here is my code :
However the code below works well :
I dont get why the two should be any different … aren’t we trying to do the exact same thing ?
here is my code :
However the code below works well :
I dont get why the two should be any different … aren’t we trying to do the exact same thing ?
You can do learn.model.movie_bias.weight.squeeze() for the embedding model.
Embedding class has a weight parameter that contains the matrix values in form of tensor.
Thanks man, it worked like a charm.Could you have a look at this too hahah FastAi + Kaggle/External Datasets