--- license: mit datasets: - deepghs/anime_classification metrics: - accuracy pipeline_tag: image-classification tags: - art --- The model used to predict the types of anime images, which includes the following four categories: * 3D: Images rendered in 3D, including Mikumikudance, Koikatsu, etc. * Bangumi: Screenshots from anime videos. * Comic: Images of manga that contain a significant amount of text or panel sequences. * Illustration: General anime illustrations. * Not Painting: (Only available in new models) Any content that cannot be called a painting, such as artist promotional posts, game screenshots, chat logs, etc. | Name | FLOPS | Params | Accuracy | AUC | Confusion | Labels | |:-----------------------------:|:-------:|:--------:|:----------:|:------:|:---------------------------------------------------------------------------------------------------------------------------:|:--------------------------------------------------------:| | mobilenetv3_v1.5_dist | 0.63G | 4.18M | 94.56% | 0.9946 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_v1.5_dist/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | caformer_s36_v1.5_focal | 22.10G | 37.22M | 95.06% | 0.9957 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_v1.5_focal/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | caformer_s36_v1.4_focal_fixed | 22.10G | 37.22M | 96.21% | 0.9971 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_v1.4_focal_fixed/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | caformer_s36_v1.4_focal_fp32 | 22.10G | 37.22M | 95.98% | 0.9969 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_v1.4_focal_fp32/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | mobilenetv3_v1.4_dist | 0.63G | 4.18M | 94.77% | 0.9950 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_v1.4_dist/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | caformer_s36_v1.4_focal | 22.10G | 37.22M | 95.82% | 0.9967 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_v1.4_focal/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | mobilenetv3_v1.3_dist | 0.63G | 4.18M | 96.41% | 0.9973 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_v1.3_dist/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | caformer_s36_v1.3_focal | 22.10G | 37.22M | 97.16% | 0.9982 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_v1.3_focal/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | mobilenetv3_v1.2_dist | 0.63G | 4.18M | 96.53% | 0.9972 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_v1.2_dist/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | caformer_s36_v1.2_focal | 22.10G | 37.22M | 97.23% | 0.9982 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_v1.2_focal/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | caformer_s36_v1.1_focal | 22.10G | 37.22M | 95.99% | 0.9967 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_v1.1_focal/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | mobilenetv3_v1_dist | 0.63G | 4.18M | 94.04% | 0.9928 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_v1_dist/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | caformer_s36_v1 | 22.10G | 37.22M | 94.72% | 0.9934 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_v1/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration`, `not_painting` | | mobilenetv3_dist | 0.63G | 4.18M | 91.98% | 0.9879 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_dist/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration` | | mobilenetv3_sce_dist | 0.63G | 4.18M | 92.35% | 0.9854 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_sce_dist/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration` | | caformer_s36_plus | 22.10G | 37.22M | 93.47% | 0.9891 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_plus/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration` | | mobilenetv3 | 0.63G | 4.18M | 88.96% | N/A | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration` | | mobilevitv2_150 | 9.09G | 9.79M | 88.21% | N/A | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilevitv2_150/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration` | | mobilenetv3_sce | 0.63G | 4.18M | 89.92% | 0.9786 | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_sce/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration` | | caformer_s36 | 22.10G | 37.22M | 88.19% | N/A | [confusion](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36/plot_confusion.png) | `3d`, `bangumi`, `comic`, `illustration` | | Model | FLOPs | Accuracy | Confusion Matrix | Description | |:--------------------:|:------:|:--------:|:-------------------------------------------------------------------------------------------------------------------------:|----------------------------------------------------------------------------------| | caformer_s36 | 22.10G | 88.19% | [Confusion Matrix](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36/plot_confusion.png) | Model: caformer_s36 from timm | | caformer_s36_plus | 22.10G | 93.47% | [Confusion Matrix](https://huggingface.co/deepghs/anime_classification/blob/main/caformer_s36_plus/plot_confusion.png) | Model: caformer_s36.sail_in22k_ft_in1k_384 pratrained from timm | | mobilenetv3 | 0.63G | 88.96% | [Confusion Matrix](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3/plot_confusion.png) | Model: mobilenetv3_large_100 from timm | | mobilenetv3_dist | 0.63G | 91.98% | [Confusion Matrix](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_dist/plot_confusion.png) | Distrillated from caformer_s36_plus, using mobilenetv3_large_100 with focal loss | | mobilenetv3_sce | 0.63G | 89.92% | [Confusion Matrix](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_sce/plot_confusion.png) | Model: mobilenetv3_large_100 from timm, use SCELoss as loss function | | mobilenetv3_sce_dist | 0.63G | 92.35% | [Confusion Matrix](https://huggingface.co/deepghs/anime_classification/blob/main/mobilenetv3_sce_dist/plot_confusion.png) | Distrillated from caformer_s36_plus, using mobilenetv3_large_100 with SCELoss | | mobilevitv2_150 | 9.09G | 88.21% | [Confusion Matrix](https://huggingface.co/deepghs/anime_classification/blob/main/mobilevitv2_150/plot_confusion.png) | Model: mobilevitv2_150 from timm |