dronefreak/bdd100k-weather-efficientvit_b0

EfficientViT-B0 Finetuned on BDD100K Weather Classification

Fine-tuned EfficientViT-B0 image classifier on the BDD100K Weather Classification dataset, trained and evaluated as part of BDD100K-Toolkit, a dependency-clean toolkit for preparing BDD100K, training models on it and evaluating them with the same metrics on the same splits.

7-class weather classification (clear / partly cloudy / overcast / rainy / snowy / foggy / unknown) derived from BDD100K's per-image attributes.weather field. Unofficial task; follows the Kaggle dataset of the same name.

---

Usage

import timm, torch
from huggingface_hub import hf_hub_download
from PIL import Image
from torchvision import transforms as T

ckpt = torch.load(
    hf_hub_download("dronefreak/bdd100k-weather-efficientvit_b0", "best.pt"), map_location="cpu", weights_only=True
)
model = timm.create_model(
    ckpt["model_name"], pretrained=False, num_classes=len(ckpt["class_names"])
)
model.load_state_dict(ckpt["state_dict"])
model.eval()
prep = T.Compose(
    [T.Resize((ckpt["imgsz"],) * 2), T.ToTensor(), T.Normalize(ckpt["mean"], ckpt["std"])]
)
with torch.no_grad():
    probs = model(prep(Image.open("street.jpg").convert("RGB"))[None]).softmax(1)[0]
print(ckpt["class_names"][probs.argmax()], f"{probs.max():.1%}")

---

Results

Evaluated on the test split (10000 images).

Metric	Value
Macro F1	65.10%
Balanced accuracy	63.62%
Macro precision	67.11%
Macro recall	63.62%
Top-1 accuracy	82.79%
Top-5 accuracy	99.79%

Per class

Class	Precision	Recall	F1	Test images
clear	89.67%	93.49%	91.54%	5346
foggy	0.00%	0.00%	0.00%	13 (few)
overcast	69.24%	65.78%	67.47%	1239
partly cloudy	68.19%	64.50%	66.30%	738
rainy	86.10%	69.65%	77.00%	738
snowy	86.24%	72.56%	78.81%	769
unknown	70.34%	79.34%	74.57%	1157

This checkpoint classifies none of the test images of foggy (13 images) correctly, so treat it as unsupported.

---

Model Zoo

All runs below were evaluated on the same test split, sorted by top-1 accuracy.

Model	Top-1	Macro F1	Balanced acc	Macro precision
ConvNeXt-Atto	83.00%	67.44%	65.25%	81.38%
EfficientViT-B0	82.79%	65.10%	63.62%	67.11%
ResNet-18	82.19%	64.30%	62.97%	66.08%
MobileNetV4-Conv-Small	82.16%	66.51%	64.37%	80.44%
YOLO26n	82.04%	64.09%	62.46%	66.57%
YOLO11n	81.32%	62.91%	61.19%	65.68%
YOLOv8n	81.22%	63.07%	61.42%	65.52%

---

Training

Setting	Value
Epochs (max)	50
Epochs (trained)	17
Best epoch (best.pt)	7
best.pt chosen by	macro_f1 on the valid split
Early stopping patience	10
Batch size	128
Image size	224
Optimizer	auto, resolved to AdamW (peak lr 3e-04)
Weights	EMA

---

Dataset

dronefreak/BDD100K-Weather-Classification holds the prepared splits these models were trained and evaluated on.

---

Limitations

• Unofficial task: labels are BDD100K's per-image attributes, not a benchmark with a public leaderboard, so scores are only comparable with other models evaluated on this split.
• Not the official test set: test here is BDD100K's official validation split (the official test labels are not released) and valid is a seeded 15% slice of the official train split.
• Heavily imbalanced: rare classes have very few test images, so their per-class scores are noisy. Prefer macro F1 over accuracy.
• Images are US dashcam frames; generalization to other regions or camera setups is untested.
• BDD100K is released for non-commercial research and education. Check its license before any use of these weights beyond research.
• clear, partly cloudy and overcast are subjective annotator calls, so expect label noise between neighbouring classes.

---

License

The weights are released under the license in the metadata above. They were trained on BDD100K, which is free for non-commercial research and education; commercial use needs separate permission (see https://www.bdd100k.com/). The prepared dataset on the Hub is tagged license: other, and the original BDD100K terms still apply.

---

Citation

Model

@article{cai2022efficientvit,
  title={EfficientViT: Multi-Scale Linear Attention for High-Resolution Dense Prediction},
  author={Cai, Han and Li, Junyan and Hu, Muyan and Gan, Chuang and Han, Song},
  journal={arXiv preprint arXiv:2205.14756},
  year={2022}
}

Dataset

@article{yu2018bdd100k,
  title={BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning},
  author={Yu, Fisher and Chen, Haofeng and Wang, Xin and Xian, Wenqi and Chen, Yingying and Liu, Fangchen and Madhavan, Vashisht and Darrell, Trevor},
  journal={arXiv preprint arXiv:1805.04687},
  year={2018}
}

---

Files

• best.pt
• metrics.json
• results.csv
• assets/demo_banner.mp4
• assets/demo_banner_poster.jpg
• README.md

---

Reproduce

Trained and evaluated with BDD100K-Toolkit: bdd100k-evaluate --dataset bdd100k-weather --checkpoint <weights> --data-dir <prepared dir> --split test.