SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Properly calculate the total train iterations and recalculate num epochs in no_trainer scripts (#17856)

Browse Source
This commit is contained in:
Zachary Mueller
2022-06-23 15:46:01 -04:00
committed by GitHub
parent 7c1b91281f
commit 75259b44bf
12 changed files with 70 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
examples/pytorch/speech-pretraining/run_wav2vec2_pretraining_no_trainer.py
View File

@@ -546,8 +546,6 @@ def main():
if args.max_train_steps is None:
args.max_train_steps = args.num_train_epochs * num_update_steps_per_epoch
else:
args.num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)
lr_scheduler = get_scheduler(
name=args.lr_scheduler_type,
@@ -556,6 +554,9 @@ def main():
num_training_steps=args.max_train_steps,
)
# Afterwards we recalculate our number of training epochs
args.num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)
# 5. Train
total_batch_size = args.per_device_train_batch_size * accelerator.num_processes * args.gradient_accumulation_steps