id
int64
0
328k
repository_name
stringlengths
7
58
file_path
stringlengths
9
302
class_name
stringlengths
5
256
human_written_code
stringlengths
16
2.16M
class_skeleton
stringlengths
18
1.49M
total_program_units
int64
1
1.76k
total_doc_str
int64
0
771
AvgCountLine
float64
0
7.89k
AvgCountLineBlank
float64
0
297
AvgCountLineCode
float64
0
7.89k
AvgCountLineComment
float64
0
7.89k
AvgCyclomatic
float64
0
130
CommentToCodeRatio
float64
0
168
CountClassBase
float64
0
40
CountClassCoupled
float64
0
583
CountClassCoupledModified
float64
0
575
CountClassDerived
float64
0
5.35k
CountDeclInstanceMethod
float64
0
529
CountDeclInstanceVariable
float64
0
296
CountDeclMethod
float64
0
599
CountDeclMethodAll
float64
0
1.12k
CountLine
float64
1
40.4k
CountLineBlank
float64
0
8.16k
CountLineCode
float64
1
25.7k
CountLineCodeDecl
float64
1
8.15k
CountLineCodeExe
float64
0
24.2k
CountLineComment
float64
0
16.5k
CountStmt
float64
1
9.71k
CountStmtDecl
float64
1
8.15k
CountStmtExe
float64
0
9.69k
MaxCyclomatic
float64
0
759
MaxInheritanceTree
float64
0
16
MaxNesting
float64
0
34
SumCyclomatic
float64
0
2.9k
4,700
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/pvt_v2/modeling_pvt_v2.py
transformers.models.pvt_v2.modeling_pvt_v2.PvtV2EncoderLayer
from ...modeling_layers import GradientCheckpointingLayer import torch from .configuration_pvt_v2 import PvtV2Config from torch import nn class PvtV2EncoderLayer(GradientCheckpointingLayer): def __init__(self, config: PvtV2Config, layer_idx: int): super().__init__() self.patch_embedding = PvtV2OverlapPatchEmbeddings(config=config, layer_idx=layer_idx) drop_path_decays = torch.linspace(0, config.drop_path_rate, sum(config.depths), device='cpu').tolist() block_layers = [] for block_idx in range(config.depths[layer_idx]): block_layers.append(PvtV2BlockLayer(config=config, layer_idx=layer_idx, drop_path=drop_path_decays[sum(config.depths[:layer_idx]) + block_idx])) self.blocks = nn.ModuleList(block_layers) self.layer_norm = nn.LayerNorm(config.hidden_sizes[layer_idx], eps=config.layer_norm_eps) def forward(self, hidden_states, output_attentions): all_self_attentions = () if output_attentions else None hidden_states, height, width = self.patch_embedding(hidden_states) for block in self.blocks: layer_outputs = block(hidden_states, height, width, output_attentions) hidden_states = layer_outputs[0] if output_attentions: all_self_attentions += (layer_outputs[1],) hidden_states = self.layer_norm(hidden_states) outputs = (hidden_states,) if output_attentions: outputs += (all_self_attentions,) return (outputs, height, width)
class PvtV2EncoderLayer(GradientCheckpointingLayer): def __init__(self, config: PvtV2Config, layer_idx: int): pass def forward(self, hidden_states, output_attentions): pass
3
0
21
2
16
3
4
0.19
1
6
3
0
2
3
2
12
43
5
32
14
29
6
23
14
20
5
1
2
7
4,701
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/pvt_v2/modeling_pvt_v2.py
transformers.models.pvt_v2.modeling_pvt_v2.PvtV2ForImageClassification
from ...utils import auto_docstring, logging from torch import nn from .configuration_pvt_v2 import PvtV2Config from ...modeling_outputs import BackboneOutput, BaseModelOutput, ImageClassifierOutput import torch from typing import Optional, Union @auto_docstring(custom_intro='\n Pvt-v2 Model transformer with an image classification head on top (a linear layer on top of the final hidden state\n of the [CLS] token) e.g. for ImageNet.\n ') class PvtV2ForImageClassification(PvtV2PreTrainedModel): def __init__(self, config: PvtV2Config) -> None: super().__init__(config) self.num_labels = config.num_labels self.pvt_v2 = PvtV2Model(config) self.classifier = nn.Linear(config.hidden_sizes[-1], config.num_labels) if config.num_labels > 0 else nn.Identity() self.post_init() @auto_docstring def forward(self, pixel_values: Optional[torch.Tensor], labels: Optional[torch.Tensor]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None) -> Union[tuple, ImageClassifierOutput]: """ 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.pvt_v2(pixel_values=pixel_values, output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=return_dict) sequence_output = outputs[0] batch_size = sequence_output.shape[0] sequence_output = sequence_output.permute(0, 2, 3, 1) sequence_output = sequence_output.reshape(batch_size, -1, self.config.hidden_sizes[-1]) sequence_output = sequence_output.mean(dim=1) logits = self.classifier(sequence_output) loss = None if labels is not None: loss = self.loss_function(labels, logits, self.config) 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)
@auto_docstring(custom_intro='\n Pvt-v2 Model transformer with an image classification head on top (a linear layer on top of the final hidden state\n of the [CLS] token) e.g. for ImageNet.\n ') class PvtV2ForImageClassification(PvtV2PreTrainedModel): def __init__(self, config: PvtV2Config) -> None: pass @auto_docstring def forward(self, pixel_values: Optional[torch.Tensor], labels: Optional[torch.Tensor]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None) -> Union[tuple, ImageClassifierOutput]: ''' 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). ''' pass
5
1
41
6
30
6
7
0.16
1
8
3
0
2
3
2
3
91
13
67
21
50
11
36
13
33
12
2
3
14
4,702
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/pvt_v2/modeling_pvt_v2.py
transformers.models.pvt_v2.modeling_pvt_v2.PvtV2Model
from ...utils import auto_docstring, logging import torch from .configuration_pvt_v2 import PvtV2Config from ...modeling_outputs import BackboneOutput, BaseModelOutput, ImageClassifierOutput from typing import Optional, Union @auto_docstring class PvtV2Model(PvtV2PreTrainedModel): def __init__(self, config: PvtV2Config): super().__init__(config) self.config = config self.encoder = PvtV2Encoder(config) self.post_init() 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) @auto_docstring def forward(self, pixel_values: torch.FloatTensor, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None) -> Union[tuple, BaseModelOutput]: 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 encoder_outputs = self.encoder(pixel_values=pixel_values, output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=return_dict) sequence_output = encoder_outputs[0] if not return_dict: return (sequence_output,) + encoder_outputs[1:] return BaseModelOutput(last_hidden_state=sequence_output, hidden_states=encoder_outputs.hidden_states, attentions=encoder_outputs.attentions)
@auto_docstring class PvtV2Model(PvtV2PreTrainedModel): def __init__(self, config: PvtV2Config): pass 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 ''' pass @auto_docstring def forward(self, pixel_values: torch.FloatTensor, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None) -> Union[tuple, BaseModelOutput]: pass
6
1
15
2
11
2
3
0.14
1
5
3
1
3
2
3
4
56
7
43
16
25
6
18
9
14
5
2
1
8
4,703
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/pvt_v2/modeling_pvt_v2.py
transformers.models.pvt_v2.modeling_pvt_v2.PvtV2OverlapPatchEmbeddings
from torch import nn from .configuration_pvt_v2 import PvtV2Config class PvtV2OverlapPatchEmbeddings(nn.Module): """Image to Patch Embedding""" def __init__(self, config: PvtV2Config, layer_idx: int): super().__init__() patch_size = config.patch_sizes[layer_idx] patch_size = (patch_size, patch_size) if isinstance(patch_size, int) else patch_size stride = config.strides[layer_idx] num_channels = config.num_channels if layer_idx == 0 else config.hidden_sizes[layer_idx - 1] hidden_size = config.hidden_sizes[layer_idx] self.patch_size = patch_size self.proj = nn.Conv2d(num_channels, hidden_size, kernel_size=patch_size, stride=stride, padding=(patch_size[0] // 2, patch_size[1] // 2)) self.layer_norm = nn.LayerNorm(hidden_size, eps=config.layer_norm_eps) def forward(self, pixel_values): embeddings = self.proj(pixel_values) _, _, height, width = embeddings.shape embeddings = embeddings.flatten(2).transpose(1, 2) embeddings = self.layer_norm(embeddings) return (embeddings, height, width)
class PvtV2OverlapPatchEmbeddings(nn.Module): '''Image to Patch Embedding''' def __init__(self, config: PvtV2Config, layer_idx: int): pass def forward(self, pixel_values): pass
3
1
11
0
11
0
2
0.04
1
3
1
0
2
3
2
12
26
2
23
12
20
1
17
12
14
3
1
0
4
4,704
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/pvt_v2/modeling_pvt_v2.py
transformers.models.pvt_v2.modeling_pvt_v2.PvtV2PreTrainedModel
import math from ...utils import auto_docstring, logging from typing import Optional, Union from torch import nn from .configuration_pvt_v2 import PvtV2Config from ...modeling_utils import PreTrainedModel @auto_docstring class PvtV2PreTrainedModel(PreTrainedModel): config: PvtV2Config base_model_prefix = 'pvt_v2' main_input_name = 'pixel_values' supports_gradient_checkpointing = True def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNorm]) -> None: """Initialize the weights""" if isinstance(module, nn.Linear): module.weight.data = nn.init.trunc_normal_(module.weight.data, mean=0.0, std=self.config.initializer_range) 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, nn.Conv2d): fan_out = module.kernel_size[0] * module.kernel_size[1] * module.out_channels fan_out //= module.groups module.weight.data.normal_(0, math.sqrt(2.0 / fan_out)) if module.bias is not None: module.bias.data.zero_()
@auto_docstring class PvtV2PreTrainedModel(PreTrainedModel): def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNorm]) -> None: '''Initialize the weights''' pass
3
1
17
0
14
3
6
0.37
1
0
0
2
1
0
1
1
28
2
19
7
17
7
17
7
15
6
1
2
6
4,705
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/pvt_v2/modeling_pvt_v2.py
transformers.models.pvt_v2.modeling_pvt_v2.PvtV2SelfAttention
from torch import nn import torch from .configuration_pvt_v2 import PvtV2Config from ...pytorch_utils import find_pruneable_heads_and_indices, prune_linear_layer import math class PvtV2SelfAttention(nn.Module): """Efficient self-attention mechanism.""" def __init__(self, config: PvtV2Config, hidden_size: int, num_attention_heads: int, spatial_reduction_ratio: int): super().__init__() self.linear_attention = config.linear_attention self.pruned_heads = set() self.hidden_size = hidden_size self.num_attention_heads = num_attention_heads if self.hidden_size % self.num_attention_heads != 0: raise ValueError(f'The hidden size ({self.hidden_size}) is not a multiple of the number of attention heads ({self.num_attention_heads})') self.attention_head_size = int(self.hidden_size / self.num_attention_heads) self.all_head_size = self.num_attention_heads * self.attention_head_size self.query = nn.Linear(self.hidden_size, self.all_head_size, bias=config.qkv_bias) self.key = nn.Linear(self.hidden_size, self.all_head_size, bias=config.qkv_bias) self.value = nn.Linear(self.hidden_size, self.all_head_size, bias=config.qkv_bias) self.attn_drop = nn.Dropout(config.attention_probs_dropout_prob) self.proj = nn.Linear(self.hidden_size, self.hidden_size) self.proj_drop = nn.Dropout(config.hidden_dropout_prob) self.spatial_reduction_ratio = spatial_reduction_ratio if self.linear_attention: self.pool = nn.AdaptiveAvgPool2d(7) self.spatial_reduction = nn.Conv2d(self.hidden_size, self.hidden_size, kernel_size=1, stride=1) self.layer_norm = nn.LayerNorm(self.hidden_size, eps=config.layer_norm_eps) self.act = nn.GELU() elif spatial_reduction_ratio > 1: self.spatial_reduction = nn.Conv2d(self.hidden_size, self.hidden_size, kernel_size=spatial_reduction_ratio, stride=spatial_reduction_ratio) self.layer_norm = nn.LayerNorm(self.hidden_size, eps=config.layer_norm_eps) def transpose_for_scores(self, hidden_states) -> torch.Tensor: new_shape = hidden_states.size()[:-1] + (self.num_attention_heads, self.attention_head_size) hidden_states = hidden_states.view(new_shape) return hidden_states.permute(0, 2, 1, 3) def forward(self, hidden_states: torch.Tensor, height: int, width: int, output_attentions: bool=False) -> tuple[torch.Tensor]: batch_size, seq_len, num_channels = hidden_states.shape query_layer = self.transpose_for_scores(self.query(hidden_states)) if self.linear_attention: hidden_states = hidden_states.permute(0, 2, 1).reshape(batch_size, num_channels, height, width) hidden_states = self.spatial_reduction(self.pool(hidden_states)).reshape(batch_size, num_channels, -1).permute(0, 2, 1) hidden_states = self.act(self.layer_norm(hidden_states)) elif self.spatial_reduction_ratio > 1: hidden_states = hidden_states.permute(0, 2, 1).reshape(batch_size, num_channels, height, width) hidden_states = self.spatial_reduction(hidden_states).reshape(batch_size, num_channels, -1).permute(0, 2, 1) hidden_states = self.layer_norm(hidden_states) key_layer = self.transpose_for_scores(self.key(hidden_states)) value_layer = self.transpose_for_scores(self.value(hidden_states)) attention_scores = torch.matmul(query_layer, key_layer.transpose(-1, -2)) attention_scores = attention_scores / math.sqrt(self.attention_head_size) attention_probs = nn.functional.softmax(attention_scores, dim=-1) attention_probs = self.attn_drop(attention_probs) context_layer = (attention_probs @ value_layer).transpose(1, 2).reshape(batch_size, seq_len, num_channels) context_layer = self.proj(context_layer) context_layer = self.proj_drop(context_layer) outputs = (context_layer, attention_probs) if output_attentions else (context_layer,) return outputs def prune_heads(self, heads): if len(heads) == 0: return heads, index = find_pruneable_heads_and_indices(heads, self.num_attention_heads, self.attention_head_size, self.pruned_heads) self.query = prune_linear_layer(self.query, index) self.key = prune_linear_layer(self.key, index) self.value = prune_linear_layer(self.value, index) self.proj = prune_linear_layer(self.proj, index, dim=1) self.num_attention_heads = self.num_attention_heads - len(heads) self.all_head_size = self.attention_head_size * self.num_attention_heads self.pruned_heads = self.pruned_heads.union(heads)
class PvtV2SelfAttention(nn.Module): '''Efficient self-attention mechanism.''' def __init__(self, config: PvtV2Config, hidden_size: int, num_attention_heads: int, spatial_reduction_ratio: int): pass def transpose_for_scores(self, hidden_states) -> torch.Tensor: pass def forward(self, hidden_states: torch.Tensor, height: int, width: int, output_attentions: bool=False) -> tuple[torch.Tensor]: pass def prune_heads(self, heads): pass
5
1
25
4
20
2
3
0.09
1
7
1
0
4
17
4
14
105
18
80
38
69
7
61
32
56
4
1
1
11
4,706
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/configuration_qwen2.py
transformers.models.qwen2.configuration_qwen2.Qwen2Config
from ...configuration_utils import PretrainedConfig, layer_type_validation from ...modeling_rope_utils import rope_config_validation class Qwen2Config(PretrainedConfig): """ This is the configuration class to store the configuration of a [`Qwen2Model`]. It is used to instantiate a Qwen2 model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of Qwen2-7B-beta [Qwen/Qwen2-7B-beta](https://huggingface.co/Qwen/Qwen2-7B-beta). 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 151936): Vocabulary size of the Qwen2 model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`Qwen2Model`] hidden_size (`int`, *optional*, defaults to 4096): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 22016): Dimension of the MLP representations. num_hidden_layers (`int`, *optional*, defaults to 32): Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 32): Number of attention heads for each attention layer in the Transformer encoder. num_key_value_heads (`int`, *optional*, defaults to 32): This is the number of key_value heads that should be used to implement Grouped Query Attention. If `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out [this paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `32`. hidden_act (`str` or `function`, *optional*, defaults to `"silu"`): The non-linear activation function (function or string) in the decoder. max_position_embeddings (`int`, *optional*, defaults to 32768): The maximum sequence length that this model might ever be used with. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. rms_norm_eps (`float`, *optional*, defaults to 1e-06): The epsilon used by the rms normalization layers. 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`. tie_word_embeddings (`bool`, *optional*, defaults to `False`): Whether the model's input and output word embeddings should be tied. rope_theta (`float`, *optional*, defaults to 10000.0): The base period of the RoPE embeddings. rope_scaling (`Dict`, *optional*): Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value accordingly. Expected contents: `rope_type` (`str`): The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope', 'llama3'], with 'default' being the original RoPE implementation. `factor` (`float`, *optional*): Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In most scaling types, a `factor` of x will enable the model to handle sequences of length x * original maximum pre-trained length. `original_max_position_embeddings` (`int`, *optional*): Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during pretraining. `attention_factor` (`float`, *optional*): Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention computation. If unspecified, it defaults to value recommended by the implementation, using the `factor` field to infer the suggested value. `beta_fast` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear ramp function. If unspecified, it defaults to 32. `beta_slow` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear ramp function. If unspecified, it defaults to 1. `short_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to short contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `long_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to long contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `low_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE `high_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE use_sliding_window (`bool`, *optional*, defaults to `False`): Whether to use sliding window attention. sliding_window (`int`, *optional*, defaults to 4096): Sliding window attention (SWA) window size. If not specified, will default to `4096`. max_window_layers (`int`, *optional*, defaults to 28): The number of layers using full attention. The first `max_window_layers` layers will use full attention, while any additional layer afterwards will use SWA (Sliding Window Attention). layer_types (`list`, *optional*): Attention pattern for each layer. attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. ```python >>> from transformers import Qwen2Model, Qwen2Config >>> # Initializing a Qwen2 style configuration >>> configuration = Qwen2Config() >>> # Initializing a model from the Qwen2-7B style configuration >>> model = Qwen2Model(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = 'qwen2' keys_to_ignore_at_inference = ['past_key_values'] base_model_tp_plan = {'layers.*.self_attn.q_proj': 'colwise', 'layers.*.self_attn.k_proj': 'colwise', 'layers.*.self_attn.v_proj': 'colwise', 'layers.*.self_attn.o_proj': 'rowwise', 'layers.*.mlp.gate_proj': 'colwise', 'layers.*.mlp.up_proj': 'colwise', 'layers.*.mlp.down_proj': 'rowwise'} base_model_pp_plan = {'embed_tokens': (['input_ids'], ['inputs_embeds']), 'layers': (['hidden_states', 'attention_mask'], ['hidden_states']), 'norm': (['hidden_states'], ['hidden_states'])} def __init__(self, vocab_size=151936, hidden_size=4096, intermediate_size=22016, num_hidden_layers=32, num_attention_heads=32, num_key_value_heads=32, hidden_act='silu', max_position_embeddings=32768, initializer_range=0.02, rms_norm_eps=1e-06, use_cache=True, tie_word_embeddings=False, rope_theta=10000.0, rope_scaling=None, use_sliding_window=False, sliding_window=4096, max_window_layers=28, layer_types=None, attention_dropout=0.0, **kwargs): self.vocab_size = vocab_size self.max_position_embeddings = max_position_embeddings self.hidden_size = hidden_size self.intermediate_size = intermediate_size self.num_hidden_layers = num_hidden_layers self.num_attention_heads = num_attention_heads self.use_sliding_window = use_sliding_window self.sliding_window = sliding_window if self.use_sliding_window else None self.max_window_layers = max_window_layers if num_key_value_heads is None: num_key_value_heads = num_attention_heads self.num_key_value_heads = num_key_value_heads self.hidden_act = hidden_act self.initializer_range = initializer_range self.rms_norm_eps = rms_norm_eps self.use_cache = use_cache self.rope_theta = rope_theta self.rope_scaling = rope_scaling self.attention_dropout = attention_dropout if self.rope_scaling is not None and 'type' in self.rope_scaling: self.rope_scaling['rope_type'] = self.rope_scaling['type'] rope_config_validation(self) self.layer_types = layer_types if self.layer_types is None: self.layer_types = ['sliding_attention' if self.sliding_window is not None and i >= self.max_window_layers else 'full_attention' for i in range(self.num_hidden_layers)] layer_type_validation(self.layer_types, self.num_hidden_layers) super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)
class Qwen2Config(PretrainedConfig): ''' This is the configuration class to store the configuration of a [`Qwen2Model`]. It is used to instantiate a Qwen2 model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of Qwen2-7B-beta [Qwen/Qwen2-7B-beta](https://huggingface.co/Qwen/Qwen2-7B-beta). 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 151936): Vocabulary size of the Qwen2 model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`Qwen2Model`] hidden_size (`int`, *optional*, defaults to 4096): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 22016): Dimension of the MLP representations. num_hidden_layers (`int`, *optional*, defaults to 32): Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 32): Number of attention heads for each attention layer in the Transformer encoder. num_key_value_heads (`int`, *optional*, defaults to 32): This is the number of key_value heads that should be used to implement Grouped Query Attention. If `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out [this paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `32`. hidden_act (`str` or `function`, *optional*, defaults to `"silu"`): The non-linear activation function (function or string) in the decoder. max_position_embeddings (`int`, *optional*, defaults to 32768): The maximum sequence length that this model might ever be used with. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. rms_norm_eps (`float`, *optional*, defaults to 1e-06): The epsilon used by the rms normalization layers. 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`. tie_word_embeddings (`bool`, *optional*, defaults to `False`): Whether the model's input and output word embeddings should be tied. rope_theta (`float`, *optional*, defaults to 10000.0): The base period of the RoPE embeddings. rope_scaling (`Dict`, *optional*): Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value accordingly. Expected contents: `rope_type` (`str`): The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope', 'llama3'], with 'default' being the original RoPE implementation. `factor` (`float`, *optional*): Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In most scaling types, a `factor` of x will enable the model to handle sequences of length x * original maximum pre-trained length. `original_max_position_embeddings` (`int`, *optional*): Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during pretraining. `attention_factor` (`float`, *optional*): Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention computation. If unspecified, it defaults to value recommended by the implementation, using the `factor` field to infer the suggested value. `beta_fast` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear ramp function. If unspecified, it defaults to 32. `beta_slow` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear ramp function. If unspecified, it defaults to 1. `short_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to short contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `long_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to long contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `low_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE `high_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE use_sliding_window (`bool`, *optional*, defaults to `False`): Whether to use sliding window attention. sliding_window (`int`, *optional*, defaults to 4096): Sliding window attention (SWA) window size. If not specified, will default to `4096`. max_window_layers (`int`, *optional*, defaults to 28): The number of layers using full attention. The first `max_window_layers` layers will use full attention, while any additional layer afterwards will use SWA (Sliding Window Attention). layer_types (`list`, *optional*): Attention pattern for each layer. attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. ```python >>> from transformers import Qwen2Model, Qwen2Config >>> # Initializing a Qwen2 style configuration >>> configuration = Qwen2Config() >>> # Initializing a model from the Qwen2-7B style configuration >>> model = Qwen2Model(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```''' def __init__(self, vocab_size=151936, hidden_size=4096, intermediate_size=22016, num_hidden_layers=32, num_attention_heads=32, num_key_value_heads=32, hidden_act='silu', max_position_embeddings=32768, initializer_range=0.02, rms_norm_eps=1e-06, use_cache=True, tie_word_embeddings=False, rope_theta=10000.0, rope_scaling=None, use_sliding_window=False, sliding_window=4096, max_window_layers=28, layer_types=None, attention_dropout=0.0, **kwargs): pass
2
1
54
3
48
3
4
1.65
1
1
0
0
1
17
1
1
172
13
60
43
37
99
28
22
26
4
1
1
4
4,707
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2Attention
from typing import Callable, Optional, Union import torch from ...processing_utils import Unpack from ...cache_utils import Cache, DynamicCache from ...modeling_flash_attention_utils import FlashAttentionKwargs from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel from ...utils.deprecation import deprecate_kwarg from torch import nn from .configuration_qwen2 import Qwen2Config class Qwen2Attention(nn.Module): """Multi-headed attention from 'Attention Is All You Need' paper""" def __init__(self, config: Qwen2Config, layer_idx: int): super().__init__() self.config = config self.layer_idx = layer_idx self.head_dim = getattr(config, 'head_dim', config.hidden_size // config.num_attention_heads) self.num_key_value_groups = config.num_attention_heads // config.num_key_value_heads self.scaling = self.head_dim ** (-0.5) self.attention_dropout = config.attention_dropout self.is_causal = True self.q_proj = nn.Linear(config.hidden_size, config.num_attention_heads * self.head_dim, bias=True) self.k_proj = nn.Linear(config.hidden_size, config.num_key_value_heads * self.head_dim, bias=True) self.v_proj = nn.Linear(config.hidden_size, config.num_key_value_heads * self.head_dim, bias=True) self.o_proj = nn.Linear(config.num_attention_heads * self.head_dim, config.hidden_size, bias=False) self.sliding_window = config.sliding_window if config.layer_types[layer_idx] == 'sliding_attention' else None @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, position_embeddings: tuple[torch.Tensor, torch.Tensor], attention_mask: Optional[torch.Tensor], past_key_values: Optional[Cache]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.Tensor, Optional[torch.Tensor]]: input_shape = hidden_states.shape[:-1] hidden_shape = (*input_shape, -1, self.head_dim) query_states = self.q_proj(hidden_states).view(hidden_shape).transpose(1, 2) key_states = self.k_proj(hidden_states).view(hidden_shape).transpose(1, 2) value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2) cos, sin = position_embeddings query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin) if past_key_values is not None: cache_kwargs = {'sin': sin, 'cos': cos, 'cache_position': cache_position} key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs) attention_interface: Callable = eager_attention_forward if self.config._attn_implementation != 'eager': attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation] attn_output, attn_weights = attention_interface(self, query_states, key_states, value_states, attention_mask, dropout=0.0 if not self.training else self.attention_dropout, scaling=self.scaling, sliding_window=self.sliding_window, **kwargs) attn_output = attn_output.reshape(*input_shape, -1).contiguous() attn_output = self.o_proj(attn_output) return (attn_output, attn_weights)
class Qwen2Attention(nn.Module): '''Multi-headed attention from 'Attention Is All You Need' paper''' def __init__(self, config: Qwen2Config, layer_idx: int): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, position_embeddings: tuple[torch.Tensor, torch.Tensor], attention_mask: Optional[torch.Tensor], past_key_values: Optional[Cache]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.Tensor, Optional[torch.Tensor]]: pass
4
1
35
4
31
1
4
0.05
1
6
3
0
2
11
2
12
74
9
63
32
52
3
37
24
34
6
1
2
7
4,708
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2DecoderLayer
import torch from typing import Callable, Optional, Union from ...utils.deprecation import deprecate_kwarg from ...modeling_layers import GenericForQuestionAnswering, GenericForSequenceClassification, GenericForTokenClassification, GradientCheckpointingLayer from ...cache_utils import Cache, DynamicCache from ...processing_utils import Unpack from .configuration_qwen2 import Qwen2Config from ...utils import TransformersKwargs, auto_docstring, can_return_tuple class Qwen2DecoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2Config, layer_idx: int): super().__init__() self.hidden_size = config.hidden_size self.self_attn = Qwen2Attention(config=config, layer_idx=layer_idx) self.mlp = Qwen2MLP(config) self.input_layernorm = Qwen2RMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.post_attention_layernorm = Qwen2RMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.attention_type = config.layer_types[layer_idx] @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, use_cache: Optional[bool]=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[TransformersKwargs]) -> torch.Tensor: residual = hidden_states hidden_states = self.input_layernorm(hidden_states) hidden_states, _ = self.self_attn(hidden_states=hidden_states, attention_mask=attention_mask, position_ids=position_ids, past_key_values=past_key_values, use_cache=use_cache, cache_position=cache_position, position_embeddings=position_embeddings, **kwargs) hidden_states = residual + hidden_states residual = hidden_states hidden_states = self.post_attention_layernorm(hidden_states) hidden_states = self.mlp(hidden_states) hidden_states = residual + hidden_states return hidden_states
class Qwen2DecoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2Config, layer_idx: int): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, use_cache: Optional[bool]=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[TransformersKwargs]) -> torch.Tensor: pass
4
0
27
3
23
2
2
0.06
1
10
6
0
2
5
2
12
55
6
47
22
33
3
23
11
20
2
1
1
4
4,709
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2ForCausalLM
from ...generation import GenerationMixin from typing import Callable, Optional, Union from ...cache_utils import Cache, DynamicCache from ...utils import TransformersKwargs, auto_docstring, can_return_tuple from torch import nn from ...modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast import torch from ...processing_utils import Unpack @auto_docstring class Qwen2ForCausalLM(Qwen2PreTrainedModel, GenerationMixin): _tied_weights_keys = ['lm_head.weight'] _tp_plan = {'lm_head': 'colwise_rep'} _pp_plan = {'lm_head': (['hidden_states'], ['logits'])} def __init__(self, config): super().__init__(config) self.model = Qwen2Model(config) self.vocab_size = config.vocab_size self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False) self.post_init() @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None, logits_to_keep: Union[int, torch.Tensor]=0, **kwargs: Unpack[TransformersKwargs]) -> CausalLMOutputWithPast: """ Example: ```python >>> from transformers import AutoTokenizer, Qwen2ForCausalLM >>> model = Qwen2ForCausalLM.from_pretrained("meta-qwen2/Qwen2-2-7b-hf") >>> tokenizer = AutoTokenizer.from_pretrained("meta-qwen2/Qwen2-2-7b-hf") >>> prompt = "Hey, are you conscious? Can you talk to me?" >>> inputs = tokenizer(prompt, return_tensors="pt") >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "Hey, are you conscious? Can you talk to me?\\nI'm not conscious, but I can talk to you." ```""" outputs: BaseModelOutputWithPast = self.model(input_ids=input_ids, attention_mask=attention_mask, position_ids=position_ids, past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache, cache_position=cache_position, **kwargs) hidden_states = outputs.last_hidden_state slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep logits = self.lm_head(hidden_states[:, slice_indices, :]) loss = None if labels is not None: loss = self.loss_function(logits=logits, labels=labels, vocab_size=self.config.vocab_size, **kwargs) return CausalLMOutputWithPast(loss=loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions)
@auto_docstring class Qwen2ForCausalLM(Qwen2PreTrainedModel, GenerationMixin): def __init__(self, config): pass @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None, logits_to_keep: Union[int, torch.Tensor]=0, **kwargs: Unpack[TransformersKwargs]) -> CausalLMOutputWithPast: ''' Example: ```python >>> from transformers import AutoTokenizer, Qwen2ForCausalLM >>> model = Qwen2ForCausalLM.from_pretrained("meta-qwen2/Qwen2-2-7b-hf") >>> tokenizer = AutoTokenizer.from_pretrained("meta-qwen2/Qwen2-2-7b-hf") >>> prompt = "Hey, are you conscious? Can you talk to me?" >>> inputs = tokenizer(prompt, return_tensors="pt") >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you." ```''' pass
6
1
14
2
9
4
2
0.38
2
9
4
0
8
3
8
9
123
21
74
36
47
28
36
20
27
8
2
1
15
4,710
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2ForQuestionAnswering
from ...modeling_layers import GenericForQuestionAnswering, GenericForSequenceClassification, GenericForTokenClassification, GradientCheckpointingLayer class Qwen2ForQuestionAnswering(GenericForQuestionAnswering, Qwen2PreTrainedModel): base_model_prefix = 'transformer'
class Qwen2ForQuestionAnswering(GenericForQuestionAnswering, Qwen2PreTrainedModel): pass
1
0
18
2
13
3
2
0.2
1
5
3
0
4
2
4
5
77
11
55
28
36
11
26
14
21
5
2
1
8
4,711
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2ForSequenceClassification
from ...modeling_layers import GenericForQuestionAnswering, GenericForSequenceClassification, GenericForTokenClassification, GradientCheckpointingLayer class Qwen2ForSequenceClassification(GenericForSequenceClassification, Qwen2PreTrainedModel): pass
class Qwen2ForSequenceClassification(GenericForSequenceClassification, Qwen2PreTrainedModel): pass
1
0
21
2
17
2
3
0.11
1
7
3
0
4
3
4
5
90
11
71
31
53
8
36
18
31
9
2
1
12
4,712
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2ForTokenClassification
from ...modeling_layers import GenericForQuestionAnswering, GenericForSequenceClassification, GenericForTokenClassification, GradientCheckpointingLayer class Qwen2ForTokenClassification(GenericForTokenClassification, Qwen2PreTrainedModel): pass
class Qwen2ForTokenClassification(GenericForTokenClassification, Qwen2PreTrainedModel): pass
1
0
17
1
14
2
3
0.11
1
5
2
0
4
4
4
5
79
8
64
28
41
7
29
15
24
5
2
1
10
4,713
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2MLP
from torch import nn from ...activations import ACT2FN class Qwen2MLP(nn.Module): def __init__(self, config): super().__init__() self.config = config self.hidden_size = config.hidden_size self.intermediate_size = config.intermediate_size self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False) self.act_fn = ACT2FN[config.hidden_act] def forward(self, x): down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x)) return down_proj
class Qwen2MLP(nn.Module): def __init__(self, config): pass def forward(self, x): pass
3
0
6
0
6
0
1
0
1
1
0
0
2
7
2
12
14
1
13
11
10
0
13
11
10
1
1
0
2
4,714
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2Model
from ...processing_utils import Unpack from ...utils.generic import check_model_inputs import torch from torch import nn from ...modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast from ...utils import TransformersKwargs, auto_docstring, can_return_tuple from ...cache_utils import Cache, DynamicCache from ...masking_utils import create_causal_mask, create_sliding_window_causal_mask from typing import Callable, Optional, Union from .configuration_qwen2 import Qwen2Config @auto_docstring class Qwen2Model(Qwen2PreTrainedModel): def __init__(self, config: Qwen2Config): super().__init__(config) self.padding_idx = config.pad_token_id self.vocab_size = config.vocab_size self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx) self.layers = nn.ModuleList([Qwen2DecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]) self.norm = Qwen2RMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.rotary_emb = Qwen2RotaryEmbedding(config=config) self.gradient_checkpointing = False self.has_sliding_layers = 'sliding_attention' in self.config.layer_types self.post_init() @check_model_inputs @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[TransformersKwargs]) -> BaseModelOutputWithPast: if (input_ids is None) ^ (inputs_embeds is not None): raise ValueError('You must specify exactly one of input_ids or inputs_embeds') if inputs_embeds is None: inputs_embeds = self.embed_tokens(input_ids) if use_cache and past_key_values is None: past_key_values = DynamicCache(config=self.config) if cache_position is None: past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0 cache_position = torch.arange(past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device) if position_ids is None: position_ids = cache_position.unsqueeze(0) if not isinstance((causal_mask_mapping := attention_mask), dict): mask_kwargs = {'config': self.config, 'input_embeds': inputs_embeds, 'attention_mask': attention_mask, 'cache_position': cache_position, 'past_key_values': past_key_values, 'position_ids': position_ids} causal_mask_mapping = {'full_attention': create_causal_mask(**mask_kwargs)} if self.has_sliding_layers: causal_mask_mapping['sliding_attention'] = create_sliding_window_causal_mask(**mask_kwargs) hidden_states = inputs_embeds position_embeddings = self.rotary_emb(hidden_states, position_ids) for decoder_layer in self.layers[:self.config.num_hidden_layers]: hidden_states = decoder_layer(hidden_states, attention_mask=causal_mask_mapping[decoder_layer.attention_type], position_ids=position_ids, past_key_values=past_key_values, use_cache=use_cache, cache_position=cache_position, position_embeddings=position_embeddings, **kwargs) hidden_states = self.norm(hidden_states) return BaseModelOutputWithPast(last_hidden_state=hidden_states, past_key_values=past_key_values if use_cache else None)
@auto_docstring class Qwen2Model(Qwen2PreTrainedModel): def __init__(self, config: Qwen2Config): pass @check_model_inputs @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[TransformersKwargs]) -> BaseModelOutputWithPast: pass
6
0
40
5
30
6
6
0.22
1
16
10
0
5
7
6
7
257
34
184
65
146
40
89
34
82
21
2
2
37
4,715
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2PreTrainedModel
from ...utils import TransformersKwargs, auto_docstring, can_return_tuple from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel from .configuration_qwen2 import Qwen2Config @auto_docstring class Qwen2PreTrainedModel(PreTrainedModel): config: Qwen2Config base_model_prefix = 'model' supports_gradient_checkpointing = True _no_split_modules = ['Qwen2DecoderLayer'] _skip_keys_device_placement = ['past_key_values'] _supports_flash_attn = True _supports_sdpa = True _supports_flex_attn = True _can_compile_fullgraph = True _supports_attention_backend = True _can_record_outputs = {'hidden_states': Qwen2DecoderLayer, 'attentions': Qwen2Attention}
@auto_docstring class Qwen2PreTrainedModel(PreTrainedModel): pass
2
0
10
0
10
0
5
0
1
0
0
5
1
0
1
1
24
1
23
15
21
0
22
15
20
5
1
2
5
4,716
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2RMSNorm
from torch import nn import torch from ...integrations import use_kernel_forward_from_hub @use_kernel_forward_from_hub('RMSNorm') class Qwen2RMSNorm(nn.Module): def __init__(self, hidden_size, eps: float=1e-06) -> None: """ Qwen2RMSNorm is equivalent to T5LayerNorm """ super().__init__() self.weight = nn.Parameter(torch.ones(hidden_size)) self.variance_epsilon = eps def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: input_dtype = hidden_states.dtype hidden_states = hidden_states.to(torch.float32) variance = hidden_states.pow(2).mean(-1, keepdim=True) hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon) return self.weight * hidden_states.to(input_dtype) def extra_repr(self): return f'{tuple(self.weight.shape)}, eps={self.variance_epsilon}'
@use_kernel_forward_from_hub('RMSNorm') class Qwen2RMSNorm(nn.Module): def __init__(self, hidden_size, eps: float=1e-06) -> None: ''' Qwen2RMSNorm is equivalent to T5LayerNorm ''' pass def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: pass def extra_repr(self): pass
5
1
5
0
4
1
1
0.23
1
2
0
0
3
2
3
13
18
2
13
8
9
3
13
8
9
1
1
0
3
4,717
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modeling_qwen2.py
transformers.models.qwen2.modeling_qwen2.Qwen2RotaryEmbedding
from ...modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update import torch from .configuration_qwen2 import Qwen2Config from torch import nn class Qwen2RotaryEmbedding(nn.Module): inv_freq: torch.Tensor def __init__(self, config: Qwen2Config, device=None): super().__init__() if hasattr(config, 'rope_scaling') and isinstance(config.rope_scaling, dict): self.rope_type = config.rope_scaling.get('rope_type', config.rope_scaling.get('type')) else: self.rope_type = 'default' self.max_seq_len_cached = config.max_position_embeddings self.original_max_seq_len = config.max_position_embeddings self.config = config self.rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type] inv_freq, self.attention_scaling = self.rope_init_fn(self.config, device) self.register_buffer('inv_freq', inv_freq, persistent=False) self.original_inv_freq = self.inv_freq @torch.no_grad() @dynamic_rope_update def forward(self, x, position_ids): inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1).to(x.device) position_ids_expanded = position_ids[:, None, :].float() device_type = x.device.type if isinstance(x.device.type, str) and x.device.type != 'mps' else 'cpu' with torch.autocast(device_type=device_type, enabled=False): freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2) emb = torch.cat((freqs, freqs), dim=-1) cos = emb.cos() * self.attention_scaling sin = emb.sin() * self.attention_scaling return (cos.to(dtype=x.dtype), sin.to(dtype=x.dtype))
class Qwen2RotaryEmbedding(nn.Module): def __init__(self, config: Qwen2Config, device=None): pass @torch.no_grad() @dynamic_rope_update def forward(self, x, position_ids): pass
5
0
18
2
13
5
3
0.35
1
4
1
0
3
7
3
13
59
8
40
21
35
14
38
20
34
3
1
1
8
4,718
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modular_qwen2.py
transformers.models.qwen2.modular_qwen2.Qwen2Attention
from ..llama.modeling_llama import LlamaAttention, LlamaDecoderLayer, LlamaForCausalLM, LlamaForQuestionAnswering, LlamaForSequenceClassification, LlamaForTokenClassification, LlamaMLP, LlamaPreTrainedModel, apply_rotary_pos_emb, eager_attention_forward from typing import Callable, Optional from ...processing_utils import Unpack from ...cache_utils import Cache, DynamicCache from ...utils.deprecation import deprecate_kwarg from torch import nn from ...modeling_utils import ALL_ATTENTION_FUNCTIONS from .configuration_qwen2 import Qwen2Config from ...modeling_flash_attention_utils import FlashAttentionKwargs import torch class Qwen2Attention(LlamaAttention): def __init__(self, config: Qwen2Config, layer_idx: int): super().__init__(config, layer_idx) self.q_proj = nn.Linear(config.hidden_size, config.num_attention_heads * self.head_dim, bias=True) self.k_proj = nn.Linear(config.hidden_size, config.num_key_value_heads * self.head_dim, bias=True) self.v_proj = nn.Linear(config.hidden_size, config.num_key_value_heads * self.head_dim, bias=True) self.o_proj = nn.Linear(config.num_attention_heads * self.head_dim, config.hidden_size, bias=False) self.sliding_window = config.sliding_window if config.layer_types[layer_idx] == 'sliding_attention' else None @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, position_embeddings: tuple[torch.Tensor, torch.Tensor], attention_mask: Optional[torch.Tensor], past_key_values: Optional[Cache]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.Tensor, Optional[torch.Tensor]]: input_shape = hidden_states.shape[:-1] hidden_shape = (*input_shape, -1, self.head_dim) query_states = self.q_proj(hidden_states).view(hidden_shape).transpose(1, 2) key_states = self.k_proj(hidden_states).view(hidden_shape).transpose(1, 2) value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2) cos, sin = position_embeddings query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin) if past_key_values is not None: cache_kwargs = {'sin': sin, 'cos': cos, 'cache_position': cache_position} key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs) attention_interface: Callable = eager_attention_forward if self.config._attn_implementation != 'eager': attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation] attn_output, attn_weights = attention_interface(self, query_states, key_states, value_states, attention_mask, dropout=0.0 if not self.training else self.attention_dropout, scaling=self.scaling, sliding_window=self.sliding_window, **kwargs) attn_output = attn_output.reshape(*input_shape, -1).contiguous() attn_output = self.o_proj(attn_output) return (attn_output, attn_weights)
class Qwen2Attention(LlamaAttention): def __init__(self, config: Qwen2Config, layer_idx: int): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, position_embeddings: tuple[torch.Tensor, torch.Tensor], attention_mask: Optional[torch.Tensor], past_key_values: Optional[Cache]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.Tensor, Optional[torch.Tensor]]: pass
4
0
32
4
28
1
4
0.04
1
6
3
0
2
7
2
14
65
8
56
27
45
2
30
17
27
6
2
2
7
4,719
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modular_qwen2.py
transformers.models.qwen2.modular_qwen2.Qwen2DecoderLayer
from ..llama.modeling_llama import LlamaAttention, LlamaDecoderLayer, LlamaForCausalLM, LlamaForQuestionAnswering, LlamaForSequenceClassification, LlamaForTokenClassification, LlamaMLP, LlamaPreTrainedModel, apply_rotary_pos_emb, eager_attention_forward from .configuration_qwen2 import Qwen2Config class Qwen2DecoderLayer(LlamaDecoderLayer): def __init__(self, config: Qwen2Config, layer_idx: int): super().__init__(config=config, layer_idx=layer_idx) self.attention_type = config.layer_types[layer_idx]
class Qwen2DecoderLayer(LlamaDecoderLayer): def __init__(self, config: Qwen2Config, layer_idx: int): pass
2
0
9
0
9
0
2
0
1
5
3
0
1
2
1
13
10
0
10
4
8
0
7
4
5
2
2
1
2
4,720
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modular_qwen2.py
transformers.models.qwen2.modular_qwen2.Qwen2ForCausalLM
