An open API service indexing awesome lists of open source software.

https://github.com/yukunj/mytorch

This is the a small Numpy-based deep learning library implementation, mimicing PyTorch
https://github.com/yukunj/mytorch

Last synced: 6 months ago
JSON representation

This is the a small Numpy-based deep learning library implementation, mimicing PyTorch

Awesome Lists containing this project

README

          

## MyTorch -- A Small Deep Learning Library Implementation

This is a small NumPy-based deep learning library implementation, mimicing PyTorch. It originates from the course [11-785 Introduction to Deep Learning](https://deeplearning.cs.cmu.edu/F21/index.html) at Carnegie Mellon Fall 2021.

---

### Content Covered

+ [Activation: (identity, sigmoid, tanh, ReLU)](src/activation.py)
+ [Batchnorm](src/batchnorm.py)
+ [SoftmaxCrossEntropy](src/loss.py)
+ [Linear Layer](src/linear.py)
+ [Flatten Layer](src/flatten.py)
+ [Multi-layer Perceptron](src/mlp.py)
+ [Convolution Layer: (1D, 2D, with/without dilation and upsampling)](src/conv.py)
+ [CNN+MLP model](src/cnn_mlp.py)
+ [Recurrent Unit Cell](src/rnn_cell.py)
+ [Gate Recurrent Unit Cell](src/gru_cell.py)
+ [Search: (greedy Search, beam Search)](src/search.py)
+ [Connectionist Temporal Classification Cell (CTC)](src/ctc.py)
+ [CTC Loss](src/ctc_loss.py)
+ [Locked Dropout Layer](src/locked_dropout.py)
+ [Embedding Dropout Layer](src/embedding_dropout.py)
+ [Language Model (prediction, generation)](src/language_model.py)