README.md

metadata

library_name: transformers
license: apache-2.0
base_model: SmilingWolf/wd-swinv2-tagger-v3
inference: false
tags:
  - wd-tagger

WD SwinV2 Tagger v3 with 🤗 transformers

Converted from SmilingWolf/wd-swinv2-tagger-v3 to transformers library format.

Example

Pipeline

from transformers import pipeline

pipe = pipeline(
    "image-classification",
    model="p1atdev/wd-swinv2-tagger-v3-hf",
    trust_remote_code=True,
)

print(pipe("sample.webp", top_k=15))
#[{'label': '1girl', 'score': 0.9973934888839722},
# {'label': 'solo', 'score': 0.9719744324684143},
# {'label': 'dress', 'score': 0.9539461135864258},
# {'label': 'hat', 'score': 0.9511678218841553},
# {'label': 'outdoors', 'score': 0.9438753128051758},
# {'label': 'sky', 'score': 0.9195725917816162},
# {'label': 'sitting', 'score': 0.9178725481033325},
# {'label': 'looking up', 'score': 0.9122412800788879},
# {'label': 'short hair', 'score': 0.8630313873291016},
# {'label': 'cloud', 'score': 0.8609118461608887},
# {'label': 'brown hair', 'score': 0.7723952531814575},
# {'label': 'short sleeves', 'score': 0.7649227380752563},
# {'label': 'day', 'score': 0.7641971111297607},
# {'label': 'rating:general', 'score': 0.7605368494987488},
# {'label': 'white dress', 'score': 0.7596388459205627}]

AutoModel

from PIL import Image

import numpy as np
import torch

from transformers import (
    AutoImageProcessor,
    AutoModelForImageClassification,
)

MODEL_NAME = "p1atdev/wd-swinv2-tagger-v3-hf"

model = AutoModelForImageClassification.from_pretrained(
    MODEL_NAME,
)
processor = AutoImageProcessor.from_pretrained(MODEL_NAME, trust_remote_code=True)

image = Image.open("sample.webp")
inputs = processor.preprocess(image, return_tensors="pt")

with torch.no_grad():
  outputs = model(**inputs.to(model.device, model.dtype))
logits = torch.sigmoid(outputs.logits[0]) # take the first logits

# get probabilities
results = {model.config.id2label[i]: logit.float() for i, logit in enumerate(logits)}
results = {
    k: v for k, v in sorted(results.items(), key=lambda item: item[1], reverse=True) if v > 0.35 # 35% threshold
}
print(results)  # rating tags and character tags are also included
#{'1girl': tensor(0.9974),
# 'solo': tensor(0.9720),
# 'dress': tensor(0.9539),
# 'hat': tensor(0.9512),
# 'outdoors': tensor(0.9439),
# 'sky': tensor(0.9196),
# 'sitting': tensor(0.9179),
# 'looking up': tensor(0.9122),
# 'short hair': tensor(0.8630),
# 'cloud': tensor(0.8609),
# 'brown hair': tensor(0.7724),
# 'short sleeves': tensor(0.7649),
# 'day': tensor(0.7642),
# 'rating:general': tensor(0.7605),
# ...

Labels

All of rating tags have prefix rating: and character tags have prefix character:.

• Rating tags: rating:general, rating:sensitive, ...
• Character tags: character:frieren, character:hatsune miku, ...