This repository contains an implementation of ensemble deep learning models to forecast or predict stock price. We used Alpha Vantage API to pull stock data(open,high,low,close,volume) and scraped news headlines from inshorts to perform sentiment analysis.
The code and the images of this repository are free to use as regulated by the licence and subject to proper attribution:
Shah, Raj and Tambe, Ashutosh and Bhatt, Tej and Rote, Uday, Real-Time Stock Market Forecasting using Ensemble Deep Learning and Rainbow DQN.
Available at SSRN: https://ssrn.com/abstract=3586788 or http://dx.doi.org/10.2139/ssrn.3586788.
It would be a better idea to create a conda environment and work in isolation
conda create -n envname python=3.6.8 anaconda
conda activate envname
Use conda deactivate
to deactivate the environment
git clone --depth 1 https://github.com/THINK989/Real-Time-Stock-Market-Prediction-using-Ensemble-DL-and-Rainbow-DQN.git && cd Real-Time-Stock-Market-Prediction-using-Ensemble-DL-and-Rainbow-DQN
pip install -r requirements.txt
To run animate.py
you require a free Alpha Vantage API Key.
Enter the key in key=''
parameter inside the animate.py file
ts = TimeSeries(key='',output_format='pandas')
To vizualize the forecast. Remember the data pulled by the API will not update the plot if the market is closed.
python animate.py
To get heatmap visualization for correlation analysis on ^NSEI(Nifty50)
python heatmap.py
This repository is distributed under MIT License
and
respective LICENSE under directories
forecast/LICENSE
or
rainbow/LICENSE
for thier independent code usage.