Downloads · 30 days
0
tms-rl-traffic-controller
Kunalsinghh/tms-rl-traffic-controller
tms-rl-traffic-controller is a reinforcement learning model from Kunalsinghh. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Deep Q-Network (DQN) based agents for adaptive traffic signal control.
Access
Public
Updated Dec 24, 2025
Repo size
39.5 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt61 MB · 97%
At a glance
- Task
- Reinforcement Learning
- License
- apache-2.0
- Access
- Public
- Created
- Dec 24, 2025
- Updated
- Dec 24, 2025
- SHA
- c2154048
Model card
From the Hugging Face model README
TMS2 - RL Traffic Management Models
RL Traffic Signal Controller
Deep Q-Network (DQN) based agents for adaptive traffic signal control.
Variants:
- v2: Baseline stable model optimized for throughput
- v3: Eco-friendly model with emissions optimization (14% CO2 reduction)
- v4: Challenging scenarios with incidents and demand spikes
- v5/final: Curriculum learning for generalization
Architecture:
- Dueling Double DQN with soft target updates
- State space: 10-12 dimensions (queues, phase, time, scenarios)
- Action space: 2 (keep phase / change phase)
Model Description
These models are part of the Traffic Management System 2 (TMS2) project, an intelligent traffic control system using deep learning and reinforcement learning.
Training Details
- Framework: PyTorch
- Training Platform: Google Colab (T4 GPU)
- Training Date: December 2025
Usage
import torch
# Load model
model = torch.load('model.pt')
model.eval()
# Inference
with torch.no_grad():
output = model(input_tensor)
License
Apache 2.0
- Task
- Reinforcement Learning
- License
- apache-2.0
- Created
- Dec 24, 2025
- Updated
- Dec 24, 2025