Anime Classifier just-a-try/cls-dbv4-full.rotate.mobilenetv4_conv_small_050

Model Details

• Model Type: Image Classification
• Model Stats:
  • Params: 963.1K
  • FLOPs / MACs: 380.8M / 185.3M
  • Image size: train = 384 x 384, test = 384 x 384
• Dataset: animetimm/danbooru-wdtagger-v4-w640-ws-full
  • Classes: r0, r90, r180, r270

Results

Metrics

#	Acc / Top-2	Macro (F1/P/R/AUC)	Micro (F1/P/R/AUC)
Validation	97.86% / 99.54%	0.979 / 0.979 / 0.979 / 0.999	0.979 / 0.979 / 0.979 / 0.999
Test	97.86% / 99.57%	0.979 / 0.979 / 0.979 / 0.999	0.979 / 0.979 / 0.979 / 0.999

Plots

#	Confusion	P/R	F1
Validation
Test

How to Use

We provided a sample image for our code samples, you can find it here.

Use Transformers And Torch

Install dghs-imgutils, timm and other necessary requirements with the following command

pip install 'dghs-imgutils>=0.19.0' torch huggingface_hub timm pillow 'transformers>=4.57.1'

After that you can load this model with timm library, and use it for train, validation and test, with the following code

import torch
from imgutils.data import load_image
from transformers import AutoImageProcessor, AutoModel

processor = AutoImageProcessor.from_pretrained('just-a-try/cls-dbv4-full.rotate.mobilenetv4_conv_small_050', trust_remote_code=True)
model = AutoModel.from_pretrained('just-a-try/cls-dbv4-full.rotate.mobilenetv4_conv_small_050', trust_remote_code=True, use_infer_head=True)
model.eval()

image = load_image('https://huggingface.co/just-a-try/cls-dbv4-full.rotate.mobilenetv4_conv_small_050/resolve/main/sample.webp', mode='RGB', force_background='white')
input_ = processor(image)['pixel_values']
# input_, shape: torch.Size([1, 3, 384, 384]), dtype: torch.float32
classes = model.config.classes
# ['r0', 'r90', 'r180', 'r270']

with torch.no_grad():
    output = model(input_)
# output, shape: torch.Size([1, 4]), dtype: torch.float32

print(dict(zip(classes, output[0].tolist())))
# {'r0': 0.0009245745022781193,
#  'r90': 0.001419938518665731,
#  'r180': 0.95500648021698,
#  'r270': 0.04264896735548973}

Citation

@misc{cls_dbv4_full_rotate_mobilenetv4_conv_small_050,
  title        = {Anime Classifier just-a-try/cls-dbv4-full.rotate.mobilenetv4_conv_small_050},
  author       = {narugo1992 and Deep Generative anime Hobbyist Syndicate (DeepGHS)},
  year         = {2025},
  howpublished = {\url{https://huggingface.co/just-a-try/cls-dbv4-full.rotate.mobilenetv4_conv_small_050}},
  note         = {A anime-style image classification model for classification task with 4 classes (r0, r90, r180, r270), trained on anime dataset dbv4-full (\url{https://huggingface.co/datasets/animetimm/danbooru-wdtagger-v4-w640-ws-full}). Model parameters: 963.1K, FLOPs: 380.8M, input resolution: 384×384.},
  license      = {mit}
}