drl-candlesticks-trader

This repository is for the code of paper "Automated Cryptocurrency Trading Approach Using Ensemble Deep Reinforcement Learning: Learn to Understand Candlesticks"

Paper Link

Despite its high risk, cryptocurrency has gained popularity as a successful trading option. Cryptocurrencies are digital assets that fluctuate dramatically in a market that operates for 24 hours. Developing trading bots using machine learning-based artificial intelligence (AI) approaches has recently received considerable attention. Previous studies have used machine learning techniques to predict financial market trends or make trading decisions, primarily using numeric data extracted from candlesticks. However, these data often ignore the temporal and spatial information present in candlesticks, resulting in a poor understanding of their significance. In this study, we used multi-resolution candlestick images that contain temporal and spatial information on prices. The goal of this study was to compare the performance of raw numeric data and candlestick image data to optimize trading strategies and maximize returns...

Image download

Candlesticks image data download: https://drive.google.com/file/d/1qX8K8CRnS3lmhA0Trd1Yk-i7_Cdchwbx/view?usp=sharing
Copy the unzipped folder "candlesticks" to folder "images"

Directories of folders are

-- drl-candlesticks-trader
 |
 -- code
  |
  -- libs
 |
 -- data
 |
 -- images
  |
  -- candlesticks
 |
 -- results
  |
  -- test
   |
   -- 2022-06-19
   |
   -- 2022-08-15
  |
  -- train
  |
  -- valid
 |
 -- runs
 |
 -- weights

Run Example

Train all PPO agents in GPU with multi-resolution raw numerical data
Type following command in the terminal: $ python3 code/main.py -t train -g ppo -r multi

Validate the first DQN agent in GPU with candlestick image
Type following code in the terminal: $ python3 code/main.py -t valid -g dqn -f 0

Code for generating candelsticks images

The code for generating candelsticks images are uploaded to the folder of "code -> preprocessing" You can write your methods in the class of Plotchart to generate different images according your design

License

MIT