Downloads · 30 days
0
JcProg/pedsense-yolo26m-detector-jaad-20e
pedsense-yolo26m-detector-jaad-20e is a object detection model from JcProg. Use it when you need objects located in an image. The card lists the license as agpl-3.0.
A single-class pedestrian detector fine-tuned on the JAAD (Joint Attention in Autonomous Driving) dataset using the PedSense-AI framework. Continued from the 10-epoch checkpoint for an additional 10 epochs.
Downloads · 30 days
0
Access
Public
Updated Mar 3, 2026
Repo size
88.6 MB
Likes
0
Public
Click a slice to open those files.
.pt88 MB · 99%
From the Hugging Face model README
A single-class pedestrian detector fine-tuned on the JAAD (Joint Attention in Autonomous Driving) dataset using the PedSense-AI framework. Continued from the 10-epoch checkpoint for an additional 10 epochs.
| Property | Value |
|---|---|
| Architecture | YOLO26m |
| Task | Object detection (1 class: pedestrian) |
| Dataset | JAAD |
| Total Epochs | 20 (10 initial + 10 resumed) |
| Image Size | 640x640 |
| Batch Size | 16 |
| Train/Val Split | 80/20 (video-level) |
| Metric | Value |
|---|---|
| Precision | 0.860 |
| Recall | 0.721 |
| mAP@50 | 0.807 |
| mAP@50-95 | 0.497 |
from ultralytics import YOLO
model = YOLO("weights/best.pt")
results = model("image.jpg")
The model was trained using the PedSense-AI framework:
# Initial 10 epochs
uv run pedsense preprocess frames
uv run pedsense train -m yolo-detector -n my_detector -e 10 --yolo-variant yolo26m
# Resumed for 10 more epochs
uv run pedsense resume
AGPL-3.0