riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-07-31 02:02:21 +06:00
Code Issues Actions 23 Packages Projects Releases Wiki Activity

update modularqwen2

Browse Source
This commit is contained in:
Arthur 2025-06-30 15:22:39 +02:00
parent 3caf7d76a0
commit 113219becd
1 changed files with 3 additions and 41 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
src/transformers/models/qwen2/modular_qwen2.py
View File

@ -12,7 +12,7 @@ from ...modeling_outputs import (
)
from ...modeling_utils import ALL_ATTENTION_FUNCTIONS
from ...processing_utils import Unpack
from ...utils import auto_docstring, can_return_tuple, logging
from ...utils import auto_docstring, check_model_inputs, logging
from ..llama.modeling_llama import (
LlamaAttention,
LlamaDecoderLayer,
@ -109,7 +109,7 @@ class Qwen2Model(MistralModel):
super().__init__(config)
self.has_sliding_layers = "sliding_attention" in self.config.layer_types
@can_return_tuple
@check_model_inputs
@auto_docstring
def forward(
self,
@ -119,30 +119,12 @@ class Qwen2Model(MistralModel):
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,
cache_position: Optional[torch.LongTensor] = None,
**flash_attn_kwargs: Unpack[FlashAttentionKwargs],
) -> BaseModelOutputWithPast:
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
)
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 and use_cache:
logger.warning_once(
"`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`."
)
use_cache = False
# TODO (joao): remove this exception in v4.56 -- it exists for users that try to pass a legacy cache
if not isinstance(past_key_values, (type(None), Cache)):
raise ValueError("The `past_key_values` should be either a `Cache` object or `None`.")
if inputs_embeds is None:
inputs_embeds = self.embed_tokens(input_ids)
@ -181,42 +163,22 @@ class Qwen2Model(MistralModel):
# create position embeddings to be shared across the decoder layers
position_embeddings = self.rotary_emb(hidden_states, position_ids)
# decoder layers
all_hidden_states = () if output_hidden_states else None
all_self_attns = () if output_attentions else None
for decoder_layer in self.layers[: self.config.num_hidden_layers]:
if output_hidden_states:
all_hidden_states += (hidden_states,)
layer_outputs = decoder_layer(
hidden_states = decoder_layer(
hidden_states,
attention_mask=causal_mask_mapping[decoder_layer.attention_type],
position_ids=position_ids,
past_key_value=past_key_values,
output_attentions=output_attentions,
use_cache=use_cache,
cache_position=cache_position,
position_embeddings=position_embeddings,
**flash_attn_kwargs,
)
hidden_states = layer_outputs[0]
if output_attentions:
all_self_attns += (layer_outputs[1],)
hidden_states = self.norm(hidden_states)
# add hidden states from the last decoder layer
if output_hidden_states:
all_hidden_states += (hidden_states,)
return BaseModelOutputWithPast(
last_hidden_state=hidden_states,
past_key_values=past_key_values if use_cache else None,
hidden_states=all_hidden_states,
attentions=all_self_attns,
)