mirror of
https://github.com/huggingface/transformers.git
synced 2025-07-31 02:02:21 +06:00
Remove padding_masks from gpt_bigcode
. (#27348)
Update modeling_gpt_bigcode.py
This commit is contained in:
parent
8c91f15ae5
commit
cc9f27bb1e
@ -235,16 +235,10 @@ class GPTBigCodeAttention(nn.Module):
|
||||
encoder_attention_mask: Optional[torch.Tensor] = None,
|
||||
use_cache: Optional[bool] = False,
|
||||
output_attentions: Optional[bool] = False,
|
||||
**kwargs,
|
||||
) -> Union[
|
||||
Tuple[torch.Tensor, Optional[torch.Tensor]],
|
||||
Tuple[torch.Tensor, Optional[torch.Tensor], Tuple[torch.Tensor, ...]],
|
||||
]:
|
||||
if "padding_mask" in kwargs:
|
||||
logger.warning_once(
|
||||
"Passing `padding_mask` is deprecated and will be removed in v4.37. Please make sure use `attention_mask` instead.`"
|
||||
)
|
||||
|
||||
if encoder_hidden_states is not None:
|
||||
if not hasattr(self, "q_attn") or not self.is_cross_attention:
|
||||
raise ValueError(
|
||||
@ -308,19 +302,10 @@ class GPTBigCodeFlashAttention2(GPTBigCodeAttention):
|
||||
encoder_attention_mask: Optional[torch.Tensor] = None,
|
||||
use_cache: Optional[bool] = False,
|
||||
output_attentions: Optional[bool] = False,
|
||||
**kwargs,
|
||||
) -> Union[
|
||||
Tuple[torch.Tensor, Optional[torch.Tensor]],
|
||||
Tuple[torch.Tensor, Optional[torch.Tensor], Tuple[torch.Tensor, ...]],
|
||||
]:
|
||||
if "padding_mask" in kwargs:
|
||||
logger.warning_once(
|
||||
"Passing `padding_mask` is deprecated and will be removed in v4.37. Please make sure use `attention_mask` instead.`"
|
||||
)
|
||||
|
||||
# overwrite attention_mask with padding_mask
|
||||
attention_mask = kwargs.pop("padding_mask")
|
||||
|
||||
if encoder_hidden_states is not None:
|
||||
if not hasattr(self, "q_attn") or not self.is_cross_attention:
|
||||
raise ValueError(
|
||||
|
Loading…
Reference in New Issue
Block a user