Downloads · 30 days
0
Tombiczek/dqn-sumo-belgradzka-ken
dqn-sumo-belgradzka-ken is a reinforcement learning model from Tombiczek. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. It is set up for sumo-rl. The card lists the license as other.
This project contains a Deep Q-Network (DQN) model trained to control traffic lights at the Belgradzka-KEN intersection in Warsaw.
Downloads · 30 days
0
Access
Public
Updated Aug 4, 2026
Repo size
8.6 MB
Likes
0
Public
Click a slice to open those files.
.gif5.4 MB · 64%
From the Hugging Face model README
This project contains a Deep Q-Network (DQN) model trained to control traffic lights at the Belgradzka-KEN intersection in Warsaw.
The model learns signal switching behavior in a SUMO simulation to improve traffic flow metrics such as queue length and delay.
dqn_fixed_lr_5e-4_225k_final.zip)This is the final selected model. Compared with the unstable variant, it uses a lower learning rate and more frequent target-network updates, which gave more stable behavior and better final metrics.
training:
total_timesteps: 225_000
seed: 42
learning_rate: 0.0005
buffer_size: 50000
learning_starts: 1000
batch_size: 64
gamma: 0.99
train_freq: 1
target_update_interval: 500
exploration_fraction: 0.2
exploration_initial_eps: 1.0
exploration_final_eps: 0.05
device: auto
dqn_fixed_tui_1e3_225k_unstable.zip)This model was intentionally kept for comparison. With a higher learning rate and a slower target-network update schedule, training was less stable and produced worse traffic control quality.
training:
total_timesteps: 225_000
seed: 42
learning_rate: 0.001
buffer_size: 50000
learning_starts: 1000
batch_size: 64
gamma: 0.99
train_freq: 1
target_update_interval: 1000
exploration_fraction: 0.2
exploration_initial_eps: 1.0
exploration_final_eps: 0.05
device: auto
GUI recording showing how the trained model controls traffic lights:

This plot shows performance trends for models trained with different numbers of steps on random traffic with fixed car arrival interval:

This plot shows performance trends for models trained with different numbers of steps on random traffic with random car arrival interval, intended to better simulate real-world traffic:

This plot compares how different hyperparameter settings affect model performance, tested on:

Two models were compared:
target_update_interval = 1000 (instead of 500)The comparison highlights the importance of choosing proper hyperparameters.


If you want to train, evaluate, or use these models yourself, see the repository: