Development on this library has slowed down, in favor of working on TensorTrade - a framework for trading with RL: https://github.com/notadamking/tensortrade
If you'd like to learn more about how we created this agent, check out the Medium article: https://towardsdatascience.com/creating-bitcoin-trading-bots-that-dont-lose-money-2e7165fb0b29
Later, we optimized this repo using feature engineering, statistical modeling, and Bayesian optimization, check it out: https://towardsdatascience.com/using-reinforcement-learning-to-trade-bitcoin-for-massive-profit-b69d0e8f583b
Discord server: https://discord.gg/ZZ7BGWh
Data sets: https://www.cryptodatadownload.com/data/northamerican/
sudo lspci | grep -i --color 'vga\|3d\|2d' | grep -i nvidia
If this returns anything, then you should have an nVIDIA card.
The first thing you will need to do to get started is install the requirements. If your system has an nVIDIA GPU that you should start by using:
pip install -r requirements.txt
More information regarding how you can take advantage of your GPU while using docker: https://github.com/NVIDIA/nvidia-docker
If you have another type of GPU or you simply want to use your CPU, use:
pip install -r requirements.no-gpu.txt
Update your current static files, that are used by default:
python ./cli.py update-static-data
Afterwards you can simply see the currently available options:
python ./cli.py --help
or simply run the project with default options:
python ./cli.py optimize
If you have a standard set of configs you want to run the trader against, you can specify a config file to load configuration from. Rename config/config.ini.dist to config/config.ini and run
python ./cli.py --from-config config/config.ini optimize
python ./cli.py optimize
Start the vagrant box using:
Code will be located at /vagrant. Play and/or test with whatever package you wish. Note: With vagrant you cannot take full advantage of your GPU, so is mainly for testing purposes
If you want to run everything within a docker container, then just use:
./run-with-docker (cpu|gpu) (yes|no) optimize
python ./ cli.py --params-db-path "postgres://rl_trader:[email protected]" optimize
The database and it's data are pesisted under
If you want to spin a docker test environment:
./run-with-docker (cpu|gpu) (yes|no)
If you want to run existing tests, then just use:
conda create --name rltrader python=3.6.8 pip git conda activate rltrader conda install tensorflow-gpu git clone https://github.com/notadamking/RLTrader pip install -r RLTrader/requirements.txt
While you could just let the agent train and run with the default PPO2 hyper-parameters, your agent would likely not be very profitable. The
stable-baselines library provides a great set of default parameters that work for most problem domains, but we need to better.
To do this, you will need to run
This can take a while (hours to days depending on your hardware setup), but over time it will print to the console as trials are completed. Once a trial is completed, it will be stored in
./data/params.db, an SQLite database, from which we can pull hyper-parameters to train our agent.
From there, agents will be trained using the best set of hyper-parameters, and later tested on completely new data to verify the generalization of the algorithm.
Feel free to ask any questions in the Discord!
Enter and run the following snippet in the first cell to load RLTrader into a Google Colab environment. Don't forget to set hardware acceleration to GPU to speed up training!
!git init && git remote add origin https://github.com/notadamking/RLTrader.git && git pull origin master
!pip install -r requirements.txt
Normally this is caused by missing mpi module. You should install it according to your platorm.
Contributions are encouraged and I will always do my best to get them implemented into the library ASAP. This project is meant to grow as the community around it grows. Let me know if there is anything that you would like to see in the future or if there is anything you feel is missing.
Working on your first Pull Request? You can learn how from this free series How to Contribute to an Open Source Project on GitHub
Want to show your support for this project and help it grow?
Head over to the successor framework: https://github.com/notadamking/tensortrade
When I run python ./optimize.py, got "Internal: Blas GEMM launch failed". How can I fix it ?
qs.report.html() function signature doesn't support
file as an argument.
Call results in:
"TypeError: html() got an unexpected keyword argument 'file'". The correct argument is `output`.
def html(returns, benchmark=None, rf=0., grayscale=False,
title='Strategy Tearsheet', output=None, compounded=True,
figfmt='svg', template_path=None, match_dates=False):
with open(output, 'w', encoding='utf-8') as f:
AttributeError: module 'contextlib' has no attribute 'nullcontext'
//It was confusing having all the code in one line and it is easy to copy /paste
conda create --name rltrader python=3.6.8 pip git
conda activate rltrader
conda install tensorflow-gpu
git clone https://github.com/notadamking/RLTrader
pip install -r RLTrader/requirements.txt
Hi! I am try use RLTrader, but this error blocked me: ModuleNotFoundError: No module named 'tensorflow.contrib'. ENV: Google Collab with GPU.
This steps make an error: 1) !git init && git remote add origin https://github.com/notadamking/RLTrader.git && git pull origin master 2) !pip install -r requirements.txt 3) !python ./cli.py update-static-data
i don't have nvidia :| soooo, i used vagrant install and test.... BUT...
Testing with vagrant
Start the vagrant box using:
vagrant up (required vagrant up --provider virtualbox)
also.. the vagrant file needed -y after the apt lines starting at 53
vm_config.vm.provision "default setup", type: "shell", inline: <<SCRIPT apt update apt install mpich libpq-dev python3-pip -y <--- added this DEBIAN_FRONTEND=noninteractive apt install python3-pip -y <--- would probably remove this line after moving pip3 up a line pip3 install -r /vagrant/requirements.no-gpu.txt
i don't know if it fully installed pip3 after making the virtual machine...
[email protected]:~$ pip3 install -r /vagrant/requirements.no-gpu.txt
Command 'pip3' not found, but can be installed with:
apt install python3-pip Please ask your administrator.
sooo ill move pip3 up to line 55 inthe vagrant file