This repository is for my learning purpose only. It exploits basic things in Pytorch such as
- What's a Tensor.grad_fn() do ?,
- How to print the various modules inside a parent model ?
- How to set a particular layer's or a whole set of layers' weights or bias to a particular number or to a particular distribution ?
- Loading weights from one model to another model ?
By skipping intermediate layers ?
- What is a Forward-Hook ? Why do we need it ?
- What is a Backward-Hook ? Why do we need it ?
- Still exploring what can we do with PyTorch......