from ..llama.modeling_llama import LlamaAttention, LlamaDecoderLayer, LlamaForCausalLM, LlamaForQuestionAnswering, LlamaForSequenceClassification, LlamaForTokenClassification, LlamaMLP, LlamaPreTrainedModel, apply_rotary_pos_emb, eager_attention_forward class Qwen2ForCausalLM(LlamaForCausalLM): pass
class Qwen2ForCausalLM(LlamaForCausalLM): pass
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
9
2
0
2
1
1
0
2
1
1
0
3
0
0
4,721
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modular_qwen2.py
transformers.models.qwen2.modular_qwen2.Qwen2ForQuestionAnswering
from ..llama.modeling_llama import LlamaAttention, LlamaDecoderLayer, LlamaForCausalLM, LlamaForQuestionAnswering, LlamaForSequenceClassification, LlamaForTokenClassification, LlamaMLP, LlamaPreTrainedModel, apply_rotary_pos_emb, eager_attention_forward class Qwen2ForQuestionAnswering(LlamaForQuestionAnswering): pass
class Qwen2ForQuestionAnswering(LlamaForQuestionAnswering): pass
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
5
2
0
2
1
1
0
2
1
1
0
3
0
0
4,722
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modular_qwen2.py
transformers.models.qwen2.modular_qwen2.Qwen2ForSequenceClassification
from ..llama.modeling_llama import LlamaAttention, LlamaDecoderLayer, LlamaForCausalLM, LlamaForQuestionAnswering, LlamaForSequenceClassification, LlamaForTokenClassification, LlamaMLP, LlamaPreTrainedModel, apply_rotary_pos_emb, eager_attention_forward class Qwen2ForSequenceClassification(LlamaForSequenceClassification): pass
class Qwen2ForSequenceClassification(LlamaForSequenceClassification): pass
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
5
2
0
2
1
1
0
2
1
1
0
3
0
0
4,723
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modular_qwen2.py
transformers.models.qwen2.modular_qwen2.Qwen2ForTokenClassification
from ..llama.modeling_llama import LlamaAttention, LlamaDecoderLayer, LlamaForCausalLM, LlamaForQuestionAnswering, LlamaForSequenceClassification, LlamaForTokenClassification, LlamaMLP, LlamaPreTrainedModel, apply_rotary_pos_emb, eager_attention_forward class Qwen2ForTokenClassification(LlamaForTokenClassification): pass
class Qwen2ForTokenClassification(LlamaForTokenClassification): pass
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
5
2
0
2
1
1
0
2
1
1
0
3
0
0
4,724
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modular_qwen2.py
transformers.models.qwen2.modular_qwen2.Qwen2MLP
from torch import nn from ..llama.modeling_llama import LlamaAttention, LlamaDecoderLayer, LlamaForCausalLM, LlamaForQuestionAnswering, LlamaForSequenceClassification, LlamaForTokenClassification, LlamaMLP, LlamaPreTrainedModel, apply_rotary_pos_emb, eager_attention_forward class Qwen2MLP(LlamaMLP): def __init__(self, config): super().__init__(config) self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
class Qwen2MLP(LlamaMLP): def __init__(self, config): pass
2
0
5
0
5
0
1
0
1
1
0
0
1
5
1
13
6
0
6
5
4
0
6
5
4
1
2
0
1
4,725
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/modular_qwen2.py
transformers.models.qwen2.modular_qwen2.Qwen2Model
import torch from ...modeling_outputs import BaseModelOutputWithPast from ...utils import TransformersKwargs, auto_docstring, logging from ..mistral.modeling_mistral import MistralModel from ...cache_utils import Cache, DynamicCache from ...masking_utils import create_causal_mask, create_sliding_window_causal_mask from ...processing_utils import Unpack from typing import Callable, Optional from ...utils.generic import check_model_inputs from .configuration_qwen2 import Qwen2Config class Qwen2Model(MistralModel): def __init__(self, config: Qwen2Config): super().__init__(config) self.has_sliding_layers = 'sliding_attention' in self.config.layer_types @check_model_inputs @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[TransformersKwargs]) -> BaseModelOutputWithPast: if (input_ids is None) ^ (inputs_embeds is not None): raise ValueError('You must specify exactly one of input_ids or inputs_embeds') if inputs_embeds is None: inputs_embeds = self.embed_tokens(input_ids) if use_cache and past_key_values is None: past_key_values = DynamicCache(config=self.config) if cache_position is None: past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0 cache_position = torch.arange(past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device) if position_ids is None: position_ids = cache_position.unsqueeze(0) if not isinstance((causal_mask_mapping := attention_mask), dict): mask_kwargs = {'config': self.config, 'input_embeds': inputs_embeds, 'attention_mask': attention_mask, 'cache_position': cache_position, 'past_key_values': past_key_values, 'position_ids': position_ids} causal_mask_mapping = {'full_attention': create_causal_mask(**mask_kwargs)} if self.has_sliding_layers: causal_mask_mapping['sliding_attention'] = create_sliding_window_causal_mask(**mask_kwargs) hidden_states = inputs_embeds position_embeddings = self.rotary_emb(hidden_states, position_ids) for decoder_layer in self.layers[:self.config.num_hidden_layers]: hidden_states = decoder_layer(hidden_states, attention_mask=causal_mask_mapping[decoder_layer.attention_type], position_ids=position_ids, past_key_values=past_key_values, use_cache=use_cache, cache_position=cache_position, position_embeddings=position_embeddings, **kwargs) hidden_states = self.norm(hidden_states) return BaseModelOutputWithPast(last_hidden_state=hidden_states, past_key_values=past_key_values if use_cache else None)
class Qwen2Model(MistralModel): def __init__(self, config: Qwen2Config): pass @check_model_inputs @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[TransformersKwargs]) -> BaseModelOutputWithPast: pass
5
0
0
0
0
0
0
0
1
0
0
0
0
0
0
7
2
0
2
1
1
0
2
1
1
0
3
0
0
4,726
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/tokenization_qwen2.py
transformers.models.qwen2.tokenization_qwen2.Qwen2Tokenizer
from typing import Optional import os from ...tokenization_utils import AddedToken, PreTrainedTokenizer import unicodedata import regex as re import json class Qwen2Tokenizer(PreTrainedTokenizer): """ Construct a Qwen2 tokenizer. Based on byte-level Byte-Pair-Encoding. Same with GPT2Tokenizer, this tokenizer has been trained to treat spaces like parts of the tokens so a word will be encoded differently whether it is at the beginning of the sentence (without space) or not: ```python >>> from transformers import Qwen2Tokenizer >>> tokenizer = Qwen2Tokenizer.from_pretrained("Qwen/Qwen-tokenizer") >>> tokenizer("Hello world")["input_ids"] [9707, 1879] >>> tokenizer(" Hello world")["input_ids"] [21927, 1879] ``` This is expected. You should not use GPT2Tokenizer instead, because of the different pretokenization rules. 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`): 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. unk_token (`str`, *optional*, defaults to `"<|endoftext|>"`): 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. bos_token (`str`, *optional*): The beginning of sequence token. Not applicable for this tokenizer. eos_token (`str`, *optional*, defaults to `"<|endoftext|>"`): The end of sequence token. pad_token (`str`, *optional*, defaults to `"<|endoftext|>"`): The token used for padding, for example when batching sequences of different lengths. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not the model should cleanup the spaces that were added when splitting the input text during the tokenization process. Not applicable to this tokenizer, since tokenization does not add spaces. split_special_tokens (`bool`, *optional*, defaults to `False`): Whether or not the special tokens should be split during the tokenization process. The default behavior is to not split special tokens. This means that if `<|endoftext|>` is the `eos_token`, then `tokenizer.tokenize("<|endoftext|>") = ['<|endoftext|>`]. Otherwise, if `split_special_tokens=True`, then `tokenizer.tokenize("<|endoftext|>")` will be give `['<', '|', 'endo', 'ft', 'ext', '|', '>']`. This argument is only supported for `slow` tokenizers for the moment. """ vocab_files_names = VOCAB_FILES_NAMES model_input_names = ['input_ids', 'attention_mask'] def __init__(self, vocab_file, merges_file, errors='replace', unk_token='<|endoftext|>', bos_token=None, eos_token='<|endoftext|>', pad_token='<|endoftext|>', clean_up_tokenization_spaces=False, split_special_tokens=False, **kwargs): bos_token = AddedToken(bos_token, lstrip=False, rstrip=False, special=True, normalized=False) if isinstance(bos_token, str) else bos_token eos_token = AddedToken(eos_token, lstrip=False, rstrip=False, special=True, normalized=False) if isinstance(eos_token, str) else eos_token unk_token = AddedToken(unk_token, lstrip=False, rstrip=False, special=True, normalized=False) if isinstance(unk_token, str) else unk_token pad_token = AddedToken(pad_token, lstrip=False, rstrip=False, special=True, normalized=False) if isinstance(pad_token, str) else pad_token with open(vocab_file, encoding='utf-8') as vocab_handle: self.encoder = json.load(vocab_handle) self.decoder = {v: k for k, v in self.encoder.items()} self.errors = errors self.byte_encoder = bytes_to_unicode() self.byte_decoder = {v: k for k, v in self.byte_encoder.items()} bpe_merges = [] with open(merges_file, encoding='utf-8') as merges_handle: for i, line in enumerate(merges_handle): line = line.strip() if i == 0 and line.startswith('#version:') or not line: continue bpe_merges.append(tuple(line.split())) self.bpe_ranks = dict(zip(bpe_merges, range(len(bpe_merges)))) self.cache = {} self.pat = re.compile(PRETOKENIZE_REGEX) if kwargs.get('add_prefix_space', False): logger.warning_once(f'{self.__class__.__name} does not support `add_prefix_space`, setting it to True has no effect.') super().__init__(errors=errors, bos_token=bos_token, eos_token=eos_token, pad_token=pad_token, unk_token=unk_token, clean_up_tokenization_spaces=clean_up_tokenization_spaces, split_special_tokens=split_special_tokens, **kwargs) @property def vocab_size(self) -> int: return len(self.encoder) def get_vocab(self): return dict(self.encoder, **self.added_tokens_encoder) def bpe(self, token): if token in self.cache: return self.cache[token] word = tuple(token) pairs = get_pairs(word) if not pairs: return token while True: bigram = min(pairs, key=lambda pair: self.bpe_ranks.get(pair, float('inf'))) if bigram not in self.bpe_ranks: break first, second = bigram new_word = [] i = 0 while i < len(word): try: j = word.index(first, i) except ValueError: new_word.extend(word[i:]) break else: new_word.extend(word[i:j]) i = j if word[i] == first and i < len(word) - 1 and (word[i + 1] == second): new_word.append(first + second) i += 2 else: new_word.append(word[i]) i += 1 new_word = tuple(new_word) word = new_word if len(word) == 1: break else: pairs = get_pairs(word) word = ' '.join(word) self.cache[token] = word return word def _tokenize(self, text): """Tokenize a string.""" bpe_tokens = [] for token in re.findall(self.pat, text): token = ''.join((self.byte_encoder[b] for b in token.encode('utf-8'))) bpe_tokens.extend((bpe_token for bpe_token in self.bpe(token).split(' '))) return bpe_tokens def _convert_token_to_id(self, token): """Converts a token (str) in an id using the vocab.""" return self.encoder.get(token, self.encoder.get(self.unk_token)) def _convert_id_to_token(self, index): """Converts an index (integer) in a token (str) using the vocab.""" return self.decoder.get(index) def convert_tokens_to_string(self, tokens): """Converts a sequence of tokens (string) in a single string.""" text = ''.join(tokens) text = bytearray([self.byte_decoder[c] for c in text]).decode('utf-8', errors=self.errors) return text def decode(self, token_ids, skip_special_tokens: bool=False, clean_up_tokenization_spaces: Optional[bool]=False, spaces_between_special_tokens: bool=False, **kwargs) -> str: return super().decode(token_ids, skip_special_tokens=skip_special_tokens, clean_up_tokenization_spaces=clean_up_tokenization_spaces, spaces_between_special_tokens=spaces_between_special_tokens, **kwargs) 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 vocab_file = os.path.join(save_directory, (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file']) merge_file = os.path.join(save_directory, (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['merges_file']) with open(vocab_file, 'w', encoding='utf-8') as f: f.write(json.dumps(self.encoder, indent=2, sort_keys=True, ensure_ascii=False) + '\n') index = 0 with open(merge_file, 'w', encoding='utf-8') as writer: writer.write('#version: 0.2\n') for bpe_tokens, token_index in sorted(self.bpe_ranks.items(), key=lambda kv: kv[1]): if index != token_index: logger.warning(f'Saving vocabulary to {merge_file}: BPE merge indices are not consecutive. Please check that the tokenizer is not corrupted!') index = token_index writer.write(' '.join(bpe_tokens) + '\n') index += 1 return (vocab_file, merge_file) def prepare_for_tokenization(self, text, **kwargs): text = unicodedata.normalize('NFC', text) return (text, kwargs)
class Qwen2Tokenizer(PreTrainedTokenizer): ''' Construct a Qwen2 tokenizer. Based on byte-level Byte-Pair-Encoding. Same with GPT2Tokenizer, this tokenizer has been trained to treat spaces like parts of the tokens so a word will be encoded differently whether it is at the beginning of the sentence (without space) or not: ```python >>> from transformers import Qwen2Tokenizer >>> tokenizer = Qwen2Tokenizer.from_pretrained("Qwen/Qwen-tokenizer") >>> tokenizer("Hello world")["input_ids"] [9707, 1879] >>> tokenizer(" Hello world")["input_ids"] [21927, 1879] ``` This is expected. You should not use GPT2Tokenizer instead, because of the different pretokenization rules. 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`): 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. unk_token (`str`, *optional*, defaults to `"<|endoftext|>"`): 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. bos_token (`str`, *optional*): The beginning of sequence token. Not applicable for this tokenizer. eos_token (`str`, *optional*, defaults to `"<|endoftext|>"`): The end of sequence token. pad_token (`str`, *optional*, defaults to `"<|endoftext|>"`): The token used for padding, for example when batching sequences of different lengths. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not the model should cleanup the spaces that were added when splitting the input text during the tokenization process. Not applicable to this tokenizer, since tokenization does not add spaces. split_special_tokens (`bool`, *optional*, defaults to `False`): Whether or not the special tokens should be split during the tokenization process. The default behavior is to not split special tokens. This means that if `<|endoftext|>` is the `eos_token`, then `tokenizer.tokenize("<|endoftext|>") = ['<|endoftext|>`]. Otherwise, if `split_special_tokens=True`, then `tokenizer.tokenize("<|endoftext|>")` will be give `['<', '|', 'endo', 'ft', 'ext', '|', '>']`. This argument is only supported for `slow` tokenizers for the moment. ''' def __init__(self, vocab_file, merges_file, errors='replace', unk_token='<|endoftext|>', bos_token=None, eos_token='<|endoftext|>', pad_token='<|endoftext|>', clean_up_tokenization_spaces=False, split_special_tokens=False, **kwargs): pass @property def vocab_size(self) -> int: pass def get_vocab(self): pass def bpe(self, token): pass def _tokenize(self, text): '''Tokenize a string.''' pass def _convert_token_to_id(self, token): '''Converts a token (str) in an id using the vocab.''' pass def _convert_id_to_token(self, index): '''Converts an index (integer) in a token (str) using the vocab.''' pass def convert_tokens_to_string(self, tokens): '''Converts a sequence of tokens (string) in a single string.''' pass def decode(self, token_ids, skip_special_tokens: bool=False, clean_up_tokenization_spaces: Optional[bool]=False, spaces_between_special_tokens: bool=False, **kwargs) -> str: pass def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str]=None) -> tuple[str]: pass def prepare_for_tokenization(self, text, **kwargs): pass
13
5
17
1
15
1
3
0.38
1
12
0
0
11
8
11
100
257
29
168
62
136
64
104
38
92
9
3
3
32
4,727
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2/tokenization_qwen2_fast.py
transformers.models.qwen2.tokenization_qwen2_fast.Qwen2TokenizerFast
from ...tokenization_utils import AddedToken from typing import Optional from ...tokenization_utils_fast import PreTrainedTokenizerFast from .tokenization_qwen2 import Qwen2Tokenizer class Qwen2TokenizerFast(PreTrainedTokenizerFast): """ Construct a "fast" Qwen2 tokenizer (backed by HuggingFace's *tokenizers* library). Based on byte-level Byte-Pair-Encoding. Same with GPT2Tokenizer, this tokenizer has been trained to treat spaces like parts of the tokens so a word will be encoded differently whether it is at the beginning of the sentence (without space) or not: ```python >>> from transformers import Qwen2TokenizerFast >>> tokenizer = Qwen2TokenizerFast.from_pretrained("Qwen/Qwen-tokenizer") >>> tokenizer("Hello world")["input_ids"] [9707, 1879] >>> tokenizer(" Hello world")["input_ids"] [21927, 1879] ``` This is expected. 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`, *optional*): Path to the vocabulary file. merges_file (`str`, *optional*): Path to the merges file. tokenizer_file (`str`, *optional*): Path to [tokenizers](https://github.com/huggingface/tokenizers) file (generally has a .json extension) that contains everything needed to load the tokenizer. unk_token (`str`, *optional*, defaults to `"<|endoftext|>"`): 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. Not applicable to this tokenizer. bos_token (`str`, *optional*): The beginning of sequence token. Not applicable for this tokenizer. eos_token (`str`, *optional*, defaults to `"<|endoftext|>"`): The end of sequence token. pad_token (`str`, *optional*, defaults to `"<|endoftext|>"`): The token used for padding, for example when batching sequences of different lengths. """ vocab_files_names = VOCAB_FILES_NAMES model_input_names = ['input_ids', 'attention_mask'] slow_tokenizer_class = Qwen2Tokenizer def __init__(self, vocab_file=None, merges_file=None, tokenizer_file=None, unk_token='<|endoftext|>', bos_token=None, eos_token='<|endoftext|>', pad_token='<|endoftext|>', **kwargs): bos_token = AddedToken(bos_token, lstrip=False, rstrip=False, special=True, normalized=False) if isinstance(bos_token, str) else bos_token eos_token = AddedToken(eos_token, lstrip=False, rstrip=False, special=True, normalized=False) if isinstance(eos_token, str) else eos_token unk_token = AddedToken(unk_token, lstrip=False, rstrip=False, special=True, normalized=False) if isinstance(unk_token, str) else unk_token pad_token = AddedToken(pad_token, lstrip=False, rstrip=False, special=True, normalized=False) if isinstance(pad_token, str) else pad_token super().__init__(vocab_file=vocab_file, merges_file=merges_file, tokenizer_file=tokenizer_file, unk_token=unk_token, bos_token=bos_token, eos_token=eos_token, pad_token=pad_token, **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)
class Qwen2TokenizerFast(PreTrainedTokenizerFast): ''' Construct a "fast" Qwen2 tokenizer (backed by HuggingFace's *tokenizers* library). Based on byte-level Byte-Pair-Encoding. Same with GPT2Tokenizer, this tokenizer has been trained to treat spaces like parts of the tokens so a word will be encoded differently whether it is at the beginning of the sentence (without space) or not: ```python >>> from transformers import Qwen2TokenizerFast >>> tokenizer = Qwen2TokenizerFast.from_pretrained("Qwen/Qwen-tokenizer") >>> tokenizer("Hello world")["input_ids"] [9707, 1879] >>> tokenizer(" Hello world")["input_ids"] [21927, 1879] ``` This is expected. 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`, *optional*): Path to the vocabulary file. merges_file (`str`, *optional*): Path to the merges file. tokenizer_file (`str`, *optional*): Path to [tokenizers](https://github.com/huggingface/tokenizers) file (generally has a .json extension) that contains everything needed to load the tokenizer. unk_token (`str`, *optional*, defaults to `"<|endoftext|>"`): 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. Not applicable to this tokenizer. bos_token (`str`, *optional*): The beginning of sequence token. Not applicable for this tokenizer. eos_token (`str`, *optional*, defaults to `"<|endoftext|>"`): The end of sequence token. pad_token (`str`, *optional*, defaults to `"<|endoftext|>"`): The token used for padding, for example when batching sequences of different lengths. ''' def __init__(self, vocab_file=None, merges_file=None, tokenizer_file=None, unk_token='<|endoftext|>', bos_token=None, eos_token='<|endoftext|>', pad_token='<|endoftext|>', **kwargs): pass def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str]=None) -> tuple[str]: pass
3
1
25
1
22
2
3
0.81
1
3
0
0
2
0
2
90
98
11
48
17
35
39
13
7
10
5
3
0
6
4,728
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/configuration_qwen2_5_vl.py
transformers.models.qwen2_5_vl.configuration_qwen2_5_vl.Qwen2_5_VLConfig
from ...configuration_utils import PretrainedConfig, layer_type_validation class Qwen2_5_VLConfig(PretrainedConfig): """ This is the configuration class to store the configuration of a [`Qwen2_5_VLModel`]. It is used to instantiate a Qwen2-VL model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of Qwen2-VL-7B-Instruct [Qwen/Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct). Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: text_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Qwen2_5_VLTextConfig`): The config object or dictionary of the text backbone. vision_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Qwen2_5_VLVisionConfig`): The config object or dictionary of the vision backbone. image_token_id (`int`, *optional*, defaults to 151655): The image token index to encode the image prompt. video_token_id (`int`, *optional*, defaults to 151656): The video token index to encode the image prompt. ```python >>> from transformers import Qwen2_5_VLForConditionalGeneration, Qwen2_5_VLConfig >>> # Initializing a Qwen2_5_VL style configuration >>> configuration = Qwen2_5_VLConfig() >>> # Initializing a model from the Qwen2-VL-7B style configuration >>> model = Qwen2_5_VLForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = 'qwen2_5_vl' sub_configs = {'vision_config': Qwen2_5_VLVisionConfig, 'text_config': Qwen2_5_VLTextConfig} keys_to_ignore_at_inference = ['past_key_values'] def __init__(self, text_config=None, vision_config=None, image_token_id=151655, video_token_id=151656, **kwargs): if isinstance(vision_config, dict): self.vision_config = self.sub_configs['vision_config'](**vision_config) elif vision_config is None: self.vision_config = self.sub_configs['vision_config']() if isinstance(text_config, dict): self.text_config = self.sub_configs['text_config'](**text_config) elif text_config is None: self.text_config = self.sub_configs['text_config'](**kwargs) self.image_token_id = image_token_id self.video_token_id = video_token_id super().__init__(**kwargs)
class Qwen2_5_VLConfig(PretrainedConfig): ''' This is the configuration class to store the configuration of a [`Qwen2_5_VLModel`]. It is used to instantiate a Qwen2-VL model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of Qwen2-VL-7B-Instruct [Qwen/Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct). Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: text_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Qwen2_5_VLTextConfig`): The config object or dictionary of the text backbone. vision_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Qwen2_5_VLVisionConfig`): The config object or dictionary of the vision backbone. image_token_id (`int`, *optional*, defaults to 151655): The image token index to encode the image prompt. video_token_id (`int`, *optional*, defaults to 151656): The video token index to encode the image prompt. ```python >>> from transformers import Qwen2_5_VLForConditionalGeneration, Qwen2_5_VLConfig >>> # Initializing a Qwen2_5_VL style configuration >>> configuration = Qwen2_5_VLConfig() >>> # Initializing a model from the Qwen2-VL-7B style configuration >>> model = Qwen2_5_VLForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```''' def __init__(self, text_config=None, vision_config=None, image_token_id=151655, video_token_id=151656, **kwargs): pass
2
1
63
5
52
6
6
1.6
1
2
0
0
1
18
1
1
183
14
65
46
41
104
34
24
32
6
1
2
6
4,729
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/configuration_qwen2_5_vl.py
transformers.models.qwen2_5_vl.configuration_qwen2_5_vl.Qwen2_5_VLVisionConfig
from ...configuration_utils import PretrainedConfig, layer_type_validation class Qwen2_5_VLVisionConfig(PretrainedConfig): model_type = 'qwen2_5_vl' base_config_key = 'vision_config' def __init__(self, depth=32, hidden_size=3584, hidden_act='silu', intermediate_size=3420, num_heads=16, in_channels=3, patch_size=14, spatial_merge_size=2, temporal_patch_size=2, tokens_per_second=4, window_size=112, out_hidden_size=3584, fullatt_block_indexes=[7, 15, 23, 31], initializer_range=0.02, **kwargs): super().__init__(**kwargs) self.depth = depth self.hidden_size = hidden_size self.hidden_act = hidden_act self.intermediate_size = intermediate_size self.num_heads = num_heads self.in_channels = in_channels self.patch_size = patch_size self.spatial_merge_size = spatial_merge_size self.temporal_patch_size = temporal_patch_size self.tokens_per_second = tokens_per_second self.window_size = window_size self.fullatt_block_indexes = fullatt_block_indexes self.out_hidden_size = out_hidden_size self.initializer_range = initializer_range
class Qwen2_5_VLVisionConfig(PretrainedConfig): def __init__(self, depth=32, hidden_size=3584, hidden_act='silu', intermediate_size=3420, num_heads=16, in_channels=3, patch_size=14, spatial_merge_size=2, temporal_patch_size=2, tokens_per_second=4, window_size=112, out_hidden_size=3584, fullatt_block_indexes=[7, 15, 23, 31], initializer_range=0.02, **kwargs): pass
2
0
32
1
31
0
1
0
1
1
0
0
1
13
1
1
36
2
34
33
16
0
18
17
16
1
1
0
1
4,730
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2MLP
import torch.nn as nn from ...activations import ACT2FN class Qwen2MLP(nn.Module): def __init__(self, config): super().__init__() self.config = config self.hidden_size = config.hidden_size self.intermediate_size = config.intermediate_size self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False) self.act_fn = ACT2FN[config.hidden_act] def forward(self, x): down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x)) return down_proj
class Qwen2MLP(nn.Module): def __init__(self, config): pass def forward(self, x): pass
3
0
6
0
6
0
1
0
1
1
0
0
2
7
2
12
14
1
13
11
10
0
13
11
10
1
1
0
2
4,731
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLAttention
import torch.nn as nn import torch from .configuration_qwen2_5_vl import Qwen2_5_VLConfig, Qwen2_5_VLTextConfig, Qwen2_5_VLVisionConfig from ...cache_utils import Cache, DynamicCache from ...processing_utils import Unpack from ...utils.deprecation import deprecate_kwarg from typing import Any, Callable, Optional, Union from ...modeling_flash_attention_utils import FlashAttentionKwargs from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel import torch.nn.functional as F class Qwen2_5_VLAttention(nn.Module): """ Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer and "Generating Long Sequences with Sparse Transformers". """ def __init__(self, config: Qwen2_5_VLTextConfig, layer_idx: Optional[int]=None): super().__init__() self.config = config self.layer_idx = layer_idx if layer_idx is None: logger.warning_once(f'Instantiating {self.__class__.__name__} without passing `layer_idx` is not recommended and will to errors during the forward call, if caching is used. Please make sure to provide a `layer_idx` when creating this class.') self.hidden_size = config.hidden_size self.num_heads = config.num_attention_heads self.head_dim = self.hidden_size // self.num_heads self.num_key_value_heads = config.num_key_value_heads self.num_key_value_groups = self.num_heads // self.num_key_value_heads self.is_causal = True self.attention_dropout = config.attention_dropout self.rope_scaling = config.rope_scaling self.scaling = self.head_dim ** (-0.5) if self.head_dim * self.num_heads != self.hidden_size: raise ValueError(f'hidden_size must be divisible by num_heads (got `hidden_size`: {self.hidden_size} and `num_heads`: {self.num_heads}).') self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=True) self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=True) self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=True) self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=False) self.sliding_window = config.sliding_window if config.layer_types[layer_idx] == 'sliding_attention' else None self.rotary_emb = Qwen2_5_VLRotaryEmbedding(config=config) @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: bsz, q_len, _ = hidden_states.size() query_states = self.q_proj(hidden_states) key_states = self.k_proj(hidden_states) value_states = self.v_proj(hidden_states) query_states = query_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) key_states = key_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) value_states = value_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) cos, sin = position_embeddings query_states, key_states = apply_multimodal_rotary_pos_emb(query_states, key_states, cos, sin, self.rope_scaling['mrope_section']) if past_key_values is not None: cache_kwargs = {'sin': sin, 'cos': cos, 'cache_position': cache_position} key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs) attention_interface: Callable = eager_attention_forward if self.config._attn_implementation != 'eager': attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation] attn_output, attn_weights = attention_interface(self, query_states, key_states, value_states, attention_mask, dropout=0.0 if not self.training else self.attention_dropout, scaling=self.scaling, sliding_window=self.sliding_window, position_ids=position_ids, **kwargs) attn_output = attn_output.reshape(bsz, q_len, -1).contiguous() attn_output = self.o_proj(attn_output) return (attn_output, attn_weights)
class Qwen2_5_VLAttention(nn.Module): ''' Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer and "Generating Long Sequences with Sparse Transformers". ''' def __init__(self, config: Qwen2_5_VLTextConfig, layer_idx: Optional[int]=None): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: pass
4
1
48
9
38
4
5
0.14
1
8
3
2
2
15
2
12
103
19
76
37
63
11
54
27
51
6
1
1
9
4,732
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLCausalLMOutputWithPast
from typing import Any, Callable, Optional, Union import torch.nn.functional as F from dataclasses import dataclass import torch.nn as nn import torch from ...utils import TransformersKwargs, auto_docstring, can_return_tuple, is_torchdynamo_compiling, logging from ...modeling_outputs import BaseModelOutputWithPast, ModelOutput from ...cache_utils import Cache, DynamicCache @dataclass @auto_docstring(custom_intro='\n Base class for Qwen2_5_VL causal language model (or autoregressive) outputs.\n ') class Qwen2_5_VLCausalLMOutputWithPast(ModelOutput): """ loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided): Language modeling loss (for next-token prediction). logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`): Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). past_key_values (`Cache`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`): It is a [`~cache_utils.Cache`] instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. """ loss: Optional[torch.FloatTensor] = None logits: Optional[torch.FloatTensor] = None past_key_values: Optional[Cache] = None hidden_states: Optional[tuple[torch.FloatTensor]] = None attentions: Optional[tuple[torch.FloatTensor]] = None rope_deltas: Optional[torch.LongTensor] = None
@dataclass @auto_docstring(custom_intro='\n Base class for Qwen2_5_VL causal language model (or autoregressive) outputs.\n ') class Qwen2_5_VLCausalLMOutputWithPast(ModelOutput): ''' loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided): Language modeling loss (for next-token prediction). logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`): Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). past_key_values (`Cache`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`): It is a [`~cache_utils.Cache`] instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. ''' pass
3
1
0
0
0
0
0
3.43
1
0
0
0
0
0
0
0
36
5
7
7
6
24
7
7
6
0
1
0
0
4,733
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLDecoderLayer
from ...modeling_layers import GradientCheckpointingLayer import torch from .configuration_qwen2_5_vl import Qwen2_5_VLConfig, Qwen2_5_VLTextConfig, Qwen2_5_VLVisionConfig from ...cache_utils import Cache, DynamicCache import torch.nn as nn from ...processing_utils import Unpack from ...utils.deprecation import deprecate_kwarg from ..qwen2.modeling_qwen2 import Qwen2RMSNorm import torch.nn.functional as F from ...modeling_flash_attention_utils import FlashAttentionKwargs from typing import Any, Callable, Optional, Union class Qwen2_5_VLDecoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2_5_VLTextConfig, layer_idx: int): super().__init__() self.hidden_size = config.hidden_size if config.use_sliding_window and config._attn_implementation != 'flash_attention_2': logger.warning_once(f'Sliding Window Attention is enabled but not implemented for `{config._attn_implementation}`; unexpected results may be encountered.') self.self_attn = Qwen2_5_VLAttention(config, layer_idx) self.mlp = Qwen2MLP(config) self.input_layernorm = Qwen2RMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.post_attention_layernorm = Qwen2RMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.attention_type = config.layer_types[layer_idx] @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: Optional[bool]=False, use_cache: Optional[bool]=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.FloatTensor, Optional[tuple[torch.FloatTensor, torch.FloatTensor]]]: """ Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`, *optional*): attention mask of size `(batch, sequence_length)` where padding elements are indicated by 0. output_attentions (`bool`, *optional*): Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. 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`). past_key_values (`Cache`, *optional*): cached past key and value projection states cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*): Indices depicting the position of the input sequence tokens in the sequence. position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*): Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`, with `head_dim` being the embedding dimension of each attention head. kwargs (`dict`, *optional*): Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code into the model """ residual = hidden_states hidden_states = self.input_layernorm(hidden_states) hidden_states, self_attn_weights = self.self_attn(hidden_states=hidden_states, attention_mask=attention_mask, position_ids=position_ids, past_key_values=past_key_values, output_attentions=output_attentions, use_cache=use_cache, cache_position=cache_position, position_embeddings=position_embeddings, **kwargs) hidden_states = residual + hidden_states residual = hidden_states hidden_states = self.post_attention_layernorm(hidden_states) hidden_states = self.mlp(hidden_states) hidden_states = residual + hidden_states outputs = (hidden_states,) if output_attentions: outputs += (self_attn_weights,) return outputs
class Qwen2_5_VLDecoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2_5_VLTextConfig, layer_idx: int): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: Optional[bool]=False, use_cache: Optional[bool]=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.FloatTensor, Optional[tuple[torch.FloatTensor, torch.FloatTensor]]]: ''' Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`, *optional*): attention mask of size `(batch, sequence_length)` where padding elements are indicated by 0. output_attentions (`bool`, *optional*): Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. 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`). past_key_values (`Cache`, *optional*): cached past key and value projection states cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*): Indices depicting the position of the input sequence tokens in the sequence. position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*): Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`, with `head_dim` being the embedding dimension of each attention head. kwargs (`dict`, *optional*): Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code into the model ''' pass
4
1
40
5
24
12
3
0.5
1
7
3
0
2
5
2
12
82
11
48
22
34
24
25
11
22
3
1
1
5
4,734
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLForConditionalGeneration
import torch.nn as nn from typing import Any, Callable, Optional, Union from ...utils import TransformersKwargs, auto_docstring, can_return_tuple, is_torchdynamo_compiling, logging import torch.nn.functional as F from ...processing_utils import Unpack import torch from ...generation import GenerationMixin from ...cache_utils import Cache, DynamicCache class Qwen2_5_VLForConditionalGeneration(Qwen2_5_VLPreTrainedModel, GenerationMixin): _checkpoint_conversion_mapping = {'^visual': 'model.visual', '^model(?!\\.(language_model|visual))': 'model.language_model'} _tied_weights_keys = ['lm_head.weight'] accepts_loss_kwargs = False def __init__(self, config): super().__init__(config) self.model = Qwen2_5_VLModel(config) self.lm_head = nn.Linear(config.text_config.hidden_size, config.text_config.vocab_size, bias=False) self.post_init() def get_input_embeddings(self): return self.model.get_input_embeddings() def set_input_embeddings(self, value): self.model.set_input_embeddings(value) def set_decoder(self, decoder): self.model.set_decoder(decoder) def get_decoder(self): return self.model.get_decoder() def get_video_features(self, pixel_values_videos: torch.FloatTensor, video_grid_thw: Optional[torch.LongTensor]=None): return self.model.get_video_features(pixel_values_videos, video_grid_thw) def get_image_features(self, pixel_values: torch.FloatTensor, image_grid_thw: Optional[torch.LongTensor]=None): return self.model.get_image_features(pixel_values, image_grid_thw) @property def language_model(self): return self.model.language_model @property def visual(self): return self.model.visual @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, logits_to_keep: Union[int, torch.Tensor]=0, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2_5_VLCausalLMOutputWithPast]: """ labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. Example: ```python >>> from PIL import Image >>> import requests >>> from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration >>> model = Qwen2_5_VLForConditionalGeneration.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") >>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") >>> messages = [ { "role": "user", "content": [ {"type": "image"}, {"type": "text", "text": "What is shown in this image?"}, ], }, ] >>> url = "https://www.ilankelman.org/stopsigns/australia.jpg" >>> image = Image.open(requests.get(url, stream=True).raw) >>> text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) >>> inputs = processor(text=[text], images=[image], vision_infos=[vision_infos]) >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "The image shows a street scene with a red stop sign in the foreground. In the background, there is a large red gate with Chinese characters ..." ```""" 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 outputs = self.model(input_ids=input_ids, pixel_values=pixel_values, pixel_values_videos=pixel_values_videos, image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, second_per_grid_ts=second_per_grid_ts, position_ids=position_ids, attention_mask=attention_mask, past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache, output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=True, cache_position=cache_position, **kwargs) hidden_states = outputs[0] slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep logits = self.lm_head(hidden_states[:, slice_indices, :]) loss = None if labels is not None: loss = self.loss_function(logits=logits, labels=labels, vocab_size=self.config.text_config.vocab_size, **kwargs) return Qwen2_5_VLCausalLMOutputWithPast(loss=loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions, rope_deltas=outputs.rope_deltas) def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, cache_position=None, position_ids=None, use_cache=True, pixel_values=None, pixel_values_videos=None, image_grid_thw=None, video_grid_thw=None, second_per_grid_ts=None, **kwargs): model_inputs = super().prepare_inputs_for_generation(input_ids, past_key_values=past_key_values, attention_mask=attention_mask, inputs_embeds=inputs_embeds, cache_position=cache_position, position_ids=position_ids, pixel_values=pixel_values, pixel_values_videos=pixel_values_videos, image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, second_per_grid_ts=second_per_grid_ts, use_cache=use_cache, **kwargs) if position_ids is None: if cache_position[0] == 0 or self.model.rope_deltas is None: vision_positions, rope_deltas = self.model.get_rope_index(model_inputs.get('input_ids', None), image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, attention_mask=attention_mask) self.model.rope_deltas = rope_deltas elif 'position_ids' in model_inputs: batch_size, seq_length = model_inputs['position_ids'].shape device = model_inputs['position_ids'].device position_ids = torch.arange(seq_length, device=device) position_ids = position_ids.view(1, 1, -1).expand(3, batch_size, -1) delta = cache_position[0] + self.model.rope_deltas delta = delta.repeat_interleave(batch_size // delta.shape[0], dim=0) vision_positions = position_ids + delta.expand_as(position_ids) text_positions = model_inputs['position_ids'][None, ...] model_inputs['position_ids'] = torch.cat([text_positions, vision_positions], dim=0) if cache_position[0] != 0: model_inputs['pixel_values'] = None model_inputs['pixel_values_videos'] = None return model_inputs def _get_image_nums_and_video_nums(self, input_ids: Optional[torch.LongTensor], inputs_embeds: Optional[torch.Tensor]=None) -> tuple[torch.Tensor, torch.Tensor]: """ Get the number of images and videos for each sample to calculate the separation length of the sample tensor. These parameters are not passed through the processor to avoid unpredictable impacts from interface modifications. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Returns: image_nums (`torch.LongTensor` of shape `(batch_size, num_images_sample)`) video_nums (`torch.LongTensor` of shape `(batch_size, num_videos_sample)`) """ image_token_id = self.config.image_token_id video_token_id = self.config.video_token_id vision_start_token_id = self.config.vision_start_token_id if inputs_embeds is not None: vision_start_mask = (inputs_embeds == self.get_input_embeddings()(torch.tensor(vision_start_token_id, dtype=torch.long, device=inputs_embeds.device)))[..., 0] image_mask = (inputs_embeds == self.get_input_embeddings()(torch.tensor(image_token_id, dtype=torch.long, device=inputs_embeds.device)))[..., 0] video_mask = (inputs_embeds == self.get_input_embeddings()(torch.tensor(video_token_id, dtype=torch.long, device=inputs_embeds.device)))[..., 0] else: vision_start_mask = input_ids == vision_start_token_id image_mask = input_ids == image_token_id video_mask = input_ids == video_token_id vision_first_mask = torch.roll(vision_start_mask, shifts=1, dims=1) image_nums = torch.sum(vision_first_mask & image_mask, dim=1) video_nums = torch.sum(vision_first_mask & video_mask, dim=1) return (image_nums, video_nums) def _expand_inputs_for_generation(self, expand_size: int=1, is_encoder_decoder: bool=False, input_ids: Optional[torch.LongTensor]=None, **model_kwargs) -> tuple[torch.LongTensor, dict[str, Any]]: if expand_size == 1: return (input_ids, model_kwargs) visual_keys = ['pixel_values', 'image_grid_thw', 'pixel_values_videos', 'video_grid_thw', 'second_per_grid_ts'] def _expand_dict_for_generation_visual(dict_to_expand): image_grid_thw = model_kwargs.get('image_grid_thw', None) video_grid_thw = model_kwargs.get('video_grid_thw', None) image_nums, video_nums = self._get_image_nums_and_video_nums(input_ids, inputs_embeds=model_kwargs.get('inputs_embeds', None)) def _repeat_interleave_samples(x, lengths, repeat_times): samples = torch.split(x, lengths) repeat_args = [repeat_times] + [1] * (x.dim() - 1) result = torch.cat([sample.repeat(*repeat_args) for sample in samples], dim=0) return result for key in dict_to_expand: if key == 'pixel_values': samples = torch.split(image_grid_thw, list(image_nums)) lengths = [torch.prod(sample, dim=1).sum() for sample in samples] dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=lengths, repeat_times=expand_size) elif key == 'image_grid_thw': lengths = list(image_nums) dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=lengths, repeat_times=expand_size) elif key == 'pixel_values_videos': samples = torch.split(video_grid_thw, list(video_nums)) lengths = [torch.prod(sample, dim=1).sum() for sample in samples] dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=lengths, repeat_times=expand_size) elif key == 'video_grid_thw': lengths = list(video_nums) dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=lengths, repeat_times=expand_size) elif key == 'second_per_grid_ts': dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=list(video_nums), repeat_times=expand_size) return dict_to_expand def _expand_dict_for_generation(dict_to_expand): for key in dict_to_expand: if key != 'cache_position' and dict_to_expand[key] is not None and isinstance(dict_to_expand[key], torch.Tensor) and (key not in visual_keys): dict_to_expand[key] = dict_to_expand[key].repeat_interleave(expand_size, dim=0) return dict_to_expand model_kwargs = _expand_dict_for_generation_visual(model_kwargs) if input_ids is not None: input_ids = input_ids.repeat_interleave(expand_size, dim=0) model_kwargs = _expand_dict_for_generation(model_kwargs) if is_encoder_decoder: if model_kwargs.get('encoder_outputs') is None: raise ValueError('If `is_encoder_decoder` is True, make sure that `encoder_outputs` is defined.') model_kwargs['encoder_outputs'] = _expand_dict_for_generation(model_kwargs['encoder_outputs']) return (input_ids, model_kwargs)
