How to execute some python code after some epoch in keras

Question

I have simple code that works fine, but i need to create something like 1% of ML done and 10% of ML done etc...

   trainX = np.array(features_data)
    trainY = np.array(labels_data)
    model = Sequential()
    model.add(Dense(10, input_dim=input_dimensions, activation='relu'))
    model.add(Dense(1))
    model.compile(loss='mean_squared_error', optimizer='adam')
    model.fit(trainX, trainY, nb_epoch=3000, batch_size=2, verbose=2)
    model.save(model_location)

Or how to execute some python code after epoch execution?

Epoch 1707/3000
0s - loss: 0.5908
Epoch 1708/3000
0s - loss: 0.4808
Epoch 1709/3000
// how to execute here some code on python
0s - loss: 0.7568
Epoch 1710/3000
0s - loss: 0.5906

Answer 1

You are looking for a Callback. It has a special method - on_epoch_end which is executed after each epoch. Documentation might show you different scenarios you could achieve using their API.