https://github.com/birdx0810/timegan-pytorch

This repository is a non-official implementation of TimeGAN (Yoon et al., NIPS2019) using PyTorch.
https://github.com/birdx0810/timegan-pytorch

deep-learning generative-adversarial-network hacktoberfest pytorch time-series

Last synced: 19 days ago
JSON representation

This repository is a non-official implementation of TimeGAN (Yoon et al., NIPS2019) using PyTorch.

• Host: GitHub
• URL: https://github.com/birdx0810/timegan-pytorch
• Owner: birdx0810
• License: mit
• Created: 2020-12-24T10:07:07.000Z (almost 4 years ago)
• Default Branch: main
• Last Pushed: 2022-07-26T05:40:09.000Z (over 2 years ago)
• Last Synced: 2023-03-04T05:26:57.031Z (over 1 year ago)
• Topics: deep-learning, generative-adversarial-network, hacktoberfest, pytorch, time-series
• Language: Jupyter Notebook
• Homepage:
• Size: 2.14 MB
• Stars: 48
• Watchers: 2
• Forks: 25
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# timegan-pytorch

This repository holds the code for the reimplementation of TimeGAN ([Yoon et al., NIPS2019](https://papers.nips.cc/paper/8789-time-series-generative-adversarial-networks)) using PyTorch. Some of the code was derived from the original implementation [here](https://github.com/jsyoon0823/TimeGAN).

> :warning: WARNING!!!

> - This implementation is written for other purposes, not for experiments in the original paper.

> - There are some known issues that I've haven't got time to resolve (see issue [#1](https://github.com/d9n13lt4n/timegan-pytorch/issues/1#issuecomment-895126605)).

## Getting Started

### Installing Requirements

This implementation assumes Python3.8 and a Linux environment with a GPU is used.

```bash

cat requirements.txt | xargs -n 1 pip install --upgrade

```

### Directory Hierarchy

```bash

data/                         # the folder holding the datasets and preprocessing files

  ├ data_preprocessing.py     # the data preprocessing functions

  └ stock.csv                 # the example stock data derived from the original repo

metrics/                      # the folder holding the metric functions for evaluating the model

  ├ dataset.py                # the dataset class for feature predicting and one-step ahead predicting

  ├ general_rnn.py            # the model for fitting the dataset during TSTR evaluation

  ├ metric_utils.py           # the main function for evaluating TSTR

  └ visualization.py          # PCA and t-SNE implementation for time series taken from the original repo

models/                       # the code for the model

output/                       # the output of the model

main.py                       # the main code for training and evaluating TSTR of the model

requirements.txt              # requirements for running code

run.sh                        # the bash script for running model

visualization.ipynb           # jupyter notebook for running visualization of original and synthetic data

```