[:6])
+
+ def _to_decimal(self, val):
+ try:
+ decimal_value = Decimal(val)
+ # See GH 662, edge case, infinite value should not be converted
+ # via `_to_decimal`
+ if not decimal_value.is_finite():
+ raise ValueError("Converted decimal value is infinite or NaN")
+ except Exception as e:
+ msg = "Could not convert %s to decimal" % val
+ six.raise_from(ValueError(msg), e)
+ else:
+ return decimal_value
+
+ # ------------------------------------------------------------------
+ # Post-Parsing construction of datetime output. These are kept as
+ # methods instead of functions for the sake of customizability via
+ # subclassing.
+
+ def _build_tzinfo(self, tzinfos, tzname, tzoffset):
+ if callable(tzinfos):
+ tzdata = tzinfos(tzname, tzoffset)
+ else:
+ tzdata = tzinfos.get(tzname)
+ # handle case where tzinfo is paased an options that returns None
+ # eg tzinfos = {'BRST' : None}
+ if isinstance(tzdata, datetime.tzinfo) or tzdata is None:
+ tzinfo = tzdata
+ elif isinstance(tzdata, text_type):
+ tzinfo = tz.tzstr(tzdata)
+ elif isinstance(tzdata, integer_types):
+ tzinfo = tz.tzoffset(tzname, tzdata)
+ else:
+ raise TypeError("Offset must be tzinfo subclass, tz string, "
+ "or int offset.")
+ return tzinfo
+
+ def _build_tzaware(self, naive, res, tzinfos):
+ if (callable(tzinfos) or (tzinfos and res.tzname in tzinfos)):
+ tzinfo = self._build_tzinfo(tzinfos, res.tzname, res.tzoffset)
+ aware = naive.replace(tzinfo=tzinfo)
+ aware = self._assign_tzname(aware, res.tzname)
+
+ elif res.tzname and res.tzname in time.tzname:
+ aware = naive.replace(tzinfo=tz.tzlocal())
+
+ # Handle ambiguous local datetime
+ aware = self._assign_tzname(aware, res.tzname)
+
+ # This is mostly relevant for winter GMT zones parsed in the UK
+ if (aware.tzname() != res.tzname and
+ res.tzname in self.info.UTCZONE):
+ aware = aware.replace(tzinfo=tz.UTC)
+
+ elif res.tzoffset == 0:
+ aware = naive.replace(tzinfo=tz.UTC)
+
+ elif res.tzoffset:
+ aware = naive.replace(tzinfo=tz.tzoffset(res.tzname, res.tzoffset))
+
+ elif not res.tzname and not res.tzoffset:
+ # i.e. no timezone information was found.
+ aware = naive
+
+ elif res.tzname:
+ # tz-like string was parsed but we don't know what to do
+ # with it
+ warnings.warn("tzname {tzname} identified but not understood. "
+ "Pass `tzinfos` argument in order to correctly "
+ "return a timezone-aware datetime. In a future "
+ "version, this will raise an "
+ "exception.".format(tzname=res.tzname),
+ category=UnknownTimezoneWarning)
+ aware = naive
+
+ return aware
+
+ def _build_naive(self, res, default):
+ repl = {}
+ for attr in ("year", "month", "day", "hour",
+ "minute", "second", "microsecond"):
+ value = getattr(res, attr)
+ if value is not None:
+ repl[attr] = value
+
+ if 'day' not in repl:
+ # If the default day exceeds the last day of the month, fall back
+ # to the end of the month.
+ cyear = default.year if res.year is None else res.year
+ cmonth = default.month if res.month is None else res.month
+ cday = default.day if res.day is None else res.day
+
+ if cday > monthrange(cyear, cmonth)[1]:
+ repl['day'] = monthrange(cyear, cmonth)[1]
+
+ naive = default.replace(**repl)
+
+ if res.weekday is not None and not res.day:
+ naive = naive + relativedelta.relativedelta(weekday=res.weekday)
+
+ return naive
+
+ def _assign_tzname(self, dt, tzname):
+ if dt.tzname() != tzname:
+ new_dt = tz.enfold(dt, fold=1)
+ if new_dt.tzname() == tzname:
+ return new_dt
+
+ return dt
+
+ def _recombine_skipped(self, tokens, skipped_idxs):
+ """
+ >>> tokens = ["foo", " ", "bar", " ", "19June2000", "baz"]
+ >>> skipped_idxs = [0, 1, 2, 5]
+ >>> _recombine_skipped(tokens, skipped_idxs)
+ ["foo bar", "baz"]
+ """
+ skipped_tokens = []
+ for i, idx in enumerate(sorted(skipped_idxs)):
+ if i > 0 and idx - 1 == skipped_idxs[i - 1]:
+ skipped_tokens[-1] = skipped_tokens[-1] + tokens[idx]
+ else:
+ skipped_tokens.append(tokens[idx])
+
+ return skipped_tokens
+
+
+DEFAULTPARSER = parser()
+
+
+def parse(timestr, parserinfo=None, **kwargs):
+ """
+
+ Parse a string in one of the supported formats, using the
+ ``parserinfo`` parameters.
+
+ :param timestr:
+ A string containing a date/time stamp.
+
+ :param parserinfo:
+ A :class:`parserinfo` object containing parameters for the parser.
+ If ``None``, the default arguments to the :class:`parserinfo`
+ constructor are used.
+
+ The ``**kwargs`` parameter takes the following keyword arguments:
+
+ :param default:
+ The default datetime object, if this is a datetime object and not
+ ``None``, elements specified in ``timestr`` replace elements in the
+ default object.
+
+ :param ignoretz:
+ If set ``True``, time zones in parsed strings are ignored and a naive
+ :class:`datetime` object is returned.
+
+ :param tzinfos:
+ Additional time zone names / aliases which may be present in the
+ string. This argument maps time zone names (and optionally offsets
+ from those time zones) to time zones. This parameter can be a
+ dictionary with timezone aliases mapping time zone names to time
+ zones or a function taking two parameters (``tzname`` and
+ ``tzoffset``) and returning a time zone.
+
+ The timezones to which the names are mapped can be an integer
+ offset from UTC in seconds or a :class:`tzinfo` object.
+
+ .. doctest::
+ :options: +NORMALIZE_WHITESPACE
+
+ >>> from dateutil.parser import parse
+ >>> from dateutil.tz import gettz
+ >>> tzinfos = {"BRST": -7200, "CST": gettz("America/Chicago")}
+ >>> parse("2012-01-19 17:21:00 BRST", tzinfos=tzinfos)
+ datetime.datetime(2012, 1, 19, 17, 21, tzinfo=tzoffset(u'BRST', -7200))
+ >>> parse("2012-01-19 17:21:00 CST", tzinfos=tzinfos)
+ datetime.datetime(2012, 1, 19, 17, 21,
+ tzinfo=tzfile('/usr/share/zoneinfo/America/Chicago'))
+
+ This parameter is ignored if ``ignoretz`` is set.
+
+ :param dayfirst:
+ Whether to interpret the first value in an ambiguous 3-integer date
+ (e.g. 01/05/09) as the day (``True``) or month (``False``). If
+ ``yearfirst`` is set to ``True``, this distinguishes between YDM and
+ YMD. If set to ``None``, this value is retrieved from the current
+ :class:`parserinfo` object (which itself defaults to ``False``).
+
+ :param yearfirst:
+ Whether to interpret the first value in an ambiguous 3-integer date
+ (e.g. 01/05/09) as the year. If ``True``, the first number is taken to
+ be the year, otherwise the last number is taken to be the year. If
+ this is set to ``None``, the value is retrieved from the current
+ :class:`parserinfo` object (which itself defaults to ``False``).
+
+ :param fuzzy:
+ Whether to allow fuzzy parsing, allowing for string like "Today is
+ January 1, 2047 at 8:21:00AM".
+
+ :param fuzzy_with_tokens:
+ If ``True``, ``fuzzy`` is automatically set to True, and the parser
+ will return a tuple where the first element is the parsed
+ :class:`datetime.datetime` datetimestamp and the second element is
+ a tuple containing the portions of the string which were ignored:
+
+ .. doctest::
+
+ >>> from dateutil.parser import parse
+ >>> parse("Today is January 1, 2047 at 8:21:00AM", fuzzy_with_tokens=True)
+ (datetime.datetime(2047, 1, 1, 8, 21), (u'Today is ', u' ', u'at '))
+
+ :return:
+ Returns a :class:`datetime.datetime` object or, if the
+ ``fuzzy_with_tokens`` option is ``True``, returns a tuple, the
+ first element being a :class:`datetime.datetime` object, the second
+ a tuple containing the fuzzy tokens.
+
+ :raises ParserError:
+ Raised for invalid or unknown string formats, if the provided
+ :class:`tzinfo` is not in a valid format, or if an invalid date would
+ be created.
+
+ :raises OverflowError:
+ Raised if the parsed date exceeds the largest valid C integer on
+ your system.
+ """
+ if parserinfo:
+ return parser(parserinfo).parse(timestr, **kwargs)
+ else:
+ return DEFAULTPARSER.parse(timestr, **kwargs)
+
+
+class _tzparser(object):
+
+ class _result(_resultbase):
+
+ __slots__ = ["stdabbr", "stdoffset", "dstabbr", "dstoffset",
+ "start", "end"]
+
+ class _attr(_resultbase):
+ __slots__ = ["month", "week", "weekday",
+ "yday", "jyday", "day", "time"]
+
+ def __repr__(self):
+ return self._repr("")
+
+ def __init__(self):
+ _resultbase.__init__(self)
+ self.start = self._attr()
+ self.end = self._attr()
+
+ def parse(self, tzstr):
+ res = self._result()
+ l = [x for x in re.split(r'([,:.]|[a-zA-Z]+|[0-9]+)',tzstr) if x]
+ used_idxs = list()
+ try:
+
+ len_l = len(l)
+
+ i = 0
+ while i < len_l:
+ # BRST+3[BRDT[+2]]
+ j = i
+ while j < len_l and not [x for x in l[j]
+ if x in "0123456789:,-+"]:
+ j += 1
+ if j != i:
+ if not res.stdabbr:
+ offattr = "stdoffset"
+ res.stdabbr = "".join(l[i:j])
+ else:
+ offattr = "dstoffset"
+ res.dstabbr = "".join(l[i:j])
+
+ for ii in range(j):
+ used_idxs.append(ii)
+ i = j
+ if (i < len_l and (l[i] in ('+', '-') or l[i][0] in
+ "0123456789")):
+ if l[i] in ('+', '-'):
+ # Yes, that's right. See the TZ variable
+ # documentation.
+ signal = (1, -1)[l[i] == '+']
+ used_idxs.append(i)
+ i += 1
+ else:
+ signal = -1
+ len_li = len(l[i])
+ if len_li == 4:
+ # -0300
+ setattr(res, offattr, (int(l[i][:2]) * 3600 +
+ int(l[i][2:]) * 60) * signal)
+ elif i + 1 < len_l and l[i + 1] == ':':
+ # -03:00
+ setattr(res, offattr,
+ (int(l[i]) * 3600 +
+ int(l[i + 2]) * 60) * signal)
+ used_idxs.append(i)
+ i += 2
+ elif len_li <= 2:
+ # -[0]3
+ setattr(res, offattr,
+ int(l[i][:2]) * 3600 * signal)
+ else:
+ return None
+ used_idxs.append(i)
+ i += 1
+ if res.dstabbr:
+ break
+ else:
+ break
+
+
+ if i < len_l:
+ for j in range(i, len_l):
+ if l[j] == ';':
+ l[j] = ','
+
+ assert l[i] == ','
+
+ i += 1
+
+ if i >= len_l:
+ pass
+ elif (8 <= l.count(',') <= 9 and
+ not [y for x in l[i:] if x != ','
+ for y in x if y not in "0123456789+-"]):
+ # GMT0BST,3,0,30,3600,10,0,26,7200[,3600]
+ for x in (res.start, res.end):
+ x.month = int(l[i])
+ used_idxs.append(i)
+ i += 2
+ if l[i] == '-':
+ value = int(l[i + 1]) * -1
+ used_idxs.append(i)
+ i += 1
+ else:
+ value = int(l[i])
+ used_idxs.append(i)
+ i += 2
+ if value:
+ x.week = value
+ x.weekday = (int(l[i]) - 1) % 7
+ else:
+ x.day = int(l[i])
+ used_idxs.append(i)
+ i += 2
+ x.time = int(l[i])
+ used_idxs.append(i)
+ i += 2
+ if i < len_l:
+ if l[i] in ('-', '+'):
+ signal = (-1, 1)[l[i] == "+"]
+ used_idxs.append(i)
+ i += 1
+ else:
+ signal = 1
+ used_idxs.append(i)
+ res.dstoffset = (res.stdoffset + int(l[i]) * signal)
+
+ # This was a made-up format that is not in normal use
+ warn(('Parsed time zone "%s"' % tzstr) +
+ 'is in a non-standard dateutil-specific format, which ' +
+ 'is now deprecated; support for parsing this format ' +
+ 'will be removed in future versions. It is recommended ' +
+ 'that you switch to a standard format like the GNU ' +
+ 'TZ variable format.', tz.DeprecatedTzFormatWarning)
+ elif (l.count(',') == 2 and l[i:].count('/') <= 2 and
+ not [y for x in l[i:] if x not in (',', '/', 'J', 'M',
+ '.', '-', ':')
+ for y in x if y not in "0123456789"]):
+ for x in (res.start, res.end):
+ if l[i] == 'J':
+ # non-leap year day (1 based)
+ used_idxs.append(i)
+ i += 1
+ x.jyday = int(l[i])
+ elif l[i] == 'M':
+ # month[-.]week[-.]weekday
+ used_idxs.append(i)
+ i += 1
+ x.month = int(l[i])
+ used_idxs.append(i)
+ i += 1
+ assert l[i] in ('-', '.')
+ used_idxs.append(i)
+ i += 1
+ x.week = int(l[i])
+ if x.week == 5:
+ x.week = -1
+ used_idxs.append(i)
+ i += 1
+ assert l[i] in ('-', '.')
+ used_idxs.append(i)
+ i += 1
+ x.weekday = (int(l[i]) - 1) % 7
+ else:
+ # year day (zero based)
+ x.yday = int(l[i]) + 1
+
+ used_idxs.append(i)
+ i += 1
+
+ if i < len_l and l[i] == '/':
+ used_idxs.append(i)
+ i += 1
+ # start time
+ len_li = len(l[i])
+ if len_li == 4:
+ # -0300
+ x.time = (int(l[i][:2]) * 3600 +
+ int(l[i][2:]) * 60)
+ elif i + 1 < len_l and l[i + 1] == ':':
+ # -03:00
+ x.time = int(l[i]) * 3600 + int(l[i + 2]) * 60
+ used_idxs.append(i)
+ i += 2
+ if i + 1 < len_l and l[i + 1] == ':':
+ used_idxs.append(i)
+ i += 2
+ x.time += int(l[i])
+ elif len_li <= 2:
+ # -[0]3
+ x.time = (int(l[i][:2]) * 3600)
+ else:
+ return None
+ used_idxs.append(i)
+ i += 1
+
+ assert i == len_l or l[i] == ','
+
+ i += 1
+
+ assert i >= len_l
+
+ except (IndexError, ValueError, AssertionError):
+ return None
+
+ unused_idxs = set(range(len_l)).difference(used_idxs)
+ res.any_unused_tokens = not {l[n] for n in unused_idxs}.issubset({",",":"})
+ return res
+
+
+DEFAULTTZPARSER = _tzparser()
+
+
+def _parsetz(tzstr):
+ return DEFAULTTZPARSER.parse(tzstr)
+
+
+class ParserError(ValueError):
+ """Exception subclass used for any failure to parse a datetime string.
+
+ This is a subclass of :py:exc:`ValueError`, and should be raised any time
+ earlier versions of ``dateutil`` would have raised ``ValueError``.
+
+ .. versionadded:: 2.8.1
+ """
+ def __str__(self):
+ try:
+ return self.args[0] % self.args[1:]
+ except (TypeError, IndexError):
+ return super(ParserError, self).__str__()
+
+ def __repr__(self):
+ args = ", ".join("'%s'" % arg for arg in self.args)
+ return "%s(%s)" % (self.__class__.__name__, args)
+
+
+class UnknownTimezoneWarning(RuntimeWarning):
+ """Raised when the parser finds a timezone it cannot parse into a tzinfo.
+
+ .. versionadded:: 2.7.0
+ """
+# vim:ts=4:sw=4:et
diff --git a/falcon/lib/python3.10/site-packages/dateutil/parser/isoparser.py b/falcon/lib/python3.10/site-packages/dateutil/parser/isoparser.py
new file mode 100644
index 0000000000000000000000000000000000000000..7060087df4776a07347cbb60127a70db393e3a65
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/parser/isoparser.py
@@ -0,0 +1,416 @@
+# -*- coding: utf-8 -*-
+"""
+This module offers a parser for ISO-8601 strings
+
+It is intended to support all valid date, time and datetime formats per the
+ISO-8601 specification.
+
+..versionadded:: 2.7.0
+"""
+from datetime import datetime, timedelta, time, date
+import calendar
+from dateutil import tz
+
+from functools import wraps
+
+import re
+import six
+
+__all__ = ["isoparse", "isoparser"]
+
+
+def _takes_ascii(f):
+ @wraps(f)
+ def func(self, str_in, *args, **kwargs):
+ # If it's a stream, read the whole thing
+ str_in = getattr(str_in, 'read', lambda: str_in)()
+
+ # If it's unicode, turn it into bytes, since ISO-8601 only covers ASCII
+ if isinstance(str_in, six.text_type):
+ # ASCII is the same in UTF-8
+ try:
+ str_in = str_in.encode('ascii')
+ except UnicodeEncodeError as e:
+ msg = 'ISO-8601 strings should contain only ASCII characters'
+ six.raise_from(ValueError(msg), e)
+
+ return f(self, str_in, *args, **kwargs)
+
+ return func
+
+
+class isoparser(object):
+ def __init__(self, sep=None):
+ """
+ :param sep:
+ A single character that separates date and time portions. If
+ ``None``, the parser will accept any single character.
+ For strict ISO-8601 adherence, pass ``'T'``.
+ """
+ if sep is not None:
+ if (len(sep) != 1 or ord(sep) >= 128 or sep in '0123456789'):
+ raise ValueError('Separator must be a single, non-numeric ' +
+ 'ASCII character')
+
+ sep = sep.encode('ascii')
+
+ self._sep = sep
+
+ @_takes_ascii
+ def isoparse(self, dt_str):
+ """
+ Parse an ISO-8601 datetime string into a :class:`datetime.datetime`.
+
+ An ISO-8601 datetime string consists of a date portion, followed
+ optionally by a time portion - the date and time portions are separated
+ by a single character separator, which is ``T`` in the official
+ standard. Incomplete date formats (such as ``YYYY-MM``) may *not* be
+ combined with a time portion.
+
+ Supported date formats are:
+
+ Common:
+
+ - ``YYYY``
+ - ``YYYY-MM``
+ - ``YYYY-MM-DD`` or ``YYYYMMDD``
+
+ Uncommon:
+
+ - ``YYYY-Www`` or ``YYYYWww`` - ISO week (day defaults to 0)
+ - ``YYYY-Www-D`` or ``YYYYWwwD`` - ISO week and day
+
+ The ISO week and day numbering follows the same logic as
+ :func:`datetime.date.isocalendar`.
+
+ Supported time formats are:
+
+ - ``hh``
+ - ``hh:mm`` or ``hhmm``
+ - ``hh:mm:ss`` or ``hhmmss``
+ - ``hh:mm:ss.ssssss`` (Up to 6 sub-second digits)
+
+ Midnight is a special case for `hh`, as the standard supports both
+ 00:00 and 24:00 as a representation. The decimal separator can be
+ either a dot or a comma.
+
+
+ .. caution::
+
+ Support for fractional components other than seconds is part of the
+ ISO-8601 standard, but is not currently implemented in this parser.
+
+ Supported time zone offset formats are:
+
+ - `Z` (UTC)
+ - `±HH:MM`
+ - `±HHMM`
+ - `±HH`
+
+ Offsets will be represented as :class:`dateutil.tz.tzoffset` objects,
+ with the exception of UTC, which will be represented as
+ :class:`dateutil.tz.tzutc`. Time zone offsets equivalent to UTC (such
+ as `+00:00`) will also be represented as :class:`dateutil.tz.tzutc`.
+
+ :param dt_str:
+ A string or stream containing only an ISO-8601 datetime string
+
+ :return:
+ Returns a :class:`datetime.datetime` representing the string.
+ Unspecified components default to their lowest value.
+
+ .. warning::
+
+ As of version 2.7.0, the strictness of the parser should not be
+ considered a stable part of the contract. Any valid ISO-8601 string
+ that parses correctly with the default settings will continue to
+ parse correctly in future versions, but invalid strings that
+ currently fail (e.g. ``2017-01-01T00:00+00:00:00``) are not
+ guaranteed to continue failing in future versions if they encode
+ a valid date.
+
+ .. versionadded:: 2.7.0
+ """
+ components, pos = self._parse_isodate(dt_str)
+
+ if len(dt_str) > pos:
+ if self._sep is None or dt_str[pos:pos + 1] == self._sep:
+ components += self._parse_isotime(dt_str[pos + 1:])
+ else:
+ raise ValueError('String contains unknown ISO components')
+
+ if len(components) > 3 and components[3] == 24:
+ components[3] = 0
+ return datetime(*components) + timedelta(days=1)
+
+ return datetime(*components)
+
+ @_takes_ascii
+ def parse_isodate(self, datestr):
+ """
+ Parse the date portion of an ISO string.
+
+ :param datestr:
+ The string portion of an ISO string, without a separator
+
+ :return:
+ Returns a :class:`datetime.date` object
+ """
+ components, pos = self._parse_isodate(datestr)
+ if pos < len(datestr):
+ raise ValueError('String contains unknown ISO ' +
+ 'components: {!r}'.format(datestr.decode('ascii')))
+ return date(*components)
+
+ @_takes_ascii
+ def parse_isotime(self, timestr):
+ """
+ Parse the time portion of an ISO string.
+
+ :param timestr:
+ The time portion of an ISO string, without a separator
+
+ :return:
+ Returns a :class:`datetime.time` object
+ """
+ components = self._parse_isotime(timestr)
+ if components[0] == 24:
+ components[0] = 0
+ return time(*components)
+
+ @_takes_ascii
+ def parse_tzstr(self, tzstr, zero_as_utc=True):
+ """
+ Parse a valid ISO time zone string.
+
+ See :func:`isoparser.isoparse` for details on supported formats.
+
+ :param tzstr:
+ A string representing an ISO time zone offset
+
+ :param zero_as_utc:
+ Whether to return :class:`dateutil.tz.tzutc` for zero-offset zones
+
+ :return:
+ Returns :class:`dateutil.tz.tzoffset` for offsets and
+ :class:`dateutil.tz.tzutc` for ``Z`` and (if ``zero_as_utc`` is
+ specified) offsets equivalent to UTC.
+ """
+ return self._parse_tzstr(tzstr, zero_as_utc=zero_as_utc)
+
+ # Constants
+ _DATE_SEP = b'-'
+ _TIME_SEP = b':'
+ _FRACTION_REGEX = re.compile(b'[\\.,]([0-9]+)')
+
+ def _parse_isodate(self, dt_str):
+ try:
+ return self._parse_isodate_common(dt_str)
+ except ValueError:
+ return self._parse_isodate_uncommon(dt_str)
+
+ def _parse_isodate_common(self, dt_str):
+ len_str = len(dt_str)
+ components = [1, 1, 1]
+
+ if len_str < 4:
+ raise ValueError('ISO string too short')
+
+ # Year
+ components[0] = int(dt_str[0:4])
+ pos = 4
+ if pos >= len_str:
+ return components, pos
+
+ has_sep = dt_str[pos:pos + 1] == self._DATE_SEP
+ if has_sep:
+ pos += 1
+
+ # Month
+ if len_str - pos < 2:
+ raise ValueError('Invalid common month')
+
+ components[1] = int(dt_str[pos:pos + 2])
+ pos += 2
+
+ if pos >= len_str:
+ if has_sep:
+ return components, pos
+ else:
+ raise ValueError('Invalid ISO format')
+
+ if has_sep:
+ if dt_str[pos:pos + 1] != self._DATE_SEP:
+ raise ValueError('Invalid separator in ISO string')
+ pos += 1
+
+ # Day
+ if len_str - pos < 2:
+ raise ValueError('Invalid common day')
+ components[2] = int(dt_str[pos:pos + 2])
+ return components, pos + 2
+
+ def _parse_isodate_uncommon(self, dt_str):
+ if len(dt_str) < 4:
+ raise ValueError('ISO string too short')
+
+ # All ISO formats start with the year
+ year = int(dt_str[0:4])
+
+ has_sep = dt_str[4:5] == self._DATE_SEP
+
+ pos = 4 + has_sep # Skip '-' if it's there
+ if dt_str[pos:pos + 1] == b'W':
+ # YYYY-?Www-?D?
+ pos += 1
+ weekno = int(dt_str[pos:pos + 2])
+ pos += 2
+
+ dayno = 1
+ if len(dt_str) > pos:
+ if (dt_str[pos:pos + 1] == self._DATE_SEP) != has_sep:
+ raise ValueError('Inconsistent use of dash separator')
+
+ pos += has_sep
+
+ dayno = int(dt_str[pos:pos + 1])
+ pos += 1
+
+ base_date = self._calculate_weekdate(year, weekno, dayno)
+ else:
+ # YYYYDDD or YYYY-DDD
+ if len(dt_str) - pos < 3:
+ raise ValueError('Invalid ordinal day')
+
+ ordinal_day = int(dt_str[pos:pos + 3])
+ pos += 3
+
+ if ordinal_day < 1 or ordinal_day > (365 + calendar.isleap(year)):
+ raise ValueError('Invalid ordinal day' +
+ ' {} for year {}'.format(ordinal_day, year))
+
+ base_date = date(year, 1, 1) + timedelta(days=ordinal_day - 1)
+
+ components = [base_date.year, base_date.month, base_date.day]
+ return components, pos
+
+ def _calculate_weekdate(self, year, week, day):
+ """
+ Calculate the day of corresponding to the ISO year-week-day calendar.
+
+ This function is effectively the inverse of
+ :func:`datetime.date.isocalendar`.
+
+ :param year:
+ The year in the ISO calendar
+
+ :param week:
+ The week in the ISO calendar - range is [1, 53]
+
+ :param day:
+ The day in the ISO calendar - range is [1 (MON), 7 (SUN)]
+
+ :return:
+ Returns a :class:`datetime.date`
+ """
+ if not 0 < week < 54:
+ raise ValueError('Invalid week: {}'.format(week))
+
+ if not 0 < day < 8: # Range is 1-7
+ raise ValueError('Invalid weekday: {}'.format(day))
+
+ # Get week 1 for the specific year:
+ jan_4 = date(year, 1, 4) # Week 1 always has January 4th in it
+ week_1 = jan_4 - timedelta(days=jan_4.isocalendar()[2] - 1)
+
+ # Now add the specific number of weeks and days to get what we want
+ week_offset = (week - 1) * 7 + (day - 1)
+ return week_1 + timedelta(days=week_offset)
+
+ def _parse_isotime(self, timestr):
+ len_str = len(timestr)
+ components = [0, 0, 0, 0, None]
+ pos = 0
+ comp = -1
+
+ if len_str < 2:
+ raise ValueError('ISO time too short')
+
+ has_sep = False
+
+ while pos < len_str and comp < 5:
+ comp += 1
+
+ if timestr[pos:pos + 1] in b'-+Zz':
+ # Detect time zone boundary
+ components[-1] = self._parse_tzstr(timestr[pos:])
+ pos = len_str
+ break
+
+ if comp == 1 and timestr[pos:pos+1] == self._TIME_SEP:
+ has_sep = True
+ pos += 1
+ elif comp == 2 and has_sep:
+ if timestr[pos:pos+1] != self._TIME_SEP:
+ raise ValueError('Inconsistent use of colon separator')
+ pos += 1
+
+ if comp < 3:
+ # Hour, minute, second
+ components[comp] = int(timestr[pos:pos + 2])
+ pos += 2
+
+ if comp == 3:
+ # Fraction of a second
+ frac = self._FRACTION_REGEX.match(timestr[pos:])
+ if not frac:
+ continue
+
+ us_str = frac.group(1)[:6] # Truncate to microseconds
+ components[comp] = int(us_str) * 10**(6 - len(us_str))
+ pos += len(frac.group())
+
+ if pos < len_str:
+ raise ValueError('Unused components in ISO string')
+
+ if components[0] == 24:
+ # Standard supports 00:00 and 24:00 as representations of midnight
+ if any(component != 0 for component in components[1:4]):
+ raise ValueError('Hour may only be 24 at 24:00:00.000')
+
+ return components
+
+ def _parse_tzstr(self, tzstr, zero_as_utc=True):
+ if tzstr == b'Z' or tzstr == b'z':
+ return tz.UTC
+
+ if len(tzstr) not in {3, 5, 6}:
+ raise ValueError('Time zone offset must be 1, 3, 5 or 6 characters')
+
+ if tzstr[0:1] == b'-':
+ mult = -1
+ elif tzstr[0:1] == b'+':
+ mult = 1
+ else:
+ raise ValueError('Time zone offset requires sign')
+
+ hours = int(tzstr[1:3])
+ if len(tzstr) == 3:
+ minutes = 0
+ else:
+ minutes = int(tzstr[(4 if tzstr[3:4] == self._TIME_SEP else 3):])
+
+ if zero_as_utc and hours == 0 and minutes == 0:
+ return tz.UTC
+ else:
+ if minutes > 59:
+ raise ValueError('Invalid minutes in time zone offset')
+
+ if hours > 23:
+ raise ValueError('Invalid hours in time zone offset')
+
+ return tz.tzoffset(None, mult * (hours * 60 + minutes) * 60)
+
+
+DEFAULT_ISOPARSER = isoparser()
+isoparse = DEFAULT_ISOPARSER.isoparse
diff --git a/falcon/lib/python3.10/site-packages/dateutil/relativedelta.py b/falcon/lib/python3.10/site-packages/dateutil/relativedelta.py
new file mode 100644
index 0000000000000000000000000000000000000000..cd323a549e0f182541ebcde2d2ea1adfbbd9701e
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/relativedelta.py
@@ -0,0 +1,599 @@
+# -*- coding: utf-8 -*-
+import datetime
+import calendar
+
+import operator
+from math import copysign
+
+from six import integer_types
+from warnings import warn
+
+from ._common import weekday
+
+MO, TU, WE, TH, FR, SA, SU = weekdays = tuple(weekday(x) for x in range(7))
+
+__all__ = ["relativedelta", "MO", "TU", "WE", "TH", "FR", "SA", "SU"]
+
+
+class relativedelta(object):
+ """
+ The relativedelta type is designed to be applied to an existing datetime and
+ can replace specific components of that datetime, or represents an interval
+ of time.
+
+ It is based on the specification of the excellent work done by M.-A. Lemburg
+ in his
+ `mx.DateTime `_ extension.
+ However, notice that this type does *NOT* implement the same algorithm as
+ his work. Do *NOT* expect it to behave like mx.DateTime's counterpart.
+
+ There are two different ways to build a relativedelta instance. The
+ first one is passing it two date/datetime classes::
+
+ relativedelta(datetime1, datetime2)
+
+ The second one is passing it any number of the following keyword arguments::
+
+ relativedelta(arg1=x,arg2=y,arg3=z...)
+
+ year, month, day, hour, minute, second, microsecond:
+ Absolute information (argument is singular); adding or subtracting a
+ relativedelta with absolute information does not perform an arithmetic
+ operation, but rather REPLACES the corresponding value in the
+ original datetime with the value(s) in relativedelta.
+
+ years, months, weeks, days, hours, minutes, seconds, microseconds:
+ Relative information, may be negative (argument is plural); adding
+ or subtracting a relativedelta with relative information performs
+ the corresponding arithmetic operation on the original datetime value
+ with the information in the relativedelta.
+
+ weekday:
+ One of the weekday instances (MO, TU, etc) available in the
+ relativedelta module. These instances may receive a parameter N,
+ specifying the Nth weekday, which could be positive or negative
+ (like MO(+1) or MO(-2)). Not specifying it is the same as specifying
+ +1. You can also use an integer, where 0=MO. This argument is always
+ relative e.g. if the calculated date is already Monday, using MO(1)
+ or MO(-1) won't change the day. To effectively make it absolute, use
+ it in combination with the day argument (e.g. day=1, MO(1) for first
+ Monday of the month).
+
+ leapdays:
+ Will add given days to the date found, if year is a leap
+ year, and the date found is post 28 of february.
+
+ yearday, nlyearday:
+ Set the yearday or the non-leap year day (jump leap days).
+ These are converted to day/month/leapdays information.
+
+ There are relative and absolute forms of the keyword
+ arguments. The plural is relative, and the singular is
+ absolute. For each argument in the order below, the absolute form
+ is applied first (by setting each attribute to that value) and
+ then the relative form (by adding the value to the attribute).
+
+ The order of attributes considered when this relativedelta is
+ added to a datetime is:
+
+ 1. Year
+ 2. Month
+ 3. Day
+ 4. Hours
+ 5. Minutes
+ 6. Seconds
+ 7. Microseconds
+
+ Finally, weekday is applied, using the rule described above.
+
+ For example
+
+ >>> from datetime import datetime
+ >>> from dateutil.relativedelta import relativedelta, MO
+ >>> dt = datetime(2018, 4, 9, 13, 37, 0)
+ >>> delta = relativedelta(hours=25, day=1, weekday=MO(1))
+ >>> dt + delta
+ datetime.datetime(2018, 4, 2, 14, 37)
+
+ First, the day is set to 1 (the first of the month), then 25 hours
+ are added, to get to the 2nd day and 14th hour, finally the
+ weekday is applied, but since the 2nd is already a Monday there is
+ no effect.
+
+ """
+
+ def __init__(self, dt1=None, dt2=None,
+ years=0, months=0, days=0, leapdays=0, weeks=0,
+ hours=0, minutes=0, seconds=0, microseconds=0,
+ year=None, month=None, day=None, weekday=None,
+ yearday=None, nlyearday=None,
+ hour=None, minute=None, second=None, microsecond=None):
+
+ if dt1 and dt2:
+ # datetime is a subclass of date. So both must be date
+ if not (isinstance(dt1, datetime.date) and
+ isinstance(dt2, datetime.date)):
+ raise TypeError("relativedelta only diffs datetime/date")
+
+ # We allow two dates, or two datetimes, so we coerce them to be
+ # of the same type
+ if (isinstance(dt1, datetime.datetime) !=
+ isinstance(dt2, datetime.datetime)):
+ if not isinstance(dt1, datetime.datetime):
+ dt1 = datetime.datetime.fromordinal(dt1.toordinal())
+ elif not isinstance(dt2, datetime.datetime):
+ dt2 = datetime.datetime.fromordinal(dt2.toordinal())
+
+ self.years = 0
+ self.months = 0
+ self.days = 0
+ self.leapdays = 0
+ self.hours = 0
+ self.minutes = 0
+ self.seconds = 0
+ self.microseconds = 0
+ self.year = None
+ self.month = None
+ self.day = None
+ self.weekday = None
+ self.hour = None
+ self.minute = None
+ self.second = None
+ self.microsecond = None
+ self._has_time = 0
+
+ # Get year / month delta between the two
+ months = (dt1.year - dt2.year) * 12 + (dt1.month - dt2.month)
+ self._set_months(months)
+
+ # Remove the year/month delta so the timedelta is just well-defined
+ # time units (seconds, days and microseconds)
+ dtm = self.__radd__(dt2)
+
+ # If we've overshot our target, make an adjustment
+ if dt1 < dt2:
+ compare = operator.gt
+ increment = 1
+ else:
+ compare = operator.lt
+ increment = -1
+
+ while compare(dt1, dtm):
+ months += increment
+ self._set_months(months)
+ dtm = self.__radd__(dt2)
+
+ # Get the timedelta between the "months-adjusted" date and dt1
+ delta = dt1 - dtm
+ self.seconds = delta.seconds + delta.days * 86400
+ self.microseconds = delta.microseconds
+ else:
+ # Check for non-integer values in integer-only quantities
+ if any(x is not None and x != int(x) for x in (years, months)):
+ raise ValueError("Non-integer years and months are "
+ "ambiguous and not currently supported.")
+
+ # Relative information
+ self.years = int(years)
+ self.months = int(months)
+ self.days = days + weeks * 7
+ self.leapdays = leapdays
+ self.hours = hours
+ self.minutes = minutes
+ self.seconds = seconds
+ self.microseconds = microseconds
+
+ # Absolute information
+ self.year = year
+ self.month = month
+ self.day = day
+ self.hour = hour
+ self.minute = minute
+ self.second = second
+ self.microsecond = microsecond
+
+ if any(x is not None and int(x) != x
+ for x in (year, month, day, hour,
+ minute, second, microsecond)):
+ # For now we'll deprecate floats - later it'll be an error.
+ warn("Non-integer value passed as absolute information. " +
+ "This is not a well-defined condition and will raise " +
+ "errors in future versions.", DeprecationWarning)
+
+ if isinstance(weekday, integer_types):
+ self.weekday = weekdays[weekday]
+ else:
+ self.weekday = weekday
+
+ yday = 0
+ if nlyearday:
+ yday = nlyearday
+ elif yearday:
+ yday = yearday
+ if yearday > 59:
+ self.leapdays = -1
+ if yday:
+ ydayidx = [31, 59, 90, 120, 151, 181, 212,
+ 243, 273, 304, 334, 366]
+ for idx, ydays in enumerate(ydayidx):
+ if yday <= ydays:
+ self.month = idx+1
+ if idx == 0:
+ self.day = yday
+ else:
+ self.day = yday-ydayidx[idx-1]
+ break
+ else:
+ raise ValueError("invalid year day (%d)" % yday)
+
+ self._fix()
+
+ def _fix(self):
+ if abs(self.microseconds) > 999999:
+ s = _sign(self.microseconds)
+ div, mod = divmod(self.microseconds * s, 1000000)
+ self.microseconds = mod * s
+ self.seconds += div * s
+ if abs(self.seconds) > 59:
+ s = _sign(self.seconds)
+ div, mod = divmod(self.seconds * s, 60)
+ self.seconds = mod * s
+ self.minutes += div * s
+ if abs(self.minutes) > 59:
+ s = _sign(self.minutes)
+ div, mod = divmod(self.minutes * s, 60)
+ self.minutes = mod * s
+ self.hours += div * s
+ if abs(self.hours) > 23:
+ s = _sign(self.hours)
+ div, mod = divmod(self.hours * s, 24)
+ self.hours = mod * s
+ self.days += div * s
+ if abs(self.months) > 11:
+ s = _sign(self.months)
+ div, mod = divmod(self.months * s, 12)
+ self.months = mod * s
+ self.years += div * s
+ if (self.hours or self.minutes or self.seconds or self.microseconds
+ or self.hour is not None or self.minute is not None or
+ self.second is not None or self.microsecond is not None):
+ self._has_time = 1
+ else:
+ self._has_time = 0
+
+ @property
+ def weeks(self):
+ return int(self.days / 7.0)
+
+ @weeks.setter
+ def weeks(self, value):
+ self.days = self.days - (self.weeks * 7) + value * 7
+
+ def _set_months(self, months):
+ self.months = months
+ if abs(self.months) > 11:
+ s = _sign(self.months)
+ div, mod = divmod(self.months * s, 12)
+ self.months = mod * s
+ self.years = div * s
+ else:
+ self.years = 0
+
+ def normalized(self):
+ """
+ Return a version of this object represented entirely using integer
+ values for the relative attributes.
+
+ >>> relativedelta(days=1.5, hours=2).normalized()
+ relativedelta(days=+1, hours=+14)
+
+ :return:
+ Returns a :class:`dateutil.relativedelta.relativedelta` object.
