Similar to how training and validation accuracy are plotted in Keras by storing them in history, I would also like to plot test accuracy over the epochs. I have used a callback defined to track test accuracy from https://github.com/keras-team/keras/issues/2548.
My code is
history = my_new_model.fit(X_train, y_train, epochs = 20, validation_split = 0.1, callbacks=[TestCallback((X_test, y_test))])
Now, I can only plot training and validation accuracy
. Getting a KeyError for test accuracy.
Can someone please help understand if the test accuracy can be plotted using the callbacks?
Thank you.
Aucun commentaire:
Enregistrer un commentaire