CNN to get denoised data from noisy data.
- Generate a CNN model in which the input is a noisy number and the output is a denoised number.
- MNIST dataset.
- Use Gaussian distribution to add noise to the input.
- Train the model using Google Colab (TPU) for 100 epochs.
- Plot the training & validation loss.
- Build a classifier based on the trained Autoencoder.
MNIST Dataset
Building Autoencoders in Keras
Denoising Autoencoder