class Qwen2_5_VLForConditionalGeneration(Qwen2_5_VLPreTrainedModel, GenerationMixin): def __init__(self, config): pass def get_input_embeddings(self): pass def set_input_embeddings(self, value): pass def set_decoder(self, decoder): pass def get_decoder(self): pass def get_video_features(self, pixel_values_videos: torch.FloatTensor, video_grid_thw: Optional[torch.LongTensor]=None): pass def get_image_features(self, pixel_values: torch.FloatTensor, image_grid_thw: Optional[torch.LongTensor]=None): pass @property def language_model(self): pass @property def visual(self): pass @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, logits_to_keep: Union[int, torch.Tensor]=0, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2_5_VLCausalLMOutputWithPast]: ''' labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. Example: ```python >>> from PIL import Image >>> import requests >>> from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration >>> model = Qwen2_5_VLForConditionalGeneration.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") >>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") >>> messages = [ { "role": "user", "content": [ {"type": "image"}, {"type": "text", "text": "What is shown in this image?"}, ], }, ] >>> url = "https://www.ilankelman.org/stopsigns/australia.jpg" >>> image = Image.open(requests.get(url, stream=True).raw) >>> text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) >>> inputs = processor(text=[text], images=[image], vision_infos=[vision_infos]) >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "The image shows a street scene with a red stop sign in the foreground. In the background, there is a large red gate with Chinese characters ..." ```''' pass def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, cache_position=None, position_ids=None, use_cache=True, pixel_values=None, pixel_values_videos=None, image_grid_thw=None, video_grid_thw=None, second_per_grid_ts=None, **kwargs): pass def _get_image_nums_and_video_nums(self, input_ids: Optional[torch.LongTensor], inputs_embeds: Optional[torch.Tensor]=None) -> tuple[torch.Tensor, torch.Tensor]: ''' Get the number of images and videos for each sample to calculate the separation length of the sample tensor. These parameters are not passed through the processor to avoid unpredictable impacts from interface modifications. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Returns: image_nums (`torch.LongTensor` of shape `(batch_size, num_images_sample)`) video_nums (`torch.LongTensor` of shape `(batch_size, num_videos_sample)`) ''' pass def _expand_inputs_for_generation(self, expand_size: int=1, is_encoder_decoder: bool=False, input_ids: Optional[torch.LongTensor]=None, **model_kwargs) -> tuple[torch.LongTensor, dict[str, Any]]: pass def _expand_dict_for_generation_visual(dict_to_expand): pass def _repeat_interleave_samples(x, lengths, repeat_times): pass def _expand_dict_for_generation_visual(dict_to_expand): pass
21
2
46
4
32
10
5
0.31
2
11
4
0
10
6
10
11
476
53
327
120
273
102
183
77
172
17
2
5
46
4,735
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLMLP
import torch.nn as nn from ...activations import ACT2FN class Qwen2_5_VLMLP(nn.Module): def __init__(self, config, bias: bool=False): super().__init__() self.hidden_size = config.hidden_size self.intermediate_size = config.intermediate_size self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=bias) self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=bias) self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=bias) self.act_fn = ACT2FN[config.hidden_act] def forward(self, hidden_state): return self.down_proj(self.act_fn(self.gate_proj(hidden_state)) * self.up_proj(hidden_state))
class Qwen2_5_VLMLP(nn.Module): def __init__(self, config, bias: bool=False): pass def forward(self, hidden_state): pass
3
0
5
0
5
0
1
0
1
2
0
0
2
6
2
12
12
1
11
9
8
0
11
9
8
1
1
0
2
4,736
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLModel
from ...processing_utils import Unpack from typing import Any, Callable, Optional, Union import torch.nn.functional as F import torch.nn as nn import torch from .configuration_qwen2_5_vl import Qwen2_5_VLConfig, Qwen2_5_VLTextConfig, Qwen2_5_VLVisionConfig from ...utils import TransformersKwargs, auto_docstring, can_return_tuple, is_torchdynamo_compiling, logging from ...cache_utils import Cache, DynamicCache @auto_docstring class Qwen2_5_VLModel(Qwen2_5_VLPreTrainedModel): base_model_prefix = '' _checkpoint_conversion_mapping = {'^model': 'language_model'} accepts_loss_kwargs = False config: Qwen2_5_VLConfig _no_split_modules = ['Qwen2_5_VLDecoderLayer', 'Qwen2_5_VLVisionBlock'] def __init__(self, config): super().__init__(config) self.visual = Qwen2_5_VisionTransformerPretrainedModel._from_config(config.vision_config) self.language_model = Qwen2_5_VLTextModel._from_config(config.text_config) self.rope_deltas = None self.post_init() def get_input_embeddings(self): return self.language_model.get_input_embeddings() def set_input_embeddings(self, value): self.language_model.set_input_embeddings(value) def set_decoder(self, decoder): self.language_model = decoder def get_decoder(self): return self.language_model def get_rope_index(self, input_ids: Optional[torch.LongTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, attention_mask: Optional[torch.Tensor]=None) -> tuple[torch.Tensor, torch.Tensor]: """ Calculate the 3D rope index based on image and video's temporal, height and width in LLM. Explanation: Each embedding sequence contains vision embedding and text embedding or just contains text embedding. For pure text embedding sequence, the rotary position embedding has no difference with modern LLMs. Examples: input_ids: [T T T T T], here T is for text. temporal position_ids: [0, 1, 2, 3, 4] height position_ids: [0, 1, 2, 3, 4] width position_ids: [0, 1, 2, 3, 4] For vision and text embedding sequence, we calculate 3D rotary position embedding for vision part and 1D rotary position embedding for text part. Examples: Temporal (Time): 3 patches, representing different segments of the video in time. Height: 2 patches, dividing each frame vertically. Width: 2 patches, dividing each frame horizontally. We also have some important parameters: fps (Frames Per Second): The video's frame rate, set to 1. This means one frame is processed each second. tokens_per_second: This is a crucial parameter. It dictates how many "time-steps" or "temporal tokens" are conceptually packed into a one-second interval of the video. In this case, we have 25 tokens per second. So each second of the video will be represented with 25 separate time points. It essentially defines the temporal granularity. temporal_patch_size: The number of frames that compose one temporal patch. Here, it's 2 frames. interval: The step size for the temporal position IDs, calculated as tokens_per_second * temporal_patch_size / fps. In this case, 25 * 2 / 1 = 50. This means that each temporal patch will be have a difference of 50 in the temporal position IDs. input_ids: [V V V V V V V V V V V V T T T T T], here V is for vision. vision temporal position_ids: [0, 0, 0, 0, 50, 50, 50, 50, 100, 100, 100, 100] vision height position_ids: [0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1] vision width position_ids: [0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1] text temporal position_ids: [101, 102, 103, 104, 105] text height position_ids: [101, 102, 103, 104, 105] text width position_ids: [101, 102, 103, 104, 105] Here we calculate the text start position_ids as the max vision position_ids plus 1. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *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**. Returns: position_ids (`torch.LongTensor` of shape `(3, batch_size, sequence_length)`) mrope_position_deltas (`torch.Tensor` of shape `(batch_size)`) """ spatial_merge_size = self.config.vision_config.spatial_merge_size image_token_id = self.config.image_token_id video_token_id = self.config.video_token_id vision_start_token_id = self.config.vision_start_token_id mrope_position_deltas = [] if input_ids is not None and (image_grid_thw is not None or video_grid_thw is not None): total_input_ids = input_ids if attention_mask is not None: attention_mask = attention_mask == 1 position_ids = torch.ones(3, input_ids.shape[0], input_ids.shape[1], dtype=input_ids.dtype, device=input_ids.device) image_index, video_index = (0, 0) for i, input_ids in enumerate(total_input_ids): if attention_mask is not None: input_ids = input_ids[attention_mask[i]] image_nums, video_nums = (0, 0) vision_start_indices = torch.argwhere(input_ids == vision_start_token_id).squeeze(1) vision_tokens = input_ids[vision_start_indices + 1] image_nums = (vision_tokens == image_token_id).sum() video_nums = (vision_tokens == video_token_id).sum() input_tokens = input_ids.tolist() llm_pos_ids_list: list = [] st = 0 remain_images, remain_videos = (image_nums, video_nums) for _ in range(image_nums + video_nums): if image_token_id in input_tokens and remain_images > 0: ed_image = input_tokens.index(image_token_id, st) else: ed_image = len(input_tokens) + 1 if video_token_id in input_tokens and remain_videos > 0: ed_video = input_tokens.index(video_token_id, st) else: ed_video = len(input_tokens) + 1 if ed_image < ed_video: t, h, w = (image_grid_thw[image_index][0], image_grid_thw[image_index][1], image_grid_thw[image_index][2]) second_per_grid_t = 0 image_index += 1 remain_images -= 1 ed = ed_image else: t, h, w = (video_grid_thw[video_index][0], video_grid_thw[video_index][1], video_grid_thw[video_index][2]) if second_per_grid_ts is not None: second_per_grid_t = second_per_grid_ts[video_index] else: second_per_grid_t = 1.0 video_index += 1 remain_videos -= 1 ed = ed_video llm_grid_t, llm_grid_h, llm_grid_w = (t.item(), h.item() // spatial_merge_size, w.item() // spatial_merge_size) text_len = ed - st st_idx = llm_pos_ids_list[-1].max() + 1 if len(llm_pos_ids_list) > 0 else 0 llm_pos_ids_list.append(torch.arange(text_len).view(1, -1).expand(3, -1) + st_idx) range_tensor = torch.arange(llm_grid_t).view(-1, 1) expanded_range = range_tensor.expand(-1, llm_grid_h * llm_grid_w) second_per_grid_t = torch.as_tensor(second_per_grid_t, dtype=range_tensor.dtype, device=range_tensor.device) time_tensor = expanded_range * second_per_grid_t * self.config.vision_config.tokens_per_second time_tensor_long = time_tensor.long() t_index = time_tensor_long.flatten() h_index = torch.arange(llm_grid_h).view(1, -1, 1).expand(llm_grid_t, -1, llm_grid_w).flatten() w_index = torch.arange(llm_grid_w).view(1, 1, -1).expand(llm_grid_t, llm_grid_h, -1).flatten() llm_pos_ids_list.append(torch.stack([t_index, h_index, w_index]) + text_len + st_idx) st = ed + llm_grid_t * llm_grid_h * llm_grid_w if st < len(input_tokens): st_idx = llm_pos_ids_list[-1].max() + 1 if len(llm_pos_ids_list) > 0 else 0 text_len = len(input_tokens) - st llm_pos_ids_list.append(torch.arange(text_len).view(1, -1).expand(3, -1) + st_idx) llm_positions = torch.cat(llm_pos_ids_list, dim=1).reshape(3, -1) if attention_mask is not None: position_ids[..., i, attention_mask[i]] = llm_positions.to(position_ids.device) else: position_ids[..., i, :] = llm_positions.to(position_ids.device) mrope_position_deltas.append(llm_positions.max() + 1 - len(total_input_ids[i])) mrope_position_deltas = torch.tensor(mrope_position_deltas).unsqueeze(1).to(device=input_ids.device) return (position_ids, mrope_position_deltas) else: if attention_mask is not None: position_ids = attention_mask.long().cumsum(-1) - 1 position_ids.masked_fill_(attention_mask == 0, 1) position_ids = position_ids.unsqueeze(0).expand(3, -1, -1).to(attention_mask.device) max_position_ids = position_ids.max(0, keepdim=False)[0].max(-1, keepdim=True)[0] mrope_position_deltas = max_position_ids + 1 - attention_mask.shape[-1] else: position_ids = torch.arange(input_ids.shape[1], device=input_ids.device).view(1, 1, -1).expand(3, input_ids.shape[0], -1) mrope_position_deltas = torch.zeros([input_ids.shape[0], 1], device=input_ids.device, dtype=input_ids.dtype) return (position_ids, mrope_position_deltas) def get_video_features(self, pixel_values_videos: torch.FloatTensor, video_grid_thw: Optional[torch.LongTensor]=None): """ Encodes videos into continuous embeddings that can be forwarded to the language model. Args: pixel_values_videos (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`): The tensors corresponding to the input videos. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. """ pixel_values_videos = pixel_values_videos.type(self.visual.dtype) video_embeds = self.visual(pixel_values_videos, grid_thw=video_grid_thw) split_sizes = (video_grid_thw.prod(-1) // self.visual.spatial_merge_size ** 2).tolist() video_embeds = torch.split(video_embeds, split_sizes) return video_embeds def get_image_features(self, pixel_values: torch.FloatTensor, image_grid_thw: Optional[torch.LongTensor]=None): """ Encodes images into continuous embeddings that can be forwarded to the language model. Args: pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`): The tensors corresponding to the input images. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. """ pixel_values = pixel_values.type(self.visual.dtype) image_embeds = self.visual(pixel_values, grid_thw=image_grid_thw) split_sizes = (image_grid_thw.prod(-1) // self.visual.spatial_merge_size ** 2).tolist() image_embeds = torch.split(image_embeds, split_sizes) return image_embeds def get_placeholder_mask(self, input_ids: torch.LongTensor, inputs_embeds: torch.FloatTensor, image_features: Optional[torch.FloatTensor]=None, video_features: Optional[torch.FloatTensor]=None): """ Obtains multimodal placeholder mask from `input_ids` or `inputs_embeds`, and checks that the placeholder token count is equal to the length of multimodal features. If the lengths are different, an error is raised. """ if input_ids is None: special_image_mask = inputs_embeds == self.get_input_embeddings()(torch.tensor(self.config.image_token_id, dtype=torch.long, device=inputs_embeds.device)) special_image_mask = special_image_mask.all(-1) special_video_mask = inputs_embeds == self.get_input_embeddings()(torch.tensor(self.config.video_token_id, dtype=torch.long, device=inputs_embeds.device)) special_video_mask = special_video_mask.all(-1) else: special_image_mask = input_ids == self.config.image_token_id special_video_mask = input_ids == self.config.video_token_id n_image_tokens = special_image_mask.sum() special_image_mask = special_image_mask.unsqueeze(-1).expand_as(inputs_embeds).to(inputs_embeds.device) if image_features is not None and inputs_embeds[special_image_mask].numel() != image_features.numel(): raise ValueError(f'Image features and image tokens do not match: tokens: {n_image_tokens}, features {image_features.shape[0]}') n_video_tokens = special_video_mask.sum() special_video_mask = special_video_mask.unsqueeze(-1).expand_as(inputs_embeds).to(inputs_embeds.device) if video_features is not None and inputs_embeds[special_video_mask].numel() != video_features.numel(): raise ValueError(f'Videos features and video tokens do not match: tokens: {n_video_tokens}, features {video_features.shape[0]}') return (special_image_mask, special_video_mask) @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2_5_VLModelOutputWithPast]: """ image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. """ 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 inputs_embeds is None: inputs_embeds = self.get_input_embeddings()(input_ids) if pixel_values is not None: image_embeds = self.get_image_features(pixel_values, image_grid_thw) image_embeds = torch.cat(image_embeds, dim=0).to(inputs_embeds.device, inputs_embeds.dtype) image_mask, _ = self.get_placeholder_mask(input_ids, inputs_embeds=inputs_embeds, image_features=image_embeds) inputs_embeds = inputs_embeds.masked_scatter(image_mask, image_embeds) if pixel_values_videos is not None: video_embeds = self.get_video_features(pixel_values_videos, video_grid_thw) video_embeds = torch.cat(video_embeds, dim=0).to(inputs_embeds.device, inputs_embeds.dtype) _, video_mask = self.get_placeholder_mask(input_ids, inputs_embeds=inputs_embeds, video_features=video_embeds) inputs_embeds = inputs_embeds.masked_scatter(video_mask, video_embeds) if position_ids is None: prefill_compiled_stage = is_torchdynamo_compiling() and (input_ids is not None and input_ids.shape[1] != 1 or (inputs_embeds is not None and inputs_embeds.shape[1] != 1)) prefill_noncompiled_stage = not is_torchdynamo_compiling() and (cache_position is not None and cache_position[0] == 0 or (past_key_values is None or past_key_values.get_seq_length() == 0)) if (prefill_compiled_stage or prefill_noncompiled_stage) or self.rope_deltas is None: position_ids, rope_deltas = self.get_rope_index(input_ids, image_grid_thw, video_grid_thw, second_per_grid_ts=second_per_grid_ts, attention_mask=attention_mask) self.rope_deltas = rope_deltas else: batch_size, seq_length, _ = inputs_embeds.shape position_ids = torch.arange(seq_length, device=inputs_embeds.device) position_ids = position_ids.view(1, 1, -1).expand(3, batch_size, -1) if cache_position is not None: delta = (cache_position[0] + self.rope_deltas).to(inputs_embeds.device) else: delta = torch.zeros((batch_size, seq_length), device=inputs_embeds.device) delta = delta.repeat_interleave(batch_size // delta.shape[0], dim=1) position_ids = position_ids + delta.to(position_ids.device) outputs = self.language_model(input_ids=None, position_ids=position_ids, attention_mask=attention_mask, past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache, output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=True, cache_position=cache_position, **kwargs) output = Qwen2_5_VLModelOutputWithPast(last_hidden_state=outputs.last_hidden_state, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions, rope_deltas=self.rope_deltas) return output if return_dict else output.to_tuple()
@auto_docstring class Qwen2_5_VLModel(Qwen2_5_VLPreTrainedModel): def __init__(self, config): pass def get_input_embeddings(self): pass def set_input_embeddings(self, value): pass def set_decoder(self, decoder): pass def get_decoder(self): pass def get_rope_index(self, input_ids: Optional[torch.LongTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, attention_mask: Optional[torch.Tensor]=None) -> tuple[torch.Tensor, torch.Tensor]: ''' Calculate the 3D rope index based on image and video's temporal, height and width in LLM. Explanation: Each embedding sequence contains vision embedding and text embedding or just contains text embedding. For pure text embedding sequence, the rotary position embedding has no difference with modern LLMs. Examples: input_ids: [T T T T T], here T is for text. temporal position_ids: [0, 1, 2, 3, 4] height position_ids: [0, 1, 2, 3, 4] width position_ids: [0, 1, 2, 3, 4] For vision and text embedding sequence, we calculate 3D rotary position embedding for vision part and 1D rotary position embedding for text part. Examples: Temporal (Time): 3 patches, representing different segments of the video in time. Height: 2 patches, dividing each frame vertically. Width: 2 patches, dividing each frame horizontally. We also have some important parameters: fps (Frames Per Second): The video's frame rate, set to 1. This means one frame is processed each second. tokens_per_second: This is a crucial parameter. It dictates how many "time-steps" or "temporal tokens" are conceptually packed into a one-second interval of the video. In this case, we have 25 tokens per second. So each second of the video will be represented with 25 separate time points. It essentially defines the temporal granularity. temporal_patch_size: The number of frames that compose one temporal patch. Here, it's 2 frames. interval: The step size for the temporal position IDs, calculated as tokens_per_second * temporal_patch_size / fps. In this case, 25 * 2 / 1 = 50. This means that each temporal patch will be have a difference of 50 in the temporal position IDs. input_ids: [V V V V V V V V V V V V T T T T T], here V is for vision. vision temporal position_ids: [0, 0, 0, 0, 50, 50, 50, 50, 100, 100, 100, 100] vision height position_ids: [0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1] vision width position_ids: [0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1] text temporal position_ids: [101, 102, 103, 104, 105] text height position_ids: [101, 102, 103, 104, 105] text width position_ids: [101, 102, 103, 104, 105] Here we calculate the text start position_ids as the max vision position_ids plus 1. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *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**. Returns: position_ids (`torch.LongTensor` of shape `(3, batch_size, sequence_length)`) mrope_position_deltas (`torch.Tensor` of shape `(batch_size)`) ''' pass def get_video_features(self, pixel_values_videos: torch.FloatTensor, video_grid_thw: Optional[torch.LongTensor]=None): ''' Encodes videos into continuous embeddings that can be forwarded to the language model. Args: pixel_values_videos (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`): The tensors corresponding to the input videos. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. ''' pass def get_image_features(self, pixel_values: torch.FloatTensor, image_grid_thw: Optional[torch.LongTensor]=None): ''' Encodes images into continuous embeddings that can be forwarded to the language model. Args: pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`): The tensors corresponding to the input images. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. ''' pass def get_placeholder_mask(self, input_ids: torch.LongTensor, inputs_embeds: torch.FloatTensor, image_features: Optional[torch.FloatTensor]=None, video_features: Optional[torch.FloatTensor]=None): ''' Obtains multimodal placeholder mask from `input_ids` or `inputs_embeds`, and checks that the placeholder token count is equal to the length of multimodal features. If the lengths are different, an error is raised. ''' pass @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2_5_VLModelOutputWithPast]: ''' image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. ''' pass
13
5
47
5
35
7
8
0.2
1
17
10
0
5
9
6
7
290
34
214
71
177
43
107
40
100
25
2
3
45
4,737
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLPatchMerger
import torch import torch.nn as nn from ..qwen2.modeling_qwen2 import Qwen2RMSNorm import torch.nn.functional as F class Qwen2_5_VLPatchMerger(nn.Module): def __init__(self, dim: int, context_dim: int, spatial_merge_size: int=2) -> None: super().__init__() self.hidden_size = context_dim * spatial_merge_size ** 2 self.ln_q = Qwen2RMSNorm(context_dim, eps=1e-06) self.mlp = nn.Sequential(nn.Linear(self.hidden_size, self.hidden_size), nn.GELU(), nn.Linear(self.hidden_size, dim)) def forward(self, x: torch.Tensor) -> torch.Tensor: x = self.mlp(self.ln_q(x).view(-1, self.hidden_size)) return x
class Qwen2_5_VLPatchMerger(nn.Module): def __init__(self, dim: int, context_dim: int, spatial_merge_size: int=2) -> None: pass def forward(self, x: torch.Tensor) -> torch.Tensor: pass
3
0
6
0
6
0
1
0
1
4
1
0
2
3
2
12
14
1
13
6
10
0
9
6
6
1
1
0
2
4,738
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLPreTrainedModel
from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel from ...utils import TransformersKwargs, auto_docstring, can_return_tuple, is_torchdynamo_compiling, logging from .configuration_qwen2_5_vl import Qwen2_5_VLConfig, Qwen2_5_VLTextConfig, Qwen2_5_VLVisionConfig @auto_docstring class Qwen2_5_VLPreTrainedModel(PreTrainedModel): config: Qwen2_5_VLConfig base_model_prefix = 'model' supports_gradient_checkpointing = True _no_split_modules = ['Qwen2_5_VLDecoderLayer', 'Qwen2_5_VLVisionBlock'] _skip_keys_device_placement = 'past_key_values' _supports_flash_attn = True _supports_sdpa = True _can_compile_fullgraph = True _supports_attention_backend = True
@auto_docstring class Qwen2_5_VLPreTrainedModel(PreTrainedModel): pass
2
0
10
0
10
0
5
0.05
1
0
0
3
1
0
1
1
21
1
20
12
18
1
19
12
17
5
1
2
5
4,739
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLRotaryEmbedding
import torch.nn as nn from ...modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update import torch from .configuration_qwen2_5_vl import Qwen2_5_VLConfig, Qwen2_5_VLTextConfig, Qwen2_5_VLVisionConfig import torch.nn.functional as F class Qwen2_5_VLRotaryEmbedding(nn.Module): inv_freq: torch.Tensor def __init__(self, config: Qwen2_5_VLTextConfig, device=None): super().__init__() if hasattr(config, 'rope_scaling') and config.rope_scaling is not None: self.rope_type = config.rope_scaling.get('rope_type', config.rope_scaling.get('type')) else: self.rope_type = 'default' self.max_seq_len_cached = config.max_position_embeddings self.original_max_seq_len = config.max_position_embeddings self.config = config self.rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type] inv_freq, self.attention_scaling = self.rope_init_fn(self.config, device) self.register_buffer('inv_freq', inv_freq, persistent=False) self.original_inv_freq = self.inv_freq @torch.no_grad() @dynamic_rope_update def forward(self, x, position_ids): inv_freq_expanded = self.inv_freq[None, None, :, None].float().expand(3, position_ids.shape[1], -1, 1) position_ids_expanded = position_ids[:, :, None, :].float() device_type = x.device.type if isinstance(x.device.type, str) and x.device.type != 'mps' else 'cpu' with torch.autocast(device_type=device_type, enabled=False): freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(2, 3) emb = torch.cat((freqs, freqs), dim=-1) cos = emb.cos() * self.attention_scaling sin = emb.sin() * self.attention_scaling return (cos.to(dtype=x.dtype), sin.to(dtype=x.dtype))
class Qwen2_5_VLRotaryEmbedding(nn.Module): def __init__(self, config: Qwen2_5_VLTextConfig, device=None): pass @torch.no_grad() @dynamic_rope_update def forward(self, x, position_ids): pass
5
0
18
2
13
5
3
0.34
1
4
1
0
3
7
3
13
59
8
41
21
36
14
37
20
33
3
1
1
8
4,740
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLVisionAttention
from typing import Any, Callable, Optional, Union from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel import torch.nn.functional as F import torch.nn as nn import torch from .configuration_qwen2_5_vl import Qwen2_5_VLConfig, Qwen2_5_VLTextConfig, Qwen2_5_VLVisionConfig class Qwen2_5_VLVisionAttention(nn.Module): def __init__(self, config: Qwen2_5_VLVisionConfig) -> None: super().__init__() self.dim = config.hidden_size self.num_heads = config.num_heads self.head_dim = self.dim // self.num_heads self.num_key_value_groups = 1 self.qkv = nn.Linear(self.dim, self.dim * 3, bias=True) self.proj = nn.Linear(self.dim, self.dim) self.scaling = self.head_dim ** (-0.5) self.config = config self.attention_dropout = 0.0 self.is_causal = False def forward(self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: Optional[torch.Tensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs) -> torch.Tensor: seq_length = hidden_states.shape[0] query_states, key_states, value_states = self.qkv(hidden_states).reshape(seq_length, 3, self.num_heads, -1).permute(1, 0, 2, 3).unbind(0) cos, sin = position_embeddings query_states, key_states = apply_rotary_pos_emb_vision(query_states, key_states, cos, sin) query_states = query_states.transpose(0, 1).unsqueeze(0) key_states = key_states.transpose(0, 1).unsqueeze(0) value_states = value_states.transpose(0, 1).unsqueeze(0) attention_interface: Callable = eager_attention_forward if self.config._attn_implementation != 'eager': attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation] if self.config._attn_implementation == 'flash_attention_2': max_seqlen = (cu_seqlens[1:] - cu_seqlens[:-1]).max() attn_output, _ = attention_interface(self, query_states, key_states, value_states, attention_mask=None, scaling=self.scaling, dropout=0.0 if not self.training else self.attention_dropout, cu_seq_lens_q=cu_seqlens, cu_seq_lens_k=cu_seqlens, max_length_q=max_seqlen, max_length_k=max_seqlen, is_causal=False, **kwargs) else: lengths = cu_seqlens[1:] - cu_seqlens[:-1] splits = [torch.split(tensor, lengths.tolist(), dim=2) for tensor in (query_states, key_states, value_states)] attn_outputs = [attention_interface(self, q, k, v, attention_mask=None, scaling=self.scaling, dropout=0.0 if not self.training else self.attention_dropout, is_causal=False, **kwargs)[0] for q, k, v in zip(*splits)] attn_output = torch.cat(attn_outputs, dim=1) attn_output = attn_output.reshape(seq_length, -1).contiguous() attn_output = self.proj(attn_output) return attn_output
class Qwen2_5_VLVisionAttention(nn.Module): def __init__(self, config: Qwen2_5_VLVisionConfig) -> None: pass def forward(self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: Optional[torch.Tensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs) -> torch.Tensor: pass
3
0
16
1
15
0
2
0
1
4
0
0
2
4
2
12
33
3
30
15
25
0
26
13
23
2
1
1
3
4,741
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VLVisionBlock
import torch.nn as nn import torch from ..qwen2.modeling_qwen2 import Qwen2RMSNorm from ...modeling_layers import GradientCheckpointingLayer from typing import Any, Callable, Optional, Union import torch.nn.functional as F class Qwen2_5_VLVisionBlock(GradientCheckpointingLayer): def __init__(self, config, attn_implementation: str='sdpa') -> None: super().__init__() self.norm1 = Qwen2RMSNorm(config.hidden_size, eps=1e-06) self.norm2 = Qwen2RMSNorm(config.hidden_size, eps=1e-06) self.attn = Qwen2_5_VLVisionAttention(config=config) self.mlp = Qwen2_5_VLMLP(config, bias=True) def forward(self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: Optional[torch.Tensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs) -> torch.Tensor: hidden_states = hidden_states + self.attn(self.norm1(hidden_states), cu_seqlens=cu_seqlens, rotary_pos_emb=rotary_pos_emb, position_embeddings=position_embeddings, **kwargs) hidden_states = hidden_states + self.mlp(self.norm2(hidden_states)) return hidden_states
class Qwen2_5_VLVisionBlock(GradientCheckpointingLayer): def __init__(self, config, attn_implementation: str='sdpa') -> None: pass def forward(self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: Optional[torch.Tensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs) -> torch.Tensor: pass
3
0
8
0
8
0
1
0
1
5
2
0
2
4
2
12
18
1
17
7
14
0
11
7
8
1
1
0
2
4,742
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VisionPatchEmbed
import torch import torch.nn as nn import torch.nn.functional as F class Qwen2_5_VisionPatchEmbed(nn.Module): def __init__(self, patch_size: int=14, temporal_patch_size: int=2, in_channels: int=3, embed_dim: int=1152) -> None: super().__init__() self.patch_size = patch_size self.temporal_patch_size = temporal_patch_size self.in_channels = in_channels self.embed_dim = embed_dim kernel_size = [temporal_patch_size, patch_size, patch_size] self.proj = nn.Conv3d(in_channels, embed_dim, kernel_size=kernel_size, stride=kernel_size, bias=False) def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: target_dtype = self.proj.weight.dtype hidden_states = hidden_states.view(-1, self.in_channels, self.temporal_patch_size, self.patch_size, self.patch_size) hidden_states = self.proj(hidden_states.to(dtype=target_dtype)).view(-1, self.embed_dim) return hidden_states
class Qwen2_5_VisionPatchEmbed(nn.Module): def __init__(self, patch_size: int=14, temporal_patch_size: int=2, in_channels: int=3, embed_dim: int=1152) -> None: pass def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: pass
3
0
11
1
11
0
1
0
1
3
0
0
2
5
2
12
24
2
22
16
13
0
14
10
11
1
1
0
2
4,743
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VisionRotaryEmbedding
import torch.nn as nn import torch.nn.functional as F import torch class Qwen2_5_VisionRotaryEmbedding(nn.Module): inv_freq: torch.Tensor def __init__(self, dim: int, theta: float=10000.0) -> None: super().__init__() inv_freq = 1.0 / theta ** (torch.arange(0, dim, 2, dtype=torch.float) / dim) self.register_buffer('inv_freq', inv_freq, persistent=False) def forward(self, seqlen: int) -> torch.Tensor: seq = torch.arange(seqlen, device=self.inv_freq.device, dtype=self.inv_freq.dtype) freqs = torch.outer(seq, self.inv_freq) return freqs
class Qwen2_5_VisionRotaryEmbedding(nn.Module): def __init__(self, dim: int, theta: float=10000.0) -> None: pass def forward(self, seqlen: int) -> torch.Tensor: pass
3
0
4
0
4
0
1
0
1
4
0
0
2
0
2
12
10
1
9
6
6
0
9
6
6
1
1
0
2
4,744
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modeling_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modeling_qwen2_5_vl.Qwen2_5_VisionTransformerPretrainedModel
from .configuration_qwen2_5_vl import Qwen2_5_VLConfig, Qwen2_5_VLTextConfig, Qwen2_5_VLVisionConfig import torch.nn as nn import torch.nn.functional as F import torch class Qwen2_5_VisionTransformerPretrainedModel(Qwen2_5_VLPreTrainedModel): config: Qwen2_5_VLVisionConfig _no_split_modules = ['Qwen2_5_VLVisionBlock'] def __init__(self, config, *inputs, **kwargs) -> None: super().__init__(config, *inputs, **kwargs) self.spatial_merge_size = config.spatial_merge_size self.patch_size = config.patch_size self.fullatt_block_indexes = config.fullatt_block_indexes self.window_size = config.window_size self.spatial_merge_unit = self.spatial_merge_size * self.spatial_merge_size self.patch_embed = Qwen2_5_VisionPatchEmbed(patch_size=config.patch_size, temporal_patch_size=config.temporal_patch_size, in_channels=config.in_channels, embed_dim=config.hidden_size) head_dim = config.hidden_size // config.num_heads self.rotary_pos_emb = Qwen2_5_VisionRotaryEmbedding(head_dim // 2) self.blocks = nn.ModuleList([Qwen2_5_VLVisionBlock(config) for _ in range(config.depth)]) self.merger = Qwen2_5_VLPatchMerger(dim=config.out_hidden_size, context_dim=config.hidden_size, spatial_merge_size=config.spatial_merge_size) self.gradient_checkpointing = False def rot_pos_emb(self, grid_thw): pos_ids = [] for t, h, w in grid_thw: hpos_ids = torch.arange(h).unsqueeze(1).expand(-1, w) hpos_ids = hpos_ids.reshape(h // self.spatial_merge_size, self.spatial_merge_size, w // self.spatial_merge_size, self.spatial_merge_size) hpos_ids = hpos_ids.permute(0, 2, 1, 3) hpos_ids = hpos_ids.flatten() wpos_ids = torch.arange(w).unsqueeze(0).expand(h, -1) wpos_ids = wpos_ids.reshape(h // self.spatial_merge_size, self.spatial_merge_size, w // self.spatial_merge_size, self.spatial_merge_size) wpos_ids = wpos_ids.permute(0, 2, 1, 3) wpos_ids = wpos_ids.flatten() pos_ids.append(torch.stack([hpos_ids, wpos_ids], dim=-1).repeat(t, 1)) pos_ids = torch.cat(pos_ids, dim=0) max_grid_size = grid_thw[:, 1:].max() rotary_pos_emb_full = self.rotary_pos_emb(max_grid_size) rotary_pos_emb = rotary_pos_emb_full[pos_ids].flatten(1) return rotary_pos_emb def get_window_index(self, grid_thw): window_index: list = [] cu_window_seqlens: list = [0] window_index_id = 0 vit_merger_window_size = self.window_size // self.spatial_merge_size // self.patch_size for grid_t, grid_h, grid_w in grid_thw: llm_grid_h, llm_grid_w = (grid_h // self.spatial_merge_size, grid_w // self.spatial_merge_size) index = torch.arange(grid_t * llm_grid_h * llm_grid_w).reshape(grid_t, llm_grid_h, llm_grid_w) pad_h = vit_merger_window_size - llm_grid_h % vit_merger_window_size pad_w = vit_merger_window_size - llm_grid_w % vit_merger_window_size num_windows_h = (llm_grid_h + pad_h) // vit_merger_window_size num_windows_w = (llm_grid_w + pad_w) // vit_merger_window_size index_padded = F.pad(index, (0, pad_w, 0, pad_h), 'constant', -100) index_padded = index_padded.reshape(grid_t, num_windows_h, vit_merger_window_size, num_windows_w, vit_merger_window_size) index_padded = index_padded.permute(0, 1, 3, 2, 4).reshape(grid_t, num_windows_h * num_windows_w, vit_merger_window_size, vit_merger_window_size) seqlens = (index_padded != -100).sum([2, 3]).reshape(-1) index_padded = index_padded.reshape(-1) index_new = index_padded[index_padded != -100] window_index.append(index_new + window_index_id) cu_seqlens_tmp = seqlens.cumsum(0) * self.spatial_merge_unit + cu_window_seqlens[-1] cu_window_seqlens.extend(cu_seqlens_tmp.tolist()) window_index_id += (grid_t * llm_grid_h * llm_grid_w).item() window_index = torch.cat(window_index, dim=0) return (window_index, cu_window_seqlens) def forward(self, hidden_states: torch.Tensor, grid_thw: torch.Tensor, **kwargs) -> torch.Tensor: """ Args: hidden_states (`torch.Tensor` of shape `(seq_len, hidden_size)`): The final hidden states of the model. grid_thw (`torch.Tensor` of shape `(num_images_or_videos, 3)`): The temporal, height and width of feature shape of each image in LLM. Returns: `torch.Tensor`: hidden_states. """ hidden_states = self.patch_embed(hidden_states) rotary_pos_emb = self.rot_pos_emb(grid_thw) window_index, cu_window_seqlens = self.get_window_index(grid_thw) cu_window_seqlens = torch.tensor(cu_window_seqlens, device=hidden_states.device, dtype=grid_thw.dtype if torch.jit.is_tracing() else torch.int32) cu_window_seqlens = torch.unique_consecutive(cu_window_seqlens) seq_len, _ = hidden_states.size() hidden_states = hidden_states.reshape(seq_len // self.spatial_merge_unit, self.spatial_merge_unit, -1) hidden_states = hidden_states[window_index, :, :] hidden_states = hidden_states.reshape(seq_len, -1) rotary_pos_emb = rotary_pos_emb.reshape(seq_len // self.spatial_merge_unit, self.spatial_merge_unit, -1) rotary_pos_emb = rotary_pos_emb[window_index, :, :] rotary_pos_emb = rotary_pos_emb.reshape(seq_len, -1) emb = torch.cat((rotary_pos_emb, rotary_pos_emb), dim=-1) position_embeddings = (emb.cos(), emb.sin()) cu_seqlens = torch.repeat_interleave(grid_thw[:, 1] * grid_thw[:, 2], grid_thw[:, 0]).cumsum(dim=0, dtype=grid_thw.dtype if torch.jit.is_tracing() else torch.int32) cu_seqlens = F.pad(cu_seqlens, (1, 0), value=0) for layer_num, blk in enumerate(self.blocks): if layer_num in self.fullatt_block_indexes: cu_seqlens_now = cu_seqlens else: cu_seqlens_now = cu_window_seqlens hidden_states = blk(hidden_states, cu_seqlens=cu_seqlens_now, position_embeddings=position_embeddings, **kwargs) hidden_states = self.merger(hidden_states) reverse_indices = torch.argsort(window_index) hidden_states = hidden_states[reverse_indices, :] return hidden_states
class Qwen2_5_VisionTransformerPretrainedModel(Qwen2_5_VLPreTrainedModel): def __init__(self, config, *inputs, **kwargs) -> None: pass def rot_pos_emb(self, grid_thw): pass def get_window_index(self, grid_thw): pass def forward(self, hidden_states: torch.Tensor, grid_thw: torch.Tensor, **kwargs) -> torch.Tensor: ''' Args: hidden_states (`torch.Tensor` of shape `(seq_len, hidden_size)`): The final hidden states of the model. grid_thw (`torch.Tensor` of shape `(num_images_or_videos, 3)`): The temporal, height and width of feature shape of each image in LLM. Returns: `torch.Tensor`: hidden_states. ''' pass
5
1
39
3
33
3
3
0.1
1
9
4
0
4
10
4
5
162
16
133
48
128
13
83
47
78
6
2
2
11
4,745
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLCausalLMOutputWithPast
from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding class Qwen2_5_VLCausalLMOutputWithPast(Qwen2VLCausalLMOutputWithPast): pass
class Qwen2_5_VLCausalLMOutputWithPast(Qwen2VLCausalLMOutputWithPast): pass
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
2
0
2
1
1
0
2
1
1
0
2
0
0
4,746
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLConfig
from transformers.models.qwen2_vl.configuration_qwen2_vl import Qwen2VLConfig, Qwen2VLTextConfig class Qwen2_5_VLConfig(Qwen2VLConfig): model_type = 'qwen2_5_vl' sub_configs = {'vision_config': Qwen2_5_VLVisionConfig, 'text_config': Qwen2_5_VLTextConfig}
class Qwen2_5_VLConfig(Qwen2VLConfig): pass
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
3
0
3
3
2
0
3
3
2
0
2
0
0
4,747
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLForConditionalGeneration
from typing import Optional, Union import torch.nn as nn import torch.nn.functional as F from ...processing_utils import MultiModalData, ProcessingKwargs, Unpack, VideosKwargs from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding import torch from ...cache_utils import Cache class Qwen2_5_VLForConditionalGeneration(Qwen2VLForConditionalGeneration): accepts_loss_kwargs = False def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, logits_to_keep: Union[int, torch.Tensor]=0, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2_5_VLCausalLMOutputWithPast]: """ labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. Example: ```python >>> from PIL import Image >>> import requests >>> from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration >>> model = Qwen2_5_VLForConditionalGeneration.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") >>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") >>> messages = [ { "role": "user", "content": [ {"type": "image"}, {"type": "text", "text": "What is shown in this image?"}, ], }, ] >>> url = "https://www.ilankelman.org/stopsigns/australia.jpg" >>> image = Image.open(requests.get(url, stream=True).raw) >>> text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) >>> inputs = processor(text=[text], images=[image], vision_infos=[vision_infos]) >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "The image shows a street scene with a red stop sign in the foreground. In the background, there is a large red gate with Chinese characters ..." ```""" 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 outputs = self.model(input_ids=input_ids, pixel_values=pixel_values, pixel_values_videos=pixel_values_videos, image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, second_per_grid_ts=second_per_grid_ts, position_ids=position_ids, attention_mask=attention_mask, past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache, output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=True, cache_position=cache_position, **kwargs) hidden_states = outputs[0] slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep logits = self.lm_head(hidden_states[:, slice_indices, :]) loss = None if labels is not None: loss = self.loss_function(logits=logits, labels=labels, vocab_size=self.config.text_config.vocab_size, **kwargs) return Qwen2_5_VLCausalLMOutputWithPast(loss=loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions, rope_deltas=outputs.rope_deltas) def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, cache_position=None, position_ids=None, use_cache=True, pixel_values=None, pixel_values_videos=None, image_grid_thw=None, video_grid_thw=None, second_per_grid_ts=None, **kwargs): model_inputs = super().prepare_inputs_for_generation(input_ids, past_key_values=past_key_values, attention_mask=attention_mask, inputs_embeds=inputs_embeds, cache_position=cache_position, position_ids=position_ids, pixel_values=pixel_values, pixel_values_videos=pixel_values_videos, image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, second_per_grid_ts=second_per_grid_ts, use_cache=use_cache, **kwargs) if position_ids is None: if cache_position[0] == 0 or self.model.rope_deltas is None: vision_positions, rope_deltas = self.model.get_rope_index(model_inputs.get('input_ids', None), image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, attention_mask=attention_mask) self.model.rope_deltas = rope_deltas elif 'position_ids' in model_inputs: batch_size, seq_length = model_inputs['position_ids'].shape device = model_inputs['position_ids'].device position_ids = torch.arange(seq_length, device=device) position_ids = position_ids.view(1, 1, -1).expand(3, batch_size, -1) delta = cache_position[0] + self.model.rope_deltas delta = delta.repeat_interleave(batch_size // delta.shape[0], dim=0) vision_positions = position_ids + delta.expand_as(position_ids) text_positions = model_inputs['position_ids'][None, ...] model_inputs['position_ids'] = torch.cat([text_positions, vision_positions], dim=0) if cache_position[0] != 0: model_inputs['pixel_values'] = None model_inputs['pixel_values_videos'] = None return model_inputs
class Qwen2_5_VLForConditionalGeneration(Qwen2VLForConditionalGeneration): def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, logits_to_keep: Union[int, torch.Tensor]=0, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2_5_VLCausalLMOutputWithPast]: ''' labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. Example: ```python >>> from PIL import Image >>> import requests >>> from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration >>> model = Qwen2_5_VLForConditionalGeneration.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") >>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct") >>> messages = [ { "role": "user", "content": [ {"type": "image"}, {"type": "text", "text": "What is shown in this image?"}, ], }, ] >>> url = "https://www.ilankelman.org/stopsigns/australia.jpg" >>> image = Image.open(requests.get(url, stream=True).raw) >>> text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) >>> inputs = processor(text=[text], images=[image], vision_infos=[vision_infos]) >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "The image shows a street scene with a red stop sign in the foreground. In the background, there is a large red gate with Chinese characters ..." ```''' pass def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, cache_position=None, position_ids=None, use_cache=True, pixel_values=None, pixel_values_videos=None, image_grid_thw=None, video_grid_thw=None, second_per_grid_ts=None, **kwargs): pass
3
1
110
11
76
25
10
0.33
1
10
3
0
4
3
4
15
448
46
307
109
261
100
165
67
160
17
3
5
40
4,748
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLMLP
from ...activations import ACT2FN import torch.nn as nn class Qwen2_5_VLMLP(nn.Module): def __init__(self, config, bias: bool=False): super().__init__() self.hidden_size = config.hidden_size self.intermediate_size = config.intermediate_size self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=bias) self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=bias) self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=bias) self.act_fn = ACT2FN[config.hidden_act] def forward(self, hidden_state): return self.down_proj(self.act_fn(self.gate_proj(hidden_state)) * self.up_proj(hidden_state))
class Qwen2_5_VLMLP(nn.Module): def __init__(self, config, bias: bool=False): pass def forward(self, hidden_state): pass
3
0
5
0
5
0
1
0
1
2
0
0
2
6
2
12
12
1
11
9
8
0
11
9
8
1
1
0
2
4,749
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLModel
