animetimm
/

swinv2_base_window8_256.dbv4a-fullxx-cls

Image Classification

Model card Files Files and versions

Metrics Training metrics Community

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

Anime Classifier swinv2_base_window8_256.dbv4a-fullxx-cls

Model Details

Model Type: Single-Label Image classification / feature backbone
Model Stats:
- Params: 96.6M
- FLOPs / MACs: 121.6G / 60.7G
- Image size: train = 448 x 448, test = 448 x 448
Dataset: deepghs/danbooru-wdtagger-v4a-w640-ws-fullxx-cls
- Labels Count: 9453

Results

#	Top-1	Top-5	Macro (F1/P/R)	Micro (F1/P/R)
Validation	90.49%	95.44%	0.881 / 0.901 / 0.885	0.905 / 0.905 / 0.905
Test	90.36%	95.34%	0.880 / 0.901 / 0.884	0.904 / 0.904 / 0.904

You can find label list in selected_tags.csv.

Downloads last month: 0

Safetensors

Model size

96.6M params

Tensor type

F32

·

Inference Providers NEW

Image Classification

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for animetimm/swinv2_base_window8_256.dbv4a-fullxx-cls

Base model

animetimm/swinv2_base_window8_256.dbv4a-fullxx

Quantized

(1)

this model