GitHub - coolport/trafficrl: Traffic light optimization - DQN reinforcement learning, Prophet forecasting, SUMO simulations

Dependencies

curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

git clone https://github.com/coolport/thesis
cd thesis

uv sync
uv run python src/trainer.py --agent dqn --episodes 10

Name		Name	Last commit message	Last commit date
Latest commit History 68 Commits
data		data
docs		docs
gui		gui
models		models
src		src
sumo		sumo
.gitignore		.gitignore
.python-version		.python-version
GEMINI.md		GEMINI.md
LICENSE		LICENSE
README.md		README.md
aadt.pdf		aadt.pdf
decision_log.csv		decision_log.csv
foi_transcript.csv		foi_transcript.csv
output.txt		output.txt
pyproject.toml		pyproject.toml
results_1000eps_chap3.csv		results_1000eps_chap3.csv
results_100eps_chap3.bak		results_100eps_chap3.bak
uv.lock		uv.lock