from ...processing_utils import MultiModalData, ProcessingKwargs, Unpack, VideosKwargs from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding import torch.nn.functional as F from typing import Optional, Union import torch from ...cache_utils import Cache from ...utils import is_torchdynamo_compiling, logging import torch.nn as nn class Qwen2_5_VLModel(Qwen2VLModel): config: Qwen2_5_VLConfig base_model_prefix = '' _no_split_modules = ['Qwen2_5_VLDecoderLayer', 'Qwen2_5_VLVisionBlock'] accepts_loss_kwargs = False def __init__(self, config): super().__init__(config) self.visual = Qwen2_5_VisionTransformerPretrainedModel._from_config(config.vision_config) def get_rope_index(self, input_ids: Optional[torch.LongTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, attention_mask: Optional[torch.Tensor]=None) -> tuple[torch.Tensor, torch.Tensor]: """ Calculate the 3D rope index based on image and video's temporal, height and width in LLM. Explanation: Each embedding sequence contains vision embedding and text embedding or just contains text embedding. For pure text embedding sequence, the rotary position embedding has no difference with modern LLMs. Examples: input_ids: [T T T T T], here T is for text. temporal position_ids: [0, 1, 2, 3, 4] height position_ids: [0, 1, 2, 3, 4] width position_ids: [0, 1, 2, 3, 4] For vision and text embedding sequence, we calculate 3D rotary position embedding for vision part and 1D rotary position embedding for text part. Examples: Temporal (Time): 3 patches, representing different segments of the video in time. Height: 2 patches, dividing each frame vertically. Width: 2 patches, dividing each frame horizontally. We also have some important parameters: fps (Frames Per Second): The video's frame rate, set to 1. This means one frame is processed each second. tokens_per_second: This is a crucial parameter. It dictates how many "time-steps" or "temporal tokens" are conceptually packed into a one-second interval of the video. In this case, we have 25 tokens per second. So each second of the video will be represented with 25 separate time points. It essentially defines the temporal granularity. temporal_patch_size: The number of frames that compose one temporal patch. Here, it's 2 frames. interval: The step size for the temporal position IDs, calculated as tokens_per_second * temporal_patch_size / fps. In this case, 25 * 2 / 1 = 50. This means that each temporal patch will be have a difference of 50 in the temporal position IDs. input_ids: [V V V V V V V V V V V V T T T T T], here V is for vision. vision temporal position_ids: [0, 0, 0, 0, 50, 50, 50, 50, 100, 100, 100, 100] vision height position_ids: [0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1] vision width position_ids: [0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1] text temporal position_ids: [101, 102, 103, 104, 105] text height position_ids: [101, 102, 103, 104, 105] text width position_ids: [101, 102, 103, 104, 105] Here we calculate the text start position_ids as the max vision position_ids plus 1. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *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**. Returns: position_ids (`torch.LongTensor` of shape `(3, batch_size, sequence_length)`) mrope_position_deltas (`torch.Tensor` of shape `(batch_size)`) """ spatial_merge_size = self.config.vision_config.spatial_merge_size image_token_id = self.config.image_token_id video_token_id = self.config.video_token_id vision_start_token_id = self.config.vision_start_token_id mrope_position_deltas = [] if input_ids is not None and (image_grid_thw is not None or video_grid_thw is not None): total_input_ids = input_ids if attention_mask is not None: attention_mask = attention_mask == 1 position_ids = torch.ones(3, input_ids.shape[0], input_ids.shape[1], dtype=input_ids.dtype, device=input_ids.device) image_index, video_index = (0, 0) for i, input_ids in enumerate(total_input_ids): if attention_mask is not None: input_ids = input_ids[attention_mask[i]] image_nums, video_nums = (0, 0) vision_start_indices = torch.argwhere(input_ids == vision_start_token_id).squeeze(1) vision_tokens = input_ids[vision_start_indices + 1] image_nums = (vision_tokens == image_token_id).sum() video_nums = (vision_tokens == video_token_id).sum() input_tokens = input_ids.tolist() llm_pos_ids_list: list = [] st = 0 remain_images, remain_videos = (image_nums, video_nums) for _ in range(image_nums + video_nums): if image_token_id in input_tokens and remain_images > 0: ed_image = input_tokens.index(image_token_id, st) else: ed_image = len(input_tokens) + 1 if video_token_id in input_tokens and remain_videos > 0: ed_video = input_tokens.index(video_token_id, st) else: ed_video = len(input_tokens) + 1 if ed_image < ed_video: t, h, w = (image_grid_thw[image_index][0], image_grid_thw[image_index][1], image_grid_thw[image_index][2]) second_per_grid_t = 0 image_index += 1 remain_images -= 1 ed = ed_image else: t, h, w = (video_grid_thw[video_index][0], video_grid_thw[video_index][1], video_grid_thw[video_index][2]) if second_per_grid_ts is not None: second_per_grid_t = second_per_grid_ts[video_index] else: second_per_grid_t = 1.0 video_index += 1 remain_videos -= 1 ed = ed_video llm_grid_t, llm_grid_h, llm_grid_w = (t.item(), h.item() // spatial_merge_size, w.item() // spatial_merge_size) text_len = ed - st st_idx = llm_pos_ids_list[-1].max() + 1 if len(llm_pos_ids_list) > 0 else 0 llm_pos_ids_list.append(torch.arange(text_len).view(1, -1).expand(3, -1) + st_idx) range_tensor = torch.arange(llm_grid_t).view(-1, 1) expanded_range = range_tensor.expand(-1, llm_grid_h * llm_grid_w) second_per_grid_t = torch.as_tensor(second_per_grid_t, dtype=range_tensor.dtype, device=range_tensor.device) time_tensor = expanded_range * second_per_grid_t * self.config.vision_config.tokens_per_second time_tensor_long = time_tensor.long() t_index = time_tensor_long.flatten() h_index = torch.arange(llm_grid_h).view(1, -1, 1).expand(llm_grid_t, -1, llm_grid_w).flatten() w_index = torch.arange(llm_grid_w).view(1, 1, -1).expand(llm_grid_t, llm_grid_h, -1).flatten() llm_pos_ids_list.append(torch.stack([t_index, h_index, w_index]) + text_len + st_idx) st = ed + llm_grid_t * llm_grid_h * llm_grid_w if st < len(input_tokens): st_idx = llm_pos_ids_list[-1].max() + 1 if len(llm_pos_ids_list) > 0 else 0 text_len = len(input_tokens) - st llm_pos_ids_list.append(torch.arange(text_len).view(1, -1).expand(3, -1) + st_idx) llm_positions = torch.cat(llm_pos_ids_list, dim=1).reshape(3, -1) if attention_mask is not None: position_ids[..., i, attention_mask[i]] = llm_positions.to(position_ids.device) else: position_ids[..., i, :] = llm_positions.to(position_ids.device) mrope_position_deltas.append(llm_positions.max() + 1 - len(total_input_ids[i])) mrope_position_deltas = torch.tensor(mrope_position_deltas).unsqueeze(1).to(device=input_ids.device) return (position_ids, mrope_position_deltas) else: if attention_mask is not None: position_ids = attention_mask.long().cumsum(-1) - 1 position_ids.masked_fill_(attention_mask == 0, 1) position_ids = position_ids.unsqueeze(0).expand(3, -1, -1).to(attention_mask.device) max_position_ids = position_ids.max(0, keepdim=False)[0].max(-1, keepdim=True)[0] mrope_position_deltas = max_position_ids + 1 - attention_mask.shape[-1] else: position_ids = torch.arange(input_ids.shape[1], device=input_ids.device).view(1, 1, -1).expand(3, input_ids.shape[0], -1) mrope_position_deltas = torch.zeros([input_ids.shape[0], 1], device=input_ids.device, dtype=input_ids.dtype) return (position_ids, mrope_position_deltas) def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2_5_VLModelOutputWithPast]: """ image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. """ 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 inputs_embeds is None: inputs_embeds = self.get_input_embeddings()(input_ids) if pixel_values is not None: image_embeds = self.get_image_features(pixel_values, image_grid_thw) image_embeds = torch.cat(image_embeds, dim=0).to(inputs_embeds.device, inputs_embeds.dtype) image_mask, _ = self.get_placeholder_mask(input_ids, inputs_embeds=inputs_embeds, image_features=image_embeds) inputs_embeds = inputs_embeds.masked_scatter(image_mask, image_embeds) if pixel_values_videos is not None: video_embeds = self.get_video_features(pixel_values_videos, video_grid_thw) video_embeds = torch.cat(video_embeds, dim=0).to(inputs_embeds.device, inputs_embeds.dtype) _, video_mask = self.get_placeholder_mask(input_ids, inputs_embeds=inputs_embeds, video_features=video_embeds) inputs_embeds = inputs_embeds.masked_scatter(video_mask, video_embeds) if position_ids is None: prefill_compiled_stage = is_torchdynamo_compiling() and (input_ids is not None and input_ids.shape[1] != 1 or (inputs_embeds is not None and inputs_embeds.shape[1] != 1)) prefill_noncompiled_stage = not is_torchdynamo_compiling() and (cache_position is not None and cache_position[0] == 0 or (past_key_values is None or past_key_values.get_seq_length() == 0)) if (prefill_compiled_stage or prefill_noncompiled_stage) or self.rope_deltas is None: position_ids, rope_deltas = self.get_rope_index(input_ids, image_grid_thw, video_grid_thw, second_per_grid_ts=second_per_grid_ts, attention_mask=attention_mask) self.rope_deltas = rope_deltas else: batch_size, seq_length, _ = inputs_embeds.shape position_ids = torch.arange(seq_length, device=inputs_embeds.device) position_ids = position_ids.view(1, 1, -1).expand(3, batch_size, -1) if cache_position is not None: delta = (cache_position[0] + self.rope_deltas).to(inputs_embeds.device) else: delta = torch.zeros((batch_size, seq_length), device=inputs_embeds.device) delta = delta.repeat_interleave(batch_size // delta.shape[0], dim=1) position_ids = position_ids + delta.to(position_ids.device) outputs = self.language_model(input_ids=None, position_ids=position_ids, attention_mask=attention_mask, past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache, output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=True, cache_position=cache_position, **kwargs) output = Qwen2_5_VLModelOutputWithPast(last_hidden_state=outputs.last_hidden_state, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions, rope_deltas=self.rope_deltas) return output if return_dict else output.to_tuple()
class Qwen2_5_VLModel(Qwen2VLModel): def __init__(self, config): pass def get_rope_index(self, input_ids: Optional[torch.LongTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, attention_mask: Optional[torch.Tensor]=None) -> tuple[torch.Tensor, torch.Tensor]: ''' Calculate the 3D rope index based on image and video's temporal, height and width in LLM. Explanation: Each embedding sequence contains vision embedding and text embedding or just contains text embedding. For pure text embedding sequence, the rotary position embedding has no difference with modern LLMs. Examples: input_ids: [T T T T T], here T is for text. temporal position_ids: [0, 1, 2, 3, 4] height position_ids: [0, 1, 2, 3, 4] width position_ids: [0, 1, 2, 3, 4] For vision and text embedding sequence, we calculate 3D rotary position embedding for vision part and 1D rotary position embedding for text part. Examples: Temporal (Time): 3 patches, representing different segments of the video in time. Height: 2 patches, dividing each frame vertically. Width: 2 patches, dividing each frame horizontally. We also have some important parameters: fps (Frames Per Second): The video's frame rate, set to 1. This means one frame is processed each second. tokens_per_second: This is a crucial parameter. It dictates how many "time-steps" or "temporal tokens" are conceptually packed into a one-second interval of the video. In this case, we have 25 tokens per second. So each second of the video will be represented with 25 separate time points. It essentially defines the temporal granularity. temporal_patch_size: The number of frames that compose one temporal patch. Here, it's 2 frames. interval: The step size for the temporal position IDs, calculated as tokens_per_second * temporal_patch_size / fps. In this case, 25 * 2 / 1 = 50. This means that each temporal patch will be have a difference of 50 in the temporal position IDs. input_ids: [V V V V V V V V V V V V T T T T T], here V is for vision. vision temporal position_ids: [0, 0, 0, 0, 50, 50, 50, 50, 100, 100, 100, 100] vision height position_ids: [0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1] vision width position_ids: [0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1] text temporal position_ids: [101, 102, 103, 104, 105] text height position_ids: [101, 102, 103, 104, 105] text width position_ids: [101, 102, 103, 104, 105] Here we calculate the text start position_ids as the max vision position_ids plus 1. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *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**. Returns: position_ids (`torch.LongTensor` of shape `(3, batch_size, sequence_length)`) mrope_position_deltas (`torch.Tensor` of shape `(batch_size)`) ''' pass def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, second_per_grid_ts: Optional[torch.Tensor]=None, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2_5_VLModelOutputWithPast]: ''' image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. second_per_grid_ts (`torch.Tensor` of shape `(num_videos)`, *optional*): The time interval (in seconds) for each grid along the temporal dimension in the 3D position IDs. ''' pass
4
2
0
0
0
0
0
0
1
0
0
0
0
0
0
7
2
0
2
1
1
0
2
1
1
0
3
0
0
4,750
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLPatchMerger
from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding class Qwen2_5_VLPatchMerger(PatchMerger): def __init__(self, dim: int, context_dim: int, spatial_merge_size: int=2) -> None: super().__init__(dim, context_dim, spatial_merge_size) self.ln_q = Qwen2RMSNorm(context_dim, eps=1e-06)
class Qwen2_5_VLPatchMerger(PatchMerger): def __init__(self, dim: int, context_dim: int, spatial_merge_size: int=2) -> None: pass
2
0
3
0
3
0
1
0
1
3
0
0
1
1
1
13
4
0
4
3
2
0
4
3
2
1
2
0
1
4,751
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLPreTrainedModel
from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding class Qwen2_5_VLPreTrainedModel(Qwen2VLPreTrainedModel): pass
class Qwen2_5_VLPreTrainedModel(Qwen2VLPreTrainedModel): pass
1
0
0
0
0
0
0
0
1
0
0
1
0
0
0
1
2
0
2
1
1
0
2
1
1
0
2
0
0
4,752
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLProcessor
from transformers.models.qwen2_vl.processing_qwen2_vl import Qwen2VLImagesKwargs, Qwen2VLProcessor from ...tokenization_utils_base import PreTokenizedInput, TextInput import numpy as np from ...feature_extraction_utils import BatchFeature from ...processing_utils import MultiModalData, ProcessingKwargs, Unpack, VideosKwargs from ...video_utils import VideoInput from ...image_utils import ImageInput from typing import Optional, Union class Qwen2_5_VLProcessor(Qwen2VLProcessor): """ Constructs a Qwen2.5-VL processor which wraps a Qwen2.5-VL image processor and a Qwen2 tokenizer into a single processor. [`Qwen2_5_VLProcessor`] offers all the functionalities of [`Qwen2VLImageProcessor`] and [`Qwen2TokenizerFast`]. See the [`~Qwen2_5_VLProcessor.__call__`] and [`~Qwen2_5_VLProcessor.decode`] for more information. Args: image_processor ([`Qwen2VLImageProcessor`], *optional*): The image processor is a required input. tokenizer ([`Qwen2TokenizerFast`], *optional*): The tokenizer is a required input. video_processor ([`Qwen2_5_VLVideoProcessor`], *optional*): The video processor is a required input. chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string. """ image_processor_class = 'AutoImageProcessor' @property def model_input_names(self): tokenizer_input_names = self.tokenizer.model_input_names image_processor_input_names = self.image_processor.model_input_names names_from_processor = list(dict.fromkeys(tokenizer_input_names + image_processor_input_names)) return names_from_processor + ['second_per_grid_ts'] def __call__(self, images: Optional[ImageInput]=None, text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]=None, videos: Optional[VideoInput]=None, **kwargs: Unpack[Qwen2_5_VLProcessorKwargs]) -> BatchFeature: """ Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text` and `kwargs` arguments to Qwen2TokenizerFast's [`~Qwen2TokenizerFast.__call__`] if `text` is not `None` to encode the text. To prepare the vision inputs, this method forwards the `vision_infos` and `kwargs` arguments to Qwen2VLImageProcessor's [`~Qwen2VLImageProcessor.__call__`] if `vision_infos` is not `None`. Args: images (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `list[PIL.Image.Image]`, `list[np.ndarray]`, `list[torch.Tensor]`): The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch tensor. Both channels-first and channels-last formats are supported. text (`str`, `list[str]`, `list[list[str]]`): The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set `is_split_into_words=True` (to lift the ambiguity with a batch of sequences). videos (`np.ndarray`, `torch.Tensor`, `list[np.ndarray]`, `list[torch.Tensor]`): The image or batch of videos to be prepared. Each video can be a 4D NumPy array or PyTorch tensor, or a nested list of 3D frames. Both channels-first and channels-last formats are supported. return_tensors (`str` or [`~utils.TensorType`], *optional*): If set, will return tensors of a particular framework. Acceptable values are: - `'pt'`: Return PyTorch `torch.Tensor` objects. - `'np'`: Return NumPy `np.ndarray` objects. Returns: [`BatchFeature`]: A [`BatchFeature`] with the following fields: - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`. - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not `None`). - **pixel_values** -- Pixel values to be fed to a model. Returned when `images` is not `None`. - **pixel_values_videos** -- Pixel values of videos to be fed to a model. Returned when `videos` is not `None`. - **image_grid_thw** -- List of image 3D grid in LLM. Returned when `images` is not `None`. - **video_grid_thw** -- List of video 3D grid in LLM. Returned when `videos` is not `None`. - **second_per_grid_ts** -- List of video seconds per time grid. Returned when `videos` is not `None`. """ output_kwargs = self._merge_kwargs(Qwen2_5_VLProcessorKwargs, tokenizer_init_kwargs=self.tokenizer.init_kwargs, **kwargs) image_inputs = videos_inputs = {} if images is not None: image_inputs = self.image_processor(images=images, **output_kwargs['images_kwargs']) image_grid_thw = image_inputs['image_grid_thw'] if videos is not None: fps = output_kwargs['videos_kwargs'].get('fps', 2.0) videos_inputs = self.video_processor(videos=videos, **output_kwargs['videos_kwargs']) video_grid_thw = videos_inputs['video_grid_thw'] if isinstance(fps, (int, float)): second_per_grid_ts = [self.video_processor.temporal_patch_size / fps] * len(video_grid_thw) elif hasattr(fps, '__len__') and len(fps) == len(video_grid_thw): second_per_grid_ts = [self.video_processor.temporal_patch_size / tmp for tmp in fps] else: raise ValueError(f"The length of fps ({(len(fps) if hasattr(fps, '__len__') else fps)}) must be equal to the length of video_grid_thw ({len(video_grid_thw)}) or fps should be a single number.") videos_inputs.update({'second_per_grid_ts': second_per_grid_ts}) if not isinstance(text, list): text = [text] text = text.copy() if images is not None: merge_length = self.image_processor.merge_size ** 2 index = 0 for i in range(len(text)): while self.image_token in text[i]: num_image_tokens = image_grid_thw[index].prod() // merge_length text[i] = text[i].replace(self.image_token, '<|placeholder|>' * num_image_tokens, 1) index += 1 text[i] = text[i].replace('<|placeholder|>', self.image_token) if videos is not None: merge_length = self.video_processor.merge_size ** 2 index = 0 for i in range(len(text)): while self.video_token in text[i]: num_video_tokens = video_grid_thw[index].prod() // merge_length text[i] = text[i].replace(self.video_token, '<|placeholder|>' * num_video_tokens, 1) index += 1 text[i] = text[i].replace('<|placeholder|>', self.video_token) return_tensors = output_kwargs['text_kwargs'].pop('return_tensors', None) return_mm_token_type_ids = output_kwargs['text_kwargs'].pop('return_mm_token_type_ids', None) text_inputs = self.tokenizer(text, **output_kwargs['text_kwargs']) self._check_special_mm_tokens(text, text_inputs, modalities=['image', 'video']) if return_mm_token_type_ids: array_ids = np.array(text_inputs['input_ids']) mm_token_type_ids = np.zeros_like(text_inputs['input_ids']) mm_token_type_ids[array_ids == self.image_token_id] = 1 text_inputs['mm_token_type_ids'] = mm_token_type_ids.tolist() return BatchFeature(data={**text_inputs, **image_inputs, **videos_inputs}, tensor_type=return_tensors) def _get_num_multimodal_tokens(self, image_sizes=None, video_sizes=None, **kwargs): """ Computes the number of placeholder tokens needed for multimodal inputs with the given sizes. Args: image_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (height, width) per each image. video_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (num_frames, height, width) per each video. Returns: `MultiModalData`: A `MultiModalData` object holding number of tokens per each of the provided input modalities, along with other useful data. """ vision_data = {} if image_sizes is not None: images_kwargs = Qwen2_5_VLProcessorKwargs._defaults.get('images_kwargs', {}) images_kwargs.update(kwargs) merge_size = images_kwargs.get('merge_size', None) or self.image_processor.merge_size num_image_patches = [self.image_processor.get_number_of_image_patches(*image_size, images_kwargs) for image_size in image_sizes] num_image_tokens = [num_patches // merge_size ** 2 for num_patches in num_image_patches] vision_data.update({'num_image_tokens': num_image_tokens, 'num_image_patches': num_image_patches}) if video_sizes is not None: videos_kwargs = Qwen2_5_VLProcessorKwargs._defaults.get('videos_kwargs', {}) videos_kwargs.update(kwargs) num_video_patches = [self.video_processor.get_number_of_video_patches(*video_size, videos_kwargs) for video_size in video_sizes] num_video_tokens = [num_patches // merge_size ** 2 for num_patches in num_video_patches] vision_data['num_video_tokens'] = num_video_tokens return MultiModalData(**vision_data)
class Qwen2_5_VLProcessor(Qwen2VLProcessor): ''' Constructs a Qwen2.5-VL processor which wraps a Qwen2.5-VL image processor and a Qwen2 tokenizer into a single processor. [`Qwen2_5_VLProcessor`] offers all the functionalities of [`Qwen2VLImageProcessor`] and [`Qwen2TokenizerFast`]. See the [`~Qwen2_5_VLProcessor.__call__`] and [`~Qwen2_5_VLProcessor.decode`] for more information. Args: image_processor ([`Qwen2VLImageProcessor`], *optional*): The image processor is a required input. tokenizer ([`Qwen2TokenizerFast`], *optional*): The tokenizer is a required input. video_processor ([`Qwen2_5_VLVideoProcessor`], *optional*): The video processor is a required input. chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string. ''' @property def model_input_names(self): pass def __call__(self, images: Optional[ImageInput]=None, text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]=None, videos: Optional[VideoInput]=None, **kwargs: Unpack[Qwen2_5_VLProcessorKwargs]) -> BatchFeature: ''' Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text` and `kwargs` arguments to Qwen2TokenizerFast's [`~Qwen2TokenizerFast.__call__`] if `text` is not `None` to encode the text. To prepare the vision inputs, this method forwards the `vision_infos` and `kwargs` arguments to Qwen2VLImageProcessor's [`~Qwen2VLImageProcessor.__call__`] if `vision_infos` is not `None`. Args: images (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `list[PIL.Image.Image]`, `list[np.ndarray]`, `list[torch.Tensor]`): The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch tensor. Both channels-first and channels-last formats are supported. text (`str`, `list[str]`, `list[list[str]]`): The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set `is_split_into_words=True` (to lift the ambiguity with a batch of sequences). videos (`np.ndarray`, `torch.Tensor`, `list[np.ndarray]`, `list[torch.Tensor]`): The image or batch of videos to be prepared. Each video can be a 4D NumPy array or PyTorch tensor, or a nested list of 3D frames. Both channels-first and channels-last formats are supported. return_tensors (`str` or [`~utils.TensorType`], *optional*): If set, will return tensors of a particular framework. Acceptable values are: - `'pt'`: Return PyTorch `torch.Tensor` objects. - `'np'`: Return NumPy `np.ndarray` objects. Returns: [`BatchFeature`]: A [`BatchFeature`] with the following fields: - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`. - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not `None`). - **pixel_values** -- Pixel values to be fed to a model. Returned when `images` is not `None`. - **pixel_values_videos** -- Pixel values of videos to be fed to a model. Returned when `videos` is not `None`. - **image_grid_thw** -- List of image 3D grid in LLM. Returned when `images` is not `None`. - **video_grid_thw** -- List of video 3D grid in LLM. Returned when `videos` is not `None`. - **second_per_grid_ts** -- List of video seconds per time grid. Returned when `videos` is not `None`. ''' pass def _get_num_multimodal_tokens(self, image_sizes=None, video_sizes=None, **kwargs): ''' Computes the number of placeholder tokens needed for multimodal inputs with the given sizes. Args: image_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (height, width) per each image. video_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (num_frames, height, width) per each video. Returns: `MultiModalData`: A `MultiModalData` object holding number of tokens per each of the provided input modalities, along with other useful data. ''' pass
5
3
107
11
62
34
13
0.72
1
7
2
0
1
2
1
24
123
13
64
22
56
46
40
14
38
13
3
3
13
4,753
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLProcessorKwargs
from ...processing_utils import MultiModalData, ProcessingKwargs, Unpack, VideosKwargs class Qwen2_5_VLProcessorKwargs(ProcessingKwargs, total=False): images_kwargs: Qwen2_5_VLImagesKwargs videos_kwargs: Qwen2_5_VLVideosProcessorKwargs _defaults = {'text_kwargs': {'padding': False, 'return_mm_token_type_ids': False}}
class Qwen2_5_VLProcessorKwargs(ProcessingKwargs, total=False): pass
1
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
8
0
8
2
7
0
3
2
2
0
3
0
0
4,754
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLVideosProcessorKwargs
from typing import Optional, Union from ...processing_utils import MultiModalData, ProcessingKwargs, Unpack, VideosKwargs class Qwen2_5_VLVideosProcessorKwargs(VideosKwargs, total=False): fps: Union[list[float], float]
class Qwen2_5_VLVideosProcessorKwargs(VideosKwargs, total=False): pass
1
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
2
0
2
1
1
0
2
1
1
0
2
0
0
4,755
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLVisionAttention
from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding class Qwen2_5_VLVisionAttention(VisionAttention): def __init__(self, config: Qwen2_5_VLVisionConfig) -> None: super().__init__(config) self.dim = config.hidden_size
class Qwen2_5_VLVisionAttention(VisionAttention): def __init__(self, config: Qwen2_5_VLVisionConfig) -> None: pass
2
0
0
0
0
0
0
0
1
0
0
0
0
0
0
12
2
0
2
1
1
0
2
1
1
0
2
0
0
4,756
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLVisionBlock
from ...modeling_layers import GradientCheckpointingLayer from typing import Optional, Union from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding import torch import torch.nn as nn import torch.nn.functional as F class Qwen2_5_VLVisionBlock(GradientCheckpointingLayer): def __init__(self, config, attn_implementation: str='sdpa') -> None: super().__init__() self.norm1 = Qwen2RMSNorm(config.hidden_size, eps=1e-06) self.norm2 = Qwen2RMSNorm(config.hidden_size, eps=1e-06) self.attn = Qwen2_5_VLVisionAttention(config=config) self.mlp = Qwen2_5_VLMLP(config, bias=True) def forward(self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: Optional[torch.Tensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs) -> torch.Tensor: hidden_states = hidden_states + self.attn(self.norm1(hidden_states), cu_seqlens=cu_seqlens, rotary_pos_emb=rotary_pos_emb, position_embeddings=position_embeddings, **kwargs) hidden_states = hidden_states + self.mlp(self.norm2(hidden_states)) return hidden_states
class Qwen2_5_VLVisionBlock(GradientCheckpointingLayer): def __init__(self, config, attn_implementation: str='sdpa') -> None: pass def forward(self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor, rotary_pos_emb: Optional[torch.Tensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs) -> torch.Tensor: pass
3
0
8
0
8
0
1
0
1
5
1
0
2
4
2
12
18
1
17
7
14
0
11
7
8
1
1
0
2
4,757
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VLVisionConfig
from ...configuration_utils import PretrainedConfig class Qwen2_5_VLVisionConfig(PretrainedConfig): model_type = 'qwen2_5_vl' base_config_key = 'vision_config' def __init__(self, depth=32, hidden_size=3584, hidden_act='silu', intermediate_size=3420, num_heads=16, in_channels=3, patch_size=14, spatial_merge_size=2, temporal_patch_size=2, tokens_per_second=4, window_size=112, out_hidden_size=3584, fullatt_block_indexes=[7, 15, 23, 31], initializer_range=0.02, **kwargs): super().__init__(**kwargs) self.depth = depth self.hidden_size = hidden_size self.hidden_act = hidden_act self.intermediate_size = intermediate_size self.num_heads = num_heads self.in_channels = in_channels self.patch_size = patch_size self.spatial_merge_size = spatial_merge_size self.temporal_patch_size = temporal_patch_size self.tokens_per_second = tokens_per_second self.window_size = window_size self.fullatt_block_indexes = fullatt_block_indexes self.out_hidden_size = out_hidden_size self.initializer_range = initializer_range
class Qwen2_5_VLVisionConfig(PretrainedConfig): def __init__(self, depth=32, hidden_size=3584, hidden_act='silu', intermediate_size=3420, num_heads=16, in_channels=3, patch_size=14, spatial_merge_size=2, temporal_patch_size=2, tokens_per_second=4, window_size=112, out_hidden_size=3584, fullatt_block_indexes=[7, 15, 23, 31], initializer_range=0.02, **kwargs): pass
2
0
32
1
31
0
1
0
1
1
0
0
1
13
1
33
36
2
34
33
16
0
18
17
16
1
2
0
1
4,758
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VisionPatchEmbed
from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding class Qwen2_5_VisionPatchEmbed(PatchEmbed): pass
class Qwen2_5_VisionPatchEmbed(PatchEmbed): pass
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
12
2
0
2
1
1
0
2
1
1
0
2
0
0
4,759
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VisionRotaryEmbedding
from transformers.models.qwen2_vl.modeling_qwen2_vl import PatchEmbed, PatchMerger, Qwen2RMSNorm, Qwen2VLCausalLMOutputWithPast, Qwen2VLForConditionalGeneration, Qwen2VLModel, Qwen2VLModelOutputWithPast, Qwen2VLPreTrainedModel, TransformersKwargs, VisionAttention, VisionRotaryEmbedding class Qwen2_5_VisionRotaryEmbedding(VisionRotaryEmbedding): pass
class Qwen2_5_VisionRotaryEmbedding(VisionRotaryEmbedding): pass
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
12
2
0
2
1
1
0
2
1
1
0
2
0
0
4,760
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py
transformers.models.qwen2_5_vl.modular_qwen2_5_vl.Qwen2_5_VisionTransformerPretrainedModel
import torch.nn as nn import torch.nn.functional as F import torch class Qwen2_5_VisionTransformerPretrainedModel(Qwen2_5_VLPreTrainedModel): config: Qwen2_5_VLVisionConfig _no_split_modules = ['Qwen2_5_VLVisionBlock'] def __init__(self, config, *inputs, **kwargs) -> None: super().__init__(config, *inputs, **kwargs) self.spatial_merge_size = config.spatial_merge_size self.patch_size = config.patch_size self.fullatt_block_indexes = config.fullatt_block_indexes self.window_size = config.window_size self.spatial_merge_unit = self.spatial_merge_size * self.spatial_merge_size self.patch_embed = Qwen2_5_VisionPatchEmbed(patch_size=config.patch_size, temporal_patch_size=config.temporal_patch_size, in_channels=config.in_channels, embed_dim=config.hidden_size) head_dim = config.hidden_size // config.num_heads self.rotary_pos_emb = Qwen2_5_VisionRotaryEmbedding(head_dim // 2) self.blocks = nn.ModuleList([Qwen2_5_VLVisionBlock(config) for _ in range(config.depth)]) self.merger = Qwen2_5_VLPatchMerger(dim=config.out_hidden_size, context_dim=config.hidden_size, spatial_merge_size=config.spatial_merge_size) self.gradient_checkpointing = False def rot_pos_emb(self, grid_thw): pos_ids = [] for t, h, w in grid_thw: hpos_ids = torch.arange(h).unsqueeze(1).expand(-1, w) hpos_ids = hpos_ids.reshape(h // self.spatial_merge_size, self.spatial_merge_size, w // self.spatial_merge_size, self.spatial_merge_size) hpos_ids = hpos_ids.permute(0, 2, 1, 3) hpos_ids = hpos_ids.flatten() wpos_ids = torch.arange(w).unsqueeze(0).expand(h, -1) wpos_ids = wpos_ids.reshape(h // self.spatial_merge_size, self.spatial_merge_size, w // self.spatial_merge_size, self.spatial_merge_size) wpos_ids = wpos_ids.permute(0, 2, 1, 3) wpos_ids = wpos_ids.flatten() pos_ids.append(torch.stack([hpos_ids, wpos_ids], dim=-1).repeat(t, 1)) pos_ids = torch.cat(pos_ids, dim=0) max_grid_size = grid_thw[:, 1:].max() rotary_pos_emb_full = self.rotary_pos_emb(max_grid_size) rotary_pos_emb = rotary_pos_emb_full[pos_ids].flatten(1) return rotary_pos_emb def get_window_index(self, grid_thw): window_index: list = [] cu_window_seqlens: list = [0] window_index_id = 0 vit_merger_window_size = self.window_size // self.spatial_merge_size // self.patch_size for grid_t, grid_h, grid_w in grid_thw: llm_grid_h, llm_grid_w = (grid_h // self.spatial_merge_size, grid_w // self.spatial_merge_size) index = torch.arange(grid_t * llm_grid_h * llm_grid_w).reshape(grid_t, llm_grid_h, llm_grid_w) pad_h = vit_merger_window_size - llm_grid_h % vit_merger_window_size pad_w = vit_merger_window_size - llm_grid_w % vit_merger_window_size num_windows_h = (llm_grid_h + pad_h) // vit_merger_window_size num_windows_w = (llm_grid_w + pad_w) // vit_merger_window_size index_padded = F.pad(index, (0, pad_w, 0, pad_h), 'constant', -100) index_padded = index_padded.reshape(grid_t, num_windows_h, vit_merger_window_size, num_windows_w, vit_merger_window_size) index_padded = index_padded.permute(0, 1, 3, 2, 4).reshape(grid_t, num_windows_h * num_windows_w, vit_merger_window_size, vit_merger_window_size) seqlens = (index_padded != -100).sum([2, 3]).reshape(-1) index_padded = index_padded.reshape(-1) index_new = index_padded[index_padded != -100] window_index.append(index_new + window_index_id) cu_seqlens_tmp = seqlens.cumsum(0) * self.spatial_merge_unit + cu_window_seqlens[-1] cu_window_seqlens.extend(cu_seqlens_tmp.tolist()) window_index_id += (grid_t * llm_grid_h * llm_grid_w).item() window_index = torch.cat(window_index, dim=0) return (window_index, cu_window_seqlens) def forward(self, hidden_states: torch.Tensor, grid_thw: torch.Tensor, **kwargs) -> torch.Tensor: """ Args: hidden_states (`torch.Tensor` of shape `(seq_len, hidden_size)`): The final hidden states of the model. grid_thw (`torch.Tensor` of shape `(num_images_or_videos, 3)`): The temporal, height and width of feature shape of each image in LLM. Returns: `torch.Tensor`: hidden_states. """ hidden_states = self.patch_embed(hidden_states) rotary_pos_emb = self.rot_pos_emb(grid_thw) window_index, cu_window_seqlens = self.get_window_index(grid_thw) cu_window_seqlens = torch.tensor(cu_window_seqlens, device=hidden_states.device, dtype=grid_thw.dtype if torch.jit.is_tracing() else torch.int32) cu_window_seqlens = torch.unique_consecutive(cu_window_seqlens) seq_len, _ = hidden_states.size() hidden_states = hidden_states.reshape(seq_len // self.spatial_merge_unit, self.spatial_merge_unit, -1) hidden_states = hidden_states[window_index, :, :] hidden_states = hidden_states.reshape(seq_len, -1) rotary_pos_emb = rotary_pos_emb.reshape(seq_len // self.spatial_merge_unit, self.spatial_merge_unit, -1) rotary_pos_emb = rotary_pos_emb[window_index, :, :] rotary_pos_emb = rotary_pos_emb.reshape(seq_len, -1) emb = torch.cat((rotary_pos_emb, rotary_pos_emb), dim=-1) position_embeddings = (emb.cos(), emb.sin()) cu_seqlens = torch.repeat_interleave(grid_thw[:, 1] * grid_thw[:, 2], grid_thw[:, 0]).cumsum(dim=0, dtype=grid_thw.dtype if torch.jit.is_tracing() else torch.int32) cu_seqlens = F.pad(cu_seqlens, (1, 0), value=0) for layer_num, blk in enumerate(self.blocks): if layer_num in self.fullatt_block_indexes: cu_seqlens_now = cu_seqlens else: cu_seqlens_now = cu_window_seqlens hidden_states = blk(hidden_states, cu_seqlens=cu_seqlens_now, position_embeddings=position_embeddings, **kwargs) hidden_states = self.merger(hidden_states) reverse_indices = torch.argsort(window_index) hidden_states = hidden_states[reverse_indices, :] return hidden_states
class Qwen2_5_VisionTransformerPretrainedModel(Qwen2_5_VLPreTrainedModel): def __init__(self, config, *inputs, **kwargs) -> None: pass def rot_pos_emb(self, grid_thw): pass def get_window_index(self, grid_thw): pass def forward(self, hidden_states: torch.Tensor, grid_thw: torch.Tensor, **kwargs) -> torch.Tensor: ''' Args: hidden_states (`torch.Tensor` of shape `(seq_len, hidden_size)`): The final hidden states of the model. grid_thw (`torch.Tensor` of shape `(num_images_or_videos, 3)`): The temporal, height and width of feature shape of each image in LLM. Returns: `torch.Tensor`: hidden_states. ''' pass
5
1
39
3
33
3
3
0.1
1
9
4
0
4
10
4
5
162
16
133
48
128
13
83
47
78
6
3
2
11
4,761
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/processing_qwen2_5_vl.py
transformers.models.qwen2_5_vl.processing_qwen2_5_vl.Qwen2_5_VLProcessor
from ...video_utils import VideoInput import numpy as np from typing import Optional, Union from ...tokenization_utils_base import PreTokenizedInput, TextInput from ...feature_extraction_utils import BatchFeature from ...image_utils import ImageInput from ...processing_utils import ImagesKwargs, MultiModalData, ProcessingKwargs, ProcessorMixin, Unpack, VideosKwargs class Qwen2_5_VLProcessor(ProcessorMixin): """ Constructs a Qwen2.5-VL processor which wraps a Qwen2.5-VL image processor and a Qwen2 tokenizer into a single processor. [`Qwen2_5_VLProcessor`] offers all the functionalities of [`Qwen2VLImageProcessor`] and [`Qwen2TokenizerFast`]. See the [`~Qwen2_5_VLProcessor.__call__`] and [`~Qwen2_5_VLProcessor.decode`] for more information. Args: image_processor ([`Qwen2VLImageProcessor`], *optional*): The image processor is a required input. tokenizer ([`Qwen2TokenizerFast`], *optional*): The tokenizer is a required input. video_processor ([`Qwen2_5_VLVideoProcessor`], *optional*): The video processor is a required input. chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string. """ attributes = ['image_processor', 'tokenizer', 'video_processor'] image_processor_class = 'AutoImageProcessor' video_processor_class = 'AutoVideoProcessor' tokenizer_class = ('Qwen2Tokenizer', 'Qwen2TokenizerFast') def __init__(self, image_processor=None, tokenizer=None, video_processor=None, chat_template=None, **kwargs): self.image_token = '<|image_pad|>' if not hasattr(tokenizer, 'image_token') else tokenizer.image_token self.video_token = '<|video_pad|>' if not hasattr(tokenizer, 'video_token') else tokenizer.video_token self.image_token_id = tokenizer.image_token_id if getattr(tokenizer, 'image_token_id', None) else tokenizer.convert_tokens_to_ids(self.image_token) self.video_token_id = tokenizer.video_token_id if getattr(tokenizer, 'video_token_id', None) else tokenizer.convert_tokens_to_ids(self.video_token) super().__init__(image_processor, tokenizer, video_processor, chat_template=chat_template) def __call__(self, images: Optional[ImageInput]=None, text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]=None, videos: Optional[VideoInput]=None, **kwargs: Unpack[Qwen2_5_VLProcessorKwargs]) -> BatchFeature: """ Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text` and `kwargs` arguments to Qwen2TokenizerFast's [`~Qwen2TokenizerFast.__call__`] if `text` is not `None` to encode the text. To prepare the vision inputs, this method forwards the `vision_infos` and `kwargs` arguments to Qwen2VLImageProcessor's [`~Qwen2VLImageProcessor.__call__`] if `vision_infos` is not `None`. Args: images (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `list[PIL.Image.Image]`, `list[np.ndarray]`, `list[torch.Tensor]`): The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch tensor. Both channels-first and channels-last formats are supported. text (`str`, `list[str]`, `list[list[str]]`): The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set `is_split_into_words=True` (to lift the ambiguity with a batch of sequences). videos (`np.ndarray`, `torch.Tensor`, `list[np.ndarray]`, `list[torch.Tensor]`): The image or batch of videos to be prepared. Each video can be a 4D NumPy array or PyTorch tensor, or a nested list of 3D frames. Both channels-first and channels-last formats are supported. return_tensors (`str` or [`~utils.TensorType`], *optional*): If set, will return tensors of a particular framework. Acceptable values are: - `'pt'`: Return PyTorch `torch.Tensor` objects. - `'np'`: Return NumPy `np.ndarray` objects. Returns: [`BatchFeature`]: A [`BatchFeature`] with the following fields: - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`. - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not `None`). - **pixel_values** -- Pixel values to be fed to a model. Returned when `images` is not `None`. - **pixel_values_videos** -- Pixel values of videos to be fed to a model. Returned when `videos` is not `None`. - **image_grid_thw** -- List of image 3D grid in LLM. Returned when `images` is not `None`. - **video_grid_thw** -- List of video 3D grid in LLM. Returned when `videos` is not `None`. - **second_per_grid_ts** -- List of video seconds per time grid. Returned when `videos` is not `None`. """ output_kwargs = self._merge_kwargs(Qwen2_5_VLProcessorKwargs, tokenizer_init_kwargs=self.tokenizer.init_kwargs, **kwargs) image_inputs = videos_inputs = {} if images is not None: image_inputs = self.image_processor(images=images, **output_kwargs['images_kwargs']) image_grid_thw = image_inputs['image_grid_thw'] if videos is not None: fps = output_kwargs['videos_kwargs'].get('fps', 2.0) videos_inputs = self.video_processor(videos=videos, **output_kwargs['videos_kwargs']) video_grid_thw = videos_inputs['video_grid_thw'] if isinstance(fps, (int, float)): second_per_grid_ts = [self.video_processor.temporal_patch_size / fps] * len(video_grid_thw) elif hasattr(fps, '__len__') and len(fps) == len(video_grid_thw): second_per_grid_ts = [self.video_processor.temporal_patch_size / tmp for tmp in fps] else: raise ValueError(f"The length of fps ({(len(fps) if hasattr(fps, '__len__') else fps)}) must be equal to the length of video_grid_thw ({len(video_grid_thw)}) or fps should be a single number.") videos_inputs.update({'second_per_grid_ts': second_per_grid_ts}) if not isinstance(text, list): text = [text] text = text.copy() if images is not None: merge_length = self.image_processor.merge_size ** 2 index = 0 for i in range(len(text)): while self.image_token in text[i]: num_image_tokens = image_grid_thw[index].prod() // merge_length text[i] = text[i].replace(self.image_token, '<|placeholder|>' * num_image_tokens, 1) index += 1 text[i] = text[i].replace('<|placeholder|>', self.image_token) if videos is not None: merge_length = self.video_processor.merge_size ** 2 index = 0 for i in range(len(text)): while self.video_token in text[i]: num_video_tokens = video_grid_thw[index].prod() // merge_length text[i] = text[i].replace(self.video_token, '<|placeholder|>' * num_video_tokens, 1) index += 1 text[i] = text[i].replace('<|placeholder|>', self.video_token) return_tensors = output_kwargs['text_kwargs'].pop('return_tensors', None) return_mm_token_type_ids = output_kwargs['text_kwargs'].pop('return_mm_token_type_ids', None) text_inputs = self.tokenizer(text, **output_kwargs['text_kwargs']) self._check_special_mm_tokens(text, text_inputs, modalities=['image', 'video']) if return_mm_token_type_ids: array_ids = np.array(text_inputs['input_ids']) mm_token_type_ids = np.zeros_like(text_inputs['input_ids']) mm_token_type_ids[array_ids == self.image_token_id] = 1 text_inputs['mm_token_type_ids'] = mm_token_type_ids.tolist() return BatchFeature(data={**text_inputs, **image_inputs, **videos_inputs}, tensor_type=return_tensors) def _get_num_multimodal_tokens(self, image_sizes=None, video_sizes=None, **kwargs): """ Computes the number of placeholder tokens needed for multimodal inputs with the given sizes. Args: image_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (height, width) per each image. video_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (num_frames, height, width) per each video. Returns: `MultiModalData`: A `MultiModalData` object holding number of tokens per each of the provided input modalities, along with other useful data. """ vision_data = {} if image_sizes is not None: images_kwargs = Qwen2_5_VLProcessorKwargs._defaults.get('images_kwargs', {}) images_kwargs.update(kwargs) merge_size = images_kwargs.get('merge_size', None) or self.image_processor.merge_size num_image_patches = [self.image_processor.get_number_of_image_patches(*image_size, images_kwargs) for image_size in image_sizes] num_image_tokens = [num_patches // merge_size ** 2 for num_patches in num_image_patches] vision_data.update({'num_image_tokens': num_image_tokens, 'num_image_patches': num_image_patches}) if video_sizes is not None: videos_kwargs = Qwen2_5_VLProcessorKwargs._defaults.get('videos_kwargs', {}) videos_kwargs.update(kwargs) num_video_patches = [self.video_processor.get_number_of_video_patches(*video_size, videos_kwargs) for video_size in video_sizes] num_video_tokens = [num_patches // merge_size ** 2 for num_patches in num_video_patches] vision_data['num_video_tokens'] = num_video_tokens return MultiModalData(**vision_data) def post_process_image_text_to_text(self, generated_outputs, skip_special_tokens=True, clean_up_tokenization_spaces=False, **kwargs): """ Post-process the output of the model to decode the text. Args: generated_outputs (`torch.Tensor` or `np.ndarray`): The output of the model `generate` function. The output is expected to be a tensor of shape `(batch_size, sequence_length)` or `(sequence_length,)`. skip_special_tokens (`bool`, *optional*, defaults to `True`): Whether or not to remove special tokens in the output. Argument passed to the tokenizer's `batch_decode` method. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not to clean up the tokenization spaces. Argument passed to the tokenizer's `batch_decode` method. **kwargs: Additional arguments to be passed to the tokenizer's `batch_decode method`. Returns: `list[str]`: The decoded text. """ return self.tokenizer.batch_decode(generated_outputs, skip_special_tokens=skip_special_tokens, clean_up_tokenization_spaces=clean_up_tokenization_spaces, **kwargs) @property def model_input_names(self): tokenizer_input_names = self.tokenizer.model_input_names image_processor_input_names = self.image_processor.model_input_names names_from_processor = list(dict.fromkeys(tokenizer_input_names + image_processor_input_names)) return names_from_processor + ['second_per_grid_ts']