+ """
+ # Cascade remainders down (rounding each to roughly nearest microsecond)
+ days = int(self.days)
+
+ hours_f = round(self.hours + 24 * (self.days - days), 11)
+ hours = int(hours_f)
+
+ minutes_f = round(self.minutes + 60 * (hours_f - hours), 10)
+ minutes = int(minutes_f)
+
+ seconds_f = round(self.seconds + 60 * (minutes_f - minutes), 8)
+ seconds = int(seconds_f)
+
+ microseconds = round(self.microseconds + 1e6 * (seconds_f - seconds))
+
+ # Constructor carries overflow back up with call to _fix()
+ return self.__class__(years=self.years, months=self.months,
+ days=days, hours=hours, minutes=minutes,
+ seconds=seconds, microseconds=microseconds,
+ leapdays=self.leapdays, year=self.year,
+ month=self.month, day=self.day,
+ weekday=self.weekday, hour=self.hour,
+ minute=self.minute, second=self.second,
+ microsecond=self.microsecond)
+
+ def __add__(self, other):
+ if isinstance(other, relativedelta):
+ return self.__class__(years=other.years + self.years,
+ months=other.months + self.months,
+ days=other.days + self.days,
+ hours=other.hours + self.hours,
+ minutes=other.minutes + self.minutes,
+ seconds=other.seconds + self.seconds,
+ microseconds=(other.microseconds +
+ self.microseconds),
+ leapdays=other.leapdays or self.leapdays,
+ year=(other.year if other.year is not None
+ else self.year),
+ month=(other.month if other.month is not None
+ else self.month),
+ day=(other.day if other.day is not None
+ else self.day),
+ weekday=(other.weekday if other.weekday is not None
+ else self.weekday),
+ hour=(other.hour if other.hour is not None
+ else self.hour),
+ minute=(other.minute if other.minute is not None
+ else self.minute),
+ second=(other.second if other.second is not None
+ else self.second),
+ microsecond=(other.microsecond if other.microsecond
+ is not None else
+ self.microsecond))
+ if isinstance(other, datetime.timedelta):
+ return self.__class__(years=self.years,
+ months=self.months,
+ days=self.days + other.days,
+ hours=self.hours,
+ minutes=self.minutes,
+ seconds=self.seconds + other.seconds,
+ microseconds=self.microseconds + other.microseconds,
+ leapdays=self.leapdays,
+ year=self.year,
+ month=self.month,
+ day=self.day,
+ weekday=self.weekday,
+ hour=self.hour,
+ minute=self.minute,
+ second=self.second,
+ microsecond=self.microsecond)
+ if not isinstance(other, datetime.date):
+ return NotImplemented
+ elif self._has_time and not isinstance(other, datetime.datetime):
+ other = datetime.datetime.fromordinal(other.toordinal())
+ year = (self.year or other.year)+self.years
+ month = self.month or other.month
+ if self.months:
+ assert 1 <= abs(self.months) <= 12
+ month += self.months
+ if month > 12:
+ year += 1
+ month -= 12
+ elif month < 1:
+ year -= 1
+ month += 12
+ day = min(calendar.monthrange(year, month)[1],
+ self.day or other.day)
+ repl = {"year": year, "month": month, "day": day}
+ for attr in ["hour", "minute", "second", "microsecond"]:
+ value = getattr(self, attr)
+ if value is not None:
+ repl[attr] = value
+ days = self.days
+ if self.leapdays and month > 2 and calendar.isleap(year):
+ days += self.leapdays
+ ret = (other.replace(**repl)
+ + datetime.timedelta(days=days,
+ hours=self.hours,
+ minutes=self.minutes,
+ seconds=self.seconds,
+ microseconds=self.microseconds))
+ if self.weekday:
+ weekday, nth = self.weekday.weekday, self.weekday.n or 1
+ jumpdays = (abs(nth) - 1) * 7
+ if nth > 0:
+ jumpdays += (7 - ret.weekday() + weekday) % 7
+ else:
+ jumpdays += (ret.weekday() - weekday) % 7
+ jumpdays *= -1
+ ret += datetime.timedelta(days=jumpdays)
+ return ret
+
+ def __radd__(self, other):
+ return self.__add__(other)
+
+ def __rsub__(self, other):
+ return self.__neg__().__radd__(other)
+
+ def __sub__(self, other):
+ if not isinstance(other, relativedelta):
+ return NotImplemented # In case the other object defines __rsub__
+ return self.__class__(years=self.years - other.years,
+ months=self.months - other.months,
+ days=self.days - other.days,
+ hours=self.hours - other.hours,
+ minutes=self.minutes - other.minutes,
+ seconds=self.seconds - other.seconds,
+ microseconds=self.microseconds - other.microseconds,
+ leapdays=self.leapdays or other.leapdays,
+ year=(self.year if self.year is not None
+ else other.year),
+ month=(self.month if self.month is not None else
+ other.month),
+ day=(self.day if self.day is not None else
+ other.day),
+ weekday=(self.weekday if self.weekday is not None else
+ other.weekday),
+ hour=(self.hour if self.hour is not None else
+ other.hour),
+ minute=(self.minute if self.minute is not None else
+ other.minute),
+ second=(self.second if self.second is not None else
+ other.second),
+ microsecond=(self.microsecond if self.microsecond
+ is not None else
+ other.microsecond))
+
+ def __abs__(self):
+ return self.__class__(years=abs(self.years),
+ months=abs(self.months),
+ days=abs(self.days),
+ hours=abs(self.hours),
+ minutes=abs(self.minutes),
+ seconds=abs(self.seconds),
+ microseconds=abs(self.microseconds),
+ leapdays=self.leapdays,
+ year=self.year,
+ month=self.month,
+ day=self.day,
+ weekday=self.weekday,
+ hour=self.hour,
+ minute=self.minute,
+ second=self.second,
+ microsecond=self.microsecond)
+
+ def __neg__(self):
+ return self.__class__(years=-self.years,
+ months=-self.months,
+ days=-self.days,
+ hours=-self.hours,
+ minutes=-self.minutes,
+ seconds=-self.seconds,
+ microseconds=-self.microseconds,
+ leapdays=self.leapdays,
+ year=self.year,
+ month=self.month,
+ day=self.day,
+ weekday=self.weekday,
+ hour=self.hour,
+ minute=self.minute,
+ second=self.second,
+ microsecond=self.microsecond)
+
+ def __bool__(self):
+ return not (not self.years and
+ not self.months and
+ not self.days and
+ not self.hours and
+ not self.minutes and
+ not self.seconds and
+ not self.microseconds and
+ not self.leapdays and
+ self.year is None and
+ self.month is None and
+ self.day is None and
+ self.weekday is None and
+ self.hour is None and
+ self.minute is None and
+ self.second is None and
+ self.microsecond is None)
+ # Compatibility with Python 2.x
+ __nonzero__ = __bool__
+
+ def __mul__(self, other):
+ try:
+ f = float(other)
+ except TypeError:
+ return NotImplemented
+
+ return self.__class__(years=int(self.years * f),
+ months=int(self.months * f),
+ days=int(self.days * f),
+ hours=int(self.hours * f),
+ minutes=int(self.minutes * f),
+ seconds=int(self.seconds * f),
+ microseconds=int(self.microseconds * f),
+ leapdays=self.leapdays,
+ year=self.year,
+ month=self.month,
+ day=self.day,
+ weekday=self.weekday,
+ hour=self.hour,
+ minute=self.minute,
+ second=self.second,
+ microsecond=self.microsecond)
+
+ __rmul__ = __mul__
+
+ def __eq__(self, other):
+ if not isinstance(other, relativedelta):
+ return NotImplemented
+ if self.weekday or other.weekday:
+ if not self.weekday or not other.weekday:
+ return False
+ if self.weekday.weekday != other.weekday.weekday:
+ return False
+ n1, n2 = self.weekday.n, other.weekday.n
+ if n1 != n2 and not ((not n1 or n1 == 1) and (not n2 or n2 == 1)):
+ return False
+ return (self.years == other.years and
+ self.months == other.months and
+ self.days == other.days and
+ self.hours == other.hours and
+ self.minutes == other.minutes and
+ self.seconds == other.seconds and
+ self.microseconds == other.microseconds and
+ self.leapdays == other.leapdays and
+ self.year == other.year and
+ self.month == other.month and
+ self.day == other.day and
+ self.hour == other.hour and
+ self.minute == other.minute and
+ self.second == other.second and
+ self.microsecond == other.microsecond)
+
+ def __hash__(self):
+ return hash((
+ self.weekday,
+ self.years,
+ self.months,
+ self.days,
+ self.hours,
+ self.minutes,
+ self.seconds,
+ self.microseconds,
+ self.leapdays,
+ self.year,
+ self.month,
+ self.day,
+ self.hour,
+ self.minute,
+ self.second,
+ self.microsecond,
+ ))
+
+ def __ne__(self, other):
+ return not self.__eq__(other)
+
+ def __div__(self, other):
+ try:
+ reciprocal = 1 / float(other)
+ except TypeError:
+ return NotImplemented
+
+ return self.__mul__(reciprocal)
+
+ __truediv__ = __div__
+
+ def __repr__(self):
+ l = []
+ for attr in ["years", "months", "days", "leapdays",
+ "hours", "minutes", "seconds", "microseconds"]:
+ value = getattr(self, attr)
+ if value:
+ l.append("{attr}={value:+g}".format(attr=attr, value=value))
+ for attr in ["year", "month", "day", "weekday",
+ "hour", "minute", "second", "microsecond"]:
+ value = getattr(self, attr)
+ if value is not None:
+ l.append("{attr}={value}".format(attr=attr, value=repr(value)))
+ return "{classname}({attrs})".format(classname=self.__class__.__name__,
+ attrs=", ".join(l))
+
+
+def _sign(x):
+ return int(copysign(1, x))
+
+# vim:ts=4:sw=4:et
diff --git a/falcon/lib/python3.10/site-packages/dateutil/rrule.py b/falcon/lib/python3.10/site-packages/dateutil/rrule.py
new file mode 100644
index 0000000000000000000000000000000000000000..571a0d2bc886a7ea4c06196b2f52e740c2ed6e9f
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/rrule.py
@@ -0,0 +1,1737 @@
+# -*- coding: utf-8 -*-
+"""
+The rrule module offers a small, complete, and very fast, implementation of
+the recurrence rules documented in the
+`iCalendar RFC `_,
+including support for caching of results.
+"""
+import calendar
+import datetime
+import heapq
+import itertools
+import re
+import sys
+from functools import wraps
+# For warning about deprecation of until and count
+from warnings import warn
+
+from six import advance_iterator, integer_types
+
+from six.moves import _thread, range
+
+from ._common import weekday as weekdaybase
+
+try:
+ from math import gcd
+except ImportError:
+ from fractions import gcd
+
+__all__ = ["rrule", "rruleset", "rrulestr",
+ "YEARLY", "MONTHLY", "WEEKLY", "DAILY",
+ "HOURLY", "MINUTELY", "SECONDLY",
+ "MO", "TU", "WE", "TH", "FR", "SA", "SU"]
+
+# Every mask is 7 days longer to handle cross-year weekly periods.
+M366MASK = tuple([1]*31+[2]*29+[3]*31+[4]*30+[5]*31+[6]*30 +
+ [7]*31+[8]*31+[9]*30+[10]*31+[11]*30+[12]*31+[1]*7)
+M365MASK = list(M366MASK)
+M29, M30, M31 = list(range(1, 30)), list(range(1, 31)), list(range(1, 32))
+MDAY366MASK = tuple(M31+M29+M31+M30+M31+M30+M31+M31+M30+M31+M30+M31+M31[:7])
+MDAY365MASK = list(MDAY366MASK)
+M29, M30, M31 = list(range(-29, 0)), list(range(-30, 0)), list(range(-31, 0))
+NMDAY366MASK = tuple(M31+M29+M31+M30+M31+M30+M31+M31+M30+M31+M30+M31+M31[:7])
+NMDAY365MASK = list(NMDAY366MASK)
+M366RANGE = (0, 31, 60, 91, 121, 152, 182, 213, 244, 274, 305, 335, 366)
+M365RANGE = (0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334, 365)
+WDAYMASK = [0, 1, 2, 3, 4, 5, 6]*55
+del M29, M30, M31, M365MASK[59], MDAY365MASK[59], NMDAY365MASK[31]
+MDAY365MASK = tuple(MDAY365MASK)
+M365MASK = tuple(M365MASK)
+
+FREQNAMES = ['YEARLY', 'MONTHLY', 'WEEKLY', 'DAILY', 'HOURLY', 'MINUTELY', 'SECONDLY']
+
+(YEARLY,
+ MONTHLY,
+ WEEKLY,
+ DAILY,
+ HOURLY,
+ MINUTELY,
+ SECONDLY) = list(range(7))
+
+# Imported on demand.
+easter = None
+parser = None
+
+
+class weekday(weekdaybase):
+ """
+ This version of weekday does not allow n = 0.
+ """
+ def __init__(self, wkday, n=None):
+ if n == 0:
+ raise ValueError("Can't create weekday with n==0")
+
+ super(weekday, self).__init__(wkday, n)
+
+
+MO, TU, WE, TH, FR, SA, SU = weekdays = tuple(weekday(x) for x in range(7))
+
+
+def _invalidates_cache(f):
+ """
+ Decorator for rruleset methods which may invalidate the
+ cached length.
+ """
+ @wraps(f)
+ def inner_func(self, *args, **kwargs):
+ rv = f(self, *args, **kwargs)
+ self._invalidate_cache()
+ return rv
+
+ return inner_func
+
+
+class rrulebase(object):
+ def __init__(self, cache=False):
+ if cache:
+ self._cache = []
+ self._cache_lock = _thread.allocate_lock()
+ self._invalidate_cache()
+ else:
+ self._cache = None
+ self._cache_complete = False
+ self._len = None
+
+ def __iter__(self):
+ if self._cache_complete:
+ return iter(self._cache)
+ elif self._cache is None:
+ return self._iter()
+ else:
+ return self._iter_cached()
+
+ def _invalidate_cache(self):
+ if self._cache is not None:
+ self._cache = []
+ self._cache_complete = False
+ self._cache_gen = self._iter()
+
+ if self._cache_lock.locked():
+ self._cache_lock.release()
+
+ self._len = None
+
+ def _iter_cached(self):
+ i = 0
+ gen = self._cache_gen
+ cache = self._cache
+ acquire = self._cache_lock.acquire
+ release = self._cache_lock.release
+ while gen:
+ if i == len(cache):
+ acquire()
+ if self._cache_complete:
+ break
+ try:
+ for j in range(10):
+ cache.append(advance_iterator(gen))
+ except StopIteration:
+ self._cache_gen = gen = None
+ self._cache_complete = True
+ break
+ release()
+ yield cache[i]
+ i += 1
+ while i < self._len:
+ yield cache[i]
+ i += 1
+
+ def __getitem__(self, item):
+ if self._cache_complete:
+ return self._cache[item]
+ elif isinstance(item, slice):
+ if item.step and item.step < 0:
+ return list(iter(self))[item]
+ else:
+ return list(itertools.islice(self,
+ item.start or 0,
+ item.stop or sys.maxsize,
+ item.step or 1))
+ elif item >= 0:
+ gen = iter(self)
+ try:
+ for i in range(item+1):
+ res = advance_iterator(gen)
+ except StopIteration:
+ raise IndexError
+ return res
+ else:
+ return list(iter(self))[item]
+
+ def __contains__(self, item):
+ if self._cache_complete:
+ return item in self._cache
+ else:
+ for i in self:
+ if i == item:
+ return True
+ elif i > item:
+ return False
+ return False
+
+ # __len__() introduces a large performance penalty.
+ def count(self):
+ """ Returns the number of recurrences in this set. It will have go
+ through the whole recurrence, if this hasn't been done before. """
+ if self._len is None:
+ for x in self:
+ pass
+ return self._len
+
+ def before(self, dt, inc=False):
+ """ Returns the last recurrence before the given datetime instance. The
+ inc keyword defines what happens if dt is an occurrence. With
+ inc=True, if dt itself is an occurrence, it will be returned. """
+ if self._cache_complete:
+ gen = self._cache
+ else:
+ gen = self
+ last = None
+ if inc:
+ for i in gen:
+ if i > dt:
+ break
+ last = i
+ else:
+ for i in gen:
+ if i >= dt:
+ break
+ last = i
+ return last
+
+ def after(self, dt, inc=False):
+ """ Returns the first recurrence after the given datetime instance. The
+ inc keyword defines what happens if dt is an occurrence. With
+ inc=True, if dt itself is an occurrence, it will be returned. """
+ if self._cache_complete:
+ gen = self._cache
+ else:
+ gen = self
+ if inc:
+ for i in gen:
+ if i >= dt:
+ return i
+ else:
+ for i in gen:
+ if i > dt:
+ return i
+ return None
+
+ def xafter(self, dt, count=None, inc=False):
+ """
+ Generator which yields up to `count` recurrences after the given
+ datetime instance, equivalent to `after`.
+
+ :param dt:
+ The datetime at which to start generating recurrences.
+
+ :param count:
+ The maximum number of recurrences to generate. If `None` (default),
+ dates are generated until the recurrence rule is exhausted.
+
+ :param inc:
+ If `dt` is an instance of the rule and `inc` is `True`, it is
+ included in the output.
+
+ :yields: Yields a sequence of `datetime` objects.
+ """
+
+ if self._cache_complete:
+ gen = self._cache
+ else:
+ gen = self
+
+ # Select the comparison function
+ if inc:
+ comp = lambda dc, dtc: dc >= dtc
+ else:
+ comp = lambda dc, dtc: dc > dtc
+
+ # Generate dates
+ n = 0
+ for d in gen:
+ if comp(d, dt):
+ if count is not None:
+ n += 1
+ if n > count:
+ break
+
+ yield d
+
+ def between(self, after, before, inc=False, count=1):
+ """ Returns all the occurrences of the rrule between after and before.
+ The inc keyword defines what happens if after and/or before are
+ themselves occurrences. With inc=True, they will be included in the
+ list, if they are found in the recurrence set. """
+ if self._cache_complete:
+ gen = self._cache
+ else:
+ gen = self
+ started = False
+ l = []
+ if inc:
+ for i in gen:
+ if i > before:
+ break
+ elif not started:
+ if i >= after:
+ started = True
+ l.append(i)
+ else:
+ l.append(i)
+ else:
+ for i in gen:
+ if i >= before:
+ break
+ elif not started:
+ if i > after:
+ started = True
+ l.append(i)
+ else:
+ l.append(i)
+ return l
+
+
+class rrule(rrulebase):
+ """
+ That's the base of the rrule operation. It accepts all the keywords
+ defined in the RFC as its constructor parameters (except byday,
+ which was renamed to byweekday) and more. The constructor prototype is::
+
+ rrule(freq)
+
+ Where freq must be one of YEARLY, MONTHLY, WEEKLY, DAILY, HOURLY, MINUTELY,
+ or SECONDLY.
+
+ .. note::
+ Per RFC section 3.3.10, recurrence instances falling on invalid dates
+ and times are ignored rather than coerced:
+
+ Recurrence rules may generate recurrence instances with an invalid
+ date (e.g., February 30) or nonexistent local time (e.g., 1:30 AM
+ on a day where the local time is moved forward by an hour at 1:00
+ AM). Such recurrence instances MUST be ignored and MUST NOT be
+ counted as part of the recurrence set.
+
+ This can lead to possibly surprising behavior when, for example, the
+ start date occurs at the end of the month:
+
+ >>> from dateutil.rrule import rrule, MONTHLY
+ >>> from datetime import datetime
+ >>> start_date = datetime(2014, 12, 31)
+ >>> list(rrule(freq=MONTHLY, count=4, dtstart=start_date))
+ ... # doctest: +NORMALIZE_WHITESPACE
+ [datetime.datetime(2014, 12, 31, 0, 0),
+ datetime.datetime(2015, 1, 31, 0, 0),
+ datetime.datetime(2015, 3, 31, 0, 0),
+ datetime.datetime(2015, 5, 31, 0, 0)]
+
+ Additionally, it supports the following keyword arguments:
+
+ :param dtstart:
+ The recurrence start. Besides being the base for the recurrence,
+ missing parameters in the final recurrence instances will also be
+ extracted from this date. If not given, datetime.now() will be used
+ instead.
+ :param interval:
+ The interval between each freq iteration. For example, when using
+ YEARLY, an interval of 2 means once every two years, but with HOURLY,
+ it means once every two hours. The default interval is 1.
+ :param wkst:
+ The week start day. Must be one of the MO, TU, WE constants, or an
+ integer, specifying the first day of the week. This will affect
+ recurrences based on weekly periods. The default week start is got
+ from calendar.firstweekday(), and may be modified by
+ calendar.setfirstweekday().
+ :param count:
+ If given, this determines how many occurrences will be generated.
+
+ .. note::
+ As of version 2.5.0, the use of the keyword ``until`` in conjunction
+ with ``count`` is deprecated, to make sure ``dateutil`` is fully
+ compliant with `RFC-5545 Sec. 3.3.10 `_. Therefore, ``until`` and ``count``
+ **must not** occur in the same call to ``rrule``.
+ :param until:
+ If given, this must be a datetime instance specifying the upper-bound
+ limit of the recurrence. The last recurrence in the rule is the greatest
+ datetime that is less than or equal to the value specified in the
+ ``until`` parameter.
+
+ .. note::
+ As of version 2.5.0, the use of the keyword ``until`` in conjunction
+ with ``count`` is deprecated, to make sure ``dateutil`` is fully
+ compliant with `RFC-5545 Sec. 3.3.10 `_. Therefore, ``until`` and ``count``
+ **must not** occur in the same call to ``rrule``.
+ :param bysetpos:
+ If given, it must be either an integer, or a sequence of integers,
+ positive or negative. Each given integer will specify an occurrence
+ number, corresponding to the nth occurrence of the rule inside the
+ frequency period. For example, a bysetpos of -1 if combined with a
+ MONTHLY frequency, and a byweekday of (MO, TU, WE, TH, FR), will
+ result in the last work day of every month.
+ :param bymonth:
+ If given, it must be either an integer, or a sequence of integers,
+ meaning the months to apply the recurrence to.
+ :param bymonthday:
+ If given, it must be either an integer, or a sequence of integers,
+ meaning the month days to apply the recurrence to.
+ :param byyearday:
+ If given, it must be either an integer, or a sequence of integers,
+ meaning the year days to apply the recurrence to.
+ :param byeaster:
+ If given, it must be either an integer, or a sequence of integers,
+ positive or negative. Each integer will define an offset from the
+ Easter Sunday. Passing the offset 0 to byeaster will yield the Easter
+ Sunday itself. This is an extension to the RFC specification.
+ :param byweekno:
+ If given, it must be either an integer, or a sequence of integers,
+ meaning the week numbers to apply the recurrence to. Week numbers
+ have the meaning described in ISO8601, that is, the first week of
+ the year is that containing at least four days of the new year.
+ :param byweekday:
+ If given, it must be either an integer (0 == MO), a sequence of
+ integers, one of the weekday constants (MO, TU, etc), or a sequence
+ of these constants. When given, these variables will define the
+ weekdays where the recurrence will be applied. It's also possible to
+ use an argument n for the weekday instances, which will mean the nth
+ occurrence of this weekday in the period. For example, with MONTHLY,
+ or with YEARLY and BYMONTH, using FR(+1) in byweekday will specify the
+ first friday of the month where the recurrence happens. Notice that in
+ the RFC documentation, this is specified as BYDAY, but was renamed to
+ avoid the ambiguity of that keyword.
+ :param byhour:
+ If given, it must be either an integer, or a sequence of integers,
+ meaning the hours to apply the recurrence to.
+ :param byminute:
+ If given, it must be either an integer, or a sequence of integers,
+ meaning the minutes to apply the recurrence to.
+ :param bysecond:
+ If given, it must be either an integer, or a sequence of integers,
+ meaning the seconds to apply the recurrence to.
+ :param cache:
+ If given, it must be a boolean value specifying to enable or disable
+ caching of results. If you will use the same rrule instance multiple
+ times, enabling caching will improve the performance considerably.
+ """
+ def __init__(self, freq, dtstart=None,
+ interval=1, wkst=None, count=None, until=None, bysetpos=None,
+ bymonth=None, bymonthday=None, byyearday=None, byeaster=None,
+ byweekno=None, byweekday=None,
+ byhour=None, byminute=None, bysecond=None,
+ cache=False):
+ super(rrule, self).__init__(cache)
+ global easter
+ if not dtstart:
+ if until and until.tzinfo:
+ dtstart = datetime.datetime.now(tz=until.tzinfo).replace(microsecond=0)
+ else:
+ dtstart = datetime.datetime.now().replace(microsecond=0)
+ elif not isinstance(dtstart, datetime.datetime):
+ dtstart = datetime.datetime.fromordinal(dtstart.toordinal())
+ else:
+ dtstart = dtstart.replace(microsecond=0)
+ self._dtstart = dtstart
+ self._tzinfo = dtstart.tzinfo
+ self._freq = freq
+ self._interval = interval
+ self._count = count
+
+ # Cache the original byxxx rules, if they are provided, as the _byxxx
+ # attributes do not necessarily map to the inputs, and this can be
+ # a problem in generating the strings. Only store things if they've
+ # been supplied (the string retrieval will just use .get())
+ self._original_rule = {}
+
+ if until and not isinstance(until, datetime.datetime):
+ until = datetime.datetime.fromordinal(until.toordinal())
+ self._until = until
+
+ if self._dtstart and self._until:
+ if (self._dtstart.tzinfo is not None) != (self._until.tzinfo is not None):
+ # According to RFC5545 Section 3.3.10:
+ # https://tools.ietf.org/html/rfc5545#section-3.3.10
+ #
+ # > If the "DTSTART" property is specified as a date with UTC
+ # > time or a date with local time and time zone reference,
+ # > then the UNTIL rule part MUST be specified as a date with
+ # > UTC time.
+ raise ValueError(
+ 'RRULE UNTIL values must be specified in UTC when DTSTART '
+ 'is timezone-aware'
+ )
+
+ if count is not None and until:
+ warn("Using both 'count' and 'until' is inconsistent with RFC 5545"
+ " and has been deprecated in dateutil. Future versions will "
+ "raise an error.", DeprecationWarning)
+
+ if wkst is None:
+ self._wkst = calendar.firstweekday()
+ elif isinstance(wkst, integer_types):
+ self._wkst = wkst
+ else:
+ self._wkst = wkst.weekday
+
+ if bysetpos is None:
+ self._bysetpos = None
+ elif isinstance(bysetpos, integer_types):
+ if bysetpos == 0 or not (-366 <= bysetpos <= 366):
+ raise ValueError("bysetpos must be between 1 and 366, "
+ "or between -366 and -1")
+ self._bysetpos = (bysetpos,)
+ else:
+ self._bysetpos = tuple(bysetpos)
+ for pos in self._bysetpos:
+ if pos == 0 or not (-366 <= pos <= 366):
+ raise ValueError("bysetpos must be between 1 and 366, "
+ "or between -366 and -1")
+
+ if self._bysetpos:
+ self._original_rule['bysetpos'] = self._bysetpos
+
+ if (byweekno is None and byyearday is None and bymonthday is None and
+ byweekday is None and byeaster is None):
+ if freq == YEARLY:
+ if bymonth is None:
+ bymonth = dtstart.month
+ self._original_rule['bymonth'] = None
+ bymonthday = dtstart.day
+ self._original_rule['bymonthday'] = None
+ elif freq == MONTHLY:
+ bymonthday = dtstart.day
+ self._original_rule['bymonthday'] = None
+ elif freq == WEEKLY:
+ byweekday = dtstart.weekday()
+ self._original_rule['byweekday'] = None
+
+ # bymonth
+ if bymonth is None:
+ self._bymonth = None
+ else:
+ if isinstance(bymonth, integer_types):
+ bymonth = (bymonth,)
+
+ self._bymonth = tuple(sorted(set(bymonth)))
+
+ if 'bymonth' not in self._original_rule:
+ self._original_rule['bymonth'] = self._bymonth
+
+ # byyearday
+ if byyearday is None:
+ self._byyearday = None
+ else:
+ if isinstance(byyearday, integer_types):
+ byyearday = (byyearday,)
+
+ self._byyearday = tuple(sorted(set(byyearday)))
+ self._original_rule['byyearday'] = self._byyearday
+
+ # byeaster
+ if byeaster is not None:
+ if not easter:
+ from dateutil import easter
+ if isinstance(byeaster, integer_types):
+ self._byeaster = (byeaster,)
+ else:
+ self._byeaster = tuple(sorted(byeaster))
+
+ self._original_rule['byeaster'] = self._byeaster
+ else:
+ self._byeaster = None
+
+ # bymonthday
+ if bymonthday is None:
+ self._bymonthday = ()
+ self._bynmonthday = ()
+ else:
+ if isinstance(bymonthday, integer_types):
+ bymonthday = (bymonthday,)
+
+ bymonthday = set(bymonthday) # Ensure it's unique
+
+ self._bymonthday = tuple(sorted(x for x in bymonthday if x > 0))
+ self._bynmonthday = tuple(sorted(x for x in bymonthday if x < 0))
+
+ # Storing positive numbers first, then negative numbers
+ if 'bymonthday' not in self._original_rule:
+ self._original_rule['bymonthday'] = tuple(
+ itertools.chain(self._bymonthday, self._bynmonthday))
+
+ # byweekno
+ if byweekno is None:
+ self._byweekno = None
+ else:
+ if isinstance(byweekno, integer_types):
+ byweekno = (byweekno,)
+
+ self._byweekno = tuple(sorted(set(byweekno)))
+
+ self._original_rule['byweekno'] = self._byweekno
+
+ # byweekday / bynweekday
+ if byweekday is None:
+ self._byweekday = None
+ self._bynweekday = None
+ else:
+ # If it's one of the valid non-sequence types, convert to a
+ # single-element sequence before the iterator that builds the
+ # byweekday set.
+ if isinstance(byweekday, integer_types) or hasattr(byweekday, "n"):
+ byweekday = (byweekday,)
+
+ self._byweekday = set()
+ self._bynweekday = set()
+ for wday in byweekday:
+ if isinstance(wday, integer_types):
+ self._byweekday.add(wday)
+ elif not wday.n or freq > MONTHLY:
+ self._byweekday.add(wday.weekday)
+ else:
+ self._bynweekday.add((wday.weekday, wday.n))
+
+ if not self._byweekday:
+ self._byweekday = None
+ elif not self._bynweekday:
+ self._bynweekday = None
+
+ if self._byweekday is not None:
+ self._byweekday = tuple(sorted(self._byweekday))
+ orig_byweekday = [weekday(x) for x in self._byweekday]
+ else:
+ orig_byweekday = ()
+
+ if self._bynweekday is not None:
+ self._bynweekday = tuple(sorted(self._bynweekday))
+ orig_bynweekday = [weekday(*x) for x in self._bynweekday]
+ else:
+ orig_bynweekday = ()
+
+ if 'byweekday' not in self._original_rule:
+ self._original_rule['byweekday'] = tuple(itertools.chain(
+ orig_byweekday, orig_bynweekday))
+
+ # byhour
+ if byhour is None:
+ if freq < HOURLY:
+ self._byhour = {dtstart.hour}
+ else:
+ self._byhour = None
+ else:
+ if isinstance(byhour, integer_types):
+ byhour = (byhour,)
+
+ if freq == HOURLY:
+ self._byhour = self.__construct_byset(start=dtstart.hour,
+ byxxx=byhour,
+ base=24)
+ else:
+ self._byhour = set(byhour)
+
+ self._byhour = tuple(sorted(self._byhour))
+ self._original_rule['byhour'] = self._byhour
+
+ # byminute
+ if byminute is None:
+ if freq < MINUTELY:
+ self._byminute = {dtstart.minute}
+ else:
+ self._byminute = None
+ else:
+ if isinstance(byminute, integer_types):
+ byminute = (byminute,)
+
+ if freq == MINUTELY:
+ self._byminute = self.__construct_byset(start=dtstart.minute,
+ byxxx=byminute,
+ base=60)
+ else:
+ self._byminute = set(byminute)
+
+ self._byminute = tuple(sorted(self._byminute))
+ self._original_rule['byminute'] = self._byminute
+
+ # bysecond
+ if bysecond is None:
+ if freq < SECONDLY:
+ self._bysecond = ((dtstart.second,))
+ else:
+ self._bysecond = None
+ else:
+ if isinstance(bysecond, integer_types):
+ bysecond = (bysecond,)
+
+ self._bysecond = set(bysecond)
+
+ if freq == SECONDLY:
+ self._bysecond = self.__construct_byset(start=dtstart.second,
+ byxxx=bysecond,
+ base=60)
+ else:
+ self._bysecond = set(bysecond)
+
+ self._bysecond = tuple(sorted(self._bysecond))
+ self._original_rule['bysecond'] = self._bysecond
+
+ if self._freq >= HOURLY:
+ self._timeset = None
+ else:
+ self._timeset = []
+ for hour in self._byhour:
+ for minute in self._byminute:
+ for second in self._bysecond:
+ self._timeset.append(
+ datetime.time(hour, minute, second,
+ tzinfo=self._tzinfo))
+ self._timeset.sort()
+ self._timeset = tuple(self._timeset)
+
+ def __str__(self):
+ """
+ Output a string that would generate this RRULE if passed to rrulestr.
+ This is mostly compatible with RFC5545, except for the
+ dateutil-specific extension BYEASTER.
+ """
+
+ output = []
+ h, m, s = [None] * 3
+ if self._dtstart:
+ output.append(self._dtstart.strftime('DTSTART:%Y%m%dT%H%M%S'))
+ h, m, s = self._dtstart.timetuple()[3:6]
+
+ parts = ['FREQ=' + FREQNAMES[self._freq]]
+ if self._interval != 1:
+ parts.append('INTERVAL=' + str(self._interval))
+
+ if self._wkst:
+ parts.append('WKST=' + repr(weekday(self._wkst))[0:2])
+
+ if self._count is not None:
+ parts.append('COUNT=' + str(self._count))
+
+ if self._until:
+ parts.append(self._until.strftime('UNTIL=%Y%m%dT%H%M%S'))
+
+ if self._original_rule.get('byweekday') is not None:
+ # The str() method on weekday objects doesn't generate
+ # RFC5545-compliant strings, so we should modify that.
+ original_rule = dict(self._original_rule)
+ wday_strings = []
+ for wday in original_rule['byweekday']:
+ if wday.n:
+ wday_strings.append('{n:+d}{wday}'.format(
+ n=wday.n,
+ wday=repr(wday)[0:2]))
+ else:
+ wday_strings.append(repr(wday))
+
+ original_rule['byweekday'] = wday_strings
+ else:
+ original_rule = self._original_rule
+
+ partfmt = '{name}={vals}'
+ for name, key in [('BYSETPOS', 'bysetpos'),
+ ('BYMONTH', 'bymonth'),
+ ('BYMONTHDAY', 'bymonthday'),
+ ('BYYEARDAY', 'byyearday'),
+ ('BYWEEKNO', 'byweekno'),
+ ('BYDAY', 'byweekday'),
+ ('BYHOUR', 'byhour'),
+ ('BYMINUTE', 'byminute'),
+ ('BYSECOND', 'bysecond'),
+ ('BYEASTER', 'byeaster')]:
+ value = original_rule.get(key)
+ if value:
+ parts.append(partfmt.format(name=name, vals=(','.join(str(v)
+ for v in value))))
+
+ output.append('RRULE:' + ';'.join(parts))
+ return '\n'.join(output)
+
+ def replace(self, **kwargs):
+ """Return new rrule with same attributes except for those attributes given new
+ values by whichever keyword arguments are specified."""
+ new_kwargs = {"interval": self._interval,
+ "count": self._count,
+ "dtstart": self._dtstart,
+ "freq": self._freq,
+ "until": self._until,
+ "wkst": self._wkst,
+ "cache": False if self._cache is None else True }
+ new_kwargs.update(self._original_rule)
+ new_kwargs.update(kwargs)
+ return rrule(**new_kwargs)
+
+ def _iter(self):
+ year, month, day, hour, minute, second, weekday, yearday, _ = \
+ self._dtstart.timetuple()
+
+ # Some local variables to speed things up a bit
+ freq = self._freq
+ interval = self._interval
+ wkst = self._wkst
+ until = self._until
+ bymonth = self._bymonth
+ byweekno = self._byweekno
+ byyearday = self._byyearday
+ byweekday = self._byweekday
+ byeaster = self._byeaster
+ bymonthday = self._bymonthday
+ bynmonthday = self._bynmonthday
+ bysetpos = self._bysetpos
+ byhour = self._byhour
+ byminute = self._byminute
+ bysecond = self._bysecond
+
+ ii = _iterinfo(self)
+ ii.rebuild(year, month)
+
+ getdayset = {YEARLY: ii.ydayset,
+ MONTHLY: ii.mdayset,
+ WEEKLY: ii.wdayset,
+ DAILY: ii.ddayset,
+ HOURLY: ii.ddayset,
+ MINUTELY: ii.ddayset,
+ SECONDLY: ii.ddayset}[freq]
+
+ if freq < HOURLY:
+ timeset = self._timeset
+ else:
+ gettimeset = {HOURLY: ii.htimeset,
+ MINUTELY: ii.mtimeset,
+ SECONDLY: ii.stimeset}[freq]
+ if ((freq >= HOURLY and
+ self._byhour and hour not in self._byhour) or
+ (freq >= MINUTELY and
+ self._byminute and minute not in self._byminute) or
+ (freq >= SECONDLY and
+ self._bysecond and second not in self._bysecond)):
+ timeset = ()
+ else:
+ timeset = gettimeset(hour, minute, second)
+
+ total = 0
+ count = self._count
+ while True:
+ # Get dayset with the right frequency
+ dayset, start, end = getdayset(year, month, day)
+
+ # Do the "hard" work ;-)
+ filtered = False
+ for i in dayset[start:end]:
+ if ((bymonth and ii.mmask[i] not in bymonth) or
+ (byweekno and not ii.wnomask[i]) or
+ (byweekday and ii.wdaymask[i] not in byweekday) or
+ (ii.nwdaymask and not ii.nwdaymask[i]) or
+ (byeaster and not ii.eastermask[i]) or
+ ((bymonthday or bynmonthday) and
+ ii.mdaymask[i] not in bymonthday and
+ ii.nmdaymask[i] not in bynmonthday) or
+ (byyearday and
+ ((i < ii.yearlen and i+1 not in byyearday and
+ -ii.yearlen+i not in byyearday) or
+ (i >= ii.yearlen and i+1-ii.yearlen not in byyearday and
+ -ii.nextyearlen+i-ii.yearlen not in byyearday)))):
+ dayset[i] = None
+ filtered = True
+
+ # Output results
+ if bysetpos and timeset:
+ poslist = []
+ for pos in bysetpos:
+ if pos < 0:
+ daypos, timepos = divmod(pos, len(timeset))
+ else:
+ daypos, timepos = divmod(pos-1, len(timeset))
+ try:
+ i = [x for x in dayset[start:end]
+ if x is not None][daypos]
+ time = timeset[timepos]
+ except IndexError:
+ pass
+ else:
+ date = datetime.date.fromordinal(ii.yearordinal+i)
+ res = datetime.datetime.combine(date, time)
+ if res not in poslist:
+ poslist.append(res)
+ poslist.sort()
+ for res in poslist:
+ if until and res > until:
+ self._len = total
+ return
+ elif res >= self._dtstart:
+ if count is not None:
+ count -= 1
+ if count < 0:
+ self._len = total
+ return
+ total += 1
+ yield res
+ else:
+ for i in dayset[start:end]:
+ if i is not None:
+ date = datetime.date.fromordinal(ii.yearordinal + i)
+ for time in timeset:
+ res = datetime.datetime.combine(date, time)
+ if until and res > until:
+ self._len = total
+ return
+ elif res >= self._dtstart:
+ if count is not None:
+ count -= 1
+ if count < 0:
+ self._len = total
+ return
+
+ total += 1
+ yield res
+
+ # Handle frequency and interval
+ fixday = False
+ if freq == YEARLY:
+ year += interval
+ if year > datetime.MAXYEAR:
+ self._len = total
+ return
+ ii.rebuild(year, month)
+ elif freq == MONTHLY:
+ month += interval
+ if month > 12:
+ div, mod = divmod(month, 12)
+ month = mod
+ year += div
+ if month == 0:
+ month = 12
+ year -= 1
+ if year > datetime.MAXYEAR:
+ self._len = total
+ return
+ ii.rebuild(year, month)
+ elif freq == WEEKLY:
+ if wkst > weekday:
+ day += -(weekday+1+(6-wkst))+self._interval*7
+ else:
+ day += -(weekday-wkst)+self._interval*7
+ weekday = wkst
+ fixday = True
+ elif freq == DAILY:
+ day += interval
+ fixday = True
+ elif freq == HOURLY:
+ if filtered:
+ # Jump to one iteration before next day
+ hour += ((23-hour)//interval)*interval
+
+ if byhour:
+ ndays, hour = self.__mod_distance(value=hour,
+ byxxx=self._byhour,
+ base=24)
+ else:
+ ndays, hour = divmod(hour+interval, 24)
+
+ if ndays:
+ day += ndays
+ fixday = True
+
+ timeset = gettimeset(hour, minute, second)
+ elif freq == MINUTELY:
+ if filtered:
+ # Jump to one iteration before next day
+ minute += ((1439-(hour*60+minute))//interval)*interval
+
+ valid = False
+ rep_rate = (24*60)
+ for j in range(rep_rate // gcd(interval, rep_rate)):
+ if byminute:
+ nhours, minute = \
+ self.__mod_distance(value=minute,
+ byxxx=self._byminute,
+ base=60)
+ else:
+ nhours, minute = divmod(minute+interval, 60)
+
+ div, hour = divmod(hour+nhours, 24)
+ if div:
+ day += div
+ fixday = True
+ filtered = False
+
+ if not byhour or hour in byhour:
+ valid = True
+ break
+
+ if not valid:
+ raise ValueError('Invalid combination of interval and ' +
+ 'byhour resulting in empty rule.')
+
+ timeset = gettimeset(hour, minute, second)
+ elif freq == SECONDLY:
+ if filtered:
+ # Jump to one iteration before next day
+ second += (((86399 - (hour * 3600 + minute * 60 + second))
+ // interval) * interval)
+
+ rep_rate = (24 * 3600)
+ valid = False
+ for j in range(0, rep_rate // gcd(interval, rep_rate)):
+ if bysecond:
+ nminutes, second = \
+ self.__mod_distance(value=second,
+ byxxx=self._bysecond,
+ base=60)
+ else:
+ nminutes, second = divmod(second+interval, 60)
+
+ div, minute = divmod(minute+nminutes, 60)
+ if div:
+ hour += div
+ div, hour = divmod(hour, 24)
+ if div:
+ day += div
+ fixday = True
+
+ if ((not byhour or hour in byhour) and
+ (not byminute or minute in byminute) and
+ (not bysecond or second in bysecond)):
+ valid = True
+ break
+
+ if not valid:
+ raise ValueError('Invalid combination of interval, ' +
+ 'byhour and byminute resulting in empty' +
+ ' rule.')
+
+ timeset = gettimeset(hour, minute, second)
+
+ if fixday and day > 28:
+ daysinmonth = calendar.monthrange(year, month)[1]
+ if day > daysinmonth:
+ while day > daysinmonth:
+ day -= daysinmonth
+ month += 1
+ if month == 13:
+ month = 1
+ year += 1
+ if year > datetime.MAXYEAR:
+ self._len = total
+ return
+ daysinmonth = calendar.monthrange(year, month)[1]
+ ii.rebuild(year, month)
+
+ def __construct_byset(self, start, byxxx, base):
+ """
+ If a `BYXXX` sequence is passed to the constructor at the same level as
+ `FREQ` (e.g. `FREQ=HOURLY,BYHOUR={2,4,7},INTERVAL=3`), there are some
+ specifications which cannot be reached given some starting conditions.
