README.md

fyp22068

This is the implementation for our FYP titled "Optimal Order Placement in Cryptocurrency Markets" for the COMP4801 course.

Group Members

1. Kritik Satija
2. Raghav Agarwal
3. Mohammad Muttasif

Contributions

Component	Tools Used	Contributor(s)
Exchange Simulator	Python	Kritik, Raghav
Deep Q-Network algorithm	Python	Kritik
Microprice algorithm	Python	Raghav
Front-end	PyQt5	Muttasif

---

Usage

Environment

Create virtual environment using pipenv, conda, or any other virtual environment creator. The example below is using conda:

conda create -n crypto-market-simulator python=3.x
conda activate crypto-market-simulator
pip install -r requirements.txt

Running the simulator

Run the file market-simulator.py in the market-simulator directory to start the simulator

python market-simulator.py

Running the Reinforcement Learning model

In the reinforcement-learning directory, run the reinforcement-learning.ipynb file using Jupyter Notebook. Run all the cells

Running the Microprice algorithm

In the microprice directory, run the microprice.ipynb file using Jupyter Notebook. Run all the cells

Running Data Analytics

In the data-analytics directory, run the data_collection.ipynb file using Jupyter Notebook. Run all the cells. This will generate the required csv files for analytics.ipynb. For convenience, a csv file has been provided for analysis