/concurrent-rl-training

A concurrent and possibly highly parallel training environment for Reinforcement Learning Trading Bot

Primary LanguagePython

A Lookahead Free Concurrent Stochastic Environment

Here is a look-ahead bias-free stochastic environment for reinforcement learning. The general idea here is to give the capacity to iterate through multiple episodes at the exact same time to dynamically train a trading agent on a high variation of prices.

Stochastic Prices

How to Install

We're assuming you already have pip and python installed. Assuming you do, install the environment using the following commands:

pip install -r requirements.txt
pip install -e . # this installs the setup.py file

How to run test code

To run the example seen in the blog, use the command from the root to start the Reinforcement Learning Server.

python stochastic_agent/rl_server/server.py

You should see the RL Server start and it'll look like: RL Server

After running the reinforcement learning server, run the command:

python stochastic_agent/backtester/run.py

You'll be able to see exactly what was in the blog.

Concurrent Decision Making