This module was motivated by a unique feature set I wanted to test.
git clone [repository]cd stock_surface
#install dependencies
pip install -r /path/to/requirements.txt
#runing tests
pytest tests/
#get stock data and train a new model
python get_tickers.py
#look at how well model did for individual stock
pytest tests/test_plotting.py::test_plot_stock
#check returns for whole portfolio
pytest tests/test_backtest.py::test_on_array_of_tickers_profit
#open the csv file
vi files/testing_files/return_output.csvget_tickers.py
This file holds the main function that pulls the stock data, manipulates the feature set, and trains the model.
sample_slopes.py
This file holds the functions to perform the feature and target value manipulation on the stock data.
settings.py
Holds the settings for the back testing
support_vector.py_
This holds the class that's used for ML
plotly_web.py
Used for making a pretty 3D surface graph on Plotly
tests/test_plotting.py Used for inputting a stock ticker and having it make a graph of the algorithm's returns alongside the close prices and trade indicators.
tests/test_back_test.py
These functions are used to test how well your model did on the market data most notably: test_on_array_of_tickers_profit()
This will output a CSV file that you can compare how each stock did to just holding it.
Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
Please make sure to update tests as appropriate.