Skip to content

Fix squared optimization steps bug #280

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
tomaarsen merged 4 commits into from
Jan 23, 2023
Merged

Fix squared optimization steps bug #280

Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
a67f3d9
Fix squared optimization steps bug
twerkmeister Jan 18, 2023
e603995
removed steps_per_epoch body.fit argument
twerkmeister Jan 19, 2023
721b384
reinstated total_train_step variable
twerkmeister Jan 23, 2023
39627c4
Reformat trainer.py
tomaarsen Jan 23, 2023
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: 1 addition & 5 deletions src/setfit/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -346,8 +346,6 @@ def train(
distance_metric=self.distance_metric,
margin=self.margin,
)

train_steps = len(train_dataloader) * self.num_epochs
else:
train_examples = []

Expand All @@ -363,19 +361,17 @@ def train(

train_dataloader = DataLoader(train_examples, shuffle=True, batch_size=batch_size)
train_loss = self.loss_class(self.model.model_body)
train_steps = len(train_dataloader) * num_epochs

logger.info("***** Running training *****")
logger.info(f" Num examples = {len(train_examples)}")
logger.info(f" Num epochs = {num_epochs}")
logger.info(f" Total optimization steps = {train_steps}")
logger.info(f" Total optimization steps = {len(train_dataloader) * num_epochs}")
logger.info(f" Total train batch size = {batch_size}")

warmup_steps = math.ceil(train_steps * self.warmup_proportion)

@tomaarsen tomaarsen Jan 19, 2023 •
edited
Loading

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
warmup_steps = math.ceil(train_steps * self.warmup_proportion)
warmup_steps = math.ceil(train_steps * self.warmup_proportion)

This still uses the train_steps variable that you removed.

self.model.model_body.fit(
train_objectives=[(train_dataloader, train_loss)],
epochs=num_epochs,
steps_per_epoch=train_steps,
optimizer_params={"lr": learning_rate},
warmup_steps=warmup_steps,
show_progress_bar=show_progress_bar,
Expand Down