Skip to content

Seonu-Lim/time-series-prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Time - Series Prediction using LSTM

This repository contains codes for time series prediction by LSTM, especially when the inputs are multivariate and the outputs are univariate.

Prerequisites

  1. nvidia gpu

  2. torch == 1.6.0

  3. blitz (for bayesian lstm)

  4. Python >= 3.6

  5. ray >= 1.0.1

The required Python packages can be simply installed through:

pip install -r requirements.txt

Before running tune.py, run :

pip install "ray[tune]"

For hyperparameter tuning, run :

python tune.py --X_dir temp/X.pt --y_dir temp/y.pt --epoch 300 --config_dir temp/config.json --model GRU --n_sample 10

About

repo for price prediction

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages