Sergey O

This error appears to only appear when conda is installed. Seems conda is interfering with google colab's `%shell` magic.

Are there suggested subbatch sizes depending on the amount of memory available? Is this something that can be automatically configured?

This is an error that you'll see if the previous step failed and didn't return any models. (So none can be displayed). In this case, that looks like a large...

Do you have an example resfile?

Can you try do a factory reset and make sure all the cells are run?

Not sure I completely follow... Are you looking to rank a list of peptide candidates (not redesign them?)

Hi @mouung , AfDesign uses Jax, not tensorflow. You'll need to make sure your Jax installation can see the GPU. Can you check what jax.default_backend() returns?

Oops. I fixed this issue in the beta branch, and just pushed the fix to the main branch. Can you try again?

Hi Peldom, can you share the example you are working with? I'll investigate.

Did you change how the sequence was initialized? Or is this the result of changing the optimizer? The current optimizer was calibrated (zero initialization, dropouts, switching between model params, normalized...