Evann Courdier
Evann Courdier
I got the same curves - Did you find an explanation ? May the pytorch version make a difference ?
@lightas I contacted the owner and he said that he froze batch norm during training. Be careful however that the --freeze-bn parameter **doesn't work** currently in the code ! Since...
@lightas The freeze-bn parameter puts BatchNorn into eval mode at model initialization. But then in the training you do model.train(), which sets the BatchNorm back to training mode...
Up! I would need this feature too.
@bilucodota Here is the output of `pip freeze`. Note there is no jupyterlab_tabnine package. I temporarily removed it to get back autocompletion. (I am running jupyterlab on a remote headless...
Sorry, I just saw your issue. It looks like it could be an issue on the browser side... but without more info and testing it seems hard to debug...
Which version of the library do you have ?
Hello, I don't think the error is related to this component. Do you have other component on the server side code ? Did you try with different keys ?