Tensor Algorithmic Trader
Python package to train ML based models for automated trading strategy on cryptocurrencies. Using Python Binance API.
To start run file tensortrader/task/bactesting_task.py. Change configuration on backtesting.yml and feature_eng.yml.
- Free software: MIT license
- App: https://tensortrader-app.onrender.com/.
Install latest release then:
pip install --force-reinstall tensortrader-0.1.0-py2.py3-none-any.whl
# Then run bot for available Tickers
tensortrader --ticker BTCUSDT- Base Architechture
- Create Trading Execution class (done)
- Create Target/Stop management (done)
- Create Trading Performance Tracking Database (done)
- Adjust Getting latest Trained ML Model
- Model Improvement
- Hyperparameter Optimization for TCN Model
- Test GCN Model
- Try out log(p(t)/p(t-1)) as return metric
- Automate in the Cloud
- Front End
- Show Training Results
- Show last returns denoising values
- Show Traiding Performance
This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.