|
--- |
|
license: mit |
|
datasets: |
|
- deepghs/monochrome_danbooru |
|
pipeline_tag: image-classification |
|
library_name: dghs-imgutils |
|
tags: |
|
- art |
|
- onnx |
|
- safetensors |
|
- emb_attachments |
|
- wdtaggers |
|
--- |
|
|
|
This is an experiment on DeepGHS's new tagger attachment training framework [deepghs/emb_attachments](https://github.com/deepghs/emb_attachments). |
|
|
|
Just some technical validation, not recommended for use in production. |
|
|
|
(2025.3.7: This feature is still WIP, you can take a look at [this PR](https://github.com/deepghs/imgutils/pull/152) and [this documentation](https://dghs-imgutils.deepghs.org/dev/attachments/api_doc/tagging/wd14.html#get-wd14-tags), you will like it.) |
|
|
|
| Name | Encoder | Problem | FLOPS | Params | AUC | accuracy | loss | mAP | Labels | |
|
|:----------------:|:----------------------------------------:|:--------------:|:-------:|:--------:|:------:|:----------:|:-------:|:-----:|:------------------:| |
|
| mlp_layer1_seed1 | wdtagger:SmilingWolf/wd-swinv2-tagger-v3 | classification | 525.3K | 525.8K | 0.9948 | 0.9652 | 0.06904 | 0.995 | monochrome, normal | |
|
|