class Qwen2_5_VLProcessor(ProcessorMixin): ''' Constructs a Qwen2.5-VL processor which wraps a Qwen2.5-VL image processor and a Qwen2 tokenizer into a single processor. [`Qwen2_5_VLProcessor`] offers all the functionalities of [`Qwen2VLImageProcessor`] and [`Qwen2TokenizerFast`]. See the [`~Qwen2_5_VLProcessor.__call__`] and [`~Qwen2_5_VLProcessor.decode`] for more information. Args: image_processor ([`Qwen2VLImageProcessor`], *optional*): The image processor is a required input. tokenizer ([`Qwen2TokenizerFast`], *optional*): The tokenizer is a required input. video_processor ([`Qwen2_5_VLVideoProcessor`], *optional*): The video processor is a required input. chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string. ''' def __init__(self, image_processor=None, tokenizer=None, video_processor=None, chat_template=None, **kwargs): pass def __call__(self, images: Optional[ImageInput]=None, text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]=None, videos: Optional[VideoInput]=None, **kwargs: Unpack[Qwen2_5_VLProcessorKwargs]) -> BatchFeature: ''' Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text` and `kwargs` arguments to Qwen2TokenizerFast's [`~Qwen2TokenizerFast.__call__`] if `text` is not `None` to encode the text. To prepare the vision inputs, this method forwards the `vision_infos` and `kwargs` arguments to Qwen2VLImageProcessor's [`~Qwen2VLImageProcessor.__call__`] if `vision_infos` is not `None`. Args: images (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `list[PIL.Image.Image]`, `list[np.ndarray]`, `list[torch.Tensor]`): The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch tensor. Both channels-first and channels-last formats are supported. text (`str`, `list[str]`, `list[list[str]]`): The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set `is_split_into_words=True` (to lift the ambiguity with a batch of sequences). videos (`np.ndarray`, `torch.Tensor`, `list[np.ndarray]`, `list[torch.Tensor]`): The image or batch of videos to be prepared. Each video can be a 4D NumPy array or PyTorch tensor, or a nested list of 3D frames. Both channels-first and channels-last formats are supported. return_tensors (`str` or [`~utils.TensorType`], *optional*): If set, will return tensors of a particular framework. Acceptable values are: - `'pt'`: Return PyTorch `torch.Tensor` objects. - `'np'`: Return NumPy `np.ndarray` objects. Returns: [`BatchFeature`]: A [`BatchFeature`] with the following fields: - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`. - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not `None`). - **pixel_values** -- Pixel values to be fed to a model. Returned when `images` is not `None`. - **pixel_values_videos** -- Pixel values of videos to be fed to a model. Returned when `videos` is not `None`. - **image_grid_thw** -- List of image 3D grid in LLM. Returned when `images` is not `None`. - **video_grid_thw** -- List of video 3D grid in LLM. Returned when `videos` is not `None`. - **second_per_grid_ts** -- List of video seconds per time grid. Returned when `videos` is not `None`. ''' pass def _get_num_multimodal_tokens(self, image_sizes=None, video_sizes=None, **kwargs): ''' Computes the number of placeholder tokens needed for multimodal inputs with the given sizes. Args: image_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (height, width) per each image. video_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (num_frames, height, width) per each video. Returns: `MultiModalData`: A `MultiModalData` object holding number of tokens per each of the provided input modalities, along with other useful data. ''' pass def post_process_image_text_to_text(self, generated_outputs, skip_special_tokens=True, clean_up_tokenization_spaces=False, **kwargs): ''' Post-process the output of the model to decode the text. Args: generated_outputs (`torch.Tensor` or `np.ndarray`): The output of the model `generate` function. The output is expected to be a tensor of shape `(batch_size, sequence_length)` or `(sequence_length,)`. skip_special_tokens (`bool`, *optional*, defaults to `True`): Whether or not to remove special tokens in the output. Argument passed to the tokenizer's `batch_decode` method. clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`): Whether or not to clean up the tokenization spaces. Argument passed to the tokenizer's `batch_decode` method. **kwargs: Additional arguments to be passed to the tokenizer's `batch_decode method`. Returns: `list[str]`: The decoded text. ''' pass @property def model_input_names(self): pass
7
4
24
2
13
9
3
0.75
1
9
2
0
6
2
6
23
168
21
84
33
70
63
57
26
50
13
2
3
20
4,762
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/processing_qwen2_5_vl.py
transformers.models.qwen2_5_vl.processing_qwen2_5_vl.Qwen2_5_VLProcessorKwargs
from ...processing_utils import ImagesKwargs, MultiModalData, ProcessingKwargs, ProcessorMixin, Unpack, VideosKwargs class Qwen2_5_VLProcessorKwargs(ProcessingKwargs, total=False): images_kwargs: Qwen2_5_VLImagesKwargs videos_kwargs: Qwen2_5_VLVideosProcessorKwargs _defaults = {'text_kwargs': {'padding': False, 'return_mm_token_type_ids': False}}
class Qwen2_5_VLProcessorKwargs(ProcessingKwargs, total=False): pass
1
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
8
0
8
2
7
0
3
2
2
0
3
0
0
4,763
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_5_vl/processing_qwen2_5_vl.py
transformers.models.qwen2_5_vl.processing_qwen2_5_vl.Qwen2_5_VLVideosProcessorKwargs
from typing import Optional, Union from ...processing_utils import ImagesKwargs, MultiModalData, ProcessingKwargs, ProcessorMixin, Unpack, VideosKwargs class Qwen2_5_VLVideosProcessorKwargs(VideosKwargs, total=False): fps: Union[list[float], float]
class Qwen2_5_VLVideosProcessorKwargs(VideosKwargs, total=False): pass
1
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
2
0
2
1
1
0
2
1
1
0
2
0
0
4,764
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/configuration_qwen2_audio.py
transformers.models.qwen2_audio.configuration_qwen2_audio.Qwen2AudioConfig
from ..auto import CONFIG_MAPPING, AutoConfig from ...configuration_utils import PretrainedConfig class Qwen2AudioConfig(PretrainedConfig): """ This is the configuration class to store the configuration of a [`Qwen2AudioForConditionalGeneration`]. It is used to instantiate an Qwen2-Audio 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 Qwen2-Audio. e.g. [Qwen/Qwen2-Audio-7B](https://huggingface.co/Qwen/Qwen2-Audio-7B) Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: audio_config (`Union[AutoConfig, dict]`, *optional*, defaults to `CLIPVisionConfig`): The config object or dictionary of the audio backbone. text_config (`Union[AutoConfig, dict]`, *optional*, defaults to `LlamaConfig`): The config object or dictionary of the text backbone. audio_token_index (`int`, *optional*, defaults to 151646): The image token index to encode the image prompt. Example: ```python >>> from transformers import Qwen2AudioForConditionalGeneration, Qwen2AudioConfig, Qwen2AudioEncoderConfig, Qwen2Config >>> # Initializing a Qwen2AudioEncoder config >>> audio_config = Qwen2AudioEncoderConfig() >>> # Initializing a Qwen2 config >>> text_config = Qwen2Config() >>> # Initializing a Qwen2Audio configuration >>> configuration = Qwen2AudioConfig(audio_config, text_config) >>> # Initializing a model from the qwen2-audio style configuration >>> model = Qwen2AudioForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = 'qwen2_audio' attribute_map = {'audio_token_id': 'audio_token_index'} sub_configs = {'text_config': AutoConfig, 'audio_config': AutoConfig} def __init__(self, audio_config=None, text_config=None, audio_token_index=151646, **kwargs): self.audio_token_index = audio_token_index if isinstance(audio_config, dict): audio_config['model_type'] = audio_config.get('model_type', 'qwen2_audio_encoder') audio_config = CONFIG_MAPPING[audio_config['model_type']](**audio_config) elif audio_config is None: audio_config = CONFIG_MAPPING['qwen2_audio_encoder'](d_model=1280, encoder_attention_heads=20, encoder_ffn_dim=5120, encoder_layerdrop=0.0, encoder_layers=32, num_mel_bins=128, max_source_positions=1500, scale_embedding=False, activation_function='gelu') self.audio_config = audio_config if isinstance(text_config, dict): text_config['model_type'] = text_config.get('model_type', 'qwen2') text_config = CONFIG_MAPPING[text_config['model_type']](**text_config) elif text_config is None: text_config = CONFIG_MAPPING['qwen2']() self.text_config = text_config super().__init__(**kwargs)
class Qwen2AudioConfig(PretrainedConfig): ''' This is the configuration class to store the configuration of a [`Qwen2AudioForConditionalGeneration`]. It is used to instantiate an Qwen2-Audio 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 Qwen2-Audio. e.g. [Qwen/Qwen2-Audio-7B](https://huggingface.co/Qwen/Qwen2-Audio-7B) Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: audio_config (`Union[AutoConfig, dict]`, *optional*, defaults to `CLIPVisionConfig`): The config object or dictionary of the audio backbone. text_config (`Union[AutoConfig, dict]`, *optional*, defaults to `LlamaConfig`): The config object or dictionary of the text backbone. audio_token_index (`int`, *optional*, defaults to 151646): The image token index to encode the image prompt. Example: ```python >>> from transformers import Qwen2AudioForConditionalGeneration, Qwen2AudioConfig, Qwen2AudioEncoderConfig, Qwen2Config >>> # Initializing a Qwen2AudioEncoder config >>> audio_config = Qwen2AudioEncoderConfig() >>> # Initializing a Qwen2 config >>> text_config = Qwen2Config() >>> # Initializing a Qwen2Audio configuration >>> configuration = Qwen2AudioConfig(audio_config, text_config) >>> # Initializing a model from the qwen2-audio style configuration >>> model = Qwen2AudioForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```''' def __init__(self, audio_config=None, text_config=None, audio_token_index=151646, **kwargs): pass
2
1
38
5
33
0
7
0.78
1
2
0
0
1
3
1
1
81
17
36
13
28
28
16
7
14
7
1
1
7
4,765
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/configuration_qwen2_audio.py
transformers.models.qwen2_audio.configuration_qwen2_audio.Qwen2AudioEncoderConfig
from ...configuration_utils import PretrainedConfig class Qwen2AudioEncoderConfig(PretrainedConfig): """ This is the configuration class to store the configuration of a [`Qwen2AudioEncoder`]. It is used to instantiate a Qwen2-Audio audio encoder according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the audio encoder of the Qwen2-Audio architecture. e.g. [Qwen/Qwen2-Audio-7B](https://huggingface.co/Qwen/Qwen2-Audio-7B) Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: num_mel_bins (`int`, *optional*, defaults to 128): Number of mel features used per input features. Should correspond to the value used in the `Qwen2AudioProcessor` class. encoder_layers (`int`, *optional*, defaults to 32): Number of encoder layers. encoder_attention_heads (`int`, *optional*, defaults to 20): Number of attention heads for each attention layer in the Transformer encoder. encoder_ffn_dim (`int`, *optional*, defaults to 5120): Dimensionality of the "intermediate" (often named feed-forward) layer in encoder. encoder_layerdrop (`float`, *optional*, defaults to 0.0): The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://huggingface.co/papers/1909.11556) for more details. d_model (`int`, *optional*, defaults to 1280): Dimensionality of the layers. dropout (`float`, *optional*, defaults to 0.0): The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. activation_function (`str`, *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. activation_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for activations inside the fully connected layer. scale_embedding (`bool`, *optional*, defaults to `False`): Scale embeddings by diving by sqrt(d_model). initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. max_source_positions (`int`, *optional*, defaults to 1500): The maximum sequence length of log-mel filter-bank features that this model might ever be used with. Example: ```python >>> from transformers import Qwen2AudioEncoderConfig, Qwen2AudioEncoder >>> # Initializing a Qwen2AudioEncoderConfig >>> configuration = Qwen2AudioEncoderConfig() >>> # Initializing a Qwen2AudioEncoder (with random weights) >>> model = Qwen2AudioEncoder(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = 'qwen2_audio_encoder' def __init__(self, num_mel_bins=128, encoder_layers=32, encoder_attention_heads=20, encoder_ffn_dim=5120, encoder_layerdrop=0.0, d_model=1280, dropout=0.0, attention_dropout=0.0, activation_function='gelu', activation_dropout=0.0, scale_embedding=False, initializer_range=0.02, max_source_positions=1500, **kwargs): super().__init__(**kwargs) self.num_mel_bins = num_mel_bins self.d_model = d_model self.encoder_layers = encoder_layers self.encoder_attention_heads = encoder_attention_heads self.encoder_ffn_dim = encoder_ffn_dim self.dropout = dropout self.attention_dropout = attention_dropout self.activation_function = activation_function self.activation_dropout = activation_dropout self.encoder_layerdrop = encoder_layerdrop self.num_hidden_layers = encoder_layers self.initializer_range = initializer_range self.scale_embedding = scale_embedding self.max_source_positions = max_source_positions
class Qwen2AudioEncoderConfig(PretrainedConfig): ''' This is the configuration class to store the configuration of a [`Qwen2AudioEncoder`]. It is used to instantiate a Qwen2-Audio audio encoder according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the audio encoder of the Qwen2-Audio architecture. e.g. [Qwen/Qwen2-Audio-7B](https://huggingface.co/Qwen/Qwen2-Audio-7B) Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: num_mel_bins (`int`, *optional*, defaults to 128): Number of mel features used per input features. Should correspond to the value used in the `Qwen2AudioProcessor` class. encoder_layers (`int`, *optional*, defaults to 32): Number of encoder layers. encoder_attention_heads (`int`, *optional*, defaults to 20): Number of attention heads for each attention layer in the Transformer encoder. encoder_ffn_dim (`int`, *optional*, defaults to 5120): Dimensionality of the "intermediate" (often named feed-forward) layer in encoder. encoder_layerdrop (`float`, *optional*, defaults to 0.0): The LayerDrop probability for the encoder. See the [LayerDrop paper](see https://huggingface.co/papers/1909.11556) for more details. d_model (`int`, *optional*, defaults to 1280): Dimensionality of the layers. dropout (`float`, *optional*, defaults to 0.0): The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. activation_function (`str`, *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. activation_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for activations inside the fully connected layer. scale_embedding (`bool`, *optional*, defaults to `False`): Scale embeddings by diving by sqrt(d_model). initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. max_source_positions (`int`, *optional*, defaults to 1500): The maximum sequence length of log-mel filter-bank features that this model might ever be used with. Example: ```python >>> from transformers import Qwen2AudioEncoderConfig, Qwen2AudioEncoder >>> # Initializing a Qwen2AudioEncoderConfig >>> configuration = Qwen2AudioEncoderConfig() >>> # Initializing a Qwen2AudioEncoder (with random weights) >>> model = Qwen2AudioEncoder(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```''' def __init__(self, num_mel_bins=128, encoder_layers=32, encoder_attention_heads=20, encoder_ffn_dim=5120, encoder_layerdrop=0.0, d_model=1280, dropout=0.0, attention_dropout=0.0, activation_function='gelu', activation_dropout=0.0, scale_embedding=False, initializer_range=0.02, max_source_positions=1500, **kwargs): pass
2
1
33
1
32
1
1
1.44
1
1
0
0
1
14
1
1
93
11
34
33
16
49
18
17
16
1
1
0
1
4,766
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/modeling_qwen2_audio.py
transformers.models.qwen2_audio.modeling_qwen2_audio.Qwen2AudioAttention
from typing import Callable, Optional, Union import torch from .configuration_qwen2_audio import Qwen2AudioConfig, Qwen2AudioEncoderConfig from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel from torch import nn class Qwen2AudioAttention(nn.Module): """Multi-headed attention from 'Attention Is All You Need' paper""" def __init__(self, embed_dim: int, num_heads: int, dropout: float=0.0, is_decoder: bool=False, bias: bool=True, is_causal: bool=False, layer_idx: Optional[int]=None, config: Optional[Qwen2AudioConfig]=None): super().__init__() self.embed_dim = embed_dim self.num_heads = num_heads self.dropout = dropout self.head_dim = embed_dim // num_heads self.config = config if self.head_dim * num_heads != self.embed_dim: raise ValueError(f'embed_dim must be divisible by num_heads (got `embed_dim`: {self.embed_dim} and `num_heads`: {num_heads}).') self.scaling = self.head_dim ** (-0.5) self.is_decoder = is_decoder self.is_causal = is_causal if layer_idx is None and is_decoder: logger.warning_once(f'Instantiating a decoder {self.__class__.__name__} without passing `layer_idx` is not recommended and will to errors during the forward call, if caching is used. Please make sure to provide a `layer_idx` when creating this class.') self.layer_idx = layer_idx self.k_proj = nn.Linear(embed_dim, embed_dim, bias=False) self.v_proj = nn.Linear(embed_dim, embed_dim, bias=bias) self.q_proj = nn.Linear(embed_dim, embed_dim, bias=bias) self.out_proj = nn.Linear(embed_dim, embed_dim, bias=bias) def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int): return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous() def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, layer_head_mask: Optional[torch.Tensor]=None, output_attentions: bool=False, **kwargs) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: """Input shape: Batch x Time x Channel""" bsz, tgt_len, _ = hidden_states.size() query_states = self._shape(self.q_proj(hidden_states) * self.scaling, tgt_len, bsz) key_states = self._shape(self.k_proj(hidden_states), -1, bsz) value_states = self._shape(self.v_proj(hidden_states), -1, bsz) attention_interface: Callable = eager_attention_forward if self.config._attn_implementation != 'eager': attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation] attn_output, attn_weights = attention_interface(self, query_states, key_states, value_states, attention_mask, dropout=0.0 if not self.training else self.dropout, scaling=1.0, output_attentions=output_attentions, head_mask=layer_head_mask, **kwargs) attn_output = attn_output.reshape(bsz, tgt_len, -1).contiguous() attn_output = self.out_proj(attn_output) return (attn_output, attn_weights)
class Qwen2AudioAttention(nn.Module): '''Multi-headed attention from 'Attention Is All You Need' paper''' def __init__(self, embed_dim: int, num_heads: int, dropout: float=0.0, is_decoder: bool=False, bias: bool=True, is_causal: bool=False, layer_idx: Optional[int]=None, config: Optional[Qwen2AudioConfig]=None): pass def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int): pass def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, layer_head_mask: Optional[torch.Tensor]=None, output_attentions: bool=False, **kwargs) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: '''Input shape: Batch x Time x Channel''' pass
4
2
40
5
31
4
5
0.14
1
8
2
2
3
13
3
13
125
19
94
47
71
13
58
28
54
11
1
2
15
4,767
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/modeling_qwen2_audio.py
transformers.models.qwen2_audio.modeling_qwen2_audio.Qwen2AudioCausalLMOutputWithPast
from ...modeling_outputs import BaseModelOutput, ModelOutput from typing import Callable, Optional, Union from ...cache_utils import Cache from dataclasses import dataclass from ...utils import auto_docstring, logging import torch @dataclass @auto_docstring(custom_intro='\n Base class for Qwen2Audio causal language model (or autoregressive) outputs.\n ') class Qwen2AudioCausalLMOutputWithPast(ModelOutput): """ loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided): Language modeling loss (for next-token prediction). logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`): Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). past_key_values (`Cache`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`): Pre-computed hidden-states that can be used to speed up auto-regressive (sequential) decoding. There are two sets of pre-computed hidden-states: key and values states in the self-attention blocks. The `past_key_values` are returned when `use_cache=True` is passed or when `config.use_cache=True`. It is a [`~cache_utils.Cache`] instance. If `past_key_values` are used, the user can optionally input only the last `input_ids` (those that don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `input_ids` of shape `(batch_size, sequence_length)`. attention_mask (`torch.FloatTensor`, *optional*): Attentions mask, used to update attention mask and position_ids. """ loss: Optional[torch.FloatTensor] = None logits: Optional[torch.FloatTensor] = None past_key_values: Optional[Cache] = None hidden_states: Optional[tuple[torch.FloatTensor]] = None attentions: Optional[tuple[torch.FloatTensor]] = None attention_mask: Optional[torch.FloatTensor] = None
@dataclass @auto_docstring(custom_intro='\n Base class for Qwen2Audio causal language model (or autoregressive) outputs.\n ') class Qwen2AudioCausalLMOutputWithPast(ModelOutput): ''' loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided): Language modeling loss (for next-token prediction). logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`): Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). past_key_values (`Cache`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`): Pre-computed hidden-states that can be used to speed up auto-regressive (sequential) decoding. There are two sets of pre-computed hidden-states: key and values states in the self-attention blocks. The `past_key_values` are returned when `use_cache=True` is passed or when `config.use_cache=True`. It is a [`~cache_utils.Cache`] instance. If `past_key_values` are used, the user can optionally input only the last `input_ids` (those that don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `input_ids` of shape `(batch_size, sequence_length)`. attention_mask (`torch.FloatTensor`, *optional*): Attentions mask, used to update attention mask and position_ids. ''' pass
3
1
0
0
0
0
0
3.43
1
0
0
0
0
0
0
0
36
5
7
7
6
24
7
7
6
0
1
0
0
4,768
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/modeling_qwen2_audio.py
transformers.models.qwen2_audio.modeling_qwen2_audio.Qwen2AudioEncoder
from ...modeling_outputs import BaseModelOutput, ModelOutput import torch import math from .configuration_qwen2_audio import Qwen2AudioConfig, Qwen2AudioEncoderConfig from torch import nn from ...utils import auto_docstring, logging @auto_docstring(custom_intro='\n The audio model from Qwen2Audio without any head or projection on top.\n ') class Qwen2AudioEncoder(Qwen2AudioPreTrainedModel): """ Transformer encoder consisting of *config.encoder_layers* self attention layers. Each layer is a [`Qwen2AudioEncoderLayer`]. Args: config: Qwen2AudioEncoderConfig """ config: Qwen2AudioEncoderConfig main_input_name = 'input_features' _no_split_modules = ['Qwen2AudioEncoderLayer'] def __init__(self, config: Qwen2AudioEncoderConfig): super().__init__(config) self.dropout = config.dropout self.layerdrop = config.encoder_layerdrop embed_dim = config.d_model self.num_mel_bins = config.num_mel_bins self.padding_idx = config.pad_token_id self.max_source_positions = config.max_source_positions self.embed_scale = math.sqrt(embed_dim) if config.scale_embedding else 1.0 self.conv1 = nn.Conv1d(self.num_mel_bins, embed_dim, kernel_size=3, padding=1) self.conv2 = nn.Conv1d(embed_dim, embed_dim, kernel_size=3, stride=2, padding=1) self.embed_positions = nn.Embedding(self.max_source_positions, embed_dim) self.embed_positions.requires_grad_(False) self.layers = nn.ModuleList([Qwen2AudioEncoderLayer(config) for _ in range(config.encoder_layers)]) self.layer_norm = nn.LayerNorm(config.d_model) self.avg_pooler = nn.AvgPool1d(2, stride=2) self.gradient_checkpointing = False self.post_init() def _freeze_parameters(self): for param in self.parameters(): param.requires_grad = False self._requires_grad = False def get_input_embeddings(self) -> nn.Module: return self.conv1 def set_input_embeddings(self, value: nn.Module): self.conv1 = value def forward(self, input_features, attention_mask=None, head_mask=None, output_attentions=None, output_hidden_states=None, return_dict=None): """ Args: attention_mask (`torch.Tensor`)`, *optional*): Qwen2Audio does not support masking of the `input_features`, this argument is preserved for compatibility, but it is not used. By default the silence in the input log mel spectrogram are ignored. head_mask (`torch.Tensor` of shape `(encoder_layers, encoder_attention_heads)`, *optional*): Mask to nullify selected heads of the 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. return_dict (`bool`, *optional*): Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple. """ expected_seq_length = self.config.max_source_positions * self.conv1.stride[0] * self.conv2.stride[0] if input_features.shape[-1] != expected_seq_length: raise ValueError(f'Qwen2Audio expects the mel input features to be of length {expected_seq_length}, but found {input_features.shape[-1]}. Make sure to pad the input mel features to {expected_seq_length}.') 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 input_features = input_features.to(dtype=self.conv1.weight.dtype, device=self.conv1.weight.device) inputs_embeds = nn.functional.gelu(self.conv1(input_features)) inputs_embeds = nn.functional.gelu(self.conv2(inputs_embeds)) inputs_embeds = inputs_embeds.permute(0, 2, 1) embed_pos = self.embed_positions.weight hidden_states = inputs_embeds + embed_pos hidden_states = nn.functional.dropout(hidden_states, p=self.dropout, training=self.training) encoder_states = () if output_hidden_states else None all_attentions = () if output_attentions else None if head_mask is not None: assert head_mask.size()[0] == len(self.layers), f'The head_mask should be specified for {len(self.layers)} layers, but it is for {head_mask.size()[0]}.' for idx, encoder_layer in enumerate(self.layers): if output_hidden_states: encoder_states = encoder_states + (hidden_states,) to_drop = False if self.training: dropout_probability = torch.rand([]) if dropout_probability < self.layerdrop: to_drop = True if to_drop: layer_outputs = (None, None) else: layer_outputs = encoder_layer(hidden_states, attention_mask, layer_head_mask=head_mask[idx] if head_mask is not None else None, output_attentions=output_attentions) hidden_states = layer_outputs[0] if output_attentions: all_attentions = all_attentions + (layer_outputs[1],) hidden_states = hidden_states.permute(0, 2, 1) hidden_states = self.avg_pooler(hidden_states) hidden_states = hidden_states.permute(0, 2, 1) hidden_states = self.layer_norm(hidden_states) if output_hidden_states: encoder_states = encoder_states + (hidden_states,) if not return_dict: return tuple((v for v in [hidden_states, encoder_states, all_attentions] if v is not None)) return BaseModelOutput(last_hidden_state=hidden_states, hidden_states=encoder_states, attentions=all_attentions) def _get_feat_extract_output_lengths(self, input_lengths: torch.LongTensor): """ Computes the output length of the convolutional layers and the output length of the audio encoder """ input_lengths = (input_lengths - 1) // 2 + 1 output_lengths = (input_lengths - 2) // 2 + 1 return (input_lengths, output_lengths)
@auto_docstring(custom_intro='\n The audio model from Qwen2Audio without any head or projection on top.\n ') class Qwen2AudioEncoder(Qwen2AudioPreTrainedModel): ''' Transformer encoder consisting of *config.encoder_layers* self attention layers. Each layer is a [`Qwen2AudioEncoderLayer`]. Args: config: Qwen2AudioEncoderConfig ''' def __init__(self, config: Qwen2AudioEncoderConfig): pass def _freeze_parameters(self): pass def get_input_embeddings(self) -> nn.Module: pass def set_input_embeddings(self, value: nn.Module): pass def forward(self, input_features, attention_mask=None, head_mask=None, output_attentions=None, output_hidden_states=None, return_dict=None): ''' Args: attention_mask (`torch.Tensor`)`, *optional*): Qwen2Audio does not support masking of the `input_features`, this argument is preserved for compatibility, but it is not used. By default the silence in the input log mel spectrogram are ignored. head_mask (`torch.Tensor` of shape `(encoder_layers, encoder_attention_heads)`, *optional*): Mask to nullify selected heads of the 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. return_dict (`bool`, *optional*): Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple. ''' pass def _get_feat_extract_output_lengths(self, input_lengths: torch.LongTensor): ''' Computes the output length of the convolutional layers and the output length of the audio encoder ''' pass
8
3
26
4
17
6
4
0.41
1
8
3
0
6
14
6
7
177
29
106
45
91
43
77
37
70
19
2
3
26
4,769
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/modeling_qwen2_audio.py
transformers.models.qwen2_audio.modeling_qwen2_audio.Qwen2AudioEncoderLayer
from torch import nn from ...modeling_layers import GradientCheckpointingLayer from .configuration_qwen2_audio import Qwen2AudioConfig, Qwen2AudioEncoderConfig from ...activations import ACT2FN import torch class Qwen2AudioEncoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2AudioConfig): super().__init__() self.embed_dim = config.d_model self.self_attn = Qwen2AudioAttention(embed_dim=self.embed_dim, num_heads=config.encoder_attention_heads, dropout=config.attention_dropout, config=config) self.self_attn_layer_norm = nn.LayerNorm(self.embed_dim) self.dropout = config.dropout self.activation_fn = ACT2FN[config.activation_function] self.activation_dropout = config.activation_dropout self.fc1 = nn.Linear(self.embed_dim, config.encoder_ffn_dim) self.fc2 = nn.Linear(config.encoder_ffn_dim, self.embed_dim) self.final_layer_norm = nn.LayerNorm(self.embed_dim) def forward(self, hidden_states: torch.Tensor, attention_mask: torch.Tensor, layer_head_mask: torch.Tensor, output_attentions: bool=False) -> torch.Tensor: """ Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`): attention mask of size `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values. layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size `(encoder_attention_heads,)`. output_attentions (`bool`, *optional*): Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. """ residual = hidden_states hidden_states = self.self_attn_layer_norm(hidden_states) hidden_states, attn_weights = self.self_attn(hidden_states=hidden_states, attention_mask=attention_mask, layer_head_mask=layer_head_mask, output_attentions=output_attentions) hidden_states = nn.functional.dropout(hidden_states, p=self.dropout, training=self.training) hidden_states = residual + hidden_states residual = hidden_states hidden_states = self.final_layer_norm(hidden_states) hidden_states = self.activation_fn(self.fc1(hidden_states)) hidden_states = nn.functional.dropout(hidden_states, p=self.activation_dropout, training=self.training) hidden_states = self.fc2(hidden_states) hidden_states = nn.functional.dropout(hidden_states, p=self.dropout, training=self.training) hidden_states = residual + hidden_states if hidden_states.dtype == torch.float16: clamp_value = torch.finfo(hidden_states.dtype).max - 1000 hidden_states = torch.clamp(hidden_states, min=-clamp_value, max=clamp_value) return (hidden_states, attn_weights)
class Qwen2AudioEncoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2AudioConfig): pass def forward(self, hidden_states: torch.Tensor, attention_mask: torch.Tensor, layer_head_mask: torch.Tensor, output_attentions: bool=False) -> torch.Tensor: ''' Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`): attention mask of size `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values. layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size `(encoder_attention_heads,)`. output_attentions (`bool`, *optional*): Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. ''' pass
3
1
33
3
25
6
2
0.22
1
4
1
0
2
9
2
12
68
7
50
22
41
11
32
16
29
3
1
1
4
4,770
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/modeling_qwen2_audio.py
transformers.models.qwen2_audio.modeling_qwen2_audio.Qwen2AudioForConditionalGeneration
from ...cache_utils import Cache from torch import nn from .configuration_qwen2_audio import Qwen2AudioConfig, Qwen2AudioEncoderConfig from ...generation import GenerationMixin import torch from ...utils import auto_docstring, logging from ..auto import AutoModel, AutoModelForCausalLM from typing import Callable, Optional, Union @auto_docstring(custom_intro='\n The QWEN2AUDIO model which consists of a audio backbone and a language model.\n ') class Qwen2AudioForConditionalGeneration(Qwen2AudioPreTrainedModel, GenerationMixin): def __init__(self, config: Qwen2AudioConfig): super().__init__(config) self.audio_tower = AutoModel.from_config(config.audio_config) self.multi_modal_projector = Qwen2AudioMultiModalProjector(config) self.vocab_size = config.text_config.vocab_size self.language_model = AutoModelForCausalLM.from_config(config.text_config) if self.language_model._tied_weights_keys is not None: self._tied_weights_keys = [f'language_model.{k}' for k in self.language_model._tied_weights_keys] self.pad_token_id = self.config.pad_token_id if self.config.pad_token_id is not None else -1 self._padding_side = 'left' self.post_init() @property def padding_side(self): return self._padding_side @padding_side.setter def padding_side(self, padding_side: str): if padding_side not in ['left', 'right']: raise ValueError(f'{padding_side} is not `left` or `right`.') self._padding_side = padding_side def get_input_embeddings(self): return self.language_model.get_input_embeddings() def set_input_embeddings(self, value): self.language_model.set_input_embeddings(value) def get_output_embeddings(self): return self.language_model.get_output_embeddings() def set_output_embeddings(self, new_embeddings): self.language_model.set_output_embeddings(new_embeddings) def set_decoder(self, decoder): self.language_model.set_decoder(decoder) def get_decoder(self): return self.language_model.get_decoder() def _merge_input_ids_with_audio_features(self, audio_features, num_audio_tokens, inputs_embeds, input_ids, attention_mask, labels): """ Merge input_ids with with audio features into final embeddings Args: audio_features (`torch.Tensor` of shape `(num_audios, max_audio_tokens, embed_dim)`): All audio vectors of all audios in the batch num_audio_tokens (`torch.LongTensor` of shape `(num_audios)`): The length of audio embeddings of each audio as stacked in `audio_features` inputs_embeds (`torch.Tensor` of shape `(batch_size, sequence_length, embed_dim)`): Token embeddings before merging with audio embeddings input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Input_ids of tokens, possibly filled with audio token attention_mask (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Mask to avoid performing attention on padding token indices. labels (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) labels need to be recalculated to support training (if provided) Returns: final_embedding, final_attention_mask, final_labels, position_ids, final_input_ids Explanation: each audio has variable length embeddings, with length specified by num_audio_tokens audio_features is concatenation of all audio embed vectors task: fill each <|AUDIO|> with the correct number of audio embeddings Example: X (5 tokens), Y (3 tokens), Z (8 tokens) X, Y are in the same sequence (in-context learning) if right padding input_ids: [ a b c d e f X g h i j k Y l m o p q r Z s t u v _ _ _ _ _ _ ] input_ids should be: [ a b c d e f X X X X X g h i j k Y Y Y l m o p q r Z Z Z Z Z Z Z Z s t u v _ _ _ _ _ ] labels should be: [ a b c d e f _ _ _ _ _ g h i j k _ _ _ l m o p q r _ _ _ _ _ _ _ _ s t u v _ _ _ _ _ ] elif left padding input_ids: [ a b c d e f X g h i j k Y l m _ _ _ _ _ _ o p q r Z s t u v ] input_ids should be: [ a b c d e f X X X X X g h i j k Y Y Y l m _ _ _ _ _ o p q r Z Z Z Z Z Z Z Z s t u v ] labels should be: [ a b c d e f _ _ _ _ _ g h i j k _ _ _ l m _ _ _ _ _ o p q r _ _ _ _ _ _ _ _ s t u v ] Edge cases: * If tokens are same but audio token sizes are different, then cannot infer left or right padding ```python url1 = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/glass-breaking-151256.mp3" audio1, _ = librosa.load(BytesIO(urlopen(url1).read()), sr=processor.feature_extractor.sampling_rate) url2 = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/f2641_0_throatclearing.wav" audio2, _ = librosa.load(BytesIO(urlopen(url2).read()), sr=processor.feature_extractor.sampling_rate) prompts = [ "[INST] <|AUDIO|> What is that in this audio? [/INST]", "[INST] <|AUDIO|> What is that in this audio? [/INST]", ] inputs = processor(text=prompts, audios=[audio1, audio2], return_tensors='pt', padding=True).to("cuda") audio1 has 101 tokens, while audio2 has 72 tokens ``` input_ids: [ a b c d X g h i j Y k l m n ] where X is 3 tokens while Y is 5, this mean after merge if left-padding (batched generation) input_ids should be: [ _ _ a b c d X X X g h i j Y Y Y Y Y k l m n ] elif (right padding) (training) input_ids should be: [ a b c d X X X g h _ _ i j Y Y Y Y Y k l m n ] """ num_audios, max_audio_tokens, embed_dim = audio_features.shape audio_features_mask = torch.arange(max_audio_tokens).expand(num_audios, max_audio_tokens).to(num_audio_tokens.device) < num_audio_tokens.unsqueeze(1) masked_audio_features = audio_features[audio_features_mask].view(-1, embed_dim) batch_size, sequence_length = input_ids.shape _left_padding = torch.any(attention_mask[:, 0] == 0) _right_padding = torch.any(attention_mask[:, -1] == 0) left_padding = True if batch_size > 1: if _left_padding and (not _right_padding): left_padding = True elif not _left_padding and _right_padding: left_padding = False elif not _left_padding and (not _right_padding): left_padding = self.padding_side == 'left' else: raise ValueError(f'both side of attention_mask has zero, invalid. {attention_mask}') special_audio_token_mask = input_ids == self.config.audio_token_id num_special_audio_tokens = torch.sum(special_audio_token_mask, dim=-1) target_device = inputs_embeds.device attention_mask = attention_mask.to(target_device) input_ids = input_ids.to(target_device) num_audio_tokens = num_audio_tokens.to(target_device) batch_indices, non_audio_indices = torch.where((input_ids != self.config.audio_token_id) & (attention_mask == 1)) token_placeholder_num = torch.zeros_like(input_ids) token_placeholder_num[special_audio_token_mask] = num_audio_tokens.long() - 1 token_placeholder_num = token_placeholder_num + 1 new_token_positions = torch.cumsum(token_placeholder_num, -1) - 1 max_token_num = token_placeholder_num.sum(-1).max() nb_audio_pad = max_token_num - 1 - new_token_positions[:, -1] if left_padding: new_token_positions += nb_audio_pad[:, None] text_to_overwrite = new_token_positions[batch_indices, non_audio_indices] batch_indices, non_audio_indices, text_to_overwrite = (batch_indices.to(target_device), non_audio_indices.to(target_device), text_to_overwrite.to(target_device)) final_embedding = torch.zeros(batch_size, max_token_num, embed_dim, dtype=inputs_embeds.dtype, device=inputs_embeds.device) final_attention_mask = torch.zeros(batch_size, max_token_num, dtype=attention_mask.dtype, device=inputs_embeds.device) final_input_ids = torch.full((batch_size, max_token_num), self.pad_token_id, dtype=input_ids.dtype, device=inputs_embeds.device) final_embedding[batch_indices, text_to_overwrite] = inputs_embeds[batch_indices, non_audio_indices] final_attention_mask[batch_indices, text_to_overwrite] = attention_mask[batch_indices, non_audio_indices] final_input_ids[batch_indices, text_to_overwrite] = input_ids[batch_indices, non_audio_indices] final_labels = None if labels is not None: labels = labels.to(target_device) final_labels = torch.full_like(final_attention_mask, self.config.ignore_index).to(torch.long) final_labels[batch_indices, text_to_overwrite] = labels[batch_indices, non_audio_indices] audio_to_overwrite = torch.full((batch_size, max_token_num), True, dtype=torch.bool, device=inputs_embeds.device) audio_to_overwrite[batch_indices, text_to_overwrite] = False seq_indices = torch.arange(max_token_num).unsqueeze(0).to(target_device) seq_indices = seq_indices.expand(batch_size, max_token_num) if left_padding: max_token_num = max_token_num.to(target_device) val = max_token_num - seq_indices <= (token_placeholder_num.sum(-1) - (attention_mask == 0).long().sum(-1))[:, None] else: val = seq_indices < (token_placeholder_num.sum(-1) - (attention_mask == 0).long().sum(-1))[:, None] audio_to_overwrite &= val if audio_to_overwrite.sum() != num_audio_tokens.sum(): raise ValueError(f'The input provided to the model are wrong. The number of audio tokens is {num_special_audio_tokens} while the number of audio given to the model is {num_audios}. This prevents correct indexing and breaks batch generation.') final_embedding[audio_to_overwrite] = masked_audio_features.contiguous().reshape(-1, embed_dim).to(target_device) final_attention_mask |= audio_to_overwrite position_ids = (final_attention_mask.cumsum(-1) - 1).masked_fill_(final_attention_mask == 0, 1) return (final_embedding, final_attention_mask, final_labels, position_ids, final_input_ids) @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, input_features: Optional[torch.FloatTensor]=None, attention_mask: Optional[torch.Tensor]=None, feature_attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None) -> Union[tuple, Qwen2AudioCausalLMOutputWithPast]: """ feature_attention_mask (`torch.Tensor` of shape `(batch_size, feature_sequence_length)`): Mask to avoid performing attention on padding feature indices. 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 masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. Example: ```python >>> from io import BytesIO >>> from urllib.request import urlopen >>> import librosa >>> from transformers import AutoProcessor, Qwen2AudioForConditionalGeneration >>> model = Qwen2AudioForConditionalGeneration.from_pretrained("Qwen/Qwen2-Audio-7B") >>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2-Audio-7B") >>> prompt = "<|audio_bos|><|AUDIO|><|audio_eos|>Generate the caption in English:" >>> url = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/glass-breaking-151256.mp3" >>> audio, _ = librosa.load(BytesIO(urlopen(url).read()), sr=self.processor.feature_extractor.sampling_rate) >>> inputs = processor(text=prompt, audios=audio, return_tensors="pt") >>> # Generate >>> generate_ids = model.generate(**inputs, max_length=30) >>> processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "Generate the caption in English: Glass is breaking." ```""" 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 target_device = self.audio_tower.device if input_features is not None: input_features = input_features.to(target_device) feature_attention_mask = feature_attention_mask.to(target_device) if inputs_embeds is None: inputs_embeds = self.get_input_embeddings()(input_ids) if input_features is not None and input_ids.shape[1] != 1: audio_feat_lengths, audio_output_lengths = self.audio_tower._get_feat_extract_output_lengths(feature_attention_mask.sum(-1)) batch_size, _, max_mel_seq_len = input_features.shape max_seq_len = (max_mel_seq_len - 2) // 2 + 1 seq_range = torch.arange(0, max_seq_len, dtype=audio_feat_lengths.dtype, device=audio_feat_lengths.device).unsqueeze(0).expand(batch_size, max_seq_len) lengths_expand = audio_feat_lengths.unsqueeze(1).expand(batch_size, max_seq_len) padding_mask = seq_range >= lengths_expand audio_attention_mask_ = padding_mask.view(batch_size, 1, 1, max_seq_len).expand(batch_size, 1, max_seq_len, max_seq_len) audio_attention_mask = audio_attention_mask_.to(dtype=self.audio_tower.conv1.weight.dtype, device=self.audio_tower.conv1.weight.device) audio_attention_mask[audio_attention_mask_] = float('-inf') audio_outputs = self.audio_tower(input_features, attention_mask=audio_attention_mask) selected_audio_feature = audio_outputs.last_hidden_state audio_features = self.multi_modal_projector(selected_audio_feature) audio_tokens = input_ids == self.config.audio_token_id legacy_processing = (audio_tokens[:, :-1] & audio_tokens[:, 1:]).sum() == 0 if legacy_processing: logger.warning_once('Expanding inputs for audio tokens in Qwen2Audio should be done in processing.') inputs_embeds, attention_mask, labels, position_ids, _ = self._merge_input_ids_with_audio_features(audio_features, audio_output_lengths, inputs_embeds, input_ids, attention_mask, labels) else: num_audios, max_audio_tokens, embed_dim = audio_features.shape audio_features_mask = torch.arange(max_audio_tokens, device=audio_output_lengths.device)[None, :] audio_features_mask = audio_features_mask < audio_output_lengths[:, None] audio_features = audio_features[audio_features_mask] n_audio_tokens = (input_ids == self.config.audio_token_id).sum().item() n_audio_features = audio_features.shape[0] if n_audio_tokens != n_audio_features: raise ValueError(f'Audio features and audio tokens do not match: tokens: {n_audio_tokens}, features {n_audio_features}') special_audio_mask = (input_ids == self.config.audio_token_id).to(inputs_embeds.device) special_audio_mask = special_audio_mask.unsqueeze(-1).expand_as(inputs_embeds) audio_features = audio_features.to(inputs_embeds.device, inputs_embeds.dtype) inputs_embeds = inputs_embeds.masked_scatter(special_audio_mask, audio_features) outputs = self.language_model(attention_mask=attention_mask, position_ids=position_ids, past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache, output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=return_dict, cache_position=cache_position) logits = outputs[0] loss = None if labels is not None: if attention_mask is not None: shift_attention_mask = attention_mask[..., 1:] shift_logits = logits[..., :-1, :][shift_attention_mask.to(logits.device) != 0].contiguous() shift_labels = labels[..., 1:][shift_attention_mask.to(labels.device) != 0].contiguous() else: shift_logits = logits[..., :-1, :].contiguous() shift_labels = labels[..., 1:].contiguous() loss_fct = nn.CrossEntropyLoss() loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1).to(shift_logits.device)) if not return_dict: output = (logits,) + outputs[1:] return (loss,) + output if loss is not None else output return Qwen2AudioCausalLMOutputWithPast(loss=loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions, attention_mask=attention_mask) def prepare_inputs_for_generation(self, *args, **kwargs): input_features = kwargs.pop('input_features', None) cache_position = kwargs.get('cache_position') model_inputs = super().prepare_inputs_for_generation(*args, **kwargs) if cache_position is not None and cache_position[0] == 0: model_inputs['input_features'] = input_features return model_inputs
