yoon-llm / timeseries_seq2seq

This repo aims to be a useful collection of notebooks/code for understanding and implementing seq2seq neural networks for time series forecasting. Networks are constructed with keras/tensorflow.

TimeSeries_Seq2Seq

This repo aims to be a useful collection of notebooks/code for understanding and implementing seq2seq neural networks for time series forecasting. Networks are constructed with keras/tensorflow.

Instructions for Working With Notebooks:

Navigate to the directory you want the git repo to live in.

1. Run git clone https://github.com/JEddy92/TimeSeries_Seq2Seq.git
2. Obtain the wikipedia web traffic data from kaggle. Store it in a folder called "data" at the top level of this repo (this is where the notebooks point to when reading data).