Elsa DENAKPO

SVC fit_proba not working for unbalanced dataset

1

Hello, It seems that the fit_proba option in SVC is only usable when we have at least 5 examples by classes. In the code a a 5-folds cross validation is...

Hello, I’m trying to do pretraning on my own data. But after some number of steps (1k or more), the loss becomes NAN. I tried to reduce the learning rate...

Hello, Great work here. Can we have an overview of the final model performances? I mean what is the perplexity and the accuracy after pretraining? Thanks