ICP13 - narhirep/Python-Deep-Learning GitHub Wiki

Welcome to the In Class Programming 11:

Objective: To train the AutoEncoder to reduce the dimensionality of the input images and to recreate the images from the encoded representation.

Implementation:

1. Add one more hidden layer to autoencoder.
2. Do the prediction on the test data and then visualize one of the reconstructed version of that test data. Also, visualize the same test data before reconstruction using Matplotlib.

CODE:

OUTPUT:

3. Repeat the question 2 on the denoisening autoencoder.
4. plot loss and accuracy using the history object.

CODE:

OUTPUT:

Video: ICP13

Conclusion: In this ICP, I learnt about the AutoEncoder as well as its components Encoder and Decoder. Also, the AutoEncoder has a wide variety of applications, such as de-noising an image to improve transparency, and dimensionality reduction, which encodes an image so that it takes up less memory and can be reconstructed using the encoded representation.