+
+ This occurs whenever the interval is not coprime with the base of a
+ given unit and the difference between the starting position and the
+ ending position is not coprime with the greatest common denominator
+ between the interval and the base. For example, with a FREQ of hourly
+ starting at 17:00 and an interval of 4, the only valid values for
+ BYHOUR would be {21, 1, 5, 9, 13, 17}, because 4 and 24 are not
+ coprime.
+
+ :param start:
+ Specifies the starting position.
+ :param byxxx:
+ An iterable containing the list of allowed values.
+ :param base:
+ The largest allowable value for the specified frequency (e.g.
+ 24 hours, 60 minutes).
+
+ This does not preserve the type of the iterable, returning a set, since
+ the values should be unique and the order is irrelevant, this will
+ speed up later lookups.
+
+ In the event of an empty set, raises a :exception:`ValueError`, as this
+ results in an empty rrule.
+ """
+
+ cset = set()
+
+ # Support a single byxxx value.
+ if isinstance(byxxx, integer_types):
+ byxxx = (byxxx, )
+
+ for num in byxxx:
+ i_gcd = gcd(self._interval, base)
+ # Use divmod rather than % because we need to wrap negative nums.
+ if i_gcd == 1 or divmod(num - start, i_gcd)[1] == 0:
+ cset.add(num)
+
+ if len(cset) == 0:
+ raise ValueError("Invalid rrule byxxx generates an empty set.")
+
+ return cset
+
+ def __mod_distance(self, value, byxxx, base):
+ """
+ Calculates the next value in a sequence where the `FREQ` parameter is
+ specified along with a `BYXXX` parameter at the same "level"
+ (e.g. `HOURLY` specified with `BYHOUR`).
+
+ :param value:
+ The old value of the component.
+ :param byxxx:
+ The `BYXXX` set, which should have been generated by
+ `rrule._construct_byset`, or something else which checks that a
+ valid rule is present.
+ :param base:
+ The largest allowable value for the specified frequency (e.g.
+ 24 hours, 60 minutes).
+
+ If a valid value is not found after `base` iterations (the maximum
+ number before the sequence would start to repeat), this raises a
+ :exception:`ValueError`, as no valid values were found.
+
+ This returns a tuple of `divmod(n*interval, base)`, where `n` is the
+ smallest number of `interval` repetitions until the next specified
+ value in `byxxx` is found.
+ """
+ accumulator = 0
+ for ii in range(1, base + 1):
+ # Using divmod() over % to account for negative intervals
+ div, value = divmod(value + self._interval, base)
+ accumulator += div
+ if value in byxxx:
+ return (accumulator, value)
+
+
+class _iterinfo(object):
+ __slots__ = ["rrule", "lastyear", "lastmonth",
+ "yearlen", "nextyearlen", "yearordinal", "yearweekday",
+ "mmask", "mrange", "mdaymask", "nmdaymask",
+ "wdaymask", "wnomask", "nwdaymask", "eastermask"]
+
+ def __init__(self, rrule):
+ for attr in self.__slots__:
+ setattr(self, attr, None)
+ self.rrule = rrule
+
+ def rebuild(self, year, month):
+ # Every mask is 7 days longer to handle cross-year weekly periods.
+ rr = self.rrule
+ if year != self.lastyear:
+ self.yearlen = 365 + calendar.isleap(year)
+ self.nextyearlen = 365 + calendar.isleap(year + 1)
+ firstyday = datetime.date(year, 1, 1)
+ self.yearordinal = firstyday.toordinal()
+ self.yearweekday = firstyday.weekday()
+
+ wday = datetime.date(year, 1, 1).weekday()
+ if self.yearlen == 365:
+ self.mmask = M365MASK
+ self.mdaymask = MDAY365MASK
+ self.nmdaymask = NMDAY365MASK
+ self.wdaymask = WDAYMASK[wday:]
+ self.mrange = M365RANGE
+ else:
+ self.mmask = M366MASK
+ self.mdaymask = MDAY366MASK
+ self.nmdaymask = NMDAY366MASK
+ self.wdaymask = WDAYMASK[wday:]
+ self.mrange = M366RANGE
+
+ if not rr._byweekno:
+ self.wnomask = None
+ else:
+ self.wnomask = [0]*(self.yearlen+7)
+ # no1wkst = firstwkst = self.wdaymask.index(rr._wkst)
+ no1wkst = firstwkst = (7-self.yearweekday+rr._wkst) % 7
+ if no1wkst >= 4:
+ no1wkst = 0
+ # Number of days in the year, plus the days we got
+ # from last year.
+ wyearlen = self.yearlen+(self.yearweekday-rr._wkst) % 7
+ else:
+ # Number of days in the year, minus the days we
+ # left in last year.
+ wyearlen = self.yearlen-no1wkst
+ div, mod = divmod(wyearlen, 7)
+ numweeks = div+mod//4
+ for n in rr._byweekno:
+ if n < 0:
+ n += numweeks+1
+ if not (0 < n <= numweeks):
+ continue
+ if n > 1:
+ i = no1wkst+(n-1)*7
+ if no1wkst != firstwkst:
+ i -= 7-firstwkst
+ else:
+ i = no1wkst
+ for j in range(7):
+ self.wnomask[i] = 1
+ i += 1
+ if self.wdaymask[i] == rr._wkst:
+ break
+ if 1 in rr._byweekno:
+ # Check week number 1 of next year as well
+ # TODO: Check -numweeks for next year.
+ i = no1wkst+numweeks*7
+ if no1wkst != firstwkst:
+ i -= 7-firstwkst
+ if i < self.yearlen:
+ # If week starts in next year, we
+ # don't care about it.
+ for j in range(7):
+ self.wnomask[i] = 1
+ i += 1
+ if self.wdaymask[i] == rr._wkst:
+ break
+ if no1wkst:
+ # Check last week number of last year as
+ # well. If no1wkst is 0, either the year
+ # started on week start, or week number 1
+ # got days from last year, so there are no
+ # days from last year's last week number in
+ # this year.
+ if -1 not in rr._byweekno:
+ lyearweekday = datetime.date(year-1, 1, 1).weekday()
+ lno1wkst = (7-lyearweekday+rr._wkst) % 7
+ lyearlen = 365+calendar.isleap(year-1)
+ if lno1wkst >= 4:
+ lno1wkst = 0
+ lnumweeks = 52+(lyearlen +
+ (lyearweekday-rr._wkst) % 7) % 7//4
+ else:
+ lnumweeks = 52+(self.yearlen-no1wkst) % 7//4
+ else:
+ lnumweeks = -1
+ if lnumweeks in rr._byweekno:
+ for i in range(no1wkst):
+ self.wnomask[i] = 1
+
+ if (rr._bynweekday and (month != self.lastmonth or
+ year != self.lastyear)):
+ ranges = []
+ if rr._freq == YEARLY:
+ if rr._bymonth:
+ for month in rr._bymonth:
+ ranges.append(self.mrange[month-1:month+1])
+ else:
+ ranges = [(0, self.yearlen)]
+ elif rr._freq == MONTHLY:
+ ranges = [self.mrange[month-1:month+1]]
+ if ranges:
+ # Weekly frequency won't get here, so we may not
+ # care about cross-year weekly periods.
+ self.nwdaymask = [0]*self.yearlen
+ for first, last in ranges:
+ last -= 1
+ for wday, n in rr._bynweekday:
+ if n < 0:
+ i = last+(n+1)*7
+ i -= (self.wdaymask[i]-wday) % 7
+ else:
+ i = first+(n-1)*7
+ i += (7-self.wdaymask[i]+wday) % 7
+ if first <= i <= last:
+ self.nwdaymask[i] = 1
+
+ if rr._byeaster:
+ self.eastermask = [0]*(self.yearlen+7)
+ eyday = easter.easter(year).toordinal()-self.yearordinal
+ for offset in rr._byeaster:
+ self.eastermask[eyday+offset] = 1
+
+ self.lastyear = year
+ self.lastmonth = month
+
+ def ydayset(self, year, month, day):
+ return list(range(self.yearlen)), 0, self.yearlen
+
+ def mdayset(self, year, month, day):
+ dset = [None]*self.yearlen
+ start, end = self.mrange[month-1:month+1]
+ for i in range(start, end):
+ dset[i] = i
+ return dset, start, end
+
+ def wdayset(self, year, month, day):
+ # We need to handle cross-year weeks here.
+ dset = [None]*(self.yearlen+7)
+ i = datetime.date(year, month, day).toordinal()-self.yearordinal
+ start = i
+ for j in range(7):
+ dset[i] = i
+ i += 1
+ # if (not (0 <= i < self.yearlen) or
+ # self.wdaymask[i] == self.rrule._wkst):
+ # This will cross the year boundary, if necessary.
+ if self.wdaymask[i] == self.rrule._wkst:
+ break
+ return dset, start, i
+
+ def ddayset(self, year, month, day):
+ dset = [None] * self.yearlen
+ i = datetime.date(year, month, day).toordinal() - self.yearordinal
+ dset[i] = i
+ return dset, i, i + 1
+
+ def htimeset(self, hour, minute, second):
+ tset = []
+ rr = self.rrule
+ for minute in rr._byminute:
+ for second in rr._bysecond:
+ tset.append(datetime.time(hour, minute, second,
+ tzinfo=rr._tzinfo))
+ tset.sort()
+ return tset
+
+ def mtimeset(self, hour, minute, second):
+ tset = []
+ rr = self.rrule
+ for second in rr._bysecond:
+ tset.append(datetime.time(hour, minute, second, tzinfo=rr._tzinfo))
+ tset.sort()
+ return tset
+
+ def stimeset(self, hour, minute, second):
+ return (datetime.time(hour, minute, second,
+ tzinfo=self.rrule._tzinfo),)
+
+
+class rruleset(rrulebase):
+ """ The rruleset type allows more complex recurrence setups, mixing
+ multiple rules, dates, exclusion rules, and exclusion dates. The type
+ constructor takes the following keyword arguments:
+
+ :param cache: If True, caching of results will be enabled, improving
+ performance of multiple queries considerably. """
+
+ class _genitem(object):
+ def __init__(self, genlist, gen):
+ try:
+ self.dt = advance_iterator(gen)
+ genlist.append(self)
+ except StopIteration:
+ pass
+ self.genlist = genlist
+ self.gen = gen
+
+ def __next__(self):
+ try:
+ self.dt = advance_iterator(self.gen)
+ except StopIteration:
+ if self.genlist[0] is self:
+ heapq.heappop(self.genlist)
+ else:
+ self.genlist.remove(self)
+ heapq.heapify(self.genlist)
+
+ next = __next__
+
+ def __lt__(self, other):
+ return self.dt < other.dt
+
+ def __gt__(self, other):
+ return self.dt > other.dt
+
+ def __eq__(self, other):
+ return self.dt == other.dt
+
+ def __ne__(self, other):
+ return self.dt != other.dt
+
+ def __init__(self, cache=False):
+ super(rruleset, self).__init__(cache)
+ self._rrule = []
+ self._rdate = []
+ self._exrule = []
+ self._exdate = []
+
+ @_invalidates_cache
+ def rrule(self, rrule):
+ """ Include the given :py:class:`rrule` instance in the recurrence set
+ generation. """
+ self._rrule.append(rrule)
+
+ @_invalidates_cache
+ def rdate(self, rdate):
+ """ Include the given :py:class:`datetime` instance in the recurrence
+ set generation. """
+ self._rdate.append(rdate)
+
+ @_invalidates_cache
+ def exrule(self, exrule):
+ """ Include the given rrule instance in the recurrence set exclusion
+ list. Dates which are part of the given recurrence rules will not
+ be generated, even if some inclusive rrule or rdate matches them.
+ """
+ self._exrule.append(exrule)
+
+ @_invalidates_cache
+ def exdate(self, exdate):
+ """ Include the given datetime instance in the recurrence set
+ exclusion list. Dates included that way will not be generated,
+ even if some inclusive rrule or rdate matches them. """
+ self._exdate.append(exdate)
+
+ def _iter(self):
+ rlist = []
+ self._rdate.sort()
+ self._genitem(rlist, iter(self._rdate))
+ for gen in [iter(x) for x in self._rrule]:
+ self._genitem(rlist, gen)
+ exlist = []
+ self._exdate.sort()
+ self._genitem(exlist, iter(self._exdate))
+ for gen in [iter(x) for x in self._exrule]:
+ self._genitem(exlist, gen)
+ lastdt = None
+ total = 0
+ heapq.heapify(rlist)
+ heapq.heapify(exlist)
+ while rlist:
+ ritem = rlist[0]
+ if not lastdt or lastdt != ritem.dt:
+ while exlist and exlist[0] < ritem:
+ exitem = exlist[0]
+ advance_iterator(exitem)
+ if exlist and exlist[0] is exitem:
+ heapq.heapreplace(exlist, exitem)
+ if not exlist or ritem != exlist[0]:
+ total += 1
+ yield ritem.dt
+ lastdt = ritem.dt
+ advance_iterator(ritem)
+ if rlist and rlist[0] is ritem:
+ heapq.heapreplace(rlist, ritem)
+ self._len = total
+
+
+
+
+class _rrulestr(object):
+ """ Parses a string representation of a recurrence rule or set of
+ recurrence rules.
+
+ :param s:
+ Required, a string defining one or more recurrence rules.
+
+ :param dtstart:
+ If given, used as the default recurrence start if not specified in the
+ rule string.
+
+ :param cache:
+ If set ``True`` caching of results will be enabled, improving
+ performance of multiple queries considerably.
+
+ :param unfold:
+ If set ``True`` indicates that a rule string is split over more
+ than one line and should be joined before processing.
+
+ :param forceset:
+ If set ``True`` forces a :class:`dateutil.rrule.rruleset` to
+ be returned.
+
+ :param compatible:
+ If set ``True`` forces ``unfold`` and ``forceset`` to be ``True``.
+
+ :param ignoretz:
+ If set ``True``, time zones in parsed strings are ignored and a naive
+ :class:`datetime.datetime` object is returned.
+
+ :param tzids:
+ If given, a callable or mapping used to retrieve a
+ :class:`datetime.tzinfo` from a string representation.
+ Defaults to :func:`dateutil.tz.gettz`.
+
+ :param tzinfos:
+ Additional time zone names / aliases which may be present in a string
+ representation. See :func:`dateutil.parser.parse` for more
+ information.
+
+ :return:
+ Returns a :class:`dateutil.rrule.rruleset` or
+ :class:`dateutil.rrule.rrule`
+ """
+
+ _freq_map = {"YEARLY": YEARLY,
+ "MONTHLY": MONTHLY,
+ "WEEKLY": WEEKLY,
+ "DAILY": DAILY,
+ "HOURLY": HOURLY,
+ "MINUTELY": MINUTELY,
+ "SECONDLY": SECONDLY}
+
+ _weekday_map = {"MO": 0, "TU": 1, "WE": 2, "TH": 3,
+ "FR": 4, "SA": 5, "SU": 6}
+
+ def _handle_int(self, rrkwargs, name, value, **kwargs):
+ rrkwargs[name.lower()] = int(value)
+
+ def _handle_int_list(self, rrkwargs, name, value, **kwargs):
+ rrkwargs[name.lower()] = [int(x) for x in value.split(',')]
+
+ _handle_INTERVAL = _handle_int
+ _handle_COUNT = _handle_int
+ _handle_BYSETPOS = _handle_int_list
+ _handle_BYMONTH = _handle_int_list
+ _handle_BYMONTHDAY = _handle_int_list
+ _handle_BYYEARDAY = _handle_int_list
+ _handle_BYEASTER = _handle_int_list
+ _handle_BYWEEKNO = _handle_int_list
+ _handle_BYHOUR = _handle_int_list
+ _handle_BYMINUTE = _handle_int_list
+ _handle_BYSECOND = _handle_int_list
+
+ def _handle_FREQ(self, rrkwargs, name, value, **kwargs):
+ rrkwargs["freq"] = self._freq_map[value]
+
+ def _handle_UNTIL(self, rrkwargs, name, value, **kwargs):
+ global parser
+ if not parser:
+ from dateutil import parser
+ try:
+ rrkwargs["until"] = parser.parse(value,
+ ignoretz=kwargs.get("ignoretz"),
+ tzinfos=kwargs.get("tzinfos"))
+ except ValueError:
+ raise ValueError("invalid until date")
+
+ def _handle_WKST(self, rrkwargs, name, value, **kwargs):
+ rrkwargs["wkst"] = self._weekday_map[value]
+
+ def _handle_BYWEEKDAY(self, rrkwargs, name, value, **kwargs):
+ """
+ Two ways to specify this: +1MO or MO(+1)
+ """
+ l = []
+ for wday in value.split(','):
+ if '(' in wday:
+ # If it's of the form TH(+1), etc.
+ splt = wday.split('(')
+ w = splt[0]
+ n = int(splt[1][:-1])
+ elif len(wday):
+ # If it's of the form +1MO
+ for i in range(len(wday)):
+ if wday[i] not in '+-0123456789':
+ break
+ n = wday[:i] or None
+ w = wday[i:]
+ if n:
+ n = int(n)
+ else:
+ raise ValueError("Invalid (empty) BYDAY specification.")
+
+ l.append(weekdays[self._weekday_map[w]](n))
+ rrkwargs["byweekday"] = l
+
+ _handle_BYDAY = _handle_BYWEEKDAY
+
+ def _parse_rfc_rrule(self, line,
+ dtstart=None,
+ cache=False,
+ ignoretz=False,
+ tzinfos=None):
+ if line.find(':') != -1:
+ name, value = line.split(':')
+ if name != "RRULE":
+ raise ValueError("unknown parameter name")
+ else:
+ value = line
+ rrkwargs = {}
+ for pair in value.split(';'):
+ name, value = pair.split('=')
+ name = name.upper()
+ value = value.upper()
+ try:
+ getattr(self, "_handle_"+name)(rrkwargs, name, value,
+ ignoretz=ignoretz,
+ tzinfos=tzinfos)
+ except AttributeError:
+ raise ValueError("unknown parameter '%s'" % name)
+ except (KeyError, ValueError):
+ raise ValueError("invalid '%s': %s" % (name, value))
+ return rrule(dtstart=dtstart, cache=cache, **rrkwargs)
+
+ def _parse_date_value(self, date_value, parms, rule_tzids,
+ ignoretz, tzids, tzinfos):
+ global parser
+ if not parser:
+ from dateutil import parser
+
+ datevals = []
+ value_found = False
+ TZID = None
+
+ for parm in parms:
+ if parm.startswith("TZID="):
+ try:
+ tzkey = rule_tzids[parm.split('TZID=')[-1]]
+ except KeyError:
+ continue
+ if tzids is None:
+ from . import tz
+ tzlookup = tz.gettz
+ elif callable(tzids):
+ tzlookup = tzids
+ else:
+ tzlookup = getattr(tzids, 'get', None)
+ if tzlookup is None:
+ msg = ('tzids must be a callable, mapping, or None, '
+ 'not %s' % tzids)
+ raise ValueError(msg)
+
+ TZID = tzlookup(tzkey)
+ continue
+
+ # RFC 5445 3.8.2.4: The VALUE parameter is optional, but may be found
+ # only once.
