Giters

Yairz1 / AutoEncoder

An implementation of AE LSTM based. We test our architecture on several tasks as reconstructing synthetic time series, s&p 500 stocks, and forecasting s&p 500 stocks based on the decoded information (also known as latent space) features we extract from the AE

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

deeplearningautoencodermachinelearningtimeseriesforcasting

Results

Synthetic data reconstruction

synthetic_recon

Mnist samples reconstruction (We treat each image as a vector)

mnist recon

Real S&P 500 stock values reconstruction

ap500 recon

About

An implementation of AE LSTM based. We test our architecture on several tasks as reconstructing synthetic time series, s&p 500 stocks, and forecasting s&p 500 stocks based on the decoded information (also known as latent space) features we extract from the AE

deeplearningautoencodermachinelearningtimeseriesforcasting

Languages

Language:Python 100.0%