This is an example project on how to use RNNs in Keras for predicting the google stock price.
The datasets used in this project is Google stocks history from 3/1/2012 - 30/12/2016 for the training dataset and from 3/1/2017 - 31/1/2017 for the test dataset. The names of the file is Google_Stock_Price_Test.csv for the test dataset and Google_Stock_Price_Train.csv for the training dataset.
- Pandas 0.23.4
- Numpy 1.15.3
- Matplotlib 2.2.2
- Scikit-learn 0.20.0
- TensorFlow 1.2.0
- Keras 2.2.4
To run this project you will need some software, like Anaconda.
From Anaconda just set as working directory the folder of the project, open the GoogleStockPrediction.py and run this file or press F5.