Skip to content

Add grad_clip to static_argnames in training algorithms #918

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
copybara-service merged 1 commit into from
Jun 5, 2026
Merged

Add grad_clip to static_argnames in training algorithms #918

Show file tree
Hide file tree
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
6 changes: 5 additions & 1 deletion init2winit/trainer_lib/test_trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -652,7 +652,8 @@ def test_text_model_trainer(self):

self.assertEqual(set(df.columns.values), set(get_column_names()))

def test_trainer(self):
@parameterized.parameters(None, 1.0)
def test_trainer(self, grad_clip):
"""Test training for two epochs on MNIST with a small model."""
rng = jax.random.PRNGKey(0)

Expand All @@ -678,6 +679,9 @@ def test_trainer(self):
'valid_size': 96,
'test_size': 80,
}
if grad_clip is not None:
hparam_overrides['opt_hparams.grad_clip'] = grad_clip

input_pipeline_hps = config_dict.ConfigDict(dict(
num_tf_data_prefetches=-1,
num_device_prefetches=0,
Expand Down
1 change: 1 addition & 0 deletions init2winit/trainer_lib/training_algorithm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -902,6 +902,7 @@ def update_params(
static_argnames=(
'training_cost_fn',
'optimizer_update_fn',
'grad_clip',
),
donate_argnums=(0, 1, 2),
)
Expand Down