Gym Fx

Forex trading simulator environment for OpenAI Gym, observations contain the order status, performance and timeseries loaded from a CSV file containing rates and indicators. Work In Progress
Alternatives To Gym Fx
Project NameStarsDownloadsRepos Using ThisPackages Using ThisMost Recent CommitTotal ReleasesLatest ReleaseOpen IssuesLicenseLanguage
Rl Stock2,419
6 months ago31mitJupyter Notebook
📈 如何用深度强化学习自动炒股
Awesome Ai In Finance1,842
16 days ago1cc0-1.0
🔬 A curated list of awesome machine learning strategies & tools in financial market.
Gym Anytrading1,531
2 months ago6mitPython
The most simple, flexible, and comprehensive OpenAI Gym trading environment (Approved by OpenAI Gym)
2 years ago11lgpl-3.0Python
Scalable, event-driven, deep-learning-friendly backtesting library
6 months ago33Python
Trading Environment(OpenAI Gym) + DDQN (Keras-RL)
Trading Gym212
2 years ago15otherJupyter Notebook
This trading-gym is the first trading for agent to train with episode of short term trading itself.
Gym Fx187
5 months ago1Python
Forex trading simulator environment for OpenAI Gym, observations contain the order status, performance and timeseries loaded from a CSV file containing rates and indicators. Work In Progress
a year ago28Python
Trading Environment(OpenAI Gym) + PPO(TensorForce)
Freqtrade Gym101
2 years ago6gpl-3.0Python
A customized gym environment for developing and comparing reinforcement learning algorithms in crypto trading.
6 years ago7apache-2.0Jupyter Notebook
Alternatives To Gym Fx
Select To Compare

Alternative Project Comparisons


The Forex environment is a forex trading simulator for OpenAI Gym, allowing to test the performace of a custom trading agent. Featuring: configurable initial capital, dynamic or dataset-based spread, CSV history timeseries for trading currencies and observations for the agent, fixed or agent-controlled take-profit, stop-loss and order volume.

The environment features discrete action spaces and optionally continuous action spaces if the orders dont have fixed take-profit/stop-loss and order volume.


Step 1 - Setup Dependencies

Install Python, pip, OpenAI Gym and other dependencies:

sudo apt-get install -y python3-numpy python3-dev cmake zlib1g-dev libjpeg-dev xvfb ffmpeg libboost-all-dev libsdl2-dev python3-pip git gcc make perl

pip3 install graphviz neat-python gitpython gym neat-python matplotlib requests

Step 2 - Get gym-forex from GitHub

git clone harveybc/gym-fx

Step 3 - Set the PYTHONPATH variable

Set the PYTHONPATH environment variable, you may add the following line to the .profile file in your home directory to export on start of sessions. Replace with your username.

  • Linux:

export PYTHONPATH=/home/username/gym-fx/:${PYTHONPATH}

  • Windows:

set PYTHONPATH="c:\Users\username\gym-fx";%PYTHONPATH%

Step 4 - Setup gym-forex

cd gym-fx

python install

Step 5 - Start your agent optimizer that uses the gym-forex environment.


sh ./



Step 6 - Optional: Configure the NEAT parameters

'nano agents/config_20'

Configure in the file, the population size and other parameters according to your computing capacity or requirements, start with the defaults.

Observation Space

A concatenation of num_ticks vectors for the lastest: vector of values from timeseries, equity and its variation, order_status( -1=closed,1=opened),time_opened (normalized with max_order_time), order_profit and its variation, order_drawdown /order_volume_pips, consecutive_drawdown/max_consecutive_dd

Action Space

discrete action 0: 0=nop,1=close,2=buy,3=sell
discrete action 0 parameter: symbol
(optional) continuous action 0 parameter: percent_tp, percent_sl,percent_max

Reward Function

The reward function is the average of the area under the curve of equity and the balance variation.

MQL4 Dataset Generator

Download and install Metatrader 4.

  • Mac

Navigate to Library -> Application -> Support -> MetaTrader 4 -> Bottles -> metatrader64 -> drive_c -> Program Files(x86) -> MetaTrader 4 > MQL4.

Copy the *.mq4 files from datasets into the Scripts folder.

To run these scripts, open MT4 and in the Navigator pane, run the scripts under the "Scripts" folder. Right click the file and click Modify. Run, edit, and debug scripts here as you see fit. The .csv files generated with these scripts will appear in Files.

  • Windows

On MT4: File-> Open Data folder -> MQL4.

Popular Gym Projects
Popular Trading Projects
Popular Applications Categories
Related Searches

Get A Weekly Email With Trending Projects For These Categories
No Spam. Unsubscribe easily at any time.
Machine Learning
Deep Learning
Reinforcement Learning
Time Series
Evolutionary Algorithm
Openai Gym
Q Learning
Deep Q Network