Backtest 1000s of minute-by-minute trading algorithms for training AI with automated pricing data from: IEX, Tradier and FinViz. Datasets and trading performance automatically published to S3 for building AI training datasets for teaching DNNs how to trade. Runs on Kubernetes and docker-compose. >150 million trading history rows generated from +5000 algorithms. Heads up: Yahoo's Finance API was disabled on 2019-01-03 https://developer.yahoo.com/yql/
removed pinance that's no longer support https://github.com/neberej/pinance as yahoo + google finance are shutdown. finviz looks like it's requiring auth as well so the integration test with finviz is disabled. removed yahoo data extraction tool. relaxed pycodestyle + flake8 versions as pycodestyle actually doesn't fail outright anymore (nice). upgraded the setup.py cmdclass runner after hitting lib2to3 errors on ubuntu with python 3.8. many flake8/pycodestyle linting fixes. removed the numpy version limit which could make more problems downstream if others notice this might need to be reset. please note - i no longer use this repo as redis in kubernetes would regularly crash with rook for the pv's. i highly recommend anyone using this repo to check out numpy vectorization because it is exponentially faster to do mass data-crunching in c/c++. hope that helps
jay-johnson
commented
4 years ago
removed pinance that's no longer support https://github.com/neberej/pinance as yahoo + google finance are shutdown. finviz looks like it's requiring auth as well so the integration test with finviz is disabled. removed yahoo data extraction tool. relaxed pycodestyle + flake8 versions as pycodestyle actually doesn't fail outright anymore (nice). upgraded the setup.py cmdclass runner after hitting lib2to3 errors on ubuntu with python 3.8. many flake8/pycodestyle linting fixes. removed the numpy version limit which could make more problems downstream if others notice this might need to be reset. please note - i no longer use this repo as redis in kubernetes would regularly crash with rook for the pv's. i highly recommend anyone using this repo to check out numpy vectorization because it is exponentially faster to do mass data-crunching in c/c++. hope that helps