Farm ML Models
Model registry plus ML Testing backtest desk on one page. Hermie / classifiers stay planned — use Backtest below for forecast-model runs.
Machine Learning catalog → every ML model, library, research note and downloaded repo we find, in one list.
Resource: Machine Learning for Trading (Stefan Jansen), via Herman Trading on X
Post by Herman Trading (@RHerman) on X — text transcribed verbatim from the screenshots · View post on X
🚨 20,000+ STARS. SOMEONE BASICALLY OPEN-SOURCED A MACHINE LEARNING TRADING LAB.
I recently came across Stefan Jansen’s “Machine Learning for Trading” GitHub repo.
20.9K stars - and this thing is HUGE.
It takes you from raw financial data all the way to building, testing and deploying ML-driven trading strategies:
- → 27 chapters
- → 9 end-to-end case studies
- → Machine Learning + Deep Learning
- → Market microstructure
- → Alternative data
- → Backtesting & transaction costs
- → Portfolio construction & risk
- → Reinforcement Learning
- → Generative AI + autonomous trading agents
There are also 100+ research primers, agent skills and production Python libraries.
And the code is PUBLIC.
I haven’t gone through the whole thing yet - but this looks like a serious rabbit hole.
Has anyone here actually USED this repo?
Found anything genuinely useful for building trading systems?
Drop your experience/results below 👇
Backtest
Listed below are all the forecast models, which can be filtered by their time horizon, currency, or name. Additionally, you can sort selective columns in ascending or descending order by clicking on the column header.
Test Result
Popular Models
ML models
| Enabled | PC | Name | Group | Version | Updated | Notes |
|---|
ML tasks
| Enabled | PC | Name | Group | Version | Updated | Notes |
|---|
Add New Model Tab
Placeholder — wire when model upload is ready.
Model Comparison
Placeholder — compare champions side-by-side later.
About Us
Data Systems Global · farm ML desk.
Contact Us
Via farm ops chat.
FAQ
Backtest numbers below are sample UI until live runs are hooked up.