Created
July 14, 2019 00:17
-
-
Save ksivaman/e98d01f6bbb0a0687ec4cfbb3e582673 to your computer and use it in GitHub Desktop.
Training loop for a simple feed forward neural network.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| def train(train_X, train_Y, epochs, lr, layers=[4, 5, 1], activate=['R', 'S']): | |
| # initiation of neural netowrk parameters | |
| params_w, params_b = init(layers) | |
| losses = [] | |
| accuracies = [] | |
| # performing calculations for subsequent iterations | |
| for i in range(epochs): | |
| # step forward | |
| y_pred, activations, outputs = forward_pass(train_X, params_w, params_b, layers, activate) | |
| # monitor loss and accuracy and keep a record of them. | |
| loss = cross_entropy_loss(y_pred, train_Y) | |
| losses.append(loss) | |
| accuracy = accuracy_metric(y_pred, train_Y) | |
| accuracies.append(accuracy) | |
| # back prop to calculate the gradients | |
| gradients = backward_pass(y_pred, train_Y, activations, outputs, params_w, params_b) | |
| # update the weights and biases | |
| params_w, params_b = param_updates(params_w, params_b, gradients, lr) | |
| print('Loss for epoch {} : {}, accuracy is {}'.format(i+1, loss, accuracy)) | |
| return params_w, params_b |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment