Hello! I am running a sweeps grid search. Unfortunately only the first sweep run is an effective training, all the others have constant value for the evaluation metrics.Here some pictures. I have the feeling that it is not doing the evaluation for the 2,3,4,… model trained. I am using HF Trainer.
Could you send a link to your workspace so I can look at this further? I want to make sure that the correct config for the sweep is being sent to the WandB or look to see if there is an issue with the way that sweeps is logging your results.
