Output effective batch size and GPU memory usage in logs when using auto_find_batch_size

alvations · March 8, 2023, 1:47pm

Is there a metric definition in evaluate that uses pynvml to monitor the GPU RAM usage while we run trainer.train()?
Is there a logging function that will automatically output the effective batch size by computing the the accumulations + batch size when using auto_find_batch_size with trainer.train()?

alvations · March 13, 2023, 10:25am

Topic		Replies	Views
Trainer with adaptive batch size? Beginners	0	496	September 29, 2023
When using `auto_find_batch_size` and a new batch size is used, output seems to indicate training examples are left off from before. Not the case? 🤗Transformers	0	1381	July 2, 2023
Batch size during training vs batch size during evaluation Beginners	1	1398	August 27, 2023
Trainer will evaluate using my entire validation set(60k), which gives me cuda memory usage issue:. Is there a param that allows evaluating only on some batches in validation set? Beginners	4	922	April 29, 2022
Where to set the Evaluation Batch Size in Trainer Beginners	2	5971	June 17, 2022