timegan-pytorch

This repository holds the code for the reimplementation of TimeGAN (Yoon et al., NIPS2019) using PyTorch. Some of the code was derived from the original implementation here.

⚠️ 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).

Getting Started

Installing Requirements

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

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

Directory Hierarchy

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