Repository files navigation

AlphaPy is a machine learning framework for both speculators anddata scientists. It is written in Python mainly with thescikit-learnandpandas libraries, as well as many other helpfulpackages for feature engineering and visualization. Here are justsome of the things you can do with AlphaPy:

• Run machine learning models usingscikit-learn,Keras,xgboost,LightGBM, andCatBoost.
• Generate blended or stacked ensembles.
• Create models for analyzing the markets withMarketFlow.
• Predict sporting events withSportFlow.
• Develop trading systems and analyze portfolios usingMarketFlowand Quantopian'spyfolio.

https://www.scottfreellc.com/alphapy-pro

http://alphapy.readthedocs.io/en/latest/

You should already have pip, Python, and optionally XGBoost, LightGBM, andCatBoost installed on your system (see below). Run the following command to installAlphaPy:

pip install -U alphapy

Pyfolio is automatically installed by AlphaPy, but if you encounterthe following error when trying to create a tear sheet:

AttributeError: 'numpy.int64' object has no attribute 'to_pydatetime'

Install pyfolio with this command:

pip install git+https://github.com/quantopian/pyfolio

For Mac and Windows users, XGBoost willnot install automaticallywithpip. For instructions to install XGBoost on your specificplatform, go tohttp://xgboost.readthedocs.io/en/latest/build.html.

For instructions to install LightGBM on your specificplatform, go tohttps://lightgbm.readthedocs.io/en/latest/Installation-Guide.html.

For instructions to install CatBoost on your specificplatform, go tohttps://catboost.ai/docs/concepts/python-installation.html.

You can find an implementation of MarketFlow here:

https://www.scottfreellc.com/gamept

The official channel for support is to open an issue on Github.

http://github.com/ScottfreeLLC/AlphaPy/issues

Follow us on Twitter:

https://twitter.com/_AlphaPy_?lang=en

If you like the software, please donate:

http://alphapy.readthedocs.io/en/latest/introduction/support.html#donations