Closed rasbt closed 6 months ago
Maybe.
This can be done in one of two ways:
IOArgs
, the downside is that you'll need to update the argument validation for all scripts.PrepareArgs
and then do io: PrepareArgs = PrepareArgs()
in the prepare_*.py
scripts. From the command line it's the same but internally it's using a different class.I think it depends on how many arguments are shared across scripts. We should also consider how it will look after #950
@awaelchli Is refactoring the current prepare_*.py
scripts in scope for #950?
Yes, with https://github.com/Lightning-AI/lit-gpt/pull/950 the goal is to replace these prepare scripts (with the exception of the pretraining datasets which need to be optimized in advance). We haven't made any decisions yet and I'd like to discuss it, but my vision was that the scripts get a datamodule as input which encapsulate all data related args.
Cool. In that case https://github.com/Lightning-AI/lit-gpt/issues/954 replaces this issue
For consistency, should it be
io.destination_path
(or perhaps even betterio.out_dir
) etc. in