I want to know why the log file take up 300G disk resource.

LambdaLabsML / examples

Deep Learning Examples

MIT License

805 stars 103 forks source link

I want to know why the log file take up 300G disk resource. #24

Open Happzy-WHU opened 1 year ago

Happzy-WHU commented 1 year ago

fvbock commented 1 year ago

it contains checkpoints from the training process - each were ~13.5G in size. look for the ckpt files and drop them if you don't need any of the intermediate models