An attempt to apply machine learning for tcp congestion control. This is our CS359 computer networks project. We have made a simple network simulator to test our algorithms. First we implemented a simplified TCP Tahoe. Then we collected data from it and trained a polynomial regression model. Finally we used the model and collected the results below.
Install dependencies using:
python3 -m pip install -U -r requirements.txt
To run the model:
python3 ./src/main.py
1 host
2 hosts
3 hosts
4 hosts