Skip to content
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

Adjustments to resuming training #1406

Open
wants to merge 1 commit into
base: dev
Choose a base branch
from
Open

Adjustments to resuming training #1406

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
12 changes: 4 additions & 8 deletions train_network.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -882,10 +882,9 @@ def remove_model(old_ckpt_name):

# training loop
if initial_step > 0: # only if skip_until_initial_step is specified
for skip_epoch in range(epoch_to_start): # skip epochs
logger.info(f"skipping epoch {skip_epoch+1} because initial_step (multiplied) is {initial_step}")
initial_step -= len(train_dataloader)
global_step = initial_step
logger.info(f"skipping epoch {epoch_to_start} because initial_step (multiplied) is {initial_step}")
initial_step -= epoch_to_start * len(train_dataloader)

for epoch in range(epoch_to_start, num_train_epochs):
accelerator.print(f"\nepoch {epoch+1}/{num_train_epochs}")
Expand All @@ -897,14 +896,11 @@ def remove_model(old_ckpt_name):

skipped_dataloader = None
if initial_step > 0:
skipped_dataloader = accelerator.skip_first_batches(train_dataloader, initial_step - 1)
initial_step = 1
skipped_dataloader = accelerator.skip_first_batches(train_dataloader, initial_step)
initial_step = 0

for step, batch in enumerate(skipped_dataloader or train_dataloader):
current_step.value = global_step
if initial_step > 0:
initial_step -= 1
continue

with accelerator.accumulate(training_model):
on_step_start(text_encoder, unet)
Expand Down