+ if parm not in {"VALUE=DATE-TIME", "VALUE=DATE"}:
+ raise ValueError("unsupported parm: " + parm)
+ else:
+ if value_found:
+ msg = ("Duplicate value parameter found in: " + parm)
+ raise ValueError(msg)
+ value_found = True
+
+ for datestr in date_value.split(','):
+ date = parser.parse(datestr, ignoretz=ignoretz, tzinfos=tzinfos)
+ if TZID is not None:
+ if date.tzinfo is None:
+ date = date.replace(tzinfo=TZID)
+ else:
+ raise ValueError('DTSTART/EXDATE specifies multiple timezone')
+ datevals.append(date)
+
+ return datevals
+
+ def _parse_rfc(self, s,
+ dtstart=None,
+ cache=False,
+ unfold=False,
+ forceset=False,
+ compatible=False,
+ ignoretz=False,
+ tzids=None,
+ tzinfos=None):
+ global parser
+ if compatible:
+ forceset = True
+ unfold = True
+
+ TZID_NAMES = dict(map(
+ lambda x: (x.upper(), x),
+ re.findall('TZID=(?P[^:]+):', s)
+ ))
+ s = s.upper()
+ if not s.strip():
+ raise ValueError("empty string")
+ if unfold:
+ lines = s.splitlines()
+ i = 0
+ while i < len(lines):
+ line = lines[i].rstrip()
+ if not line:
+ del lines[i]
+ elif i > 0 and line[0] == " ":
+ lines[i-1] += line[1:]
+ del lines[i]
+ else:
+ i += 1
+ else:
+ lines = s.split()
+ if (not forceset and len(lines) == 1 and (s.find(':') == -1 or
+ s.startswith('RRULE:'))):
+ return self._parse_rfc_rrule(lines[0], cache=cache,
+ dtstart=dtstart, ignoretz=ignoretz,
+ tzinfos=tzinfos)
+ else:
+ rrulevals = []
+ rdatevals = []
+ exrulevals = []
+ exdatevals = []
+ for line in lines:
+ if not line:
+ continue
+ if line.find(':') == -1:
+ name = "RRULE"
+ value = line
+ else:
+ name, value = line.split(':', 1)
+ parms = name.split(';')
+ if not parms:
+ raise ValueError("empty property name")
+ name = parms[0]
+ parms = parms[1:]
+ if name == "RRULE":
+ for parm in parms:
+ raise ValueError("unsupported RRULE parm: "+parm)
+ rrulevals.append(value)
+ elif name == "RDATE":
+ for parm in parms:
+ if parm != "VALUE=DATE-TIME":
+ raise ValueError("unsupported RDATE parm: "+parm)
+ rdatevals.append(value)
+ elif name == "EXRULE":
+ for parm in parms:
+ raise ValueError("unsupported EXRULE parm: "+parm)
+ exrulevals.append(value)
+ elif name == "EXDATE":
+ exdatevals.extend(
+ self._parse_date_value(value, parms,
+ TZID_NAMES, ignoretz,
+ tzids, tzinfos)
+ )
+ elif name == "DTSTART":
+ dtvals = self._parse_date_value(value, parms, TZID_NAMES,
+ ignoretz, tzids, tzinfos)
+ if len(dtvals) != 1:
+ raise ValueError("Multiple DTSTART values specified:" +
+ value)
+ dtstart = dtvals[0]
+ else:
+ raise ValueError("unsupported property: "+name)
+ if (forceset or len(rrulevals) > 1 or rdatevals
+ or exrulevals or exdatevals):
+ if not parser and (rdatevals or exdatevals):
+ from dateutil import parser
+ rset = rruleset(cache=cache)
+ for value in rrulevals:
+ rset.rrule(self._parse_rfc_rrule(value, dtstart=dtstart,
+ ignoretz=ignoretz,
+ tzinfos=tzinfos))
+ for value in rdatevals:
+ for datestr in value.split(','):
+ rset.rdate(parser.parse(datestr,
+ ignoretz=ignoretz,
+ tzinfos=tzinfos))
+ for value in exrulevals:
+ rset.exrule(self._parse_rfc_rrule(value, dtstart=dtstart,
+ ignoretz=ignoretz,
+ tzinfos=tzinfos))
+ for value in exdatevals:
+ rset.exdate(value)
+ if compatible and dtstart:
+ rset.rdate(dtstart)
+ return rset
+ else:
+ return self._parse_rfc_rrule(rrulevals[0],
+ dtstart=dtstart,
+ cache=cache,
+ ignoretz=ignoretz,
+ tzinfos=tzinfos)
+
+ def __call__(self, s, **kwargs):
+ return self._parse_rfc(s, **kwargs)
+
+
+rrulestr = _rrulestr()
+
+# vim:ts=4:sw=4:et
diff --git a/falcon/lib/python3.10/site-packages/dateutil/tz/__init__.py b/falcon/lib/python3.10/site-packages/dateutil/tz/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..af1352c47292f4eebc5cae8da45641b5544558e3
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/tz/__init__.py
@@ -0,0 +1,12 @@
+# -*- coding: utf-8 -*-
+from .tz import *
+from .tz import __doc__
+
+__all__ = ["tzutc", "tzoffset", "tzlocal", "tzfile", "tzrange",
+ "tzstr", "tzical", "tzwin", "tzwinlocal", "gettz",
+ "enfold", "datetime_ambiguous", "datetime_exists",
+ "resolve_imaginary", "UTC", "DeprecatedTzFormatWarning"]
+
+
+class DeprecatedTzFormatWarning(Warning):
+ """Warning raised when time zones are parsed from deprecated formats."""
diff --git a/falcon/lib/python3.10/site-packages/dateutil/tz/__pycache__/tz.cpython-310.pyc b/falcon/lib/python3.10/site-packages/dateutil/tz/__pycache__/tz.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..f04c80c502b9cd4ac6e129863cd21b81ccabf07a
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/dateutil/tz/__pycache__/tz.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/dateutil/tz/_common.py b/falcon/lib/python3.10/site-packages/dateutil/tz/_common.py
new file mode 100644
index 0000000000000000000000000000000000000000..e6ac11831522b266114d5b68ee1da298e3aeb14a
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/tz/_common.py
@@ -0,0 +1,419 @@
+from six import PY2
+
+from functools import wraps
+
+from datetime import datetime, timedelta, tzinfo
+
+
+ZERO = timedelta(0)
+
+__all__ = ['tzname_in_python2', 'enfold']
+
+
+def tzname_in_python2(namefunc):
+ """Change unicode output into bytestrings in Python 2
+
+ tzname() API changed in Python 3. It used to return bytes, but was changed
+ to unicode strings
+ """
+ if PY2:
+ @wraps(namefunc)
+ def adjust_encoding(*args, **kwargs):
+ name = namefunc(*args, **kwargs)
+ if name is not None:
+ name = name.encode()
+
+ return name
+
+ return adjust_encoding
+ else:
+ return namefunc
+
+
+# The following is adapted from Alexander Belopolsky's tz library
+# https://github.com/abalkin/tz
+if hasattr(datetime, 'fold'):
+ # This is the pre-python 3.6 fold situation
+ def enfold(dt, fold=1):
+ """
+ Provides a unified interface for assigning the ``fold`` attribute to
+ datetimes both before and after the implementation of PEP-495.
+
+ :param fold:
+ The value for the ``fold`` attribute in the returned datetime. This
+ should be either 0 or 1.
+
+ :return:
+ Returns an object for which ``getattr(dt, 'fold', 0)`` returns
+ ``fold`` for all versions of Python. In versions prior to
+ Python 3.6, this is a ``_DatetimeWithFold`` object, which is a
+ subclass of :py:class:`datetime.datetime` with the ``fold``
+ attribute added, if ``fold`` is 1.
+
+ .. versionadded:: 2.6.0
+ """
+ return dt.replace(fold=fold)
+
+else:
+ class _DatetimeWithFold(datetime):
+ """
+ This is a class designed to provide a PEP 495-compliant interface for
+ Python versions before 3.6. It is used only for dates in a fold, so
+ the ``fold`` attribute is fixed at ``1``.
+
+ .. versionadded:: 2.6.0
+ """
+ __slots__ = ()
+
+ def replace(self, *args, **kwargs):
+ """
+ Return a datetime with the same attributes, except for those
+ attributes given new values by whichever keyword arguments are
+ specified. Note that tzinfo=None can be specified to create a naive
+ datetime from an aware datetime with no conversion of date and time
+ data.
+
+ This is reimplemented in ``_DatetimeWithFold`` because pypy3 will
+ return a ``datetime.datetime`` even if ``fold`` is unchanged.
+ """
+ argnames = (
+ 'year', 'month', 'day', 'hour', 'minute', 'second',
+ 'microsecond', 'tzinfo'
+ )
+
+ for arg, argname in zip(args, argnames):
+ if argname in kwargs:
+ raise TypeError('Duplicate argument: {}'.format(argname))
+
+ kwargs[argname] = arg
+
+ for argname in argnames:
+ if argname not in kwargs:
+ kwargs[argname] = getattr(self, argname)
+
+ dt_class = self.__class__ if kwargs.get('fold', 1) else datetime
+
+ return dt_class(**kwargs)
+
+ @property
+ def fold(self):
+ return 1
+
+ def enfold(dt, fold=1):
+ """
+ Provides a unified interface for assigning the ``fold`` attribute to
+ datetimes both before and after the implementation of PEP-495.
+
+ :param fold:
+ The value for the ``fold`` attribute in the returned datetime. This
+ should be either 0 or 1.
+
+ :return:
+ Returns an object for which ``getattr(dt, 'fold', 0)`` returns
+ ``fold`` for all versions of Python. In versions prior to
+ Python 3.6, this is a ``_DatetimeWithFold`` object, which is a
+ subclass of :py:class:`datetime.datetime` with the ``fold``
+ attribute added, if ``fold`` is 1.
+
+ .. versionadded:: 2.6.0
+ """
+ if getattr(dt, 'fold', 0) == fold:
+ return dt
+
+ args = dt.timetuple()[:6]
+ args += (dt.microsecond, dt.tzinfo)
+
+ if fold:
+ return _DatetimeWithFold(*args)
+ else:
+ return datetime(*args)
+
+
+def _validate_fromutc_inputs(f):
+ """
+ The CPython version of ``fromutc`` checks that the input is a ``datetime``
+ object and that ``self`` is attached as its ``tzinfo``.
+ """
+ @wraps(f)
+ def fromutc(self, dt):
+ if not isinstance(dt, datetime):
+ raise TypeError("fromutc() requires a datetime argument")
+ if dt.tzinfo is not self:
+ raise ValueError("dt.tzinfo is not self")
+
+ return f(self, dt)
+
+ return fromutc
+
+
+class _tzinfo(tzinfo):
+ """
+ Base class for all ``dateutil`` ``tzinfo`` objects.
+ """
+
+ def is_ambiguous(self, dt):
+ """
+ Whether or not the "wall time" of a given datetime is ambiguous in this
+ zone.
+
+ :param dt:
+ A :py:class:`datetime.datetime`, naive or time zone aware.
+
+
+ :return:
+ Returns ``True`` if ambiguous, ``False`` otherwise.
+
+ .. versionadded:: 2.6.0
+ """
+
+ dt = dt.replace(tzinfo=self)
+
+ wall_0 = enfold(dt, fold=0)
+ wall_1 = enfold(dt, fold=1)
+
+ same_offset = wall_0.utcoffset() == wall_1.utcoffset()
+ same_dt = wall_0.replace(tzinfo=None) == wall_1.replace(tzinfo=None)
+
+ return same_dt and not same_offset
+
+ def _fold_status(self, dt_utc, dt_wall):
+ """
+ Determine the fold status of a "wall" datetime, given a representation
+ of the same datetime as a (naive) UTC datetime. This is calculated based
+ on the assumption that ``dt.utcoffset() - dt.dst()`` is constant for all
+ datetimes, and that this offset is the actual number of hours separating
+ ``dt_utc`` and ``dt_wall``.
+
+ :param dt_utc:
+ Representation of the datetime as UTC
+
+ :param dt_wall:
+ Representation of the datetime as "wall time". This parameter must
+ either have a `fold` attribute or have a fold-naive
+ :class:`datetime.tzinfo` attached, otherwise the calculation may
+ fail.
+ """
+ if self.is_ambiguous(dt_wall):
+ delta_wall = dt_wall - dt_utc
+ _fold = int(delta_wall == (dt_utc.utcoffset() - dt_utc.dst()))
+ else:
+ _fold = 0
+
+ return _fold
+
+ def _fold(self, dt):
+ return getattr(dt, 'fold', 0)
+
+ def _fromutc(self, dt):
+ """
+ Given a timezone-aware datetime in a given timezone, calculates a
+ timezone-aware datetime in a new timezone.
+
+ Since this is the one time that we *know* we have an unambiguous
+ datetime object, we take this opportunity to determine whether the
+ datetime is ambiguous and in a "fold" state (e.g. if it's the first
+ occurrence, chronologically, of the ambiguous datetime).
+
+ :param dt:
+ A timezone-aware :class:`datetime.datetime` object.
+ """
+
+ # Re-implement the algorithm from Python's datetime.py
+ dtoff = dt.utcoffset()
+ if dtoff is None:
+ raise ValueError("fromutc() requires a non-None utcoffset() "
+ "result")
+
+ # The original datetime.py code assumes that `dst()` defaults to
+ # zero during ambiguous times. PEP 495 inverts this presumption, so
+ # for pre-PEP 495 versions of python, we need to tweak the algorithm.
+ dtdst = dt.dst()
+ if dtdst is None:
+ raise ValueError("fromutc() requires a non-None dst() result")
+ delta = dtoff - dtdst
+
+ dt += delta
+ # Set fold=1 so we can default to being in the fold for
+ # ambiguous dates.
+ dtdst = enfold(dt, fold=1).dst()
+ if dtdst is None:
+ raise ValueError("fromutc(): dt.dst gave inconsistent "
+ "results; cannot convert")
+ return dt + dtdst
+
+ @_validate_fromutc_inputs
+ def fromutc(self, dt):
+ """
+ Given a timezone-aware datetime in a given timezone, calculates a
+ timezone-aware datetime in a new timezone.
+
+ Since this is the one time that we *know* we have an unambiguous
+ datetime object, we take this opportunity to determine whether the
+ datetime is ambiguous and in a "fold" state (e.g. if it's the first
+ occurrence, chronologically, of the ambiguous datetime).
+
+ :param dt:
+ A timezone-aware :class:`datetime.datetime` object.
+ """
+ dt_wall = self._fromutc(dt)
+
+ # Calculate the fold status given the two datetimes.
+ _fold = self._fold_status(dt, dt_wall)
+
+ # Set the default fold value for ambiguous dates
+ return enfold(dt_wall, fold=_fold)
+
+
+class tzrangebase(_tzinfo):
+ """
+ This is an abstract base class for time zones represented by an annual
+ transition into and out of DST. Child classes should implement the following
+ methods:
+
+ * ``__init__(self, *args, **kwargs)``
+ * ``transitions(self, year)`` - this is expected to return a tuple of
+ datetimes representing the DST on and off transitions in standard
+ time.
+
+ A fully initialized ``tzrangebase`` subclass should also provide the
+ following attributes:
+ * ``hasdst``: Boolean whether or not the zone uses DST.
+ * ``_dst_offset`` / ``_std_offset``: :class:`datetime.timedelta` objects
+ representing the respective UTC offsets.
+ * ``_dst_abbr`` / ``_std_abbr``: Strings representing the timezone short
+ abbreviations in DST and STD, respectively.
+ * ``_hasdst``: Whether or not the zone has DST.
+
+ .. versionadded:: 2.6.0
+ """
+ def __init__(self):
+ raise NotImplementedError('tzrangebase is an abstract base class')
+
+ def utcoffset(self, dt):
+ isdst = self._isdst(dt)
+
+ if isdst is None:
+ return None
+ elif isdst:
+ return self._dst_offset
+ else:
+ return self._std_offset
+
+ def dst(self, dt):
+ isdst = self._isdst(dt)
+
+ if isdst is None:
+ return None
+ elif isdst:
+ return self._dst_base_offset
+ else:
+ return ZERO
+
+ @tzname_in_python2
+ def tzname(self, dt):
+ if self._isdst(dt):
+ return self._dst_abbr
+ else:
+ return self._std_abbr
+
+ def fromutc(self, dt):
+ """ Given a datetime in UTC, return local time """
+ if not isinstance(dt, datetime):
+ raise TypeError("fromutc() requires a datetime argument")
+
+ if dt.tzinfo is not self:
+ raise ValueError("dt.tzinfo is not self")
+
+ # Get transitions - if there are none, fixed offset
+ transitions = self.transitions(dt.year)
+ if transitions is None:
+ return dt + self.utcoffset(dt)
+
+ # Get the transition times in UTC
+ dston, dstoff = transitions
+
+ dston -= self._std_offset
+ dstoff -= self._std_offset
+
+ utc_transitions = (dston, dstoff)
+ dt_utc = dt.replace(tzinfo=None)
+
+ isdst = self._naive_isdst(dt_utc, utc_transitions)
+
+ if isdst:
+ dt_wall = dt + self._dst_offset
+ else:
+ dt_wall = dt + self._std_offset
+
+ _fold = int(not isdst and self.is_ambiguous(dt_wall))
+
+ return enfold(dt_wall, fold=_fold)
+
+ def is_ambiguous(self, dt):
+ """
+ Whether or not the "wall time" of a given datetime is ambiguous in this
+ zone.
+
+ :param dt:
+ A :py:class:`datetime.datetime`, naive or time zone aware.
+
+
+ :return:
+ Returns ``True`` if ambiguous, ``False`` otherwise.
+
+ .. versionadded:: 2.6.0
+ """
+ if not self.hasdst:
+ return False
+
+ start, end = self.transitions(dt.year)
+
+ dt = dt.replace(tzinfo=None)
+ return (end <= dt < end + self._dst_base_offset)
+
+ def _isdst(self, dt):
+ if not self.hasdst:
+ return False
+ elif dt is None:
+ return None
+
+ transitions = self.transitions(dt.year)
+
+ if transitions is None:
+ return False
+
+ dt = dt.replace(tzinfo=None)
+
+ isdst = self._naive_isdst(dt, transitions)
+
+ # Handle ambiguous dates
+ if not isdst and self.is_ambiguous(dt):
+ return not self._fold(dt)
+ else:
+ return isdst
+
+ def _naive_isdst(self, dt, transitions):
+ dston, dstoff = transitions
+
+ dt = dt.replace(tzinfo=None)
+
+ if dston < dstoff:
+ isdst = dston <= dt < dstoff
+ else:
+ isdst = not dstoff <= dt < dston
+
+ return isdst
+
+ @property
+ def _dst_base_offset(self):
+ return self._dst_offset - self._std_offset
+
+ __hash__ = None
+
+ def __ne__(self, other):
+ return not (self == other)
+
+ def __repr__(self):
+ return "%s(...)" % self.__class__.__name__
+
+ __reduce__ = object.__reduce__
diff --git a/falcon/lib/python3.10/site-packages/dateutil/tz/win.py b/falcon/lib/python3.10/site-packages/dateutil/tz/win.py
new file mode 100644
index 0000000000000000000000000000000000000000..cde07ba792c40903f0c334839140173b39fd8124
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/tz/win.py
@@ -0,0 +1,370 @@
+# -*- coding: utf-8 -*-
+"""
+This module provides an interface to the native time zone data on Windows,
+including :py:class:`datetime.tzinfo` implementations.
+
+Attempting to import this module on a non-Windows platform will raise an
+:py:obj:`ImportError`.
+"""
+# This code was originally contributed by Jeffrey Harris.
+import datetime
+import struct
+
+from six.moves import winreg
+from six import text_type
+
+try:
+ import ctypes
+ from ctypes import wintypes
+except ValueError:
+ # ValueError is raised on non-Windows systems for some horrible reason.
+ raise ImportError("Running tzwin on non-Windows system")
+
+from ._common import tzrangebase
+
+__all__ = ["tzwin", "tzwinlocal", "tzres"]
+
+ONEWEEK = datetime.timedelta(7)
+
+TZKEYNAMENT = r"SOFTWARE\Microsoft\Windows NT\CurrentVersion\Time Zones"
+TZKEYNAME9X = r"SOFTWARE\Microsoft\Windows\CurrentVersion\Time Zones"
+TZLOCALKEYNAME = r"SYSTEM\CurrentControlSet\Control\TimeZoneInformation"
+
+
+def _settzkeyname():
+ handle = winreg.ConnectRegistry(None, winreg.HKEY_LOCAL_MACHINE)
+ try:
+ winreg.OpenKey(handle, TZKEYNAMENT).Close()
+ TZKEYNAME = TZKEYNAMENT
+ except WindowsError:
+ TZKEYNAME = TZKEYNAME9X
+ handle.Close()
+ return TZKEYNAME
+
+
+TZKEYNAME = _settzkeyname()
+
+
+class tzres(object):
+ """
+ Class for accessing ``tzres.dll``, which contains timezone name related
+ resources.
+
+ .. versionadded:: 2.5.0
+ """
+ p_wchar = ctypes.POINTER(wintypes.WCHAR) # Pointer to a wide char
+
+ def __init__(self, tzres_loc='tzres.dll'):
+ # Load the user32 DLL so we can load strings from tzres
+ user32 = ctypes.WinDLL('user32')
+
+ # Specify the LoadStringW function
+ user32.LoadStringW.argtypes = (wintypes.HINSTANCE,
+ wintypes.UINT,
+ wintypes.LPWSTR,
+ ctypes.c_int)
+
+ self.LoadStringW = user32.LoadStringW
+ self._tzres = ctypes.WinDLL(tzres_loc)
+ self.tzres_loc = tzres_loc
+
+ def load_name(self, offset):
+ """
+ Load a timezone name from a DLL offset (integer).
+
+ >>> from dateutil.tzwin import tzres
+ >>> tzr = tzres()
+ >>> print(tzr.load_name(112))
+ 'Eastern Standard Time'
+
+ :param offset:
+ A positive integer value referring to a string from the tzres dll.
+
+ .. note::
+
+ Offsets found in the registry are generally of the form
+ ``@tzres.dll,-114``. The offset in this case is 114, not -114.
+
+ """
+ resource = self.p_wchar()
+ lpBuffer = ctypes.cast(ctypes.byref(resource), wintypes.LPWSTR)
+ nchar = self.LoadStringW(self._tzres._handle, offset, lpBuffer, 0)
+ return resource[:nchar]
+
+ def name_from_string(self, tzname_str):
+ """
+ Parse strings as returned from the Windows registry into the time zone
+ name as defined in the registry.
