Fastai/Pytorch - doesn't do "fitting" in RMarkdown

Hey all,

I am trying to copy a lesson from fast.ai -- specifically, the one from this github document (in a jupyter notebook).

My goal is to run it in an RMarkdown document, with some code chunks as R (for the data munging) and some as Python (for the fastai/pytorch code itself).

It works well until I run the learn.fit() function -- I get the error ValueError: I/O operation on closed file.

Does anyone know what might be causing this?

Thanks,

Alan

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.