说明文档

这是一个用于预测动漫风格 Stable Diffusion 生成图像是否损坏的分类器模型。

模型

名称	FLOPS	参数量	准确率	AUC	混淆矩阵	标签
caformer_s36_v0_focal	22.10G	37.21M	95.65%	0.9916	confusion	`corrupted`, `normal`
caformer_s36_v0_sce	22.10G	37.21M	95.77%	0.9894	confusion	`corrupted`, `normal`
mobilenetv3_v0_focal_dist	0.63G	4.18M	94.02%	0.9842	confusion	`corrupted`, `normal`

使用方法

你可以使用 dghs-imgutils 来调用此模型。

from imgutils.generic import classify_predict_score

classify_predict_score(
    'sample_image.png',
    repo_id='deepghs/ai_image_corrupted',
    model_name='mobilenetv3_v0_focal_dist',
)
# {'corrupted': 0.7807788848876953, 'normal': 0.2192210853099823}

注意

此模型基于 SD1.5 图像训练，由 8 个不同的基础模型生成。

解决此问题的更好方案是使用艺术风格嵌入等技术指标，一旦动漫图像的艺术风格嵌入模型完成，此模型将被弃用。

引用

@misc{Citation,
  title={AI-Corrupt Score for Anime Images},
  author={narugo1992},
  year={2023},
  howpublished={\url{https://huggingface.co/deepghs/ai_image_corrupted}}
}

deepghs/ai_image_corrupted

作者 deepghs

image-classification dghs-imgutils

↓ 0 ♥ 2

创建时间: 2023-11-08 04:20:12+00:00

更新时间: 2024-09-28 04:03:30+00:00

在 Hugging Face 上查看

文件 (38)

.gitattributes

README.md

caformer_s36_v0_focal/meta.json

caformer_s36_v0_focal/metrics.json

caformer_s36_v0_focal/model.ckpt

caformer_s36_v0_focal/model.onnx ONNX

caformer_s36_v0_focal/plot_confusion.png

caformer_s36_v0_focal/plot_f1_curve.png

caformer_s36_v0_focal/plot_p_curve.png

caformer_s36_v0_focal/plot_pr_curve.png

caformer_s36_v0_focal/plot_r_curve.png

caformer_s36_v0_focal/plot_roc_curve.png

caformer_s36_v0_focal/plot_sample_corrupted.png

caformer_s36_v0_focal/plot_sample_normal.png

caformer_s36_v0_sce/meta.json

caformer_s36_v0_sce/metrics.json

caformer_s36_v0_sce/model.ckpt

caformer_s36_v0_sce/model.onnx ONNX

caformer_s36_v0_sce/plot_confusion.png

caformer_s36_v0_sce/plot_f1_curve.png

caformer_s36_v0_sce/plot_p_curve.png

caformer_s36_v0_sce/plot_pr_curve.png

caformer_s36_v0_sce/plot_r_curve.png

caformer_s36_v0_sce/plot_roc_curve.png

caformer_s36_v0_sce/plot_sample_corrupted.png

caformer_s36_v0_sce/plot_sample_normal.png

mobilenetv3_v0_focal_dist/meta.json

mobilenetv3_v0_focal_dist/metrics.json

mobilenetv3_v0_focal_dist/model.ckpt

mobilenetv3_v0_focal_dist/model.onnx ONNX

mobilenetv3_v0_focal_dist/plot_confusion.png

mobilenetv3_v0_focal_dist/plot_f1_curve.png

mobilenetv3_v0_focal_dist/plot_p_curve.png

mobilenetv3_v0_focal_dist/plot_pr_curve.png

mobilenetv3_v0_focal_dist/plot_r_curve.png

mobilenetv3_v0_focal_dist/plot_roc_curve.png

mobilenetv3_v0_focal_dist/plot_sample_corrupted.png

mobilenetv3_v0_focal_dist/plot_sample_normal.png