/TimeSeriesForecastingNN

Using neural networks to perform time series forecasting on Indian stocks

Primary LanguagePython

Time series Analysis of Indian Stocks with LSTMs

The aim of this repo is to demonstrate the use of LSTMs as a tool to model time-series data for making stock forecasting. The data is scraped dynamically using nsepy, a library made to fetch data for Indian markets.

Finally, streamlit is used to visualize the process of training in a web app as shown above. In order to reproduce this code on your system, you need to have the following installed on your system

  • python 3.6+ (I have run on python 3.6, should work on higher versions)
  • pytorch 1.4.0+
  • pandas 1.0.1+
  • numpy 1.18.1 +
  • sklearn 0.22.1 +
  • nsepy 0.8
  • streamlit 0.69.2

Once you get these packages, you can clone this repository and run the app for yourself using the command

streamlit run forecasting.py

For understanding more on how the data is extracted and models are built, refer this medium blog post.

Thanks for your attention! Hope this repo helped you in some way :)