Aller au contenu
ModelPrésentéExterne
Télécharger

Yolo26n Coco128 Finetune

This repository contains best.pt, a YOLO26n object-detection model fine-tuned on the COCO128 dataset.

Sourcehuggingface.co/deki/yolo26n-coco128-finetune/tree/d4f684d1674aa6700d914283a566d5dd017d4052
TypeOtherVersionRevision d4f684dBClasse

Source du modèle

Type
Other
Taille du fichier
5,24 MB
Version
vRevision d4f684d
Mis à jour
1 nov. 2026
Téléchargements
26
Source
Hugging Face

Description de la source

YOLO26n COCO128 Fine-Tuned Object Detection

This repository contains best.pt, a YOLO26n object-detection model fine-tuned on the COCO128 dataset.

Training

The model was initialized from pretrained YOLO26n weights and fine-tuned on COCO128 using Ultralytics.

Training settings:

  • Base model: yolo26n.pt
  • Dataset: coco128.yaml
  • Epochs: 50
  • Image size: 512
  • Batch size: 4
  • Device: Apple MPS
  • Run name: yolo26n_coco128_finetune_v2

Results

Best validation metrics from the training run:

  • mAP50: 0.731
  • mAP50-95: 0.558

Usage

from ultralytics import YOLO

model = YOLO("best.pt")
results = model("image.jpg")
results[0].show()

Sources

1 source
Hugging FacePrincipalVersion d4f684d1674aa6700d914283a566d5dd017d4052
huggingface.co/deki/yolo26n-coco128-finetune/tree/d4f684d1674aa6700d914283a566d5dd017d4052

Vérifié 1 nov.

Artefacts du modèle

1 artefact

best.pt

pt · 5,24 MB · SHA-256 188c52723f2b…ba00 · Hugging Face

Download file

Extraits de sources

3 extraits
Readme18 aoûthuggingface.co/deki/yolo26n-coco128-finetune

--- license: agpl-3.0 library_name: ultralytics tags: - yolo - yolo26 - object-detection - coco128 - fine-tuned datasets: - coco128 --- # YOLO26n COCO128 Fine-Tuned Object Detection This repository contains `best.pt`, a YOLO26n object-detection model fine-tuned on the COCO128 dataset. ## Training The model was initialized from pretrained YOLO26n weights and fine-tuned on COCO128 using Ultralytics. Training settings: - Base model: `yolo26n.pt` - Dataset: `coco128.yaml` - Epochs: 50 - Image size: 512 - Batch size: 4 - Device: Apple MPS - Run name: `yolo26n_coco128_finetune_v2` ## Results Best validation metrics from the training run: - mAP50: 0.731 - mAP50-95: 0.558 ## Usage ```python from ultralytics import YOLO model = YOLO("best.pt") results = model("image.jpg") results[0].show()

Note1 nov.huggingface.co/deki/yolo26n-coco128-finetune

Source context: 26 downloads · 1 likes · Pipeline object-detection · Library ultralytics · Repo deki/yolo26n-coco128-finetune

Note18 aoûthuggingface.co/deki/yolo26n-coco128-finetune

Source context: 11 downloads · 0 likes · Pipeline object-detection · Library ultralytics · Repo deki/yolo26n-coco128-finetune