WUT_Computer_Science/Programming/USD
2026-02-06 22:14:54 +01:00
..
CoPO@2d526eadc7 Add 'Programming/USD/' from commit 'b10db883f21e4356d3b370201f3a87898cf1dc5a' 2026-02-06 22:14:54 +01:00
report Add 'Programming/USD/' from commit 'b10db883f21e4356d3b370201f3a87898cf1dc5a' 2026-02-06 22:14:54 +01:00
.gitignore Add 'Programming/USD/' from commit 'b10db883f21e4356d3b370201f3a87898cf1dc5a' 2026-02-06 22:14:54 +01:00
.gitmodules Add 'Programming/USD/' from commit 'b10db883f21e4356d3b370201f3a87898cf1dc5a' 2026-02-06 22:14:54 +01:00
README.md Add 'Programming/USD/' from commit 'b10db883f21e4356d3b370201f3a87898cf1dc5a' 2026-02-06 22:14:54 +01:00

USD

Task:

Zapoznaj się z MetaDrive. Wytrenuj co najmniej dwóch różnych agentów wykorzystując algorytmy wieloagentowe (MA), na co najmniej trzech różnych mapach. Omów otrzymane wyniki oraz zwizualizuj działanie wytrenowanych agentów.

Setup

conda create -n copo python=3.7
conda activate copo

# Install MetaDrive version 0.2.5
pip install git+https://github.com/metadriverse/metadrive.git@releases/0.2.5

pip install torch

git clone https://github.com/decisionforce/CoPO
cd CoPO/copo_code
pip install -e .

pip install -U ray==1.2.0 "ray[rllib]==1.2.0"
pip install -U "numpy<1.19.0"
pip uninstall opencv-python
pip uninstall opencv-python-headless
pip install opencv-python==4.5.5.64
pip install pydantic==1.9.0

How to train a RL agents

cd CoPo/copo_code/copo/
python train_all_cl.py --exp-name my_cl

Training process 4.7h

python train_all_ippo.py --exp-name my_ippo

Training process 7.3h

How to evaluate

python eval.py --root my_cl
python eval.py --root my_ippo