@auto_docstring(custom_intro='\n The QWEN2AUDIO model which consists of a audio backbone and a language model.\n ') class Qwen2AudioForConditionalGeneration(Qwen2AudioPreTrainedModel, GenerationMixin): def __init__(self, config: Qwen2AudioConfig): pass @property def padding_side(self): pass @padding_side.setter def padding_side(self): pass def get_input_embeddings(self): pass def set_input_embeddings(self, value): pass def get_output_embeddings(self): pass def set_output_embeddings(self, new_embeddings): pass def set_decoder(self, decoder): pass def get_decoder(self): pass def _merge_input_ids_with_audio_features(self, audio_features, num_audio_tokens, inputs_embeds, input_ids, attention_mask, labels): ''' Merge input_ids with with audio features into final embeddings Args: audio_features (`torch.Tensor` of shape `(num_audios, max_audio_tokens, embed_dim)`): All audio vectors of all audios in the batch num_audio_tokens (`torch.LongTensor` of shape `(num_audios)`): The length of audio embeddings of each audio as stacked in `audio_features` inputs_embeds (`torch.Tensor` of shape `(batch_size, sequence_length, embed_dim)`): Token embeddings before merging with audio embeddings input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Input_ids of tokens, possibly filled with audio token attention_mask (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Mask to avoid performing attention on padding token indices. labels (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) labels need to be recalculated to support training (if provided) Returns: final_embedding, final_attention_mask, final_labels, position_ids, final_input_ids Explanation: each audio has variable length embeddings, with length specified by num_audio_tokens audio_features is concatenation of all audio embed vectors task: fill each <|AUDIO|> with the correct number of audio embeddings Example: X (5 tokens), Y (3 tokens), Z (8 tokens) X, Y are in the same sequence (in-context learning) if right padding input_ids: [ a b c d e f X g h i j k Y l m o p q r Z s t u v _ _ _ _ _ _ ] input_ids should be: [ a b c d e f X X X X X g h i j k Y Y Y l m o p q r Z Z Z Z Z Z Z Z s t u v _ _ _ _ _ ] labels should be: [ a b c d e f _ _ _ _ _ g h i j k _ _ _ l m o p q r _ _ _ _ _ _ _ _ s t u v _ _ _ _ _ ] elif left padding input_ids: [ a b c d e f X g h i j k Y l m _ _ _ _ _ _ o p q r Z s t u v ] input_ids should be: [ a b c d e f X X X X X g h i j k Y Y Y l m _ _ _ _ _ o p q r Z Z Z Z Z Z Z Z s t u v ] labels should be: [ a b c d e f _ _ _ _ _ g h i j k _ _ _ l m _ _ _ _ _ o p q r _ _ _ _ _ _ _ _ s t u v ] Edge cases: * If tokens are same but audio token sizes are different, then cannot infer left or right padding ```python url1 = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/glass-breaking-151256.mp3" audio1, _ = librosa.load(BytesIO(urlopen(url1).read()), sr=processor.feature_extractor.sampling_rate) url2 = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/f2641_0_throatclearing.wav" audio2, _ = librosa.load(BytesIO(urlopen(url2).read()), sr=processor.feature_extractor.sampling_rate) prompts = [ "[INST] <|AUDIO|> What is that in this audio? [/INST]", "[INST] <|AUDIO|> What is that in this audio? [/INST]", ] inputs = processor(text=prompts, audios=[audio1, audio2], return_tensors='pt', padding=True).to("cuda") audio1 has 101 tokens, while audio2 has 72 tokens ``` input_ids: [ a b c d X g h i j Y k l m n ] where X is 3 tokens while Y is 5, this mean after merge if left-padding (batched generation) input_ids should be: [ _ _ a b c d X X X g h i j Y Y Y Y Y k l m n ] elif (right padding) (training) input_ids should be: [ a b c d X X X g h _ _ i j Y Y Y Y Y k l m n ] ''' pass @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, input_features: Optional[torch.FloatTensor]=None, attention_mask: Optional[torch.Tensor]=None, feature_attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, return_dict: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None) -> Union[tuple, Qwen2AudioCausalLMOutputWithPast]: ''' feature_attention_mask (`torch.Tensor` of shape `(batch_size, feature_sequence_length)`): Mask to avoid performing attention on padding feature indices. 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 masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. Example: ```python >>> from io import BytesIO >>> from urllib.request import urlopen >>> import librosa >>> from transformers import AutoProcessor, Qwen2AudioForConditionalGeneration >>> model = Qwen2AudioForConditionalGeneration.from_pretrained("Qwen/Qwen2-Audio-7B") >>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2-Audio-7B") >>> prompt = "<|audio_bos|><|AUDIO|><|audio_eos|>Generate the caption in English:" >>> url = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/glass-breaking-151256.mp3" >>> audio, _ = librosa.load(BytesIO(urlopen(url).read()), sr=self.processor.feature_extractor.sampling_rate) >>> inputs = processor(text=prompt, audios=audio, return_tensors="pt") >>> # Generate >>> generate_ids = model.generate(**inputs, max_length=30) >>> processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "Generate the caption in English: Glass is breaking." ```''' pass def prepare_inputs_for_generation(self, *args, **kwargs): pass
17
2
36
4
22
10
4
0.48
2
14
6
0
14
7
14
15
531
63
318
117
269
152
195
84
180
13
2
4
54
4,771
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/modeling_qwen2_audio.py
transformers.models.qwen2_audio.modeling_qwen2_audio.Qwen2AudioMultiModalProjector
from .configuration_qwen2_audio import Qwen2AudioConfig, Qwen2AudioEncoderConfig from torch import nn class Qwen2AudioMultiModalProjector(nn.Module): def __init__(self, config: Qwen2AudioConfig): super().__init__() self.linear = nn.Linear(config.audio_config.d_model, config.text_config.hidden_size, bias=True) def forward(self, audio_features): hidden_states = self.linear(audio_features) return hidden_states
class Qwen2AudioMultiModalProjector(nn.Module): def __init__(self, config: Qwen2AudioConfig): pass def forward(self, audio_features): pass
3
0
3
0
3
0
1
0
1
2
1
0
2
1
2
12
8
1
7
5
4
0
7
5
4
1
1
0
2
4,772
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/modeling_qwen2_audio.py
transformers.models.qwen2_audio.modeling_qwen2_audio.Qwen2AudioPreTrainedModel
from torch import nn from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel from ...utils import auto_docstring, logging from .configuration_qwen2_audio import Qwen2AudioConfig, Qwen2AudioEncoderConfig @auto_docstring class Qwen2AudioPreTrainedModel(PreTrainedModel): config: Qwen2AudioConfig base_model_prefix = 'model' supports_gradient_checkpointing = True _no_split_modules = ['Qwen2AudioAttention'] _skip_keys_device_placement = 'past_key_values' _supports_flash_attn = True _supports_sdpa = True def _init_weights(self, module): std = self.config.initializer_range if hasattr(self.config, 'initializer_range') else self.config.audio_config.initializer_range if isinstance(module, (nn.Linear, nn.Conv1d)): module.weight.data.normal_(mean=0.0, std=std) if module.bias is not None: module.bias.data.zero_() elif isinstance(module, nn.LayerNorm): module.weight.data.fill_(1.0) module.bias.data.zero_() elif isinstance(module, nn.Embedding): module.weight.data.normal_(mean=0.0, std=std) if module.padding_idx is not None: module.weight.data[module.padding_idx].zero_()
@auto_docstring class Qwen2AudioPreTrainedModel(PreTrainedModel): def _init_weights(self, module): pass
3
0
13
1
10
2
6
0.11
1
0
0
2
1
0
1
1
22
2
18
10
16
2
17
10
15
6
1
2
6
4,773
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_audio/processing_qwen2_audio.py
transformers.models.qwen2_audio.processing_qwen2_audio.Qwen2AudioProcessor
from ...tokenization_utils_base import PreTokenizedInput, TextInput from ...feature_extraction_utils import BatchFeature from typing import Union import numpy as np from ...processing_utils import ProcessingKwargs, ProcessorMixin, Unpack class Qwen2AudioProcessor(ProcessorMixin): """ Constructs a Qwen2Audio processor which wraps a Qwen2Audio feature extractor and a Qwen2Audio tokenizer into a single processor. [`Qwen2AudioProcessor`] offers all the functionalities of [`WhisperFeatureExtractor`] and [`Qwen2TokenizerFast`]. See the [`~Qwen2AudioProcessor.__call__`] and [`~Qwen2AudioProcessor.decode`] for more information. Args: feature_extractor ([`WhisperFeatureExtractor`], *optional*): The feature extractor is a required input. tokenizer ([`Qwen2TokenizerFast`], *optional*): The tokenizer is a required input. chat_template (`Optional[str]`, *optional*): The Jinja template to use for formatting the conversation. If not provided, the default chat template is used. audio_token (`str`, *optional*, defaults to `"<|AUDIO|>"`): The token to use for audio tokens. audio_bos_token (`str`, *optional*, defaults to `"<|audio_bos|>"`): The token to use for audio bos tokens. audio_eos_token (`str`, *optional*, defaults to `"<|audio_eos|>"`): The token to use for audio eos tokens. """ attributes = ['feature_extractor', 'tokenizer'] feature_extractor_class = 'WhisperFeatureExtractor' tokenizer_class = 'AutoTokenizer' def __init__(self, feature_extractor=None, tokenizer=None, chat_template=None, audio_token='<|AUDIO|>', audio_bos_token='<|audio_bos|>', audio_eos_token='<|audio_eos|>'): if chat_template is None: chat_template = self.default_chat_template self.audio_token = tokenizer.audio_token if hasattr(tokenizer, 'audio_token') else audio_token self.audio_token_id = tokenizer.convert_tokens_to_ids(self.audio_token) self.audio_bos_token = tokenizer.audio_bos_token if hasattr(tokenizer, 'audio_bos_token') else audio_bos_token self.audio_eos_token = tokenizer.audio_eos_token if hasattr(tokenizer, 'audio_eos_token') else audio_eos_token super().__init__(feature_extractor, tokenizer, chat_template=chat_template) def __call__(self, text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]=None, audio: Union[np.ndarray, list[np.ndarray]]=None, **kwargs: Unpack[Qwen2AudioProcessorKwargs]) -> BatchFeature: """ Main method to prepare for the model one or several sequences(s) and audio(s). This method forwards the `text` and `kwargs` arguments to Qwen2TokenizerFast's [`~Qwen2TokenizerFast.__call__`] if `text` is not `None` to encode the text. To prepare the audio(s), this method forwards the `audios` and `kwargs` arguments to WhisperFeatureExtractor's [`~WhisperFeatureExtractor.__call__`] if `audios` is not `None`. Please refer to the docstring of the above two methods for more information. Args: text (`str`, `list[str]`, `list[list[str]]`): The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set `is_split_into_words=True` (to lift the ambiguity with a batch of sequences). audio (`np.ndarray`, `list[np.ndarray]`): The audio or batch of audios to be prepared. Each audio can be a NumPy array. """ if text is None: raise ValueError('You need to specify `text` input to process.') elif isinstance(text, str): text = [text] elif not isinstance(text, list) and (not isinstance(text[0], str)): raise ValueError('Invalid input text. Please provide a string, or a list of strings') output_kwargs = self._merge_kwargs(Qwen2AudioProcessorKwargs, tokenizer_init_kwargs=self.tokenizer.init_kwargs, **kwargs) if audio is not None: num_audio_tokens = sum((sample.count(self.audio_token) for sample in text)) num_audios = 1 if type(audio) is np.ndarray else len(audio) if num_audio_tokens != num_audios: raise ValueError(f"Found {num_audio_tokens} {self.audio_token} token{('s' if num_audio_tokens > 1 else '')} in provided text but received {num_audios} audio{('s' if num_audios > 1 else '')}") output_kwargs['audio_kwargs']['return_attention_mask'] = True output_kwargs['audio_kwargs']['padding'] = 'max_length' audio_inputs = self.feature_extractor(audio, **output_kwargs['audio_kwargs']) audio_inputs['feature_attention_mask'] = audio_inputs.pop('attention_mask') expanded_text = [] audio_lengths = audio_inputs['feature_attention_mask'].sum(-1).tolist() for sample in text: replace_str = [] while self.audio_token in sample: audio_length = audio_lengths.pop(0) input_length = (audio_length - 1) // 2 + 1 num_audio_tokens = (input_length - 2) // 2 + 1 expanded_audio_token = self.audio_token * num_audio_tokens audio_token_start_idx = sample.find(self.audio_token) audio_token_end_idx = audio_token_start_idx + len(self.audio_token) has_bos = sample[audio_token_start_idx - len(self.audio_bos_token):audio_token_start_idx] == self.audio_bos_token has_eos = sample[audio_token_end_idx:audio_token_end_idx + len(self.audio_eos_token)] == self.audio_eos_token if not has_bos and (not has_eos): expanded_audio_token = self.audio_bos_token + expanded_audio_token + self.audio_eos_token replace_str.append(expanded_audio_token) sample = sample.replace(self.audio_token, '<placeholder>', 1) while '<placeholder>' in sample: sample = sample.replace('<placeholder>', replace_str.pop(0), 1) expanded_text.append(sample) text = expanded_text return_tensors = output_kwargs['text_kwargs'].pop('return_tensors', None) inputs = self.tokenizer(text, **output_kwargs['text_kwargs']) self._check_special_mm_tokens(text, inputs, modalities=['audio']) if audio is not None: inputs.update(audio_inputs) return BatchFeature(data={**inputs}, tensor_type=return_tensors) @property def model_input_names(self): tokenizer_input_names = self.tokenizer.model_input_names feature_extractor_input_names = self.feature_extractor.model_input_names return list(dict.fromkeys(tokenizer_input_names + feature_extractor_input_names + ['feature_attention_mask'])) @property def default_chat_template(self): """ This default vicuna template formats inputs in the form of a chat history. For each message in the chat history: * the template will output the role of the speaker followed by the content of the message. * content is a list of strings and audios. * If the content element is an audio, the template will output a sequence of <|AUDIO|> tokens Example: ```python messages = [ {'role': 'system', 'content': 'You are a helpful assistant.'}, {"role": "user", "content": [ {"type": "audio", "audio_url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/glass-breaking-151256.mp3"}, {"type": "text", "text": "What's that sound?"}, ]}, {"role": "assistant", "content": "It is the sound of glass shattering."}, {"role": "user", "content": [ {"type": "audio", "audio_url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/f2641_0_throatclearing.wav"}, {"type": "text", "text": "How about this one?"}, ]}, ] result = template.render(messages=messages, add_generation_prompt=True) ``` """ return "{% set audio_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if 'audio' in content or 'audio_url' in content or message['type'] == 'audio' or content['type'] == 'audio' %}{% set audio_count.value = audio_count.value + 1 %}Audio {{ audio_count.value }}: <|audio_bos|><|AUDIO|><|audio_eos|>\n{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
class Qwen2AudioProcessor(ProcessorMixin): ''' Constructs a Qwen2Audio processor which wraps a Qwen2Audio feature extractor and a Qwen2Audio tokenizer into a single processor. [`Qwen2AudioProcessor`] offers all the functionalities of [`WhisperFeatureExtractor`] and [`Qwen2TokenizerFast`]. See the [`~Qwen2AudioProcessor.__call__`] and [`~Qwen2AudioProcessor.decode`] for more information. Args: feature_extractor ([`WhisperFeatureExtractor`], *optional*): The feature extractor is a required input. tokenizer ([`Qwen2TokenizerFast`], *optional*): The tokenizer is a required input. chat_template (`Optional[str]`, *optional*): The Jinja template to use for formatting the conversation. If not provided, the default chat template is used. audio_token (`str`, *optional*, defaults to `"<|AUDIO|>"`): The token to use for audio tokens. audio_bos_token (`str`, *optional*, defaults to `"<|audio_bos|>"`): The token to use for audio bos tokens. audio_eos_token (`str`, *optional*, defaults to `"<|audio_eos|>"`): The token to use for audio eos tokens. ''' def __init__(self, feature_extractor=None, tokenizer=None, chat_template=None, audio_token='<|AUDIO|>', audio_bos_token='<|audio_bos|>', audio_eos_token='<|audio_eos|>'): pass def __call__(self, text: Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]=None, audio: Union[np.ndarray, list[np.ndarray]]=None, **kwargs: Unpack[Qwen2AudioProcessorKwargs]) -> BatchFeature: ''' Main method to prepare for the model one or several sequences(s) and audio(s). This method forwards the `text` and `kwargs` arguments to Qwen2TokenizerFast's [`~Qwen2TokenizerFast.__call__`] if `text` is not `None` to encode the text. To prepare the audio(s), this method forwards the `audios` and `kwargs` arguments to WhisperFeatureExtractor's [`~WhisperFeatureExtractor.__call__`] if `audios` is not `None`. Please refer to the docstring of the above two methods for more information. Args: text (`str`, `list[str]`, `list[list[str]]`): The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set `is_split_into_words=True` (to lift the ambiguity with a batch of sequences). audio (`np.ndarray`, `list[np.ndarray]`): The audio or batch of audios to be prepared. Each audio can be a NumPy array. ''' pass @property def model_input_names(self): pass @property def default_chat_template(self): ''' This default vicuna template formats inputs in the form of a chat history. For each message in the chat history: * the template will output the role of the speaker followed by the content of the message. * content is a list of strings and audios. * If the content element is an audio, the template will output a sequence of <|AUDIO|> tokens Example: ```python messages = [ {'role': 'system', 'content': 'You are a helpful assistant.'}, {"role": "user", "content": [ {"type": "audio", "audio_url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/glass-breaking-151256.mp3"}, {"type": "text", "text": "What's that sound?"}, ]}, {"role": "assistant", "content": "It is the sound of glass shattering."}, {"role": "user", "content": [ {"type": "audio", "audio_url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/f2641_0_throatclearing.wav"}, {"type": "text", "text": "How about this one?"}, ]}, ] result = template.render(messages=messages, add_generation_prompt=True) ``` ''' pass
7
3
31
3
18
10
4
0.69
1
9
1
0
6
3
6
23
217
27
113
46
89
78
58
29
51
14
2
4
23
4,774
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/configuration_qwen2_moe.py
transformers.models.qwen2_moe.configuration_qwen2_moe.Qwen2MoeConfig
from ...configuration_utils import PretrainedConfig from ...modeling_rope_utils import rope_config_validation class Qwen2MoeConfig(PretrainedConfig): """ This is the configuration class to store the configuration of a [`Qwen2MoeModel`]. It is used to instantiate a Qwen2MoE model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of [Qwen/Qwen1.5-MoE-A2.7B](https://huggingface.co/Qwen/Qwen1.5-MoE-A2.7B). 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 151936): Vocabulary size of the Qwen2MoE model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`Qwen2MoeModel`] hidden_size (`int`, *optional*, defaults to 2048): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 5632): Dimension of the MLP representations. num_hidden_layers (`int`, *optional*, defaults to 24): Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 16): Number of attention heads for each attention layer in the Transformer encoder. num_key_value_heads (`int`, *optional*, defaults to 16): This is the number of key_value heads that should be used to implement Grouped Query Attention. If `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out [this paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `32`. hidden_act (`str` or `function`, *optional*, defaults to `"silu"`): The non-linear activation function (function or string) in the decoder. max_position_embeddings (`int`, *optional*, defaults to 32768): The maximum sequence length that this model might ever be used with. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. rms_norm_eps (`float`, *optional*, defaults to 1e-06): The epsilon used by the rms normalization layers. 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`. tie_word_embeddings (`bool`, *optional*, defaults to `False`): Whether the model's input and output word embeddings should be tied. rope_theta (`float`, *optional*, defaults to 10000.0): The base period of the RoPE embeddings. rope_scaling (`Dict`, *optional*): Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value accordingly. Expected contents: `rope_type` (`str`): The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope', 'llama3'], with 'default' being the original RoPE implementation. `factor` (`float`, *optional*): Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In most scaling types, a `factor` of x will enable the model to handle sequences of length x * original maximum pre-trained length. `original_max_position_embeddings` (`int`, *optional*): Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during pretraining. `attention_factor` (`float`, *optional*): Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention computation. If unspecified, it defaults to value recommended by the implementation, using the `factor` field to infer the suggested value. `beta_fast` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear ramp function. If unspecified, it defaults to 32. `beta_slow` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear ramp function. If unspecified, it defaults to 1. `short_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to short contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `long_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to long contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `low_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE `high_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE use_sliding_window (`bool`, *optional*, defaults to `False`): Whether to use sliding window attention. sliding_window (`int`, *optional*, defaults to 4096): Sliding window attention (SWA) window size. If not specified, will default to `4096`. max_window_layers (`int`, *optional*, defaults to 28): The number of layers using full attention. The first `max_window_layers` layers will use full attention, while any additional layer afterwards will use SWA (Sliding Window Attention). attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. decoder_sparse_step (`int`, *optional*, defaults to 1): The frequency of the MoE layer. moe_intermediate_size (`int`, *optional*, defaults to 1408): Intermediate size of the routed expert. shared_expert_intermediate_size (`int`, *optional*, defaults to 5632): Intermediate size of the shared expert. num_experts_per_tok (`int`, *optional*, defaults to 4): Number of selected experts. num_experts (`int`, *optional*, defaults to 60): Number of routed experts. norm_topk_prob (`bool`, *optional*, defaults to `False`): Whether to normalize the topk probabilities. output_router_logits (`bool`, *optional*, defaults to `False`): Whether or not the router logits should be returned by the model. Enabling this will also allow the model to output the auxiliary loss, including load balancing loss and router z-loss. router_aux_loss_coef (`float`, *optional*, defaults to 0.001): The aux loss factor for the total loss. mlp_only_layers (`list[int]`, *optional*, defaults to `[]`): Indicate which layers use Qwen2MoeMLP rather than Qwen2MoeSparseMoeBlock The list contains layer index, from 0 to num_layers-1 if we have num_layers layers If `mlp_only_layers` is empty, `decoder_sparse_step` is used to determine the sparsity. qkv_bias (`bool`, *optional*, defaults to `True`): Whether to add a bias to the queries, keys and values. ```python >>> from transformers import Qwen2MoeModel, Qwen2MoeConfig >>> # Initializing a Qwen2MoE style configuration >>> configuration = Qwen2MoeConfig() >>> # Initializing a model from the Qwen1.5-MoE-A2.7B" style configuration >>> model = Qwen2MoeModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = 'qwen2_moe' keys_to_ignore_at_inference = ['past_key_values'] base_model_tp_plan = {'layers.*.self_attn.q_proj': 'colwise', 'layers.*.self_attn.k_proj': 'colwise', 'layers.*.self_attn.v_proj': 'colwise', 'layers.*.self_attn.o_proj': 'rowwise', 'layers.*.mlp.gate_proj': 'colwise', 'layers.*.mlp.up_proj': 'colwise', 'layers.*.mlp.down_proj': 'rowwise'} base_model_pp_plan = {'embed_tokens': (['input_ids'], ['inputs_embeds']), 'layers': (['hidden_states', 'attention_mask'], ['hidden_states']), 'norm': (['hidden_states'], ['hidden_states'])} def __init__(self, vocab_size=151936, hidden_size=2048, intermediate_size=5632, num_hidden_layers=24, num_attention_heads=16, num_key_value_heads=16, hidden_act='silu', max_position_embeddings=32768, initializer_range=0.02, rms_norm_eps=1e-06, use_cache=True, tie_word_embeddings=False, rope_theta=10000.0, rope_scaling=None, use_sliding_window=False, sliding_window=4096, max_window_layers=28, attention_dropout=0.0, decoder_sparse_step=1, moe_intermediate_size=1408, shared_expert_intermediate_size=5632, num_experts_per_tok=4, num_experts=60, norm_topk_prob=False, output_router_logits=False, router_aux_loss_coef=0.001, mlp_only_layers=None, qkv_bias=True, **kwargs): self.vocab_size = vocab_size self.max_position_embeddings = max_position_embeddings self.hidden_size = hidden_size self.intermediate_size = intermediate_size self.num_hidden_layers = num_hidden_layers self.num_attention_heads = num_attention_heads self.use_sliding_window = use_sliding_window self.sliding_window = sliding_window if use_sliding_window else None self.max_window_layers = max_window_layers self.num_key_value_heads = num_key_value_heads self.hidden_act = hidden_act self.initializer_range = initializer_range self.rms_norm_eps = rms_norm_eps self.use_cache = use_cache self.rope_theta = rope_theta self.rope_scaling = rope_scaling self.attention_dropout = attention_dropout if self.rope_scaling is not None and 'type' in self.rope_scaling: self.rope_scaling['rope_type'] = self.rope_scaling['type'] rope_config_validation(self) self.decoder_sparse_step = decoder_sparse_step self.moe_intermediate_size = moe_intermediate_size self.shared_expert_intermediate_size = shared_expert_intermediate_size self.num_experts_per_tok = num_experts_per_tok self.num_experts = num_experts self.norm_topk_prob = norm_topk_prob self.output_router_logits = output_router_logits self.router_aux_loss_coef = router_aux_loss_coef self.mlp_only_layers = [] if mlp_only_layers is None else mlp_only_layers self.qkv_bias = qkv_bias super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)
class Qwen2MoeConfig(PretrainedConfig): ''' This is the configuration class to store the configuration of a [`Qwen2MoeModel`]. It is used to instantiate a Qwen2MoE model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of [Qwen/Qwen1.5-MoE-A2.7B](https://huggingface.co/Qwen/Qwen1.5-MoE-A2.7B). 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 151936): Vocabulary size of the Qwen2MoE model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`Qwen2MoeModel`] hidden_size (`int`, *optional*, defaults to 2048): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 5632): Dimension of the MLP representations. num_hidden_layers (`int`, *optional*, defaults to 24): Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 16): Number of attention heads for each attention layer in the Transformer encoder. num_key_value_heads (`int`, *optional*, defaults to 16): This is the number of key_value heads that should be used to implement Grouped Query Attention. If `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out [this paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `32`. hidden_act (`str` or `function`, *optional*, defaults to `"silu"`): The non-linear activation function (function or string) in the decoder. max_position_embeddings (`int`, *optional*, defaults to 32768): The maximum sequence length that this model might ever be used with. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. rms_norm_eps (`float`, *optional*, defaults to 1e-06): The epsilon used by the rms normalization layers. 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`. tie_word_embeddings (`bool`, *optional*, defaults to `False`): Whether the model's input and output word embeddings should be tied. rope_theta (`float`, *optional*, defaults to 10000.0): The base period of the RoPE embeddings. rope_scaling (`Dict`, *optional*): Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value accordingly. Expected contents: `rope_type` (`str`): The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope', 'llama3'], with 'default' being the original RoPE implementation. `factor` (`float`, *optional*): Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In most scaling types, a `factor` of x will enable the model to handle sequences of length x * original maximum pre-trained length. `original_max_position_embeddings` (`int`, *optional*): Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during pretraining. `attention_factor` (`float`, *optional*): Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention computation. If unspecified, it defaults to value recommended by the implementation, using the `factor` field to infer the suggested value. `beta_fast` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear ramp function. If unspecified, it defaults to 32. `beta_slow` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear ramp function. If unspecified, it defaults to 1. `short_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to short contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `long_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to long contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `low_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE `high_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE use_sliding_window (`bool`, *optional*, defaults to `False`): Whether to use sliding window attention. sliding_window (`int`, *optional*, defaults to 4096): Sliding window attention (SWA) window size. If not specified, will default to `4096`. max_window_layers (`int`, *optional*, defaults to 28): The number of layers using full attention. The first `max_window_layers` layers will use full attention, while any additional layer afterwards will use SWA (Sliding Window Attention). attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. decoder_sparse_step (`int`, *optional*, defaults to 1): The frequency of the MoE layer. moe_intermediate_size (`int`, *optional*, defaults to 1408): Intermediate size of the routed expert. shared_expert_intermediate_size (`int`, *optional*, defaults to 5632): Intermediate size of the shared expert. num_experts_per_tok (`int`, *optional*, defaults to 4): Number of selected experts. num_experts (`int`, *optional*, defaults to 60): Number of routed experts. norm_topk_prob (`bool`, *optional*, defaults to `False`): Whether to normalize the topk probabilities. output_router_logits (`bool`, *optional*, defaults to `False`): Whether or not the router logits should be returned by the model. Enabling this will also allow the model to output the auxiliary loss, including load balancing loss and router z-loss. router_aux_loss_coef (`float`, *optional*, defaults to 0.001): The aux loss factor for the total loss. mlp_only_layers (`list[int]`, *optional*, defaults to `[]`): Indicate which layers use Qwen2MoeMLP rather than Qwen2MoeSparseMoeBlock The list contains layer index, from 0 to num_layers-1 if we have num_layers layers If `mlp_only_layers` is empty, `decoder_sparse_step` is used to determine the sparsity. qkv_bias (`bool`, *optional*, defaults to `True`): Whether to add a bias to the queries, keys and values. ```python >>> from transformers import Qwen2MoeModel, Qwen2MoeConfig >>> # Initializing a Qwen2MoE style configuration >>> configuration = Qwen2MoeConfig() >>> # Initializing a model from the Qwen1.5-MoE-A2.7B" style configuration >>> model = Qwen2MoeModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```''' def __init__(self, vocab_size=151936, hidden_size=2048, intermediate_size=5632, num_hidden_layers=24, num_attention_heads=16, num_key_value_heads=16, hidden_act='silu', max_position_embeddings=32768, initializer_range=0.02, rms_norm_eps=1e-06, use_cache=True, tie_word_embeddings=False, rope_theta=10000.0, rope_scaling=None, use_sliding_window=False, sliding_window=4096, max_window_layers=28, attention_dropout=0.0, decoder_sparse_step=1, moe_intermediate_size=1408, shared_expert_intermediate_size=5632, num_experts_per_tok=4, num_experts=60, norm_topk_prob=False, output_router_logits=False, router_aux_loss_coef=0.001, mlp_only_layers=None, qkv_bias=True, **kwargs): pass
2
1
70
3
64
3
4
1.58
1
1
0
0
1
26
1
1
209
13
76
61
44
120
35
31
33
4
1
1
4
4,775
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeAttention
import torch from ...cache_utils import Cache, DynamicCache, StaticCache from torch import nn from ...utils.deprecation import deprecate_kwarg from typing import Optional, Union from .configuration_qwen2_moe import Qwen2MoeConfig import torch.nn.functional as F import math class Qwen2MoeAttention(nn.Module): """ Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer and "Generating Long Sequences with Sparse Transformers". """ def __init__(self, config: Qwen2MoeConfig, layer_idx: Optional[int]=None): super().__init__() self.config = config self.layer_idx = layer_idx if layer_idx is None: logger.warning_once(f'Instantiating {self.__class__.__name__} without passing `layer_idx` is not recommended and will to errors during the forward call, if caching is used. Please make sure to provide a `layer_idx` when creating this class.') self.hidden_size = config.hidden_size self.num_heads = config.num_attention_heads self.head_dim = getattr(config, 'head_dim', config.hidden_size // config.num_attention_heads) self.num_key_value_heads = config.num_key_value_heads self.num_key_value_groups = self.num_heads // self.num_key_value_heads self.max_position_embeddings = config.max_position_embeddings self.rope_theta = config.rope_theta self.is_causal = True self.attention_dropout = config.attention_dropout if self.head_dim * self.num_heads != self.hidden_size: raise ValueError(f'hidden_size must be divisible by num_heads (got `hidden_size`: {self.hidden_size} and `num_heads`: {self.num_heads}).') self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=self.config.qkv_bias) self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=self.config.qkv_bias) self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=self.config.qkv_bias) self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=False) self.rotary_emb = Qwen2MoeRotaryEmbedding(config=self.config) @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: bsz, q_len, _ = hidden_states.size() query_states = self.q_proj(hidden_states) key_states = self.k_proj(hidden_states) value_states = self.v_proj(hidden_states) query_states = query_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) key_states = key_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) value_states = value_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) cos, sin = position_embeddings query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids) if past_key_values is not None: cache_kwargs = {'sin': sin, 'cos': cos, 'cache_position': cache_position} key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs) key_states = repeat_kv(key_states, self.num_key_value_groups) value_states = repeat_kv(value_states, self.num_key_value_groups) attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim) if attention_mask is not None: causal_mask = attention_mask[:, :, :, :key_states.shape[-2]] attn_weights = attn_weights + causal_mask attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query_states.dtype) attn_weights = nn.functional.dropout(attn_weights, p=self.attention_dropout, training=self.training) attn_output = torch.matmul(attn_weights, value_states) if attn_output.size() != (bsz, self.num_heads, q_len, self.head_dim): raise ValueError(f'`attn_output` should be of size {(bsz, self.num_heads, q_len, self.head_dim)}, but is {attn_output.size()}') attn_output = attn_output.transpose(1, 2).contiguous() attn_output = attn_output.reshape(bsz, q_len, self.hidden_size) attn_output = self.o_proj(attn_output) if not output_attentions: attn_weights = None return (attn_output, attn_weights)
class Qwen2MoeAttention(nn.Module): ''' Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer and "Generating Long Sequences with Sparse Transformers". ''' def __init__(self, config: Qwen2MoeConfig, layer_idx: Optional[int]=None): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: pass
4
1
45
8
36
3
4
0.14
1
8
3
2
2
16
2
12
98
18
73
38
60
10
53
28
50
5
1
1
8
4,776
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeDecoderLayer
from ...cache_utils import Cache, DynamicCache, StaticCache from typing import Optional, Union from ...modeling_layers import GenericForQuestionAnswering, GenericForSequenceClassification, GenericForTokenClassification, GradientCheckpointingLayer from .configuration_qwen2_moe import Qwen2MoeConfig import torch from ...utils.deprecation import deprecate_kwarg import torch.nn.functional as F class Qwen2MoeDecoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2MoeConfig, layer_idx: int): super().__init__() self.hidden_size = config.hidden_size self.self_attn = QWEN2MOE_ATTENTION_CLASSES[config._attn_implementation](config, layer_idx) if layer_idx not in config.mlp_only_layers and (config.num_experts > 0 and (layer_idx + 1) % config.decoder_sparse_step == 0): self.mlp = Qwen2MoeSparseMoeBlock(config) else: self.mlp = Qwen2MoeMLP(config, intermediate_size=config.intermediate_size) self.input_layernorm = Qwen2MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.post_attention_layernorm = Qwen2MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps) @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: Optional[bool]=False, output_router_logits: Optional[bool]=False, use_cache: Optional[bool]=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs) -> tuple[torch.FloatTensor, Optional[tuple[torch.FloatTensor, torch.FloatTensor]]]: """ Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`, *optional*): attention mask of size `(batch, sequence_length)` where padding elements are indicated by 0. 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_router_logits (`bool`, *optional*): Whether or not to return the logits of all the routers. They are useful for computing the router loss, and should not be returned during inference. 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`). past_key_values (`Cache`, *optional*): cached past key and value projection states cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*): Indices depicting the position of the input sequence tokens in the sequence. position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*): Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`, with `head_dim` being the embedding dimension of each attention head. kwargs (`dict`, *optional*): Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code into the model """ residual = hidden_states hidden_states = self.input_layernorm(hidden_states) hidden_states, self_attn_weights = self.self_attn(hidden_states=hidden_states, attention_mask=attention_mask, position_ids=position_ids, past_key_values=past_key_values, output_attentions=output_attentions, use_cache=use_cache, cache_position=cache_position, position_embeddings=position_embeddings) hidden_states = residual + hidden_states residual = hidden_states hidden_states = self.post_attention_layernorm(hidden_states) hidden_states = self.mlp(hidden_states) if isinstance(hidden_states, tuple): hidden_states, router_logits = hidden_states else: router_logits = None hidden_states = residual + hidden_states outputs = (hidden_states,) if output_attentions: outputs += (self_attn_weights,) if output_router_logits: outputs += (router_logits,) return outputs
class Qwen2MoeDecoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2MoeConfig, layer_idx: int): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: Optional[bool]=False, output_router_logits: Optional[bool]=False, use_cache: Optional[bool]=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs) -> tuple[torch.FloatTensor, Optional[tuple[torch.FloatTensor, torch.FloatTensor]]]: ''' Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`, *optional*): attention mask of size `(batch, sequence_length)` where padding elements are indicated by 0. 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_router_logits (`bool`, *optional*): Whether or not to return the logits of all the routers. They are useful for computing the router loss, and should not be returned during inference. 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`). past_key_values (`Cache`, *optional*): cached past key and value projection states cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*): Indices depicting the position of the input sequence tokens in the sequence. position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*): Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`, with `head_dim` being the embedding dimension of each attention head. kwargs (`dict`, *optional*): Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code into the model ''' pass
4
1
47
7
27
14
4
0.49
1
9
4
0
2
5
2
12
96
15
55
24
40
27
30
12
27
5
1
1
7
4,777
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeFlashAttention2
from typing import Optional, Union from ...modeling_flash_attention_utils import flash_attn_supports_top_left_mask, is_flash_attn_available from ...cache_utils import Cache, DynamicCache, StaticCache import torch from ...utils.deprecation import deprecate_kwarg import torch.nn.functional as F class Qwen2MoeFlashAttention2(Qwen2MoeAttention): """ Qwen2Moe flash attention module, following Qwen2Moe attention module. This module inherits from `Qwen2MoeAttention` as the weights of the module stays untouched. The only required change would be on the forward pass where it needs to correctly call the public API of flash attention and deal with padding tokens in case the input contains any of them. Additionally, for sliding window attention, we apply SWA only to the bottom config.max_window_layers layers. """ def __init__(self, *args, **kwargs): super().__init__(*args, **kwargs) self._flash_attn_uses_top_left_mask = flash_attn_supports_top_left_mask() @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None): bsz, q_len, _ = hidden_states.size() query_states = self.q_proj(hidden_states) key_states = self.k_proj(hidden_states) value_states = self.v_proj(hidden_states) query_states = query_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) key_states = key_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) value_states = value_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) cos, sin = position_embeddings query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin) if past_key_values is not None: cache_kwargs = {'sin': sin, 'cos': cos, 'cache_position': cache_position} key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs) key_states = repeat_kv(key_states, self.num_key_value_groups) value_states = repeat_kv(value_states, self.num_key_value_groups) dropout_rate = 0.0 if not self.training else self.attention_dropout input_dtype = query_states.dtype device_type = query_states.device.type if query_states.device.type != 'mps' else 'cpu' if input_dtype == torch.float32: if torch.is_autocast_enabled(): target_dtype = torch.get_autocast_dtype(device_type) if hasattr(torch, 'get_autocast_dtype') else torch.get_autocast_gpu_dtype() elif hasattr(self.config, '_pre_quantization_dtype'): target_dtype = self.config._pre_quantization_dtype else: target_dtype = self.q_proj.weight.dtype logger.warning_once(f'The input hidden states seems to be silently casted in float32, this might be related to the fact you have upcasted embedding or layer norm layers in float32. We will cast back the input in {target_dtype}.') query_states = query_states.to(target_dtype) key_states = key_states.to(target_dtype) value_states = value_states.to(target_dtype) query_states = query_states.transpose(1, 2) key_states = key_states.transpose(1, 2) value_states = value_states.transpose(1, 2) if self.config.use_sliding_window and getattr(self.config, 'sliding_window', None) is not None and (self.layer_idx >= self.config.max_window_layers): sliding_window = self.config.sliding_window else: sliding_window = None attn_output = _flash_attention_forward(query_states, key_states, value_states, attention_mask, q_len, position_ids=position_ids, dropout=dropout_rate, sliding_window=sliding_window, is_causal=self.is_causal, use_top_left_mask=self._flash_attn_uses_top_left_mask) attn_output = attn_output.reshape(bsz, q_len, self.hidden_size).contiguous() attn_output = self.o_proj(attn_output) if not output_attentions: attn_weights = None return (attn_output, attn_weights)
class Qwen2MoeFlashAttention2(Qwen2MoeAttention): ''' Qwen2Moe flash attention module, following Qwen2Moe attention module. This module inherits from `Qwen2MoeAttention` as the weights of the module stays untouched. The only required change would be on the forward pass where it needs to correctly call the public API of flash attention and deal with padding tokens in case the input contains any of them. Additionally, for sliding window attention, we apply SWA only to the bottom config.max_window_layers layers. ''' def __init__(self, *args, **kwargs): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None): pass
4
1
49
8
37
6
5
0.24
1
4
1
0
2
2
2
14
107
17
74
27
61
18
42
16
39
8
2
2
9
4,778
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeForCausalLM
from typing import Optional, Union from ...utils import auto_docstring, can_return_tuple, is_torch_flex_attn_available, logging import torch import torch.nn.functional as F from ...modeling_outputs import MoeCausalLMOutputWithPast, MoeModelOutputWithPast from torch import nn from ...generation import GenerationMixin from ...cache_utils import Cache, DynamicCache, StaticCache class Qwen2MoeForCausalLM(Qwen2MoePreTrainedModel, GenerationMixin): _tied_weights_keys = ['lm_head.weight'] _tp_plan = {'lm_head': 'colwise_rep'} _pp_plan = {'lm_head': (['hidden_states'], ['logits'])} def __init__(self, config): super().__init__(config) self.model = Qwen2MoeModel(config) self.vocab_size = config.vocab_size self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False) self.router_aux_loss_coef = config.router_aux_loss_coef self.num_experts = config.num_experts self.num_experts_per_tok = config.num_experts_per_tok self.post_init() @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, output_router_logits: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None, logits_to_keep: Union[int, torch.Tensor]=0, **kwargs) -> MoeCausalLMOutputWithPast: """ labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. Example: ```python >>> from transformers import AutoTokenizer, Qwen2MoeForCausalLM >>> model = Qwen2MoeForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS) >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER) >>> prompt = "Hey, are you conscious? Can you talk to me?" >>> inputs = tokenizer(prompt, return_tensors="pt") >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "Hey, are you conscious? Can you talk to me?\\nI'm not conscious, but I can talk to you." ```""" output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions output_router_logits = output_router_logits if output_router_logits is not None else self.config.output_router_logits output_hidden_states = output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states outputs: MoeModelOutputWithPast = self.model(input_ids=input_ids, attention_mask=attention_mask, position_ids=position_ids, past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache, output_attentions=output_attentions, output_hidden_states=output_hidden_states, output_router_logits=output_router_logits, cache_position=cache_position) hidden_states = outputs.last_hidden_state slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep logits = self.lm_head(hidden_states[:, slice_indices, :]) loss = None if labels is not None: loss = self.loss_function(logits, labels, self.vocab_size, **kwargs) aux_loss = None if output_router_logits: aux_loss = load_balancing_loss_func(outputs.router_logits, self.num_experts, self.num_experts_per_tok, attention_mask) if labels is not None: loss += self.router_aux_loss_coef * aux_loss.to(loss.device) return MoeCausalLMOutputWithPast(loss=loss, aux_loss=aux_loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions, router_logits=outputs.router_logits)
class Qwen2MoeForCausalLM(Qwen2MoePreTrainedModel, GenerationMixin): def __init__(self, config): pass @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, output_router_logits: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None, logits_to_keep: Union[int, torch.Tensor]=0, **kwargs) -> MoeCausalLMOutputWithPast: ''' labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. Example: ```python >>> from transformers import AutoTokenizer, Qwen2MoeForCausalLM >>> model = Qwen2MoeForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS) >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER) >>> prompt = "Hey, are you conscious? Can you talk to me?" >>> inputs = tokenizer(prompt, return_tensors="pt") >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you." ```''' pass
5
1
17
2
11
4
3
0.31
2
7
2
0
8
6
8
9
146
23
95
41
67
29
47
24
38
13
2
2
20
4,779
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeForQuestionAnswering
from ...modeling_layers import GenericForQuestionAnswering, GenericForSequenceClassification, GenericForTokenClassification, GradientCheckpointingLayer class Qwen2MoeForQuestionAnswering(GenericForQuestionAnswering, Qwen2MoePreTrainedModel): ...
