Stock Market Predictor with LSTM network. Web scraping and analyzing tools (ohlc, mean)
With this stock market predictor you will be able to analyze almost 500 companies and their future. This project includes a WebScraping tool to get the stock data from different companies.
Also you will be able to plot different finance graphs of the OHLC or the mean of the AdjClose
The program uses Python 3.7.3
There are python files and jupyter notebooks
Modules used: numpy, pandas, mpl_finance, matplotlib, keras, pickle...
The file of web scraping is StockScraper.py
-Execute the function save_sp500_tickers() to get the 500 companies tickers and save on a file format .pickle
-Execute get_data() to use the yahoo_finance API and get the data with Pandas from the different companies symbols (required first step) and save it on a folder called stock_dfs
-Finally execute compile_data() to create the csv of all the companies data for the correlation
This steps are esential for the next steps
Optional
-Use visualize_data() to see the plot of the correlation between companies (it will take some time)
The file of neural networkm predictor is is AIFinance.py
-To train and test a company you should use the function predictFuture('AAPL') You can use whatever company symbol you want
#Analyze data Use the jupyter notebook StockMarket.ipynb
You will be able to see step by step the different analyze tools and also the training of a company
The company can be change at the bottom of the notebook