keras: how to save the training history attribute of the history object

Question

In Keras, we can return the output of model.fit to a history as follows:

 history = model.fit(X_train, y_train, 
                     batch_size=batch_size, 
                     nb_epoch=nb_epoch,
                     validation_data=(X_test, y_test))

Now, how to save the history attribute of the history object to a file for further uses (e.g. draw plots of acc or loss against epochs)?

Answer 1

What I use is the following:

with open('/trainHistoryDict', 'wb') as file_pi:
    pickle.dump(history.history, file_pi)

In this way I save the history as a dictionary in case I want to plot the loss or accuracy later on. Later, when you want to load the history again, you can use:

with open('/trainHistoryDict', "rb") as file_pi:
    history = pickle.load(file_pi)

Why choose pickle over json?

The comment under this answer accurately states:

[Storing the history as json] does not work anymore in tensorflow keras. I had issues with: TypeError: Object of type 'float32' is not JSON serializable.

There are ways to tell json how to encode numpy objects, which you can learn about from this other question, so there's nothing wrong with using json in this case, it's just more complicated than simply dumping to a pickle file.