class Qwen2MoeForQuestionAnswering(GenericForQuestionAnswering, Qwen2MoePreTrainedModel): pass
1
0
18
2
13
3
2
0.22
1
5
3
0
4
2
4
5
77
11
55
28
36
12
26
14
21
5
2
1
8
4,780
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeForSequenceClassification
from ...modeling_layers import GenericForQuestionAnswering, GenericForSequenceClassification, GenericForTokenClassification, GradientCheckpointingLayer class Qwen2MoeForSequenceClassification(GenericForSequenceClassification, Qwen2MoePreTrainedModel): ...
class Qwen2MoeForSequenceClassification(GenericForSequenceClassification, Qwen2MoePreTrainedModel): pass
1
0
21
2
17
2
3
0.11
1
7
3
0
4
3
4
5
90
11
71
31
53
8
36
18
31
9
2
1
12
4,781
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeForTokenClassification
from ...modeling_layers import GenericForQuestionAnswering, GenericForSequenceClassification, GenericForTokenClassification, GradientCheckpointingLayer class Qwen2MoeForTokenClassification(GenericForTokenClassification, Qwen2MoePreTrainedModel): ...
class Qwen2MoeForTokenClassification(GenericForTokenClassification, Qwen2MoePreTrainedModel): pass
1
0
17
1
14
2
3
0.11
1
5
2
0
4
4
4
5
79
8
64
28
41
7
29
15
24
5
2
1
10
4,782
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeMLP
from ...activations import ACT2FN from torch import nn class Qwen2MoeMLP(nn.Module): def __init__(self, config, intermediate_size=None): super().__init__() self.config = config self.hidden_size = config.hidden_size self.intermediate_size = intermediate_size self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False) self.act_fn = ACT2FN[config.hidden_act] def forward(self, x): return self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
class Qwen2MoeMLP(nn.Module): def __init__(self, config, intermediate_size=None): pass def forward(self, x): pass
3
0
6
0
6
0
1
0
1
1
0
0
2
7
2
12
13
1
12
10
9
0
12
10
9
1
1
0
2
4,783
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeModel
from .configuration_qwen2_moe import Qwen2MoeConfig from typing import Optional, Union import torch from ...modeling_attn_mask_utils import AttentionMaskConverter from torch import nn from ...utils import auto_docstring, can_return_tuple, is_torch_flex_attn_available, logging import torch.nn.functional as F from ...cache_utils import Cache, DynamicCache, StaticCache from ...modeling_outputs import MoeCausalLMOutputWithPast, MoeModelOutputWithPast @auto_docstring class Qwen2MoeModel(Qwen2MoePreTrainedModel): def __init__(self, config: Qwen2MoeConfig): super().__init__(config) self.padding_idx = config.pad_token_id self.vocab_size = config.vocab_size self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx) self.layers = nn.ModuleList([Qwen2MoeDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]) self._attn_implementation = config._attn_implementation self.norm = Qwen2MoeRMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.rotary_emb = Qwen2MoeRotaryEmbedding(config=config) self.gradient_checkpointing = False self.post_init() @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, output_router_logits: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None) -> MoeModelOutputWithPast: output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions output_router_logits = output_router_logits if output_router_logits is not None else self.config.output_router_logits output_hidden_states = output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states use_cache = use_cache if use_cache is not None else self.config.use_cache if (input_ids is None) ^ (inputs_embeds is not None): raise ValueError('You must specify exactly one of input_ids or inputs_embeds') 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 if use_cache and past_key_values is None: past_key_values = DynamicCache(config=self.config) if inputs_embeds is None: inputs_embeds = self.embed_tokens(input_ids) if cache_position is None: past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0 cache_position = torch.arange(past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device) if position_ids is None: position_ids = cache_position.unsqueeze(0) causal_mask = self._update_causal_mask(attention_mask, inputs_embeds, cache_position, past_key_values, output_attentions) hidden_states = inputs_embeds position_embeddings = self.rotary_emb(hidden_states, position_ids) all_hidden_states = () if output_hidden_states else None all_self_attns = () if output_attentions else None all_router_logits = () if output_router_logits else None for decoder_layer in self.layers: if output_hidden_states: all_hidden_states += (hidden_states,) layer_outputs = decoder_layer(hidden_states, attention_mask=causal_mask, position_ids=position_ids, past_key_values=past_key_values, output_attentions=output_attentions, output_router_logits=output_router_logits, use_cache=use_cache, cache_position=cache_position, position_embeddings=position_embeddings) hidden_states = layer_outputs[0] if output_attentions: all_self_attns += (layer_outputs[1],) if output_router_logits and layer_outputs[-1] is not None: all_router_logits += (layer_outputs[-1],) hidden_states = self.norm(hidden_states) if output_hidden_states: all_hidden_states += (hidden_states,) return MoeModelOutputWithPast(last_hidden_state=hidden_states, past_key_values=past_key_values, hidden_states=all_hidden_states, attentions=all_self_attns, router_logits=all_router_logits) def _update_causal_mask(self, attention_mask: Union[torch.Tensor, 'BlockMask'], input_tensor: torch.Tensor, cache_position: torch.Tensor, past_key_values: Cache, output_attentions: bool=False): if self.config._attn_implementation == 'flash_attention_2': if attention_mask is not None and past_key_values is not None: is_padding_right = attention_mask[:, -1].sum().item() != input_tensor.size()[0] if is_padding_right: raise ValueError("You are attempting to perform batched generation with padding_side='right' this may lead to unexpected behaviour for Flash Attention version of Qwen2Moe. Make sure to call `tokenizer.padding_side = 'left'` before tokenizing the input. ") if attention_mask is not None and 0.0 in attention_mask: return attention_mask return None if self.config._attn_implementation == 'flex_attention': if isinstance(attention_mask, torch.Tensor): attention_mask = make_flex_block_causal_mask(attention_mask) return attention_mask past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0 using_static_cache = isinstance(past_key_values, StaticCache) if self.config._attn_implementation == 'sdpa' and (not using_static_cache) and (not output_attentions): if AttentionMaskConverter._ignore_causal_mask_sdpa(attention_mask, inputs_embeds=input_tensor, past_key_values_length=past_seen_tokens, sliding_window=self.config.sliding_window, is_training=self.training): return None dtype = input_tensor.dtype min_dtype = torch.finfo(dtype).min sequence_length = input_tensor.shape[1] if using_static_cache: target_length = past_key_values.get_max_cache_shape() else: target_length = attention_mask.shape[-1] if isinstance(attention_mask, torch.Tensor) else past_seen_tokens + sequence_length + 1 causal_mask = self._prepare_4d_causal_attention_mask_with_cache_position(attention_mask, sequence_length=sequence_length, target_length=target_length, dtype=dtype, cache_position=cache_position, batch_size=input_tensor.shape[0], config=self.config, past_key_values=past_key_values) if self.config._attn_implementation == 'sdpa' and attention_mask is not None and (attention_mask.device.type in ['cuda', 'xpu', 'npu']) and (not output_attentions): causal_mask = AttentionMaskConverter._unmask_unattended(causal_mask, min_dtype) return causal_mask @staticmethod def _prepare_4d_causal_attention_mask_with_cache_position(attention_mask: torch.Tensor, sequence_length: int, target_length: int, dtype: torch.dtype, cache_position: torch.Tensor, batch_size: int, config: Qwen2MoeConfig, past_key_values: Cache): """ Creates a causal 4D mask of shape `(batch_size, 1, query_length, key_value_length)` from a 2D mask of shape `(batch_size, key_value_length)`, or if the input `attention_mask` is already 4D, do nothing. Args: attention_mask (`torch.Tensor`): A 2D attention mask of shape `(batch_size, key_value_length)` or a 4D attention mask of shape `(batch_size, 1, query_length, key_value_length)`. sequence_length (`int`): The sequence length being processed. target_length (`int`): The target length: when generating with static cache, the mask should be as long as the static cache, to account for the 0 padding, the part of the cache that is not filled yet. dtype (`torch.dtype`): The dtype to use for the 4D attention mask. cache_position (`torch.Tensor`): Indices depicting the position of the input sequence tokens in the sequence. batch_size (`torch.Tensor`): Batch size. config (`Qwen2MoeConfig`): The model's configuration class past_key_values (`Cache`): The cache class that is being used currently to generate """ if attention_mask is not None and attention_mask.dim() == 4: causal_mask = attention_mask else: min_dtype = torch.finfo(dtype).min causal_mask = torch.full((sequence_length, target_length), fill_value=min_dtype, dtype=dtype, device=cache_position.device) diagonal_attend_mask = torch.arange(target_length, device=cache_position.device) > cache_position.reshape(-1, 1) text_config = config.get_text_config() if getattr(text_config, 'use_sliding_window', True) and text_config.sliding_window is not None: is_static_sliding_cache = isinstance(past_key_values, StaticCache) and all(past_key_values.is_sliding) if not is_static_sliding_cache or sequence_length > target_length: sliding_attend_mask = torch.arange(target_length, device=cache_position.device) <= cache_position.reshape(-1, 1) - text_config.sliding_window diagonal_attend_mask.bitwise_or_(sliding_attend_mask) causal_mask *= diagonal_attend_mask causal_mask = causal_mask[None, None, :, :].expand(batch_size, 1, -1, -1) if attention_mask is not None: causal_mask = causal_mask.clone() if attention_mask.shape[-1] > target_length: attention_mask = attention_mask[:, :target_length] mask_length = attention_mask.shape[-1] padding_mask = causal_mask[:, :, :, :mask_length] + attention_mask[:, None, None, :].to(causal_mask.device) padding_mask = padding_mask == 0 causal_mask[:, :, :, :mask_length] = causal_mask[:, :, :, :mask_length].masked_fill(padding_mask, min_dtype) return causal_mask
@auto_docstring class Qwen2MoeModel(Qwen2MoePreTrainedModel): def __init__(self, config: Qwen2MoeConfig): pass @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, output_router_logits: Optional[bool]=None, cache_position: Optional[torch.LongTensor]=None) -> MoeModelOutputWithPast: pass def _update_causal_mask(self, attention_mask: Union[torch.Tensor, 'BlockMask'], input_tensor: torch.Tensor, cache_position: torch.Tensor, past_key_values: Cache, output_attentions: bool=False): pass @staticmethod def _prepare_4d_causal_attention_mask_with_cache_position(attention_mask: torch.Tensor, sequence_length: int, target_length: int, dtype: torch.dtype, cache_position: torch.Tensor, batch_size: int, config: Qwen2MoeConfig, past_key_values: Cache): ''' Creates a causal 4D mask of shape `(batch_size, 1, query_length, key_value_length)` from a 2D mask of shape `(batch_size, key_value_length)`, or if the input `attention_mask` is already 4D, do nothing. Args: attention_mask (`torch.Tensor`): A 2D attention mask of shape `(batch_size, key_value_length)` or a 4D attention mask of shape `(batch_size, 1, query_length, key_value_length)`. sequence_length (`int`): The sequence length being processed. target_length (`int`): The target length: when generating with static cache, the mask should be as long as the static cache, to account for the 0 padding, the part of the cache that is not filled yet. dtype (`torch.dtype`): The dtype to use for the 4D attention mask. cache_position (`torch.Tensor`): Indices depicting the position of the input sequence tokens in the sequence. batch_size (`torch.Tensor`): Batch size. config (`Qwen2MoeConfig`): The model's configuration class past_key_values (`Cache`): The cache class that is being used currently to generate ''' pass
9
1
51
5
39
7
8
0.21
1
17
10
0
5
9
6
7
323
36
239
75
200
50
117
42
110
29
2
3
49
4,784
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoePreTrainedModel
from .configuration_qwen2_moe import Qwen2MoeConfig from ...modeling_utils import PreTrainedModel from ...utils import auto_docstring, can_return_tuple, is_torch_flex_attn_available, logging from torch import nn @auto_docstring class Qwen2MoePreTrainedModel(PreTrainedModel): config: Qwen2MoeConfig base_model_prefix = 'model' supports_gradient_checkpointing = True _no_split_modules = ['Qwen2MoeDecoderLayer'] _skip_keys_device_placement = 'past_key_values' _supports_flash_attn = True _supports_sdpa = True def _init_weights(self, module): std = self.config.initializer_range if isinstance(module, nn.Linear): module.weight.data.normal_(mean=0.0, std=std) if module.bias is not None: module.bias.data.zero_() elif isinstance(module, nn.Embedding): module.weight.data.normal_(mean=0.0, std=std) if module.padding_idx is not None: module.weight.data[module.padding_idx].zero_() elif isinstance(module, Qwen2MoeRMSNorm): module.weight.data.fill_(1.0)
@auto_docstring class Qwen2MoePreTrainedModel(PreTrainedModel): def _init_weights(self, module): pass
3
0
10
0
10
0
5
0
1
0
0
5
1
0
1
1
20
1
19
11
17
0
18
11
16
5
1
2
5
4,785
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeRMSNorm
import torch from torch import nn import torch.nn.functional as F class Qwen2MoeRMSNorm(nn.Module): def __init__(self, hidden_size, eps=1e-06): """ Qwen2MoeRMSNorm is equivalent to T5LayerNorm """ super().__init__() self.weight = nn.Parameter(torch.ones(hidden_size)) self.variance_epsilon = eps def forward(self, hidden_states): input_dtype = hidden_states.dtype hidden_states = hidden_states.to(torch.float32) variance = hidden_states.pow(2).mean(-1, keepdim=True) hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon) return self.weight * hidden_states.to(input_dtype) def extra_repr(self): return f'{tuple(self.weight.shape)}, eps={self.variance_epsilon}'
class Qwen2MoeRMSNorm(nn.Module): def __init__(self, hidden_size, eps=1e-06): ''' Qwen2MoeRMSNorm is equivalent to T5LayerNorm ''' pass def forward(self, hidden_states): pass def extra_repr(self): pass
4
1
5
0
4
1
1
0.23
1
2
0
0
3
2
3
13
18
2
13
8
9
3
13
8
9
1
1
0
3
4,786
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeRotaryEmbedding
from torch import nn import torch.nn.functional as F from .configuration_qwen2_moe import Qwen2MoeConfig import torch from ...modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update class Qwen2MoeRotaryEmbedding(nn.Module): inv_freq: torch.Tensor def __init__(self, config: Qwen2MoeConfig, device=None): super().__init__() if hasattr(config, 'rope_scaling') and isinstance(config.rope_scaling, dict): self.rope_type = config.rope_scaling.get('rope_type', config.rope_scaling.get('type')) else: self.rope_type = 'default' self.max_seq_len_cached = config.max_position_embeddings self.original_max_seq_len = config.max_position_embeddings self.config = config self.rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type] inv_freq, self.attention_scaling = self.rope_init_fn(self.config, device) self.register_buffer('inv_freq', inv_freq, persistent=False) self.original_inv_freq = self.inv_freq @torch.no_grad() @dynamic_rope_update def forward(self, x, position_ids): inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1).to(x.device) position_ids_expanded = position_ids[:, None, :].float() device_type = x.device.type if isinstance(x.device.type, str) and x.device.type != 'mps' else 'cpu' with torch.autocast(device_type=device_type, enabled=False): freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2) emb = torch.cat((freqs, freqs), dim=-1) cos = emb.cos() * self.attention_scaling sin = emb.sin() * self.attention_scaling return (cos.to(dtype=x.dtype), sin.to(dtype=x.dtype))
class Qwen2MoeRotaryEmbedding(nn.Module): def __init__(self, config: Qwen2MoeConfig, device=None): pass @torch.no_grad() @dynamic_rope_update def forward(self, x, position_ids): pass
5
0
18
2
13
5
3
0.35
1
4
1
0
3
7
3
13
59
8
40
21
35
14
38
20
34
3
1
1
8
4,787
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeSdpaAttention
from ...utils.deprecation import deprecate_kwarg from ...cache_utils import Cache, DynamicCache, StaticCache import torch from typing import Optional, Union import torch.nn.functional as F class Qwen2MoeSdpaAttention(Qwen2MoeAttention): """ Qwen2Moe attention module using torch.nn.functional.scaled_dot_product_attention. This module inherits from `Qwen2MoeAttention` as the weights of the module stays untouched. The only changes are on the forward pass to adapt to SDPA API. """ @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: if output_attentions: logger.warning_once('Qwen2MoeModel is using Qwen2MoeSdpaAttention, but `torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to the manual attention implementation, but specifying the manual implementation will be required from Transformers version v5.0.0 onwards. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.') return super().forward(hidden_states=hidden_states, attention_mask=attention_mask, position_ids=position_ids, past_key_values=past_key_values, output_attentions=output_attentions, use_cache=use_cache, cache_position=cache_position, position_embeddings=position_embeddings) bsz, q_len, _ = hidden_states.size() query_states = self.q_proj(hidden_states) key_states = self.k_proj(hidden_states) value_states = self.v_proj(hidden_states) query_states = query_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) key_states = key_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) value_states = value_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) cos, sin = position_embeddings query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin) if past_key_values is not None: cache_kwargs = {'sin': sin, 'cos': cos, 'cache_position': cache_position} key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs) key_states = repeat_kv(key_states, self.num_key_value_groups) value_states = repeat_kv(value_states, self.num_key_value_groups) causal_mask = attention_mask if attention_mask is not None: causal_mask = attention_mask[:, :, :, :key_states.shape[-2]] if query_states.device.type == 'cuda' and attention_mask is not None: query_states = query_states.contiguous() key_states = key_states.contiguous() value_states = value_states.contiguous() is_causal = bool(causal_mask is None and q_len > 1) attn_output = torch.nn.functional.scaled_dot_product_attention(query_states, key_states, value_states, attn_mask=causal_mask, dropout_p=self.attention_dropout if self.training else 0.0, is_causal=is_causal) attn_output = attn_output.transpose(1, 2).contiguous() attn_output = attn_output.view(bsz, q_len, self.hidden_size) attn_output = self.o_proj(attn_output) return (attn_output, None)
class Qwen2MoeSdpaAttention(Qwen2MoeAttention): ''' Qwen2Moe attention module using torch.nn.functional.scaled_dot_product_attention. This module inherits from `Qwen2MoeAttention` as the weights of the module stays untouched. The only changes are on the forward pass to adapt to SDPA API. ''' @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: pass
3
1
79
13
60
9
7
0.25
1
4
1
0
1
0
1
13
87
14
61
21
49
15
32
11
30
7
2
1
7
4,788
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py
transformers.models.qwen2_moe.modeling_qwen2_moe.Qwen2MoeSparseMoeBlock
import torch.nn.functional as F from torch import nn import torch class Qwen2MoeSparseMoeBlock(nn.Module): def __init__(self, config): super().__init__() self.num_experts = config.num_experts self.top_k = config.num_experts_per_tok self.norm_topk_prob = config.norm_topk_prob self.gate = nn.Linear(config.hidden_size, config.num_experts, bias=False) self.experts = nn.ModuleList([Qwen2MoeMLP(config, intermediate_size=config.moe_intermediate_size) for _ in range(self.num_experts)]) self.shared_expert = Qwen2MoeMLP(config, intermediate_size=config.shared_expert_intermediate_size) self.shared_expert_gate = torch.nn.Linear(config.hidden_size, 1, bias=False) def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: """ """ batch_size, sequence_length, hidden_dim = hidden_states.shape hidden_states = hidden_states.view(-1, hidden_dim) router_logits = self.gate(hidden_states) routing_weights = F.softmax(router_logits, dim=1, dtype=torch.float) routing_weights, selected_experts = torch.topk(routing_weights, self.top_k, dim=-1) if self.norm_topk_prob: routing_weights /= routing_weights.sum(dim=-1, keepdim=True) routing_weights = routing_weights.to(hidden_states.dtype) final_hidden_states = torch.zeros((batch_size * sequence_length, hidden_dim), dtype=hidden_states.dtype, device=hidden_states.device) expert_mask = torch.nn.functional.one_hot(selected_experts, num_classes=self.num_experts).permute(2, 1, 0) expert_hit = torch.greater(expert_mask.sum(dim=(-1, -2)), 0).nonzero() for expert_idx in expert_hit: expert_layer = self.experts[expert_idx] idx, top_x = torch.where(expert_mask[expert_idx].squeeze(0)) current_state = hidden_states[None, top_x].reshape(-1, hidden_dim) current_hidden_states = expert_layer(current_state) * routing_weights[top_x, idx, None] final_hidden_states.index_add_(0, top_x, current_hidden_states.to(hidden_states.dtype)) shared_expert_output = self.shared_expert(hidden_states) shared_expert_output = F.sigmoid(self.shared_expert_gate(hidden_states)) * shared_expert_output final_hidden_states = final_hidden_states + shared_expert_output final_hidden_states = final_hidden_states.reshape(batch_size, sequence_length, hidden_dim) return (final_hidden_states, router_logits)
class Qwen2MoeSparseMoeBlock(nn.Module): def __init__(self, config): pass def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: ''' ''' pass
3
1
29
6
18
6
2
0.33
1
5
1
0
2
7
2
12
60
12
36
23
33
12
32
22
29
3
1
1
4
4,789
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/configuration_qwen2_vl.py
transformers.models.qwen2_vl.configuration_qwen2_vl.Qwen2VLConfig
from ...configuration_utils import PretrainedConfig, layer_type_validation class Qwen2VLConfig(PretrainedConfig): """ This is the configuration class to store the configuration of a [`Qwen2VLModel`]. It is used to instantiate a Qwen2-VL model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of Qwen2-VL-7B-Instruct [Qwen/Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct). Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: text_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Qwen2_5_VLTextConfig`): The config object or dictionary of the text backbone. vision_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Qwen2_5_VLVisionConfig`): The config object or dictionary of the vision backbone. image_token_id (`int`, *optional*, defaults to 151655): The image token index to encode the image prompt. video_token_id (`int`, *optional*, defaults to 151656): The video token index to encode the image prompt. ```python >>> from transformers import Qwen2_5_VLForConditionalGeneration, Qwen2_5_VLConfig >>> # Initializing a Qwen2_5_VL style configuration >>> configuration = Qwen2_5_VLConfig() >>> # Initializing a model from the Qwen2-VL-7B style configuration >>> model = Qwen2_5_VLForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = 'qwen2_vl' sub_configs = {'vision_config': Qwen2VLVisionConfig, 'text_config': Qwen2VLTextConfig} keys_to_ignore_at_inference = ['past_key_values'] def __init__(self, text_config=None, vision_config=None, image_token_id=151655, video_token_id=151656, **kwargs): if isinstance(vision_config, dict): self.vision_config = self.sub_configs['vision_config'](**vision_config) elif vision_config is None: self.vision_config = self.sub_configs['vision_config']() if isinstance(text_config, dict): self.text_config = self.sub_configs['text_config'](**text_config) elif text_config is None: self.text_config = self.sub_configs['text_config'](**kwargs) self.image_token_id = image_token_id self.video_token_id = video_token_id super().__init__(**kwargs)
class Qwen2VLConfig(PretrainedConfig): ''' This is the configuration class to store the configuration of a [`Qwen2VLModel`]. It is used to instantiate a Qwen2-VL model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of Qwen2-VL-7B-Instruct [Qwen/Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct). Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: text_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Qwen2_5_VLTextConfig`): The config object or dictionary of the text backbone. vision_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Qwen2_5_VLVisionConfig`): The config object or dictionary of the vision backbone. image_token_id (`int`, *optional*, defaults to 151655): The image token index to encode the image prompt. video_token_id (`int`, *optional*, defaults to 151656): The video token index to encode the image prompt. ```python >>> from transformers import Qwen2_5_VLForConditionalGeneration, Qwen2_5_VLConfig >>> # Initializing a Qwen2_5_VL style configuration >>> configuration = Qwen2_5_VLConfig() >>> # Initializing a model from the Qwen2-VL-7B style configuration >>> model = Qwen2_5_VLForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```''' def __init__(self, text_config=None, vision_config=None, image_token_id=151655, video_token_id=151656, **kwargs): pass
2
1
63
5
52
6
6
1.6
1
2
0
0
1
18
1
1
183
14
65
46
41
104
34
24
32
6
1
2
6
4,790
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/configuration_qwen2_vl.py
transformers.models.qwen2_vl.configuration_qwen2_vl.Qwen2VLVisionConfig
from ...configuration_utils import PretrainedConfig, layer_type_validation class Qwen2VLVisionConfig(PretrainedConfig): model_type = 'qwen2_vl' base_config_key = 'vision_config' def __init__(self, depth=32, embed_dim=1280, hidden_size=3584, hidden_act='quick_gelu', mlp_ratio=4, num_heads=16, in_channels=3, patch_size=14, spatial_merge_size=2, temporal_patch_size=2, initializer_range=0.02, **kwargs): super().__init__(**kwargs) self.depth = depth self.embed_dim = embed_dim self.hidden_size = hidden_size self.hidden_act = hidden_act self.mlp_ratio = mlp_ratio self.num_heads = num_heads self.in_channels = in_channels self.patch_size = patch_size self.spatial_merge_size = spatial_merge_size self.temporal_patch_size = temporal_patch_size self.initializer_range = initializer_range
class Qwen2VLVisionConfig(PretrainedConfig): def __init__(self, depth=32, embed_dim=1280, hidden_size=3584, hidden_act='quick_gelu', mlp_ratio=4, num_heads=16, in_channels=3, patch_size=14, spatial_merge_size=2, temporal_patch_size=2, initializer_range=0.02, **kwargs): pass
2
0
26
1
25
0
1
0
1
1
0
0
1
10
1
1
30
2
28
27
13
0
15
14
13
1
1
0
1
4,791
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/image_processing_qwen2_vl.py
transformers.models.qwen2_vl.image_processing_qwen2_vl.Qwen2VLImageProcessor
from ...image_utils import OPENAI_CLIP_MEAN, OPENAI_CLIP_STD, ChannelDimension, ImageInput, PILImageResampling, get_image_size, infer_channel_dimension_format, is_scaled_image, make_flat_list_of_images, to_numpy_array, valid_images, validate_preprocess_arguments from ...image_processing_utils import BaseImageProcessor, BatchFeature from typing import Optional, Union from ...utils import TensorType, logging from ...video_utils import VideoInput, make_batched_videos import numpy as np from ...image_transforms import convert_to_rgb, resize, to_channel_dimension_format class Qwen2VLImageProcessor(BaseImageProcessor): """ Constructs a Qwen2-VL image processor that dynamically resizes images based on the original images. Args: do_resize (`bool`, *optional*, defaults to `True`): Whether to resize the image's (height, width) dimensions. size (`dict[str, int]`, *optional*, defaults to `{"shortest_edge": 56 * 56, "longest_edge": 28 * 28 * 1280}`): Size of the image after resizing. `shortest_edge` and `longest_edge` keys must be present. resample (`PILImageResampling`, *optional*, defaults to `Resampling.BICUBIC`): Resampling filter to use when resizing the image. do_rescale (`bool`, *optional*, defaults to `True`): Whether to rescale the image by the specified scale `rescale_factor`. rescale_factor (`int` or `float`, *optional*, defaults to `1/255`): Scale factor to use if rescaling the image. do_normalize (`bool`, *optional*, defaults to `True`): Whether to normalize the image. image_mean (`float` or `list[float]`, *optional*, defaults to `[0.48145466, 0.4578275, 0.40821073]`): Mean to use if normalizing the image. This is a float or list of floats for each channel in the image. image_std (`float` or `list[float]`, *optional*, defaults to `[0.26862954, 0.26130258, 0.27577711]`): Standard deviation to use if normalizing the image. This is a float or list of floats for each channel in the image. do_convert_rgb (`bool`, *optional*, defaults to `True`): Whether to convert the image to RGB. min_pixels (`int`, *optional*, defaults to `56 * 56`): The min pixels of the image to resize the image. max_pixels (`int`, *optional*, defaults to `28 * 28 * 1280`): The max pixels of the image to resize the image. patch_size (`int`, *optional*, defaults to 14): The spatial patch size of the vision encoder. temporal_patch_size (`int`, *optional*, defaults to 2): The temporal patch size of the vision encoder. merge_size (`int`, *optional*, defaults to 2): The merge size of the vision encoder to llm encoder. """ model_input_names = ['pixel_values', 'image_grid_thw', 'pixel_values_videos', 'video_grid_thw'] def __init__(self, do_resize: bool=True, size: Optional[dict[str, int]]=None, resample: PILImageResampling=PILImageResampling.BICUBIC, 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, do_convert_rgb: bool=True, min_pixels: Optional[int]=None, max_pixels: Optional[int]=None, patch_size: int=14, temporal_patch_size: int=2, merge_size: int=2, **kwargs) -> None: super().__init__(**kwargs) if size is not None and ('shortest_edge' not in size or 'longest_edge' not in size): raise ValueError("size must contain 'shortest_edge' and 'longest_edge' keys.") else: size = {'shortest_edge': 56 * 56, 'longest_edge': 28 * 28 * 1280} if min_pixels is not None: size['shortest_edge'] = min_pixels if max_pixels is not None: size['longest_edge'] = max_pixels self.min_pixels = size['shortest_edge'] self.max_pixels = size['longest_edge'] self.size = size self.do_resize = do_resize self.resample = resample self.do_rescale = do_rescale self.rescale_factor = rescale_factor self.do_normalize = do_normalize self.image_mean = image_mean if image_mean is not None else OPENAI_CLIP_MEAN self.image_std = image_std if image_std is not None else OPENAI_CLIP_STD self.patch_size = patch_size self.temporal_patch_size = temporal_patch_size self.merge_size = merge_size self.do_convert_rgb = do_convert_rgb def _preprocess(self, images: Union[ImageInput, VideoInput], do_resize: Optional[bool]=None, size: Optional[dict[str, int]]=None, resample: Optional[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, patch_size: Optional[int]=None, temporal_patch_size: Optional[int]=None, merge_size: Optional[int]=None, do_convert_rgb: Optional[bool]=None, data_format: Optional[ChannelDimension]=ChannelDimension.FIRST, input_data_format: Optional[Union[str, ChannelDimension]]=None): """ Preprocess an image or batch of images. Copy of the `preprocess` method from `CLIPImageProcessor`. Args: images (`ImageInput`): Image or batch of images to preprocess. Expects pixel values ranging from 0 to 255. If pixel values range from 0 to 1, set `do_rescale=False`. vision_info (`list[Dict]`, *optional*): Optional list of dictionaries containing additional information about vision inputs. do_resize (`bool`, *optional*, defaults to `self.do_resize`): Whether to resize the image. size (`dict[str, int]`, *optional*, defaults to `self.size`): Size of the image after resizing. `shortest_edge` and `longest_edge` keys must be present. resample (`PILImageResampling`, *optional*, defaults to `self.resample`): Resampling filter to use if resizing the image. This can be one of the `PILImageResampling` enums. do_rescale (`bool`, *optional*, defaults to `self.do_rescale`): Whether to rescale the image. rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`): Scale factor to use if rescaling the image. 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`): Mean to use if normalizing the image. Can be a float or a list of floats corresponding to the number of channels in the image. image_std (`float` or `list[float]`, *optional*, defaults to `self.image_std`): Standard deviation to use if normalizing the image. Can be a float or a list of floats corresponding to the number of channels in the image. patch_size (`int`, *optional*, defaults to `self.patch_size`): The spatial patch size of the vision encoder. temporal_patch_size (`int`, *optional*, defaults to `self.temporal_patch_size`): The temporal patch size of the vision encoder. merge_size (`int`, *optional*, defaults to `self.merge_size`): The merge size of the vision encoder to llm encoder. do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`): Whether to convert the image to RGB. data_format (`ChannelDimension`, *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. 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. - `"none"` or `ChannelDimension.NONE`: image in (height, width) format. """ images = make_flat_list_of_images(images) if do_convert_rgb: images = [convert_to_rgb(image) for image in images] images = [to_numpy_array(image) for image in images] if do_rescale and is_scaled_image(images[0]): 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: input_data_format = infer_channel_dimension_format(images[0]) height, width = get_image_size(images[0], channel_dim=input_data_format) resized_height, resized_width = (height, width) processed_images = [] for image in images: if do_resize: resized_height, resized_width = smart_resize(height, width, factor=patch_size * merge_size, min_pixels=size['shortest_edge'], max_pixels=size['longest_edge']) image = resize(image, size=(resized_height, resized_width), resample=resample, input_data_format=input_data_format) if do_rescale: image = self.rescale(image, scale=rescale_factor, input_data_format=input_data_format) if do_normalize: image = self.normalize(image=image, mean=image_mean, std=image_std, input_data_format=input_data_format) image = to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format) processed_images.append(image) patches = np.array(processed_images) if data_format == ChannelDimension.LAST: patches = patches.transpose(0, 3, 1, 2) if patches.shape[0] % temporal_patch_size != 0: repeats = np.repeat(patches[-1][np.newaxis], temporal_patch_size - patches.shape[0] % temporal_patch_size, axis=0) patches = np.concatenate([patches, repeats], axis=0) channel = patches.shape[1] grid_t = patches.shape[0] // temporal_patch_size grid_h, grid_w = (resized_height // patch_size, resized_width // patch_size) patches = patches.reshape(grid_t, temporal_patch_size, channel, grid_h // merge_size, merge_size, patch_size, grid_w // merge_size, merge_size, patch_size) patches = patches.transpose(0, 3, 6, 4, 7, 2, 1, 5, 8) flatten_patches = patches.reshape(grid_t * grid_h * grid_w, channel * temporal_patch_size * patch_size * patch_size) return (flatten_patches, (grid_t, grid_h, grid_w)) def preprocess(self, images: ImageInput, videos: Optional[VideoInput]=None, do_resize: Optional[bool]=None, size: Optional[dict[str, int]]=None, min_pixels: Optional[int]=None, max_pixels: Optional[int]=None, resample: Optional[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, patch_size: Optional[int]=None, temporal_patch_size: Optional[int]=None, merge_size: Optional[int]=None, do_convert_rgb: Optional[bool]=None, return_tensors: Optional[Union[str, TensorType]]=None, data_format: Optional[ChannelDimension]=ChannelDimension.FIRST, input_data_format: Optional[Union[str, ChannelDimension]]=None): """ 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`. videos (`VideoInput`): Video to preprocess. Expects a single or batch of videos with pixel values ranging from 0 to 255. If passing in videos 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`): Size of the image after resizing. Shortest edge of the image is resized to size["shortest_edge"], with the longest edge resized to keep the input aspect ratio. resample (`int`, *optional*, defaults to `self.resample`): Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. 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. 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 for normalization. Only has an effect if `do_normalize` is set to `True`. image_std (`float` or `list[float]`, *optional*, defaults to `self.image_std`): Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to `True`. min_pixels (`int`, *optional*, defaults to `self.min_pixels`): The min pixels of the image to resize the image. max_pixels (`int`, *optional*, defaults to `self.max_pixels`): The max pixels of the image to resize the image. patch_size (`int`, *optional*, defaults to `self.patch_size`): The spatial patch size of the vision encoder. temporal_patch_size (`int`, *optional*, defaults to `self.temporal_patch_size`): The temporal patch size of the vision encoder. merge_size (`int`, *optional*, defaults to `self.merge_size`): The merge size of the vision encoder to llm encoder. do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`): Whether to convert the image to RGB. return_tensors (`str` or `TensorType`, *optional*): The type of tensors to return. Can be one of: - Unset: Return a list of `np.ndarray`. - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`. - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.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. """ min_pixels = min_pixels if min_pixels is not None else self.min_pixels max_pixels = max_pixels if max_pixels is not None else self.max_pixels if size is not None: if 'shortest_edge' not in size or 'longest_edge' not in size: raise ValueError("size must contain 'shortest_edge' and 'longest_edge' keys.") min_pixels = size['shortest_edge'] elif min_pixels is not None and max_pixels is not None: size = {'shortest_edge': min_pixels, 'longest_edge': max_pixels} else: size = {**self.size} do_resize = do_resize if do_resize is not None else self.do_resize resample = resample if resample is not None else self.resample do_rescale = do_rescale if do_rescale is not None else self.do_rescale rescale_factor = rescale_factor if rescale_factor is not None else self.rescale_factor do_normalize = do_normalize if do_normalize is not None else self.do_normalize 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 patch_size = patch_size if patch_size is not None else self.patch_size temporal_patch_size = temporal_patch_size if temporal_patch_size is not None else self.temporal_patch_size merge_size = merge_size if merge_size is not None else self.merge_size do_convert_rgb = do_convert_rgb if do_convert_rgb is not None else self.do_convert_rgb if images is not None: images = self.fetch_images(images) images = make_flat_list_of_images(images) if images is not None and (not valid_images(images)): raise ValueError('Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, or torch.Tensor') validate_preprocess_arguments(rescale_factor=rescale_factor, do_normalize=do_normalize, image_mean=image_mean, image_std=image_std, do_resize=do_resize, size=size, resample=resample) data = {} if images is not None: pixel_values, vision_grid_thws = ([], []) for image in images: patches, image_grid_thw = self._preprocess(image, do_resize=do_resize, size=size, resample=resample, do_rescale=do_rescale, rescale_factor=rescale_factor, do_normalize=do_normalize, image_mean=image_mean, image_std=image_std, patch_size=patch_size, temporal_patch_size=temporal_patch_size, merge_size=merge_size, data_format=data_format, do_convert_rgb=do_convert_rgb, input_data_format=input_data_format) pixel_values.extend(patches) vision_grid_thws.append(image_grid_thw) pixel_values = np.array(pixel_values) vision_grid_thws = np.array(vision_grid_thws) data.update({'pixel_values': pixel_values, 'image_grid_thw': vision_grid_thws}) if videos is not None: logger.warning("`Qwen2VLImageProcessor` works only with image inputs and doesn't process videos anymore. This is a deprecated behavior and will be removed in v5.0. Your videos should be forwarded to `Qwen2VLVideoProcessor`. ") videos = make_batched_videos(videos) pixel_values_videos, vision_grid_thws_videos = ([], []) for images in videos: patches, video_grid_thw = self._preprocess(images, do_resize=do_resize, size=size, resample=resample, do_rescale=do_rescale, rescale_factor=rescale_factor, do_normalize=do_normalize, image_mean=image_mean, image_std=image_std, patch_size=patch_size, temporal_patch_size=temporal_patch_size, merge_size=merge_size, data_format=data_format, do_convert_rgb=do_convert_rgb, input_data_format=input_data_format) pixel_values_videos.extend(patches) vision_grid_thws_videos.append(video_grid_thw) data.update({'pixel_values_videos': np.array(pixel_values_videos), 'video_grid_thw': np.array(vision_grid_thws_videos)}) return BatchFeature(data=data, tensor_type=return_tensors) def get_number_of_image_patches(self, height: int, width: int, images_kwargs=None): """ A utility that returns number of image patches for a given image size. Args: height (`int`): Height of the input image. width (`int`): Width of the input image. images_kwargs (`dict`, *optional*) Any kwargs to override defaults of the image processor. Returns: `int`: Number of image patches per image. """ min_pixels = images_kwargs['min_pixels'] if 'min_pixels' in images_kwargs else self.size['shortest_edge'] max_pixels = images_kwargs['max_pixels'] if 'max_pixels' in images_kwargs else self.size['longest_edge'] patch_size = images_kwargs.get('patch_size', self.patch_size) merge_size = images_kwargs.get('merge_size', self.merge_size) factor = patch_size * merge_size resized_height, resized_width = smart_resize(height, width, factor, min_pixels=min_pixels, max_pixels=max_pixels) grid_h, grid_w = (resized_height // patch_size, resized_width // patch_size) return grid_h * grid_w
