So, let's define a helper function that will make us able to kick off the training process. This function will take the input images, one-hot encoding of the target classes, and the keep probability value as input. Then, it will feed these values to the computational graph and call the model optimizer:
#Define a helper function for kicking off the training processdef train(session, model_optimizer, keep_probability, in_feature_batch, target_batch):session.run(model_optimizer, feed_dict={input_images: in_feature_batch, input_images_target: target_batch, keep_prob: keep_probability})
We'll need to validate our model during different time steps in the training process, so we are going to define a helper function that will print ...
