Instructions to use deepghs/idolsankaku_tagger_with_embeddings with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- timm
How to use deepghs/idolsankaku_tagger_with_embeddings with timm:
import timm model = timm.create_model("hf_hub:deepghs/idolsankaku_tagger_with_embeddings", pretrained=True) - Notebooks
- Google Colab
- Kaggle
Idolsankaku taggers with embeddings and logits output.
How To Use
You can use these embedding-supported models with dghs-realutils library, see documentation of realutils.tagging.idolsankaku for more information.
Models
2 models exported from TIMM in total.
| Name | Params | Flops | Input Size | Features | Classes | Model | Architecture | Created At |
|---|---|---|---|---|---|---|---|---|
| idolsankaku-swinv2-tagger-v1 | 66.5M | 46.2G | 448 | 1024 | 573 | SwinTransformerV2 | swinv2_base_window8_256 | 2024-12-01 |
| idolsankaku-eva02-large-tagger-v1 | 303.1M | 697.6G | 480 | 1024 | 573 | Eva | eva02_large_patch14_448 | 2024-11-29 |
Model tree for deepghs/idolsankaku_tagger_with_embeddings
Base model
deepghs/idolsankaku-eva02-large-tagger-v1