Panjuoffset TwoClass

🚀 PanJu Offset Detection by Image

This project uses fine - tuning from google/vit - base - patch16 - 224 to detect PanJu offsets in images, aiming to provide an effective solution for image - based offset detection.

🚀 Quick Start

This model is fine - tuned from google/vit - base - patch16 - 224(https://huggingface.co/google/vit - base - patch16 - 224) for PanJu offset detection.

📦 Dataset

DatasetDict({
    train: Dataset({
        features: ['image', 'label'],
        num_rows: 329
    })
    validation: Dataset({
        features: ['image', 'label'],
        num_rows: 56
    })
})

In the training set, there are 36 break samples and 293 normal samples. In the validation set, there are 5 break samples and 51 normal samples.

💻 Usage Examples

Basic Usage

Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:

# Load image
import torch
from transformers import ViTFeatureExtractor, ViTForImageClassification,AutoModel
from PIL import Image
import requests
url='https://datasets-server.huggingface.co/assets/ShihTing/IsCausewayOffset/--/ShihTing--IsCausewayOffset/validation/0/image/image.jpg'
image = Image.open(requests.get(url, stream=True).raw)
# Load model
from transformers import AutoFeatureExtractor, AutoModelForImageClassification
device = torch.device('cpu')
extractor = AutoFeatureExtractor.from_pretrained('ShihTing/PanJuOffset_TwoClass')
model = AutoModelForImageClassification.from_pretrained('ShihTing/PanJuOffset_TwoClass')
# Predict
inputs = extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
Prob = outputs.logits.softmax(dim=-1).tolist()
print(Prob)
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])

📄 License

This project is licensed under the Apache - 2.0 license.