class Qwen2VLImageProcessor(BaseImageProcessor): ''' Constructs a Qwen2-VL image processor that dynamically resizes images based on the original images. Args: do_resize (`bool`, *optional*, defaults to `True`): Whether to resize the image's (height, width) dimensions. size (`dict[str, int]`, *optional*, defaults to `{"shortest_edge": 56 * 56, "longest_edge": 28 * 28 * 1280}`): Size of the image after resizing. `shortest_edge` and `longest_edge` keys must be present. resample (`PILImageResampling`, *optional*, defaults to `Resampling.BICUBIC`): Resampling filter to use when resizing the image. do_rescale (`bool`, *optional*, defaults to `True`): Whether to rescale the image by the specified scale `rescale_factor`. rescale_factor (`int` or `float`, *optional*, defaults to `1/255`): Scale factor to use if rescaling the image. do_normalize (`bool`, *optional*, defaults to `True`): Whether to normalize the image. image_mean (`float` or `list[float]`, *optional*, defaults to `[0.48145466, 0.4578275, 0.40821073]`): Mean to use if normalizing the image. This is a float or list of floats for each channel in the image. image_std (`float` or `list[float]`, *optional*, defaults to `[0.26862954, 0.26130258, 0.27577711]`): Standard deviation to use if normalizing the image. This is a float or list of floats for each channel in the image. do_convert_rgb (`bool`, *optional*, defaults to `True`): Whether to convert the image to RGB. min_pixels (`int`, *optional*, defaults to `56 * 56`): The min pixels of the image to resize the image. max_pixels (`int`, *optional*, defaults to `28 * 28 * 1280`): The max pixels of the image to resize the image. patch_size (`int`, *optional*, defaults to 14): The spatial patch size of the vision encoder. temporal_patch_size (`int`, *optional*, defaults to 2): The temporal patch size of the vision encoder. merge_size (`int`, *optional*, defaults to 2): The merge size of the vision encoder to llm encoder. ''' def __init__(self, do_resize: bool=True, size: Optional[dict[str, int]]=None, resample: PILImageResampling=PILImageResampling.BICUBIC, 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, do_convert_rgb: bool=True, min_pixels: Optional[int]=None, max_pixels: Optional[int]=None, patch_size: int=14, temporal_patch_size: int=2, merge_size: int=2, **kwargs) -> None: pass def _preprocess(self, images: Union[ImageInput, VideoInput], do_resize: Optional[bool]=None, size: Optional[dict[str, int]]=None, resample: Optional[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, patch_size: Optional[int]=None, temporal_patch_size: Optional[int]=None, merge_size: Optional[int]=None, do_convert_rgb: Optional[bool]=None, data_format: Optional[ChannelDimension]=ChannelDimension.FIRST, input_data_format: Optional[Union[str, ChannelDimension]]=None): ''' Preprocess an image or batch of images. Copy of the `preprocess` method from `CLIPImageProcessor`. Args: images (`ImageInput`): Image or batch of images to preprocess. Expects pixel values ranging from 0 to 255. If pixel values range from 0 to 1, set `do_rescale=False`. vision_info (`list[Dict]`, *optional*): Optional list of dictionaries containing additional information about vision inputs. do_resize (`bool`, *optional*, defaults to `self.do_resize`): Whether to resize the image. size (`dict[str, int]`, *optional*, defaults to `self.size`): Size of the image after resizing. `shortest_edge` and `longest_edge` keys must be present. resample (`PILImageResampling`, *optional*, defaults to `self.resample`): Resampling filter to use if resizing the image. This can be one of the `PILImageResampling` enums. do_rescale (`bool`, *optional*, defaults to `self.do_rescale`): Whether to rescale the image. rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`): Scale factor to use if rescaling the image. 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`): Mean to use if normalizing the image. Can be a float or a list of floats corresponding to the number of channels in the image. image_std (`float` or `list[float]`, *optional*, defaults to `self.image_std`): Standard deviation to use if normalizing the image. Can be a float or a list of floats corresponding to the number of channels in the image. patch_size (`int`, *optional*, defaults to `self.patch_size`): The spatial patch size of the vision encoder. temporal_patch_size (`int`, *optional*, defaults to `self.temporal_patch_size`): The temporal patch size of the vision encoder. merge_size (`int`, *optional*, defaults to `self.merge_size`): The merge size of the vision encoder to llm encoder. do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`): Whether to convert the image to RGB. data_format (`ChannelDimension`, *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. 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. - `"none"` or `ChannelDimension.NONE`: image in (height, width) format. ''' pass def preprocess(self, images: ImageInput, videos: Optional[VideoInput]=None, do_resize: Optional[bool]=None, size: Optional[dict[str, int]]=None, min_pixels: Optional[int]=None, max_pixels: Optional[int]=None, resample: Optional[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, patch_size: Optional[int]=None, temporal_patch_size: Optional[int]=None, merge_size: Optional[int]=None, do_convert_rgb: Optional[bool]=None, return_tensors: Optional[Union[str, TensorType]]=None, data_format: Optional[ChannelDimension]=ChannelDimension.FIRST, input_data_format: Optional[Union[str, ChannelDimension]]=None): ''' 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`. videos (`VideoInput`): Video to preprocess. Expects a single or batch of videos with pixel values ranging from 0 to 255. If passing in videos 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`): Size of the image after resizing. Shortest edge of the image is resized to size["shortest_edge"], with the longest edge resized to keep the input aspect ratio. resample (`int`, *optional*, defaults to `self.resample`): Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. 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. 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 for normalization. Only has an effect if `do_normalize` is set to `True`. image_std (`float` or `list[float]`, *optional*, defaults to `self.image_std`): Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to `True`. min_pixels (`int`, *optional*, defaults to `self.min_pixels`): The min pixels of the image to resize the image. max_pixels (`int`, *optional*, defaults to `self.max_pixels`): The max pixels of the image to resize the image. patch_size (`int`, *optional*, defaults to `self.patch_size`): The spatial patch size of the vision encoder. temporal_patch_size (`int`, *optional*, defaults to `self.temporal_patch_size`): The temporal patch size of the vision encoder. merge_size (`int`, *optional*, defaults to `self.merge_size`): The merge size of the vision encoder to llm encoder. do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`): Whether to convert the image to RGB. return_tensors (`str` or `TensorType`, *optional*): The type of tensors to return. Can be one of: - Unset: Return a list of `np.ndarray`. - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`. - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.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. ''' pass def get_number_of_image_patches(self, height: int, width: int, images_kwargs=None): ''' A utility that returns number of image patches for a given image size. Args: height (`int`): Height of the input image. width (`int`): Width of the input image. images_kwargs (`dict`, *optional*) Any kwargs to override defaults of the image processor. Returns: `int`: Number of image patches per image. ''' pass
5
4
98
6
64
28
10
0.59
1
8
2
0
3
14
3
23
330
22
194
78
145
114
89
33
85
17
3
2
30
4,792
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/image_processing_qwen2_vl_fast.py
transformers.models.qwen2_vl.image_processing_qwen2_vl_fast.Qwen2VLImageProcessorFast
from ...processing_utils import Unpack import torch from ...video_utils import VideoInput, make_batched_videos from .image_processing_qwen2_vl import smart_resize from ...image_utils import OPENAI_CLIP_MEAN, OPENAI_CLIP_STD, ChannelDimension, ImageInput, PILImageResampling, SizeDict from ...image_processing_utils_fast import BaseImageProcessorFast, DefaultFastImageProcessorKwargs, group_images_by_shape, reorder_images from ...utils import TensorType, auto_docstring, is_torchvision_v2_available, logging from ...image_processing_utils import BatchFeature from typing import Optional, Union @auto_docstring class Qwen2VLImageProcessorFast(BaseImageProcessorFast): do_resize = True resample = PILImageResampling.BICUBIC size = {'shortest_edge': 56 * 56, 'longest_edge': 28 * 28 * 1280} do_rescale = True do_normalize = True image_mean = OPENAI_CLIP_MEAN image_std = OPENAI_CLIP_STD do_convert_rgb = True patch_size = 14 temporal_patch_size = 2 merge_size = 2 min_pixels = None max_pixels = None valid_kwargs = Qwen2VLFastImageProcessorKwargs model_input_names = ['pixel_values', 'image_grid_thw', 'pixel_values_videos', 'video_grid_thw'] def __init__(self, **kwargs: Unpack[Qwen2VLFastImageProcessorKwargs]): size = kwargs.pop('size', None) min_pixels = kwargs.pop('min_pixels', None) max_pixels = kwargs.pop('max_pixels', None) size = self.size if size is None else size if min_pixels is not None: size['shortest_edge'] = min_pixels size.pop('min_pixels', None) if max_pixels is not None: size['longest_edge'] = max_pixels size.pop('max_pixels', None) if 'shortest_edge' not in size or 'longest_edge' not in size: raise ValueError("size must contain 'shortest_edge' and 'longest_edge' keys.") super().__init__(size=size, min_pixels=min_pixels, max_pixels=max_pixels, **kwargs) def _further_process_kwargs(self, size: Optional[SizeDict]=None, min_pixels: Optional[int]=None, max_pixels: Optional[int]=None, **kwargs) -> dict: """ Update kwargs that need further processing before being validated Can be overridden by subclasses to customize the processing of kwargs. """ if min_pixels is not None and max_pixels is not None: size = {'shortest_edge': min_pixels, 'longest_edge': max_pixels} elif size is not None: if 'shortest_edge' not in size or 'longest_edge' not in size: raise ValueError("size must contain 'shortest_edge' and 'longest_edge' keys.") min_pixels = size['shortest_edge'] max_pixels = size['longest_edge'] else: size = {**self.size} return super()._further_process_kwargs(size=size, min_pixels=min_pixels, max_pixels=max_pixels, **kwargs) @auto_docstring def preprocess(self, images: ImageInput, videos: Optional[VideoInput]=None, **kwargs: Unpack[Qwen2VLFastImageProcessorKwargs]) -> BatchFeature: return super().preprocess(images, videos, **kwargs) def _preprocess_image_like_inputs(self, images: ImageInput, videos: VideoInput, do_convert_rgb: bool, input_data_format: ChannelDimension, device: Optional[Union[str, 'torch.device']]=None, **kwargs: Unpack[DefaultFastImageProcessorKwargs]) -> BatchFeature: """ Preprocess image-like inputs. To be overridden by subclasses when image-like inputs other than images should be processed. It can be used for segmentation maps, depth maps, etc. """ batch_feature = BatchFeature() if images is not None: images = self._prepare_image_like_inputs(images=images, do_convert_rgb=do_convert_rgb, input_data_format=input_data_format, device=device) batch_feature = self._preprocess(images, **kwargs) if videos is not None: logger.warning("`Qwen2VLImageProcessorFast` works only with image inputs and doesn't process videos anymore. This is a deprecated behavior and will be removed in v5.0. Your videos should be forwarded to `Qwen2VLVideoProcessor`. ") videos = make_batched_videos(videos) videos = [torch.stack(self._prepare_image_like_inputs(video, do_convert_rgb, input_data_format, device)) for video in videos] video_outputs = self._preprocess(videos, **kwargs) batch_feature.update({'pixel_values_videos': video_outputs.pixel_values, 'video_grid_thw': video_outputs.image_grid_thw}) return batch_feature def _preprocess(self, images: list['torch.Tensor'], do_resize: bool, size: SizeDict, interpolation: Optional['F.InterpolationMode'], do_rescale: bool, rescale_factor: float, do_normalize: bool, image_mean: Optional[Union[float, list[float]]], image_std: Optional[Union[float, list[float]]], patch_size: int, temporal_patch_size: int, merge_size: int, disable_grouping: Optional[bool], return_tensors: Optional[Union[str, TensorType]], **kwargs): grouped_images, grouped_images_index = group_images_by_shape(images, disable_grouping=disable_grouping) resized_images_grouped = {} for shape, stacked_images in grouped_images.items(): height, width = stacked_images.shape[-2:] if do_resize: resized_height, resized_width = smart_resize(height, width, factor=patch_size * merge_size, min_pixels=size['shortest_edge'], max_pixels=size['longest_edge']) stacked_images = self.resize(image=stacked_images, size=SizeDict(height=resized_height, width=resized_width), interpolation=interpolation) resized_images_grouped[shape] = stacked_images resized_images = reorder_images(resized_images_grouped, grouped_images_index) grouped_images, grouped_images_index = group_images_by_shape(resized_images, disable_grouping=disable_grouping) processed_images_grouped = {} processed_grids = {} for shape, stacked_images in grouped_images.items(): resized_height, resized_width = stacked_images.shape[-2:] patches = self.rescale_and_normalize(stacked_images, do_rescale, rescale_factor, do_normalize, image_mean, image_std) if patches.ndim == 4: patches = patches.unsqueeze(1) if patches.shape[1] % temporal_patch_size != 0: repeats = patches[:, -1:].repeat(1, temporal_patch_size - 1, 1, 1, 1) patches = torch.cat([patches, repeats], dim=1) batch_size, grid_t, channel = patches.shape[:3] grid_t = grid_t // temporal_patch_size grid_h, grid_w = (resized_height // patch_size, resized_width // patch_size) patches = patches.view(batch_size, grid_t, temporal_patch_size, channel, grid_h // merge_size, merge_size, patch_size, grid_w // merge_size, merge_size, patch_size) patches = patches.permute(0, 1, 4, 7, 5, 8, 3, 2, 6, 9) flatten_patches = patches.reshape(batch_size, grid_t * grid_h * grid_w, channel * temporal_patch_size * patch_size * patch_size) processed_images_grouped[shape] = flatten_patches processed_grids[shape] = [[grid_t, grid_h, grid_w]] * batch_size processed_images = reorder_images(processed_images_grouped, grouped_images_index) processed_grids = reorder_images(processed_grids, grouped_images_index) pixel_values = torch.cat(processed_images, dim=0) image_grid_thw = torch.tensor(processed_grids) return BatchFeature(data={'pixel_values': pixel_values, 'image_grid_thw': image_grid_thw}, tensor_type=return_tensors) def get_number_of_image_patches(self, height: int, width: int, images_kwargs=None): """ A utility that returns number of image patches for a given image size. Note: Do not remove this method! It is used by vLLM to infer the number of patches and placeholders without an image input. Args: height (`int`): Height of the input image. width (`int`): Width of the input image. images_kwargs (`dict`, *optional*) Any kwargs to override defaults of the image processor. Returns: `int`: Number of image patches per image. """ min_pixels = images_kwargs['min_pixels'] if 'min_pixels' in images_kwargs else self.size['shortest_edge'] max_pixels = images_kwargs['max_pixels'] if 'max_pixels' in images_kwargs else self.size['longest_edge'] patch_size = images_kwargs.get('patch_size', self.patch_size) merge_size = images_kwargs.get('merge_size', self.merge_size) factor = patch_size * merge_size resized_height, resized_width = smart_resize(height, width, factor, min_pixels=min_pixels, max_pixels=max_pixels) grid_h, grid_w = (resized_height // patch_size, resized_width // patch_size) return grid_h * grid_w
@auto_docstring class Qwen2VLImageProcessorFast(BaseImageProcessorFast): def __init__(self, **kwargs: Unpack[Qwen2VLFastImageProcessorKwargs]): pass def _further_process_kwargs(self, size: Optional[SizeDict]=None, min_pixels: Optional[int]=None, max_pixels: Optional[int]=None, **kwargs) -> dict: ''' Update kwargs that need further processing before being validated Can be overridden by subclasses to customize the processing of kwargs. ''' pass @auto_docstring def preprocess(self, images: ImageInput, videos: Optional[VideoInput]=None, **kwargs: Unpack[Qwen2VLFastImageProcessorKwargs]) -> BatchFeature: pass def _preprocess_image_like_inputs(self, images: ImageInput, videos: VideoInput, do_convert_rgb: bool, input_data_format: ChannelDimension, device: Optional[Union[str, 'torch.device']]=None, **kwargs: Unpack[DefaultFastImageProcessorKwargs]) -> BatchFeature: ''' Preprocess image-like inputs. To be overridden by subclasses when image-like inputs other than images should be processed. It can be used for segmentation maps, depth maps, etc. ''' pass def _preprocess_image_like_inputs(self, images: ImageInput, videos: VideoInput, do_convert_rgb: bool, input_data_format: ChannelDimension, device: Optional[Union[str, 'torch.device']]=None, **kwargs: Unpack[DefaultFastImageProcessorKwargs]) -> BatchFeature: pass def get_number_of_image_patches(self, height: int, width: int, images_kwargs=None): ''' A utility that returns number of image patches for a given image size. Note: Do not remove this method! It is used by vLLM to infer the number of patches and placeholders without an image input. Args: height (`int`): Height of the input image. width (`int`): Width of the input image. images_kwargs (`dict`, *optional*) Any kwargs to override defaults of the image processor. Returns: `int`: Number of image patches per image. ''' pass
9
3
90
5
57
29
9
0.46
1
11
4
0
3
0
3
37
289
17
186
71
150
86
86
39
82
20
4
2
26
4,793
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py
transformers.models.qwen2_vl.modeling_qwen2_vl.PatchEmbed
import torch.nn.functional as F import torch import torch.nn as nn class PatchEmbed(nn.Module): def __init__(self, patch_size: int=14, temporal_patch_size: int=2, in_channels: int=3, embed_dim: int=1152) -> None: super().__init__() self.patch_size = patch_size self.temporal_patch_size = temporal_patch_size self.in_channels = in_channels self.embed_dim = embed_dim kernel_size = [temporal_patch_size, patch_size, patch_size] self.proj = nn.Conv3d(in_channels, embed_dim, kernel_size=kernel_size, stride=kernel_size, bias=False) def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: target_dtype = self.proj.weight.dtype hidden_states = hidden_states.view(-1, self.in_channels, self.temporal_patch_size, self.patch_size, self.patch_size) hidden_states = self.proj(hidden_states.to(dtype=target_dtype)).view(-1, self.embed_dim) return hidden_states
class PatchEmbed(nn.Module): def __init__(self, patch_size: int=14, temporal_patch_size: int=2, in_channels: int=3, embed_dim: int=1152) -> None: pass def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: pass
3
0
11
1
11
0
1
0
1
3
0
0
2
5
2
12
24
2
22
16
13
0
14
10
11
1
1
0
2
4,794
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py
transformers.models.qwen2_vl.modeling_qwen2_vl.PatchMerger
import torch.nn.functional as F from torch.nn import LayerNorm import torch import torch.nn as nn class PatchMerger(nn.Module): def __init__(self, dim: int, context_dim: int, spatial_merge_size: int=2) -> None: super().__init__() self.hidden_size = context_dim * spatial_merge_size ** 2 self.ln_q = LayerNorm(context_dim, eps=1e-06) self.mlp = nn.Sequential(nn.Linear(self.hidden_size, self.hidden_size), nn.GELU(), nn.Linear(self.hidden_size, dim)) def forward(self, x: torch.Tensor) -> torch.Tensor: x = self.mlp(self.ln_q(x).view(-1, self.hidden_size)) return x
class PatchMerger(nn.Module): def __init__(self, dim: int, context_dim: int, spatial_merge_size: int=2) -> None: pass def forward(self, x: torch.Tensor) -> torch.Tensor: pass
3
0
6
0
6
0
1
0
1
3
0
0
2
3
2
12
14
1
13
6
10
0
9
6
6
1
1
0
2
4,795
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py
transformers.models.qwen2_vl.modeling_qwen2_vl.Qwen2MLP
from ...activations import ACT2FN import torch.nn as nn class Qwen2MLP(nn.Module): def __init__(self, config): super().__init__() self.config = config self.hidden_size = config.hidden_size self.intermediate_size = config.intermediate_size self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False) self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False) self.act_fn = ACT2FN[config.hidden_act] def forward(self, x): down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x)) return down_proj
class Qwen2MLP(nn.Module): def __init__(self, config): pass def forward(self, x): pass
3
0
6
0
6
0
1
0
1
1
0
0
2
7
2
12
14
1
13
11
10
0
13
11
10
1
1
0
2
4,796
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py
transformers.models.qwen2_vl.modeling_qwen2_vl.Qwen2VLAttention
from ...utils.deprecation import deprecate_kwarg from ...modeling_flash_attention_utils import FlashAttentionKwargs from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel from typing import Any, Callable, Optional, Union import torch.nn.functional as F import torch.nn as nn from .configuration_qwen2_vl import Qwen2VLConfig, Qwen2VLTextConfig, Qwen2VLVisionConfig import torch from ...cache_utils import Cache, DynamicCache from ...processing_utils import Unpack class Qwen2VLAttention(nn.Module): """ Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer and "Generating Long Sequences with Sparse Transformers". """ def __init__(self, config: Qwen2VLTextConfig, layer_idx: Optional[int]=None): super().__init__() self.config = config self.layer_idx = layer_idx if layer_idx is None: logger.warning_once(f'Instantiating {self.__class__.__name__} without passing `layer_idx` is not recommended and will to errors during the forward call, if caching is used. Please make sure to provide a `layer_idx` when creating this class.') self.hidden_size = config.hidden_size self.num_heads = config.num_attention_heads self.head_dim = self.hidden_size // self.num_heads self.num_key_value_heads = config.num_key_value_heads self.num_key_value_groups = self.num_heads // self.num_key_value_heads self.is_causal = True self.attention_dropout = config.attention_dropout self.rope_scaling = config.rope_scaling self.scaling = self.head_dim ** (-0.5) if self.head_dim * self.num_heads != self.hidden_size: raise ValueError(f'hidden_size must be divisible by num_heads (got `hidden_size`: {self.hidden_size} and `num_heads`: {self.num_heads}).') self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=True) self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=True) self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=True) self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=False) self.sliding_window = config.sliding_window if config.layer_types[layer_idx] == 'sliding_attention' else None self.rotary_emb = Qwen2VLRotaryEmbedding(config=config) @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: bsz, q_len, _ = hidden_states.size() query_states = self.q_proj(hidden_states) key_states = self.k_proj(hidden_states) value_states = self.v_proj(hidden_states) query_states = query_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) key_states = key_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) value_states = value_states.view(bsz, q_len, -1, self.head_dim).transpose(1, 2) cos, sin = position_embeddings query_states, key_states = apply_multimodal_rotary_pos_emb(query_states, key_states, cos, sin, self.rope_scaling['mrope_section']) if past_key_values is not None: cache_kwargs = {'sin': sin, 'cos': cos, 'cache_position': cache_position} key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx, cache_kwargs) attention_interface: Callable = eager_attention_forward if self.config._attn_implementation != 'eager': attention_interface = ALL_ATTENTION_FUNCTIONS[self.config._attn_implementation] attn_output, attn_weights = attention_interface(self, query_states, key_states, value_states, attention_mask, dropout=0.0 if not self.training else self.attention_dropout, scaling=self.scaling, sliding_window=self.sliding_window, position_ids=position_ids, **kwargs) attn_output = attn_output.reshape(bsz, q_len, -1).contiguous() attn_output = self.o_proj(attn_output) return (attn_output, attn_weights)
class Qwen2VLAttention(nn.Module): ''' Multi-headed attention from 'Attention Is All You Need' paper. Modified to use sliding window attention: Longformer and "Generating Long Sequences with Sparse Transformers". ''' def __init__(self, config: Qwen2VLTextConfig, layer_idx: Optional[int]=None): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: bool=False, use_cache: bool=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.Tensor, Optional[torch.Tensor], Optional[tuple[torch.Tensor]]]: pass
4
1
48
9
38
4
5
0.14
1
8
3
2
2
15
2
12
103
19
76
37
63
11
54
27
51
6
1
1
9
4,797
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py
transformers.models.qwen2_vl.modeling_qwen2_vl.Qwen2VLCausalLMOutputWithPast
from typing import Any, Callable, Optional, Union import torch.nn.functional as F from dataclasses import dataclass import torch.nn as nn from ...utils import TransformersKwargs, auto_docstring, can_return_tuple, is_torchdynamo_compiling, logging from ...modeling_outputs import BaseModelOutputWithPast, ModelOutput import torch from ...cache_utils import Cache, DynamicCache @dataclass @auto_docstring(custom_intro='\n Base class for Qwen2VL causal language model (or autoregressive) outputs.\n ') class Qwen2VLCausalLMOutputWithPast(ModelOutput): """ loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided): Language modeling loss (for next-token prediction). logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`): Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). past_key_values (`Cache`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`): It is a [`~cache_utils.Cache`] instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. """ loss: Optional[torch.FloatTensor] = None logits: Optional[torch.FloatTensor] = None past_key_values: Optional[Cache] = None hidden_states: Optional[tuple[torch.FloatTensor]] = None attentions: Optional[tuple[torch.FloatTensor]] = None rope_deltas: Optional[torch.LongTensor] = None
@dataclass @auto_docstring(custom_intro='\n Base class for Qwen2VL causal language model (or autoregressive) outputs.\n ') class Qwen2VLCausalLMOutputWithPast(ModelOutput): ''' loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided): Language modeling loss (for next-token prediction). logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`): Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax). past_key_values (`Cache`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`): It is a [`~cache_utils.Cache`] instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. ''' pass
3
1
0
0
0
0
0
3.43
1
0
0
0
0
0
0
0
36
5
7
7
6
24
7
7
6
0
1
0
0
4,798
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py
transformers.models.qwen2_vl.modeling_qwen2_vl.Qwen2VLDecoderLayer
from ...modeling_layers import GradientCheckpointingLayer from ...utils.deprecation import deprecate_kwarg from ...modeling_flash_attention_utils import FlashAttentionKwargs from typing import Any, Callable, Optional, Union import torch.nn.functional as F from ..qwen2.modeling_qwen2 import Qwen2RMSNorm import torch.nn as nn from .configuration_qwen2_vl import Qwen2VLConfig, Qwen2VLTextConfig, Qwen2VLVisionConfig import torch from ...cache_utils import Cache, DynamicCache from ...processing_utils import Unpack class Qwen2VLDecoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2VLTextConfig, layer_idx: int): super().__init__() self.hidden_size = config.hidden_size if config.use_sliding_window and config._attn_implementation != 'flash_attention_2': logger.warning_once(f'Sliding Window Attention is enabled but not implemented for `{config._attn_implementation}`; unexpected results may be encountered.') self.self_attn = Qwen2VLAttention(config, layer_idx) self.mlp = Qwen2MLP(config) self.input_layernorm = Qwen2RMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.post_attention_layernorm = Qwen2RMSNorm(config.hidden_size, eps=config.rms_norm_eps) self.attention_type = config.layer_types[layer_idx] @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: Optional[bool]=False, use_cache: Optional[bool]=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.FloatTensor, Optional[tuple[torch.FloatTensor, torch.FloatTensor]]]: """ Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`, *optional*): attention mask of size `(batch, sequence_length)` where padding elements are indicated by 0. output_attentions (`bool`, *optional*): Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. 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`). past_key_values (`Cache`, *optional*): cached past key and value projection states cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*): Indices depicting the position of the input sequence tokens in the sequence. position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*): Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`, with `head_dim` being the embedding dimension of each attention head. kwargs (`dict`, *optional*): Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code into the model """ residual = hidden_states hidden_states = self.input_layernorm(hidden_states) hidden_states, self_attn_weights = self.self_attn(hidden_states=hidden_states, attention_mask=attention_mask, position_ids=position_ids, past_key_values=past_key_values, output_attentions=output_attentions, use_cache=use_cache, cache_position=cache_position, position_embeddings=position_embeddings, **kwargs) hidden_states = residual + hidden_states residual = hidden_states hidden_states = self.post_attention_layernorm(hidden_states) hidden_states = self.mlp(hidden_states) hidden_states = residual + hidden_states outputs = (hidden_states,) if output_attentions: outputs += (self_attn_weights,) return outputs
class Qwen2VLDecoderLayer(GradientCheckpointingLayer): def __init__(self, config: Qwen2VLTextConfig, layer_idx: int): pass @deprecate_kwarg('past_key_value', new_name='past_key_values', version='4.58') def forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, output_attentions: Optional[bool]=False, use_cache: Optional[bool]=False, cache_position: Optional[torch.LongTensor]=None, position_embeddings: Optional[tuple[torch.Tensor, torch.Tensor]]=None, **kwargs: Unpack[FlashAttentionKwargs]) -> tuple[torch.FloatTensor, Optional[tuple[torch.FloatTensor, torch.FloatTensor]]]: ''' Args: hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)` attention_mask (`torch.FloatTensor`, *optional*): attention mask of size `(batch, sequence_length)` where padding elements are indicated by 0. output_attentions (`bool`, *optional*): Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. 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`). past_key_values (`Cache`, *optional*): cached past key and value projection states cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*): Indices depicting the position of the input sequence tokens in the sequence. position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*): Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`, with `head_dim` being the embedding dimension of each attention head. kwargs (`dict`, *optional*): Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code into the model ''' pass
4
1
40
5
24
12
3
0.5
1
7
3
0
2
5
2
12
82
11
48
22
34
24
25
11
22
3
1
1
5
4,799
huggingface/pytorch-pretrained-BERT
huggingface_pytorch-pretrained-BERT/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py
transformers.models.qwen2_vl.modeling_qwen2_vl.Qwen2VLForConditionalGeneration
from typing import Any, Callable, Optional, Union import torch.nn.functional as F from ...generation import GenerationMixin import torch.nn as nn from ...utils import TransformersKwargs, auto_docstring, can_return_tuple, is_torchdynamo_compiling, logging import torch from ...cache_utils import Cache, DynamicCache from ...processing_utils import Unpack class Qwen2VLForConditionalGeneration(Qwen2VLPreTrainedModel, GenerationMixin): _checkpoint_conversion_mapping = {'^visual': 'model.visual', '^model(?!\\.(language_model|visual))': 'model.language_model'} _tied_weights_keys = ['lm_head.weight'] def __init__(self, config): super().__init__(config) self.model = Qwen2VLModel(config) self.lm_head = nn.Linear(config.text_config.hidden_size, config.text_config.vocab_size, bias=False) self.post_init() def get_input_embeddings(self): return self.model.get_input_embeddings() def set_input_embeddings(self, value): self.model.set_input_embeddings(value) def set_decoder(self, decoder): self.model.set_decoder(decoder) def get_decoder(self): return self.model.get_decoder() def get_video_features(self, pixel_values_videos: torch.FloatTensor, video_grid_thw: Optional[torch.LongTensor]=None): return self.model.get_video_features(pixel_values_videos, video_grid_thw) def get_image_features(self, pixel_values: torch.FloatTensor, image_grid_thw: Optional[torch.LongTensor]=None): return self.model.get_image_features(pixel_values, image_grid_thw) @property def language_model(self): return self.model.language_model @property def visual(self): return self.model.visual @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2VLCausalLMOutputWithPast]: """ labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. Example: ```python >>> from PIL import Image >>> import requests >>> from transformers import AutoProcessor, Qwen2VLForConditionalGeneration >>> model = Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct") >>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct") >>> messages = [ { "role": "user", "content": [ {"type": "image"}, {"type": "text", "text": "What is shown in this image?"}, ], }, ] >>> url = "https://www.ilankelman.org/stopsigns/australia.jpg" >>> image = Image.open(requests.get(url, stream=True).raw) >>> text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) >>> inputs = processor(text=[text], images=[image], vision_infos=[vision_infos]) >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "The image shows a street scene with a red stop sign in the foreground. In the background, there is a large red gate with Chinese characters ..." ```""" 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 outputs = self.model(input_ids=input_ids, pixel_values=pixel_values, pixel_values_videos=pixel_values_videos, image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, position_ids=position_ids, attention_mask=attention_mask, past_key_values=past_key_values, inputs_embeds=inputs_embeds, use_cache=use_cache, output_attentions=output_attentions, output_hidden_states=output_hidden_states, return_dict=True, cache_position=cache_position, **kwargs) hidden_states = outputs[0] logits = self.lm_head(hidden_states) loss = None if labels is not None: loss = self.loss_function(logits=logits, labels=labels, vocab_size=self.config.text_config.vocab_size, **kwargs) return Qwen2VLCausalLMOutputWithPast(loss=loss, logits=logits, past_key_values=outputs.past_key_values, hidden_states=outputs.hidden_states, attentions=outputs.attentions, rope_deltas=outputs.rope_deltas) def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, cache_position=None, position_ids=None, use_cache=True, pixel_values=None, pixel_values_videos=None, image_grid_thw=None, video_grid_thw=None, **kwargs): model_inputs = super().prepare_inputs_for_generation(input_ids, past_key_values=past_key_values, attention_mask=attention_mask, inputs_embeds=inputs_embeds, cache_position=cache_position, position_ids=position_ids, pixel_values=pixel_values, pixel_values_videos=pixel_values_videos, image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, use_cache=use_cache, **kwargs) if position_ids is None: prefill_compiled_stage = is_torchdynamo_compiling() and (input_ids is not None and input_ids.shape[1] != 1 or (inputs_embeds is not None and inputs_embeds.shape[1] != 1)) prefill_noncompiled_stage = not is_torchdynamo_compiling() and (cache_position is not None and cache_position[0] == 0 or (past_key_values is None or past_key_values.get_seq_length() == 0)) if (prefill_compiled_stage or prefill_noncompiled_stage) or self.model.rope_deltas is None: vision_positions, rope_deltas = self.model.get_rope_index(model_inputs.get('input_ids', None), image_grid_thw=image_grid_thw, video_grid_thw=video_grid_thw, attention_mask=attention_mask) self.model.rope_deltas = rope_deltas elif 'position_ids' in model_inputs: batch_size, seq_length = model_inputs['position_ids'].shape device = model_inputs['position_ids'].device position_ids = torch.arange(seq_length, device=device) position_ids = position_ids.view(1, 1, -1).expand(3, batch_size, -1) delta = cache_position[0] + self.model.rope_deltas delta = delta.repeat_interleave(batch_size // delta.shape[0], dim=0) vision_positions = position_ids + delta.expand_as(position_ids) text_positions = model_inputs['position_ids'][None, ...] model_inputs['position_ids'] = torch.cat([text_positions, vision_positions], dim=0) if model_inputs['cache_position'][0] != 0: model_inputs['pixel_values'] = None model_inputs['pixel_values_videos'] = None return model_inputs def _get_image_nums_and_video_nums(self, input_ids: Optional[torch.LongTensor], inputs_embeds: Optional[torch.Tensor]=None) -> tuple[torch.Tensor, torch.Tensor]: """ Get the number of images and videos for each sample to calculate the separation length of the sample tensor. These parameters are not passed through the processor to avoid unpredictable impacts from interface modifications. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Returns: image_nums (`torch.LongTensor` of shape `(batch_size, num_images_sample)`) video_nums (`torch.LongTensor` of shape `(batch_size, num_videos_sample)`) """ image_token_id = self.config.image_token_id video_token_id = self.config.video_token_id vision_start_token_id = self.config.vision_start_token_id if inputs_embeds is not None: vision_start_mask = (inputs_embeds == self.get_input_embeddings()(torch.tensor(vision_start_token_id, dtype=torch.long, device=inputs_embeds.device)))[..., 0] image_mask = (inputs_embeds == self.get_input_embeddings()(torch.tensor(image_token_id, dtype=torch.long, device=inputs_embeds.device)))[..., 0] video_mask = (inputs_embeds == self.get_input_embeddings()(torch.tensor(video_token_id, dtype=torch.long, device=inputs_embeds.device)))[..., 0] else: vision_start_mask = input_ids == vision_start_token_id image_mask = input_ids == image_token_id video_mask = input_ids == video_token_id vision_first_mask = torch.roll(vision_start_mask, shifts=1, dims=1) image_nums = torch.sum(vision_first_mask & image_mask, dim=1) video_nums = torch.sum(vision_first_mask & video_mask, dim=1) return (image_nums, video_nums) def _expand_inputs_for_generation(self, expand_size: int=1, is_encoder_decoder: bool=False, input_ids: Optional[torch.LongTensor]=None, **model_kwargs) -> tuple[torch.LongTensor, dict[str, Any]]: if expand_size == 1: return (input_ids, model_kwargs) visual_keys = ['pixel_values', 'image_grid_thw', 'pixel_values_videos', 'video_grid_thw', 'second_per_grid_ts'] def _expand_dict_for_generation_visual(dict_to_expand): image_grid_thw = model_kwargs.get('image_grid_thw', None) video_grid_thw = model_kwargs.get('video_grid_thw', None) image_nums, video_nums = self._get_image_nums_and_video_nums(input_ids, inputs_embeds=model_kwargs.get('inputs_embeds', None)) def _repeat_interleave_samples(x, lengths, repeat_times): samples = torch.split(x, lengths) repeat_args = [repeat_times] + [1] * (x.dim() - 1) result = torch.cat([sample.repeat(*repeat_args) for sample in samples], dim=0) return result for key in dict_to_expand: if key == 'pixel_values': samples = torch.split(image_grid_thw, list(image_nums)) lengths = [torch.prod(sample, dim=1).sum() for sample in samples] dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=lengths, repeat_times=expand_size) elif key == 'image_grid_thw': lengths = list(image_nums) dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=lengths, repeat_times=expand_size) elif key == 'pixel_values_videos': samples = torch.split(video_grid_thw, list(video_nums)) lengths = [torch.prod(sample, dim=1).sum() for sample in samples] dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=lengths, repeat_times=expand_size) elif key == 'video_grid_thw': lengths = list(video_nums) dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=lengths, repeat_times=expand_size) elif key == 'second_per_grid_ts': dict_to_expand[key] = _repeat_interleave_samples(dict_to_expand[key], lengths=list(video_nums), repeat_times=expand_size) return dict_to_expand def _expand_dict_for_generation(dict_to_expand): for key in dict_to_expand: if key != 'cache_position' and dict_to_expand[key] is not None and isinstance(dict_to_expand[key], torch.Tensor) and (key not in visual_keys): dict_to_expand[key] = dict_to_expand[key].repeat_interleave(expand_size, dim=0) return dict_to_expand model_kwargs = _expand_dict_for_generation_visual(model_kwargs) if input_ids is not None: input_ids = input_ids.repeat_interleave(expand_size, dim=0) model_kwargs = _expand_dict_for_generation(model_kwargs) if is_encoder_decoder: if model_kwargs.get('encoder_outputs') is None: raise ValueError('If `is_encoder_decoder` is True, make sure that `encoder_outputs` is defined.') model_kwargs['encoder_outputs'] = _expand_dict_for_generation(model_kwargs['encoder_outputs']) return (input_ids, model_kwargs)
class Qwen2VLForConditionalGeneration(Qwen2VLPreTrainedModel, GenerationMixin): def __init__(self, config): pass def get_input_embeddings(self): pass def set_input_embeddings(self, value): pass def set_decoder(self, decoder): pass def get_decoder(self): pass def get_video_features(self, pixel_values_videos: torch.FloatTensor, video_grid_thw: Optional[torch.LongTensor]=None): pass def get_image_features(self, pixel_values: torch.FloatTensor, image_grid_thw: Optional[torch.LongTensor]=None): pass @property def language_model(self): pass @property def visual(self): pass @can_return_tuple @auto_docstring def forward(self, input_ids: Optional[torch.LongTensor]=None, attention_mask: Optional[torch.Tensor]=None, position_ids: Optional[torch.LongTensor]=None, past_key_values: Optional[Cache]=None, inputs_embeds: Optional[torch.FloatTensor]=None, labels: Optional[torch.LongTensor]=None, use_cache: Optional[bool]=None, output_attentions: Optional[bool]=None, output_hidden_states: Optional[bool]=None, pixel_values: Optional[torch.Tensor]=None, pixel_values_videos: Optional[torch.FloatTensor]=None, image_grid_thw: Optional[torch.LongTensor]=None, video_grid_thw: Optional[torch.LongTensor]=None, rope_deltas: Optional[torch.LongTensor]=None, cache_position: Optional[torch.LongTensor]=None, **kwargs: Unpack[TransformersKwargs]) -> Union[tuple, Qwen2VLCausalLMOutputWithPast]: ''' labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*): Labels for computing the masked language modeling loss. Indices should either be in `[0, ..., config.vocab_size]` or -100 (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]`. image_grid_thw (`torch.LongTensor` of shape `(num_images, 3)`, *optional*): The temporal, height and width of feature shape of each image in LLM. video_grid_thw (`torch.LongTensor` of shape `(num_videos, 3)`, *optional*): The temporal, height and width of feature shape of each video in LLM. rope_deltas (`torch.LongTensor` of shape `(batch_size, )`, *optional*): The rope index difference between sequence length and multimodal rope. Example: ```python >>> from PIL import Image >>> import requests >>> from transformers import AutoProcessor, Qwen2VLForConditionalGeneration >>> model = Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct") >>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct") >>> messages = [ { "role": "user", "content": [ {"type": "image"}, {"type": "text", "text": "What is shown in this image?"}, ], }, ] >>> url = "https://www.ilankelman.org/stopsigns/australia.jpg" >>> image = Image.open(requests.get(url, stream=True).raw) >>> text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) >>> inputs = processor(text=[text], images=[image], vision_infos=[vision_infos]) >>> # Generate >>> generate_ids = model.generate(inputs.input_ids, max_length=30) >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] "The image shows a street scene with a red stop sign in the foreground. In the background, there is a large red gate with Chinese characters ..." ```''' pass def prepare_inputs_for_generation(self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, cache_position=None, position_ids=None, use_cache=True, pixel_values=None, pixel_values_videos=None, image_grid_thw=None, video_grid_thw=None, **kwargs): pass def _get_image_nums_and_video_nums(self, input_ids: Optional[torch.LongTensor], inputs_embeds: Optional[torch.Tensor]=None) -> tuple[torch.Tensor, torch.Tensor]: ''' Get the number of images and videos for each sample to calculate the separation length of the sample tensor. These parameters are not passed through the processor to avoid unpredictable impacts from interface modifications. Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Returns: image_nums (`torch.LongTensor` of shape `(batch_size, num_images_sample)`) video_nums (`torch.LongTensor` of shape `(batch_size, num_videos_sample)`) ''' pass def _expand_inputs_for_generation(self, expand_size: int=1, is_encoder_decoder: bool=False, input_ids: Optional[torch.LongTensor]=None, **model_kwargs) -> tuple[torch.LongTensor, dict[str, Any]]: pass def _expand_dict_for_generation_visual(dict_to_expand): pass def _repeat_interleave_samples(x, lengths, repeat_times): pass def _expand_dict_for_generation_visual(dict_to_expand): pass
21
2
42
4
30
9
5
0.31
2
11
4
0
10
6
10
11
435
45
303
107
252
93
166
67
155
17
2
4
45