Skip to content

Add default model_kwargs to Generation Tester #482

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
regisss merged 1 commit into from
Oct 24, 2023
Merged

Add default model_kwargs to Generation Tester #482

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 11 additions & 0 deletions tests/transformers/tests/generation/test_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,6 +84,11 @@ class GenerationTesterMixin:
all_generative_model_classes = ()
input_name = "input_ids"

def _update_default_model_kwargs(self, model_kwargs):
model_kwargs["limit_hpu_graphs"] = False
model_kwargs["reuse_cache"] = False
model_kwargs["bucket_size"] = -1

def _get_input_ids_and_config(self, batch_size=2):
config, inputs_dict = self.model_tester.prepare_config_and_inputs_for_common()
input_ids = inputs_dict[self.input_name]
Expand Down Expand Up @@ -268,6 +273,7 @@ def _greedy_generate(

kwargs = {}
model_kwargs = {"attention_mask": attention_mask} if attention_mask is not None else {}

output_generate = model.generate(
input_ids,
do_sample=False,
Expand All @@ -294,6 +300,7 @@ def _greedy_generate(

with torch.no_grad():
model_kwargs = {"attention_mask": attention_mask} if attention_mask is not None else {}
self._update_default_model_kwargs(model_kwargs)
output_greedy = model.greedy_search(
input_ids,
max_length=max_length,
Expand Down Expand Up @@ -423,6 +430,7 @@ def _beam_search_generate(

with torch.no_grad():
model_kwargs = {"attention_mask": attention_mask} if attention_mask is not None else {}
self._update_default_model_kwargs(model_kwargs)
output_beam_search = model.beam_search(
input_ids.repeat_interleave(beam_scorer.num_beams, dim=0),
beam_scorer,
Expand Down Expand Up @@ -552,6 +560,8 @@ def _group_beam_search_generate(

with torch.no_grad():
model_kwargs = {"attention_mask": attention_mask} if attention_mask is not None else {}
self._update_default_model_kwargs(model_kwargs)

output_group_beam_search = model.group_beam_search(
input_ids.repeat_interleave(beam_scorer.num_beams, dim=0),
beam_scorer,
Expand Down Expand Up @@ -615,6 +625,7 @@ def _constrained_beam_search_generate(

with torch.no_grad():
model_kwargs = {"attention_mask": attention_mask} if attention_mask is not None else {}
self._update_default_model_kwargs(model_kwargs)
output_group_beam_search = model.constrained_beam_search(
input_ids.repeat_interleave(constrained_beam_scorer.num_beams, dim=0),
constrained_beam_scorer,
Expand Down