riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-08-03 03:31:05 +06:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'upstream/main' into add-informer2020-model

Browse Source
This commit is contained in:
elisim 2023-03-01 18:51:33 +07:00
commit e455bbf5c7
2 changed files with 14 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/transformers/models/bart/modeling_bart.py
View File

@ -1058,6 +1058,13 @@ class BartDecoder(BartPretrainedModel):
hidden_states = nn.functional.dropout(hidden_states, p=self.dropout, training=self.training)
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
# decoder layers
all_hidden_states = () if output_hidden_states else None
all_self_attns = () if output_attentions else None
@ -1084,11 +1091,6 @@ class BartDecoder(BartPretrainedModel):
past_key_value = past_key_values[idx] if past_key_values is not None else None
if self.gradient_checkpointing and self.training:
if use_cache:
logger.warning_once(
"`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
)
use_cache = False
def create_custom_forward(module):
def custom_forward(*inputs):

12
src/transformers/models/plbart/modeling_plbart.py
View File

@ -1021,6 +1021,13 @@ class PLBartDecoder(PLBartPreTrainedModel):
hidden_states = nn.functional.dropout(hidden_states, p=self.dropout, training=self.training)
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
# decoder layers
all_hidden_states = () if output_hidden_states else None
all_self_attns = () if output_attentions else None
@ -1047,11 +1054,6 @@ class PLBartDecoder(PLBartPreTrainedModel):
past_key_value = past_key_values[idx] if past_key_values is not None else None
if self.gradient_checkpointing and self.training:
if use_cache:
logger.warning_once(
"`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
)
use_cache = False
def create_custom_forward(module):
def custom_forward(*inputs):