I would like to load the trained model and train it further with new data samples(Tabular Data). How would i do that ?
The same way it’s described for images. learn.save() and learn.load() and pass in your training data. You’d need to make sure that your new data has the same preprocessing applied to it (this is easier in fastai2 vs fastai1) as well.
Can you link to an example of loading new training data into the model, how does this deal with categorical embeddings if there might be new items that weren’t seen in the previous one?