Stored artifact is not h5 file

Hi,

I’m using tensorflow with the python callback. The callback stores several files automatically for each run:

  • saved_model.pb
  • keras_metadata.pb

I would like to automatically store the weights h5 file as well. Is there an option for this? Do I have to do it manually?
Note that I can find best-model.h5 in the wand folder of each run, but for some reason it is not uploaded to the server.

Thank you,
Arnaud

Hi @secondlayer,

Conversion to .h5 files will have to be done in your python process and then uploaded to Weights and Biases.

Hi Arnaud,

We wanted to follow up with you regarding your support request as we have not heard back from you. Please let us know if we can be of further assistance or if your issue has been resolved.

Thanks,
Ramit

Hi Arnaud, since we have not heard back from you we are going to close this request. If you would like to re-open the conversation, please let us know!

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.