Skip to content

Latest commit

 

History

History
2 lines (2 loc) · 351 Bytes

README.md

File metadata and controls

2 lines (2 loc) · 351 Bytes

Residual network for the MNIST dataset

Objective: employ the model subclassing API together with custom layers to create a residual network architecture. Train the custom model on the MNIST dataset by using a custom training loop and implementing the automatic differentiation tools in Tensorflow to calculate the gradients for backpropagation.