+
+ >>> from dateutil.tzwin import tzres
+ >>> tzr = tzres()
+ >>> print(tzr.name_from_string('@tzres.dll,-251'))
+ 'Dateline Daylight Time'
+ >>> print(tzr.name_from_string('Eastern Standard Time'))
+ 'Eastern Standard Time'
+
+ :param tzname_str:
+ A timezone name string as returned from a Windows registry key.
+
+ :return:
+ Returns the localized timezone string from tzres.dll if the string
+ is of the form `@tzres.dll,-offset`, else returns the input string.
+ """
+ if not tzname_str.startswith('@'):
+ return tzname_str
+
+ name_splt = tzname_str.split(',-')
+ try:
+ offset = int(name_splt[1])
+ except:
+ raise ValueError("Malformed timezone string.")
+
+ return self.load_name(offset)
+
+
+class tzwinbase(tzrangebase):
+ """tzinfo class based on win32's timezones available in the registry."""
+ def __init__(self):
+ raise NotImplementedError('tzwinbase is an abstract base class')
+
+ def __eq__(self, other):
+ # Compare on all relevant dimensions, including name.
+ if not isinstance(other, tzwinbase):
+ return NotImplemented
+
+ return (self._std_offset == other._std_offset and
+ self._dst_offset == other._dst_offset and
+ self._stddayofweek == other._stddayofweek and
+ self._dstdayofweek == other._dstdayofweek and
+ self._stdweeknumber == other._stdweeknumber and
+ self._dstweeknumber == other._dstweeknumber and
+ self._stdhour == other._stdhour and
+ self._dsthour == other._dsthour and
+ self._stdminute == other._stdminute and
+ self._dstminute == other._dstminute and
+ self._std_abbr == other._std_abbr and
+ self._dst_abbr == other._dst_abbr)
+
+ @staticmethod
+ def list():
+ """Return a list of all time zones known to the system."""
+ with winreg.ConnectRegistry(None, winreg.HKEY_LOCAL_MACHINE) as handle:
+ with winreg.OpenKey(handle, TZKEYNAME) as tzkey:
+ result = [winreg.EnumKey(tzkey, i)
+ for i in range(winreg.QueryInfoKey(tzkey)[0])]
+ return result
+
+ def display(self):
+ """
+ Return the display name of the time zone.
+ """
+ return self._display
+
+ def transitions(self, year):
+ """
+ For a given year, get the DST on and off transition times, expressed
+ always on the standard time side. For zones with no transitions, this
+ function returns ``None``.
+
+ :param year:
+ The year whose transitions you would like to query.
+
+ :return:
+ Returns a :class:`tuple` of :class:`datetime.datetime` objects,
+ ``(dston, dstoff)`` for zones with an annual DST transition, or
+ ``None`` for fixed offset zones.
+ """
+
+ if not self.hasdst:
+ return None
+
+ dston = picknthweekday(year, self._dstmonth, self._dstdayofweek,
+ self._dsthour, self._dstminute,
+ self._dstweeknumber)
+
+ dstoff = picknthweekday(year, self._stdmonth, self._stddayofweek,
+ self._stdhour, self._stdminute,
+ self._stdweeknumber)
+
+ # Ambiguous dates default to the STD side
+ dstoff -= self._dst_base_offset
+
+ return dston, dstoff
+
+ def _get_hasdst(self):
+ return self._dstmonth != 0
+
+ @property
+ def _dst_base_offset(self):
+ return self._dst_base_offset_
+
+
+class tzwin(tzwinbase):
+ """
+ Time zone object created from the zone info in the Windows registry
+
+ These are similar to :py:class:`dateutil.tz.tzrange` objects in that
+ the time zone data is provided in the format of a single offset rule
+ for either 0 or 2 time zone transitions per year.
+
+ :param: name
+ The name of a Windows time zone key, e.g. "Eastern Standard Time".
+ The full list of keys can be retrieved with :func:`tzwin.list`.
+ """
+
+ def __init__(self, name):
+ self._name = name
+
+ with winreg.ConnectRegistry(None, winreg.HKEY_LOCAL_MACHINE) as handle:
+ tzkeyname = text_type("{kn}\\{name}").format(kn=TZKEYNAME, name=name)
+ with winreg.OpenKey(handle, tzkeyname) as tzkey:
+ keydict = valuestodict(tzkey)
+
+ self._std_abbr = keydict["Std"]
+ self._dst_abbr = keydict["Dlt"]
+
+ self._display = keydict["Display"]
+
+ # See http://ww_winreg.jsiinc.com/SUBA/tip0300/rh0398.htm
+ tup = struct.unpack("=3l16h", keydict["TZI"])
+ stdoffset = -tup[0]-tup[1] # Bias + StandardBias * -1
+ dstoffset = stdoffset-tup[2] # + DaylightBias * -1
+ self._std_offset = datetime.timedelta(minutes=stdoffset)
+ self._dst_offset = datetime.timedelta(minutes=dstoffset)
+
+ # for the meaning see the win32 TIME_ZONE_INFORMATION structure docs
+ # http://msdn.microsoft.com/en-us/library/windows/desktop/ms725481(v=vs.85).aspx
+ (self._stdmonth,
+ self._stddayofweek, # Sunday = 0
+ self._stdweeknumber, # Last = 5
+ self._stdhour,
+ self._stdminute) = tup[4:9]
+
+ (self._dstmonth,
+ self._dstdayofweek, # Sunday = 0
+ self._dstweeknumber, # Last = 5
+ self._dsthour,
+ self._dstminute) = tup[12:17]
+
+ self._dst_base_offset_ = self._dst_offset - self._std_offset
+ self.hasdst = self._get_hasdst()
+
+ def __repr__(self):
+ return "tzwin(%s)" % repr(self._name)
+
+ def __reduce__(self):
+ return (self.__class__, (self._name,))
+
+
+class tzwinlocal(tzwinbase):
+ """
+ Class representing the local time zone information in the Windows registry
+
+ While :class:`dateutil.tz.tzlocal` makes system calls (via the :mod:`time`
+ module) to retrieve time zone information, ``tzwinlocal`` retrieves the
+ rules directly from the Windows registry and creates an object like
+ :class:`dateutil.tz.tzwin`.
+
+ Because Windows does not have an equivalent of :func:`time.tzset`, on
+ Windows, :class:`dateutil.tz.tzlocal` instances will always reflect the
+ time zone settings *at the time that the process was started*, meaning
+ changes to the machine's time zone settings during the run of a program
+ on Windows will **not** be reflected by :class:`dateutil.tz.tzlocal`.
+ Because ``tzwinlocal`` reads the registry directly, it is unaffected by
+ this issue.
+ """
+ def __init__(self):
+ with winreg.ConnectRegistry(None, winreg.HKEY_LOCAL_MACHINE) as handle:
+ with winreg.OpenKey(handle, TZLOCALKEYNAME) as tzlocalkey:
+ keydict = valuestodict(tzlocalkey)
+
+ self._std_abbr = keydict["StandardName"]
+ self._dst_abbr = keydict["DaylightName"]
+
+ try:
+ tzkeyname = text_type('{kn}\\{sn}').format(kn=TZKEYNAME,
+ sn=self._std_abbr)
+ with winreg.OpenKey(handle, tzkeyname) as tzkey:
+ _keydict = valuestodict(tzkey)
+ self._display = _keydict["Display"]
+ except OSError:
+ self._display = None
+
+ stdoffset = -keydict["Bias"]-keydict["StandardBias"]
+ dstoffset = stdoffset-keydict["DaylightBias"]
+
+ self._std_offset = datetime.timedelta(minutes=stdoffset)
+ self._dst_offset = datetime.timedelta(minutes=dstoffset)
+
+ # For reasons unclear, in this particular key, the day of week has been
+ # moved to the END of the SYSTEMTIME structure.
+ tup = struct.unpack("=8h", keydict["StandardStart"])
+
+ (self._stdmonth,
+ self._stdweeknumber, # Last = 5
+ self._stdhour,
+ self._stdminute) = tup[1:5]
+
+ self._stddayofweek = tup[7]
+
+ tup = struct.unpack("=8h", keydict["DaylightStart"])
+
+ (self._dstmonth,
+ self._dstweeknumber, # Last = 5
+ self._dsthour,
+ self._dstminute) = tup[1:5]
+
+ self._dstdayofweek = tup[7]
+
+ self._dst_base_offset_ = self._dst_offset - self._std_offset
+ self.hasdst = self._get_hasdst()
+
+ def __repr__(self):
+ return "tzwinlocal()"
+
+ def __str__(self):
+ # str will return the standard name, not the daylight name.
+ return "tzwinlocal(%s)" % repr(self._std_abbr)
+
+ def __reduce__(self):
+ return (self.__class__, ())
+
+
+def picknthweekday(year, month, dayofweek, hour, minute, whichweek):
+ """ dayofweek == 0 means Sunday, whichweek 5 means last instance """
+ first = datetime.datetime(year, month, 1, hour, minute)
+
+ # This will work if dayofweek is ISO weekday (1-7) or Microsoft-style (0-6),
+ # Because 7 % 7 = 0
+ weekdayone = first.replace(day=((dayofweek - first.isoweekday()) % 7) + 1)
+ wd = weekdayone + ((whichweek - 1) * ONEWEEK)
+ if (wd.month != month):
+ wd -= ONEWEEK
+
+ return wd
+
+
+def valuestodict(key):
+ """Convert a registry key's values to a dictionary."""
+ dout = {}
+ size = winreg.QueryInfoKey(key)[1]
+ tz_res = None
+
+ for i in range(size):
+ key_name, value, dtype = winreg.EnumValue(key, i)
+ if dtype == winreg.REG_DWORD or dtype == winreg.REG_DWORD_LITTLE_ENDIAN:
+ # If it's a DWORD (32-bit integer), it's stored as unsigned - convert
+ # that to a proper signed integer
+ if value & (1 << 31):
+ value = value - (1 << 32)
+ elif dtype == winreg.REG_SZ:
+ # If it's a reference to the tzres DLL, load the actual string
+ if value.startswith('@tzres'):
+ tz_res = tz_res or tzres()
+ value = tz_res.name_from_string(value)
+
+ value = value.rstrip('\x00') # Remove trailing nulls
+
+ dout[key_name] = value
+
+ return dout
diff --git a/falcon/lib/python3.10/site-packages/dateutil/tzwin.py b/falcon/lib/python3.10/site-packages/dateutil/tzwin.py
new file mode 100644
index 0000000000000000000000000000000000000000..cebc673e40fc376653ebf037e96f0a6d0b33e906
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/tzwin.py
@@ -0,0 +1,2 @@
+# tzwin has moved to dateutil.tz.win
+from .tz.win import *
diff --git a/falcon/lib/python3.10/site-packages/dateutil/utils.py b/falcon/lib/python3.10/site-packages/dateutil/utils.py
new file mode 100644
index 0000000000000000000000000000000000000000..dd2d245a0bebcd5fc37ac20526aabbd5358dab0e
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/dateutil/utils.py
@@ -0,0 +1,71 @@
+# -*- coding: utf-8 -*-
+"""
+This module offers general convenience and utility functions for dealing with
+datetimes.
+
+.. versionadded:: 2.7.0
+"""
+from __future__ import unicode_literals
+
+from datetime import datetime, time
+
+
+def today(tzinfo=None):
+ """
+ Returns a :py:class:`datetime` representing the current day at midnight
+
+ :param tzinfo:
+ The time zone to attach (also used to determine the current day).
+
+ :return:
+ A :py:class:`datetime.datetime` object representing the current day
+ at midnight.
+ """
+
+ dt = datetime.now(tzinfo)
+ return datetime.combine(dt.date(), time(0, tzinfo=tzinfo))
+
+
+def default_tzinfo(dt, tzinfo):
+ """
+ Sets the ``tzinfo`` parameter on naive datetimes only
+
+ This is useful for example when you are provided a datetime that may have
+ either an implicit or explicit time zone, such as when parsing a time zone
+ string.
+
+ .. doctest::
+
+ >>> from dateutil.tz import tzoffset
+ >>> from dateutil.parser import parse
+ >>> from dateutil.utils import default_tzinfo
+ >>> dflt_tz = tzoffset("EST", -18000)
+ >>> print(default_tzinfo(parse('2014-01-01 12:30 UTC'), dflt_tz))
+ 2014-01-01 12:30:00+00:00
+ >>> print(default_tzinfo(parse('2014-01-01 12:30'), dflt_tz))
+ 2014-01-01 12:30:00-05:00
+
+ :param dt:
+ The datetime on which to replace the time zone
+
+ :param tzinfo:
+ The :py:class:`datetime.tzinfo` subclass instance to assign to
+ ``dt`` if (and only if) it is naive.
+
+ :return:
+ Returns an aware :py:class:`datetime.datetime`.
+ """
+ if dt.tzinfo is not None:
+ return dt
+ else:
+ return dt.replace(tzinfo=tzinfo)
+
+
+def within_delta(dt1, dt2, delta):
+ """
+ Useful for comparing two datetimes that may have a negligible difference
+ to be considered equal.
+ """
+ delta = abs(delta)
+ difference = dt1 - dt2
+ return -delta <= difference <= delta
diff --git a/falcon/lib/python3.10/site-packages/dateutil/zoneinfo/__pycache__/__init__.cpython-310.pyc b/falcon/lib/python3.10/site-packages/dateutil/zoneinfo/__pycache__/__init__.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..98d7a78cf0302b2e2bff41da1e23b6c99878cabb
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/dateutil/zoneinfo/__pycache__/__init__.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/httpcore/_async/__pycache__/connection.cpython-310.pyc b/falcon/lib/python3.10/site-packages/httpcore/_async/__pycache__/connection.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..f70ceb4979c17ad970df5811b10338aad613c21e
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/httpcore/_async/__pycache__/connection.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/INSTALLER b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/INSTALLER
new file mode 100644
index 0000000000000000000000000000000000000000..a1b589e38a32041e49332e5e81c2d363dc418d68
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/INSTALLER
@@ -0,0 +1 @@
+pip
diff --git a/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/LICENSE b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..582ddf59e08277fe6e78cee924d2c84805fe36fe
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2020 ExecutableBookProject
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/LICENSE.markdown-it b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/LICENSE.markdown-it
new file mode 100644
index 0000000000000000000000000000000000000000..7ffa058cb78f8fb9beb974d9fd429004d2d2e585
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/LICENSE.markdown-it
@@ -0,0 +1,22 @@
+Copyright (c) 2014 Vitaly Puzrin, Alex Kocharin.
+
+Permission is hereby granted, free of charge, to any person
+obtaining a copy of this software and associated documentation
+files (the "Software"), to deal in the Software without
+restriction, including without limitation the rights to use,
+copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the
+Software is furnished to do so, subject to the following
+conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+OTHER DEALINGS IN THE SOFTWARE.
diff --git a/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/METADATA b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/METADATA
new file mode 100644
index 0000000000000000000000000000000000000000..0e0f2e852faa83601cf90ed7d400cc19a1bbe3bf
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/METADATA
@@ -0,0 +1,204 @@
+Metadata-Version: 2.1
+Name: markdown-it-py
+Version: 2.2.0
+Summary: Python port of markdown-it. Markdown parsing, done right!
+Keywords: markdown,lexer,parser,commonmark,markdown-it
+Author-email: Chris Sewell
+Requires-Python: >=3.7
+Description-Content-Type: text/markdown
+Classifier: Development Status :: 5 - Production/Stable
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.7
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: Implementation :: CPython
+Classifier: Programming Language :: Python :: Implementation :: PyPy
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: Topic :: Text Processing :: Markup
+Requires-Dist: mdurl~=0.1
+Requires-Dist: typing_extensions>=3.7.4;python_version<'3.8'
+Requires-Dist: psutil ; extra == "benchmarking"
+Requires-Dist: pytest ; extra == "benchmarking"
+Requires-Dist: pytest-benchmark ; extra == "benchmarking"
+Requires-Dist: pre-commit~=3.0 ; extra == "code_style"
+Requires-Dist: commonmark~=0.9 ; extra == "compare"
+Requires-Dist: markdown~=3.4 ; extra == "compare"
+Requires-Dist: mistletoe~=1.0 ; extra == "compare"
+Requires-Dist: mistune~=2.0 ; extra == "compare"
+Requires-Dist: panflute~=2.3 ; extra == "compare"
+Requires-Dist: linkify-it-py>=1,<3 ; extra == "linkify"
+Requires-Dist: mdit-py-plugins ; extra == "plugins"
+Requires-Dist: gprof2dot ; extra == "profiling"
+Requires-Dist: attrs ; extra == "rtd"
+Requires-Dist: myst-parser ; extra == "rtd"
+Requires-Dist: pyyaml ; extra == "rtd"
+Requires-Dist: sphinx ; extra == "rtd"
+Requires-Dist: sphinx-copybutton ; extra == "rtd"
+Requires-Dist: sphinx-design ; extra == "rtd"
+Requires-Dist: sphinx_book_theme ; extra == "rtd"
+Requires-Dist: coverage ; extra == "testing"
+Requires-Dist: pytest ; extra == "testing"
+Requires-Dist: pytest-cov ; extra == "testing"
+Requires-Dist: pytest-regressions ; extra == "testing"
+Project-URL: Documentation, https://markdown-it-py.readthedocs.io
+Project-URL: Homepage, https://github.com/executablebooks/markdown-it-py
+Provides-Extra: benchmarking
+Provides-Extra: code_style
+Provides-Extra: compare
+Provides-Extra: linkify
+Provides-Extra: plugins
+Provides-Extra: profiling
+Provides-Extra: rtd
+Provides-Extra: testing
+
+# markdown-it-py
+
+[![Github-CI][github-ci]][github-link]
+[![Coverage Status][codecov-badge]][codecov-link]
+[![PyPI][pypi-badge]][pypi-link]
+[![Conda][conda-badge]][conda-link]
+[![Code style: black][black-badge]][black-link]
+[![PyPI - Downloads][install-badge]][install-link]
+
+> Markdown parser done right.
+
+- Follows the __[CommonMark spec](http://spec.commonmark.org/)__ for baseline parsing
+- Configurable syntax: you can add new rules and even replace existing ones.
+- Pluggable: Adds syntax extensions to extend the parser (see the [plugin list][md-plugins]).
+- High speed (see our [benchmarking tests][md-performance])
+- [Safe by default][md-security]
+
+This is a Python port of [markdown-it], and some of its associated plugins.
+For more details see: .
+
+For details on [markdown-it] itself, see:
+
+- The __[Live demo](https://markdown-it.github.io)__
+- [The markdown-it README][markdown-it-readme]
+
+## Installation
+
+```bash
+conda install -c conda-forge markdown-it-py
+```
+
+or
+
+```bash
+pip install markdown-it-py[plugins]
+```
+
+or with extras
+
+```bash
+conda install -c conda-forge markdown-it-py linkify-it-py mdit-py-plugins
+pip install markdown-it-py[linkify,plugins]
+```
+
+## Usage
+
+### Python API Usage
+
+Render markdown to HTML with markdown-it-py and a custom configuration
+with and without plugins and features:
+
+```python
+from markdown_it import MarkdownIt
+from mdit_py_plugins.front_matter import front_matter_plugin
+from mdit_py_plugins.footnote import footnote_plugin
+
+md = (
+ MarkdownIt('commonmark' ,{'breaks':True,'html':True})
+ .use(front_matter_plugin)
+ .use(footnote_plugin)
+ .enable('table')
+)
+text = ("""
+---
+a: 1
+---
+
+a | b
+- | -
+1 | 2
+
+A footnote [^1]
+
+[^1]: some details
+""")
+tokens = md.parse(text)
+html_text = md.render(text)
+
+## To export the html to a file, uncomment the lines below:
+# from pathlib import Path
+# Path("output.html").write_text(html_text)
+```
+
+### Command-line Usage
+
+Render markdown to HTML with markdown-it-py from the
+command-line:
+
+```console
+usage: markdown-it [-h] [-v] [filenames [filenames ...]]
+
+Parse one or more markdown files, convert each to HTML, and print to stdout
+
+positional arguments:
+ filenames specify an optional list of files to convert
+
+optional arguments:
+ -h, --help show this help message and exit
+ -v, --version show program's version number and exit
+
+Interactive:
+
+ $ markdown-it
+ markdown-it-py [version 0.0.0] (interactive)
+ Type Ctrl-D to complete input, or Ctrl-C to exit.
+ >>> # Example
+ ... > markdown *input*
+ ...
+ Example
+
+ markdown input
+
+
+Batch:
+
+ $ markdown-it README.md README.footer.md > index.html
+
+```
+
+## References / Thanks
+
+Big thanks to the authors of [markdown-it]:
+
+- Alex Kocharin [github/rlidwka](https://github.com/rlidwka)
+- Vitaly Puzrin [github/puzrin](https://github.com/puzrin)
+
+Also [John MacFarlane](https://github.com/jgm) for his work on the CommonMark spec and reference implementations.
