#168168th of 206 in ML frameworks and librariesFor finance and business, trading, stocks and crypto, finance and accounting
finrl-trading
Weight-based quant trading framework: backtest and Alpaca execution
- Stars
- 3.9k
- License
- Apache-2.0
- Last commit
- Sep 2026
- Last release
- Mar 2026
- Language
- Python
3,018 on 22 Apr 20263,936 on 11 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
FinRL-X is a Python trading framework in which every strategy outputs a target portfolio weight vector. The same weights feed a bt-based backtest engine and Alpaca execution. It includes a data layer (Yahoo Finance, FMP, WRDS, SQLite cache), a Random Forest stock selector, PPO/SAC allocators and an adaptive multi-asset rotation strategy. It succeeds the original FinRL.
Who it is for: Quant researchers and developers building and paper-trading portfolio strategies
Strengths
- Single weight-vector interface lets selection, allocation, timing and risk modules be swapped independently
- Same weights run through backtest and Alpaca paper/live execution, with multi-account support
- Yahoo Finance works as a free default data source; FMP and WRDS are optional
- deploy.sh runs data download, backtest and paper trading in one command
Weaknesses
- Live execution is limited to Alpaca; no other brokers listed
- Reported results are the authors' own backtests plus a short paper-trading window
- Manual backtest runs need daily CSVs under data/fmp_daily/ first
- README states it is for educational and research use only
What it needs
- Docker + Compose
- Needs Alpaca API, Yahoo Finance, FMP API (optional), WRDS (optional), SQLite
- Models: Random Forest, PPO, SAC
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in ML frameworks and libraries
See all 206| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | ultralyticsPython library and CLI for training and running YOLO vision models | 94 out of 100 |
| 3 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |