Trading Bot This project is a Q-learning based bot that uses historical data to make a working model. After a model has been made the bot uses sentiment analysis of news articles as an extra data point. The bot runs on the Alpaca Stock Trading API and uses the Polygon data from Alpaca as well. Results After giving the bot a $100,000 paper account and letting it run in and around a global pandemic these were the results. Downsides I have no idea how to actually implement Q-learning correctly so it could be very wrong. Please consider reaching out if you have some experience and want to help with it! Also cannot train on old news articles, so sentiment only runs during realtime. Data All data is automatically requested from the Alpaca Barset and Polygon News Getting Started To use this script you have to go out and get a few API keys: Make a Funded Alpaca Brokerage Account Optional now: Apply for the Natural Language API After you finish with these few steps, write down your API keys, you'll need them below. The next thing you want to do is set up an Ubuntu server. I can write a detail

⚠ No verified equity curve — no track-record source connected.

No score history is stored yet — only the current score is shown.

⚠ No live verification account connected — ask for proof before buying.

Alerts on changes: coming soon

Prop-firm compatibility not provided.

Data-completeness & trust index (not a profitability rating)