This repo implements MyTorch, a deep learning library that does not use PyTorch and only uses NumPy. The purpose of this repo is to make the ideas behind autodifferentiation and backprop clear and concrete (with code). The repo is structured in a modular fashion where a deep learning system is decomposed into multiple modules like Loss
, Activation
, rnn_cell
, etc for clarity. We try to make everything as vectorized as possible since unvectorized code are usually slow and inconcise.
-
Notifications
You must be signed in to change notification settings - Fork 0
jwrh/MyTorch
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
MyTorch, a NumPy library created from scratch for simulating PyTorch's functionalities.
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published