+
+[github-ci]: https://github.com/executablebooks/markdown-it-py/workflows/Python%20package/badge.svg?branch=master
+[github-link]: https://github.com/executablebooks/markdown-it-py
+[pypi-badge]: https://img.shields.io/pypi/v/markdown-it-py.svg
+[pypi-link]: https://pypi.org/project/markdown-it-py
+[conda-badge]: https://anaconda.org/conda-forge/markdown-it-py/badges/version.svg
+[conda-link]: https://anaconda.org/conda-forge/markdown-it-py
+[codecov-badge]: https://codecov.io/gh/executablebooks/markdown-it-py/branch/master/graph/badge.svg
+[codecov-link]: https://codecov.io/gh/executablebooks/markdown-it-py
+[black-badge]: https://img.shields.io/badge/code%20style-black-000000.svg
+[black-link]: https://github.com/ambv/black
+[install-badge]: https://img.shields.io/pypi/dw/markdown-it-py?label=pypi%20installs
+[install-link]: https://pypistats.org/packages/markdown-it-py
+
+[CommonMark spec]: http://spec.commonmark.org/
+[markdown-it]: https://github.com/markdown-it/markdown-it
+[markdown-it-readme]: https://github.com/markdown-it/markdown-it/blob/master/README.md
+[md-security]: https://markdown-it-py.readthedocs.io/en/latest/other.html
+[md-performance]: https://markdown-it-py.readthedocs.io/en/latest/other.html
+[md-plugins]: https://markdown-it-py.readthedocs.io/en/latest/plugins.html
+
diff --git a/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/RECORD b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/RECORD
new file mode 100644
index 0000000000000000000000000000000000000000..b97d550b3bee80a25af6f17d997e400918b515bc
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/RECORD
@@ -0,0 +1,139 @@
+../../../bin/markdown-it,sha256=Q_YRyeJCLmZmytP58igdcusr8J2Vs9PnsU5VJ3qWR_4,232
+markdown_it/__init__.py,sha256=fLRtFqAOvMj0leE2CHPwL276J0ukSRXs3ahCqQroJA4,113
+markdown_it/__pycache__/__init__.cpython-310.pyc,,
+markdown_it/__pycache__/_compat.cpython-310.pyc,,
+markdown_it/__pycache__/_punycode.cpython-310.pyc,,
+markdown_it/__pycache__/main.cpython-310.pyc,,
+markdown_it/__pycache__/parser_block.cpython-310.pyc,,
+markdown_it/__pycache__/parser_core.cpython-310.pyc,,
+markdown_it/__pycache__/parser_inline.cpython-310.pyc,,
+markdown_it/__pycache__/renderer.cpython-310.pyc,,
+markdown_it/__pycache__/ruler.cpython-310.pyc,,
+markdown_it/__pycache__/token.cpython-310.pyc,,
+markdown_it/__pycache__/tree.cpython-310.pyc,,
+markdown_it/__pycache__/utils.cpython-310.pyc,,
+markdown_it/_compat.py,sha256=n4Zy30zDHMXj1mpjvKKS2yWWwDn84YFqKjEQibc3HJk,248
+markdown_it/_punycode.py,sha256=BOoOYeY0XaXhtj5mFOfGU5rsOVbkp0SuiWdj3ZWi91g,2317
+markdown_it/cli/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+markdown_it/cli/__pycache__/__init__.cpython-310.pyc,,
+markdown_it/cli/__pycache__/parse.cpython-310.pyc,,
+markdown_it/cli/parse.py,sha256=ZiTSx6t7nLk7rGAtIi0a02EB9sDGJn7YLjKKtufdwNA,2901
+markdown_it/common/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+markdown_it/common/__pycache__/__init__.cpython-310.pyc,,
+markdown_it/common/__pycache__/entities.cpython-310.pyc,,
+markdown_it/common/__pycache__/html_blocks.cpython-310.pyc,,
+markdown_it/common/__pycache__/html_re.cpython-310.pyc,,
+markdown_it/common/__pycache__/normalize_url.cpython-310.pyc,,
+markdown_it/common/__pycache__/utils.cpython-310.pyc,,
+markdown_it/common/entities.py,sha256=6ulEjBAWYH5lVobgpn5lChPYhPKkdPyVHpNT7O1_x90,156
+markdown_it/common/html_blocks.py,sha256=1cMBp6jIdXqCHvEs2mpJqVGqTuFi6ExL4VO754yksgU,932
+markdown_it/common/html_re.py,sha256=0q5QFkSnX_l7Ob37MDSj2UYT0onCPz_07hUod2K-a6A,929
+markdown_it/common/normalize_url.py,sha256=K5ESxTC0jbBoR6iyqM4pILDbQUKtajJkCBdSdXSQHwo,2631
+markdown_it/common/utils.py,sha256=qX-Sn0kbtOjWxl6DR3lbOWOJl6HJrURwxfmoCFpNpjY,10894
+markdown_it/helpers/__init__.py,sha256=9W7GycpZcq2up1CdVcUpdN77i9Vl4N0CT3y3qMkTjY4,253
+markdown_it/helpers/__pycache__/__init__.cpython-310.pyc,,
+markdown_it/helpers/__pycache__/parse_link_destination.cpython-310.pyc,,
+markdown_it/helpers/__pycache__/parse_link_label.cpython-310.pyc,,
+markdown_it/helpers/__pycache__/parse_link_title.cpython-310.pyc,,
+markdown_it/helpers/parse_link_destination.py,sha256=UBImo4DafhkxeuZ6tckiY7AOqp2dqE_-_2A4bTKZx0U,1953
+markdown_it/helpers/parse_link_label.py,sha256=3Nq9_GAerchVKc9MJ2uuUGXvFvP9wp6LON1KbvBvbY4,1070
+markdown_it/helpers/parse_link_title.py,sha256=wn9rChklaFuxl4USHud1NmGKffNXy2MH-1xq3K82bYo,1410
+markdown_it/main.py,sha256=5jzlmj68b6LfsLqd8LnzOZ0HvZUJ2aeo-qrzKtWMILw,12228
+markdown_it/parser_block.py,sha256=P0yCwXCj2hCjJQXRz0J4ZVGBMzG8_wymUdCMqQcIBgQ,3662
+markdown_it/parser_core.py,sha256=sKr6D6X5JubJOJVc82wIzEMRcE7GzmfOwN2Jy4yVyd8,835
+markdown_it/parser_inline.py,sha256=O1XBwD-Te0oRc6I7EOmlO6HZSU57slEo6XTn3911bAE,4130
+markdown_it/port.yaml,sha256=8_m2-Wa5qZtivaPdjrvGpM3Pekvm3vBYzI2o1hrfef8,2516
+markdown_it/presets/__init__.py,sha256=owmG4LPxdS2op0KbOwhDiGmCG0MPUz6qaVrUzFJLZoo,898
+markdown_it/presets/__pycache__/__init__.cpython-310.pyc,,
+markdown_it/presets/__pycache__/commonmark.cpython-310.pyc,,
+markdown_it/presets/__pycache__/default.cpython-310.pyc,,
+markdown_it/presets/__pycache__/zero.cpython-310.pyc,,
+markdown_it/presets/commonmark.py,sha256=Aktdrf9CJuvvLa3DCBdI_52YIVNOqALgrGWEqwzT6Jc,2809
+markdown_it/presets/default.py,sha256=lX_NmF6kbfVQs5oSTYsHquI5wqkHTTjKk-NSxcCHn4o,1765
+markdown_it/presets/zero.py,sha256=y_Pgp5VazNAlEUYUZ33wB7KwwmnjxwZ0FWyJAwYRUTk,2006
+markdown_it/py.typed,sha256=8PjyZ1aVoQpRVvt71muvuq5qE-jTFZkK-GLHkhdebmc,26
+markdown_it/renderer.py,sha256=S7gQMB8QbjsB-CEDBhDTxwgVnUGK-keqN_xenR2V0pM,10041
+markdown_it/ruler.py,sha256=e_F2f8gv94OruKpJuW386N7hWrzUYLVrdnMzF9KJXa0,8376
+markdown_it/rules_block/__init__.py,sha256=8su1tOxDw_IR9JSdgqfkGZSNdYpZZCBC9MMmFODdbmE,553
+markdown_it/rules_block/__pycache__/__init__.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/blockquote.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/code.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/fence.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/heading.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/hr.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/html_block.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/lheading.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/list.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/paragraph.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/reference.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/state_block.cpython-310.pyc,,
+markdown_it/rules_block/__pycache__/table.cpython-310.pyc,,
+markdown_it/rules_block/blockquote.py,sha256=f-EyFuXSxCYjXOB6CqJYNF2mqpZNzlIQ6bfElpKBLQY,9057
+markdown_it/rules_block/code.py,sha256=H6ipitwYLvrxmnh6ZQe8D1T0Nokq6ffTTX8-3I9P294,886
+markdown_it/rules_block/fence.py,sha256=xc37ptiY_1eLDWf9N4ADoBW1YqiOKA_sukddZ7xTg_I,2704
+markdown_it/rules_block/heading.py,sha256=tQw9YGs37bac7REQAxr_qMNEFzmOIcYcmdqpErxRle0,1879
+markdown_it/rules_block/hr.py,sha256=WzLPoDDSnE0PwqzdUKoEPQ1cv9TrZOkm6JyzW3t0CyU,1309
+markdown_it/rules_block/html_block.py,sha256=81kD0Xvp_NEV_9o8djgnIdxUg6CrKG79GRaQ-UtQBuQ,2808
+markdown_it/rules_block/lheading.py,sha256=DNjewEmj3y2Q3yf2r-kTOWskIS3_dIjm014pgLl7UjM,2798
+markdown_it/rules_block/list.py,sha256=T_YFMvNdT6HYLwRqWDcPdKPOZ-_dkMfu0_27vWbFG7s,9944
+markdown_it/rules_block/paragraph.py,sha256=7_RVdYBBQ3BxFev4jqC_81KLQWd-diMALBCQ1TT1SCw,1851
+markdown_it/rules_block/reference.py,sha256=9MO-tyW3eBh1sNO_o07LvVat1sr0BKW5ekw7jys-OV8,6323
+markdown_it/rules_block/state_block.py,sha256=e-MMQA7RZu-39Y_FixWAGIwOGgOYiVJh7Zp5C-0hJPw,7226
+markdown_it/rules_block/table.py,sha256=hz_u0uCXH70YpofuQ291pOobzZEuv3AirJF1iKODxbE,7226
+markdown_it/rules_core/__init__.py,sha256=l-DuqjBZgKDa0eS7ZUt5YbM7yKt6Q5MplwhSNhRIvu0,344
+markdown_it/rules_core/__pycache__/__init__.cpython-310.pyc,,
+markdown_it/rules_core/__pycache__/block.cpython-310.pyc,,
+markdown_it/rules_core/__pycache__/inline.cpython-310.pyc,,
+markdown_it/rules_core/__pycache__/linkify.cpython-310.pyc,,
+markdown_it/rules_core/__pycache__/normalize.cpython-310.pyc,,
+markdown_it/rules_core/__pycache__/replacements.cpython-310.pyc,,
+markdown_it/rules_core/__pycache__/smartquotes.cpython-310.pyc,,
+markdown_it/rules_core/__pycache__/state_core.cpython-310.pyc,,
+markdown_it/rules_core/block.py,sha256=HM_YYqPaNByUGa5tv9eYduTbrjQ8Vv3jYxHPhw-e7X0,413
+markdown_it/rules_core/inline.py,sha256=9oWmeBhJHE7x47oJcN9yp6UsAZtrEY_A-VmfoMvKld4,325
+markdown_it/rules_core/linkify.py,sha256=7JfV9eGJq-0FQPW6Kf9mR_brYwIrISg0b855yN4pnJ8,4833
+markdown_it/rules_core/normalize.py,sha256=qVkBO4elitPzyP_sQENho-ycUl8s4eNZ1zZrsR2AAgk,402
+markdown_it/rules_core/replacements.py,sha256=0rDO0fAkGQChNMx3-ZNtMyaHVVD68CZov5BjJi419eQ,3560
+markdown_it/rules_core/smartquotes.py,sha256=nb6GCrpGE0I2yFgpJDvxfGlrJu6csK4_LzWzN-0Omf0,7251
+markdown_it/rules_core/state_core.py,sha256=BD97qL7jZtPMOxqxB-7CxnDtyb8z6T5Tp7GUm_rjCHk,584
+markdown_it/rules_inline/__init__.py,sha256=9KvX3KO-Ql9fiffMrQVn72H9QlW6RRhZikmWeOr6Aks,649
+markdown_it/rules_inline/__pycache__/__init__.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/autolink.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/backticks.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/balance_pairs.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/emphasis.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/entity.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/escape.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/html_inline.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/image.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/link.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/newline.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/state_inline.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/strikethrough.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/text.cpython-310.pyc,,
+markdown_it/rules_inline/__pycache__/text_collapse.cpython-310.pyc,,
+markdown_it/rules_inline/autolink.py,sha256=JqMiaYT024UP6rNGDQtjIWh_1AqowscJF5L04sDjvMc,2131
+markdown_it/rules_inline/backticks.py,sha256=VDLhkLSSq3jfgumdfDi-QGt6pCmal3D1gGjB840KaGk,2112
+markdown_it/rules_inline/balance_pairs.py,sha256=ahcpUdf-0l0xNZ70i-bh4FvgoRjrnkoXJywR7qxO1rw,4062
+markdown_it/rules_inline/emphasis.py,sha256=tBDiQhxaKn0LqILs2vgiU5Y0EnW1PlKfLEvuIsw3qIo,2948
+markdown_it/rules_inline/entity.py,sha256=BJw-EgM8MWsoJ81wDOzfkPHEuUrf2H_AEtdaNK1xBjU,1653
+markdown_it/rules_inline/escape.py,sha256=vkwY9Hv2lmatf6E4IPss7kobeknefFGcxHDUsFwyPe0,1116
+markdown_it/rules_inline/html_inline.py,sha256=Cq1ZnZR9Hcl53xSDOcvLZyhSQcm6XGIW_ExLnHKsfo8,1019
+markdown_it/rules_inline/image.py,sha256=SkpGaOw6Tq1JnRRU6TQ3nG8nkO916Cuqqyt13tx6Ufk,4260
+markdown_it/rules_inline/link.py,sha256=9y9JqHILElVeKA-y6FtGn7tLURaCcwTIVQ8GU2h3mzY,4362
+markdown_it/rules_inline/newline.py,sha256=Xr8BMLY6sxYxS_Fz_J2pqNg5T7PfWBwBvuzA4RTjOfQ,1176
+markdown_it/rules_inline/state_inline.py,sha256=8o4RnePtNHTcpoK1rwF1udok0dAD4ObWMuWQi4ePOuo,5388
+markdown_it/rules_inline/strikethrough.py,sha256=bcJ2UEwhD7eJW1QIYkBbsGtF8mM23MiUI_YTjGPEX7M,3390
+markdown_it/rules_inline/text.py,sha256=ZbL8C0O1uCVHInjfuKpUjFpRXFe5e15rqSidhF2hbbE,1427
+markdown_it/rules_inline/text_collapse.py,sha256=UVGNOEsliLWmtnII3rOa1dx0KX_grRle43H_j5p9VxU,1491
+markdown_it/token.py,sha256=YyOPpk2YcxQHgRSEETRDoeicgq2KFc8jyuLRNsAsNPQ,6374
+markdown_it/tree.py,sha256=jjczB5MxFIx7KS-w14kGOTKWgMezYi2L15zO3FCgqTg,11052
+markdown_it/utils.py,sha256=flD7GSloGybEgXXVaE31KKmwAlBwSmF35d4TGmf-ur8,3262
+markdown_it_py-2.2.0.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4
+markdown_it_py-2.2.0.dist-info/LICENSE,sha256=SiJg1uLND1oVGh6G2_59PtVSseK-q_mUHBulxJy85IQ,1078
+markdown_it_py-2.2.0.dist-info/LICENSE.markdown-it,sha256=eSxIxahJoV_fnjfovPnm0d0TsytGxkKnSKCkapkZ1HM,1073
+markdown_it_py-2.2.0.dist-info/METADATA,sha256=Fb2DIBEdUs2mxdmSJJo2SvtRGDyYwNRH4Zech-Q7Dsw,6812
+markdown_it_py-2.2.0.dist-info/RECORD,,
+markdown_it_py-2.2.0.dist-info/REQUESTED,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+markdown_it_py-2.2.0.dist-info/WHEEL,sha256=rSgq_JpHF9fHR1lx53qwg_1-2LypZE_qmcuXbVUq948,81
+markdown_it_py-2.2.0.dist-info/entry_points.txt,sha256=T81l7fHQ3pllpQ4wUtQK6a8g_p6wxQbnjKVHCk2WMG4,58
diff --git a/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/REQUESTED b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/REQUESTED
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/WHEEL b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/WHEEL
new file mode 100644
index 0000000000000000000000000000000000000000..db4a255f3a271083b66f205eeaa81f74c97bb8ac
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/WHEEL
@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: flit 3.8.0
+Root-Is-Purelib: true
+Tag: py3-none-any
diff --git a/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/entry_points.txt b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/entry_points.txt
new file mode 100644
index 0000000000000000000000000000000000000000..7d829cd792a5d754844f433c6a8dd499564fdcbf
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/markdown_it_py-2.2.0.dist-info/entry_points.txt
@@ -0,0 +1,3 @@
+[console_scripts]
+markdown-it=markdown_it.cli.parse:main
+
diff --git a/falcon/lib/python3.10/site-packages/sympy/solvers/tests/__pycache__/test_solvers.cpython-310.pyc b/falcon/lib/python3.10/site-packages/sympy/solvers/tests/__pycache__/test_solvers.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..2539640ac1409fe1f35d429ed6c685544556a94f
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/sympy/solvers/tests/__pycache__/test_solvers.cpython-310.pyc
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afad2f29a076e249fa41ea0d7cbb5e86e036b8d353bbabc59cf76ab602f0a659
+size 108384
diff --git a/falcon/lib/python3.10/site-packages/sympy/solvers/tests/__pycache__/test_solveset.cpython-310.pyc b/falcon/lib/python3.10/site-packages/sympy/solvers/tests/__pycache__/test_solveset.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..7cddc6008be9b90b5f9cc9ffa1592cbfa8fa6e06
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/sympy/solvers/tests/__pycache__/test_solveset.cpython-310.pyc
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cffb50d71806115094deb9a1a7c186f4a23f5e92b8abce29126970bea29b7ed
+size 137706
diff --git a/falcon/lib/python3.10/site-packages/tests/__pycache__/__init__.cpython-310.pyc b/falcon/lib/python3.10/site-packages/tests/__pycache__/__init__.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..85dd4e78b9af042190e43694d838656eba501109
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/tests/__pycache__/__init__.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/tests/__pycache__/check_ic_formatting.cpython-310.pyc b/falcon/lib/python3.10/site-packages/tests/__pycache__/check_ic_formatting.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..e6e521be3f6bcbceda9687c1f3492bacbddf9435
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/tests/__pycache__/check_ic_formatting.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/tests/__pycache__/test_icecream.cpython-310.pyc b/falcon/lib/python3.10/site-packages/tests/__pycache__/test_icecream.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..5b30d0e289fa8d35b0706611fe482c1d9f518f46
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/tests/__pycache__/test_icecream.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/tests/__pycache__/test_install.cpython-310.pyc b/falcon/lib/python3.10/site-packages/tests/__pycache__/test_install.cpython-310.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..8e60e414e39645ab2ec018bd0beba175e31783b0
Binary files /dev/null and b/falcon/lib/python3.10/site-packages/tests/__pycache__/test_install.cpython-310.pyc differ
diff --git a/falcon/lib/python3.10/site-packages/tests/test_install.py b/falcon/lib/python3.10/site-packages/tests/test_install.py
new file mode 100644
index 0000000000000000000000000000000000000000..6860fd781ae94ae7c59b3323eaf52281d1377388
--- /dev/null
+++ b/falcon/lib/python3.10/site-packages/tests/test_install.py
@@ -0,0 +1,38 @@
+# -*- coding: utf-8 -*-
+
+#
+# IceCream - Never use print() to debug again
+#
+# Ansgar Grunseid
+# grunseid.com
+# grunseid@gmail.com
+#
+# License: MIT
+#
+
+import unittest
+
+import icecream
+from .test_icecream import (
+ disableColoring, captureStandardStreams, parseOutputIntoPairs)
+
+from .install_test_import import runMe
+
+
+class TestIceCreamInstall(unittest.TestCase):
+ def testInstall(self):
+ icecream.install()
+ with disableColoring(), captureStandardStreams() as (out, err):
+ runMe()
+ assert parseOutputIntoPairs(out, err, 1)[0][0] == ('x', '3')
+ icecream.uninstall() # Clean up builtins.
+
+ def testUninstall(self):
+ try:
+ icecream.uninstall()
+ except AttributeError: # Already uninstalled.
+ pass
+
+ # NameError: global name 'ic' is not defined.
+ with self.assertRaises(NameError):
+ runMe()