Skip to content
Discussion options

You must be logged in to vote

The --max-workers flag controls how many GPU worker processes run in parallel, not CPU cores. Each worker loads the model onto a GPU and processes files independently.

With a single GPU, --max-workers 1 is usually best (default). With 2 GPUs (like Kaggle dual T4), --max-workers 2 lets each GPU process a file simultaneously.

Setting it higher than your GPU count will not help — the workers would compete for the same GPU memory.

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by TinyRick1489
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants