vit base nsfw detector

from transformers import pipeline
from PIL import Image

img = Image.open("<path_to_image_file>")
predict = pipeline("image-classification", model="AdamCodd/vit-base-nsfw-detector")
predict(img)

from transformers import ViTImageProcessor, AutoModelForImageClassification
from PIL import Image
import requests

url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
processor = ViTImageProcessor.from_pretrained('AdamCodd/vit-base-nsfw-detector')
model = AutoModelForImageClassification.from_pretrained('AdamCodd/vit-base-nsfw-detector')
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
Predicted class: sfw

/* Instructions:
- Place this script in an HTML file using the <script type="module"> tag.

- Ensure the HTML file is served over a local or remote server (e.g., using Python's http.server, Node.js server, or similar).

- Replace 'https://example.com/path/to/image.jpg' in the classifyImage function call with the URL of the image you want to classify.

*
Example of how to include this script in HTML:

<script type="module" src="path/to/this_script.js"></script>

*
This setup ensures that the script can use imports and perform network requests without CORS issues.

*/
import { pipeline, env } from 'https://cdn.jsdelivr.net/npm/@xenova/[email protected]';

// Since we will download the model from HuggingFace Hub, we can skip the local model check
env.allowLocalModels = false;
// Load the image classification model
const classifier = await pipeline('image-classification', 'AdamCodd/vit-base-nsfw-detector');
// Function to fetch and classify an image from a URL
async function classifyImage(url) {
  try {
    const response = await fetch(url);
    if (!response.ok) throw new Error('Failed to load image');
const blob = await response.blob();
    const image = new Image();
    const imagePromise = new Promise((resolve, reject) => {
      image.onload = () => resolve(image);
      image.onerror = reject;
      image.src = URL.createObjectURL(blob);
    });
const img = await imagePromise; // Ensure the image is loaded
    const classificationResults = await classifier([img.src]); // Classify the image
    console.log('Predicted class: ', classificationResults[0].label);
  } catch (error) {
    console.error('Error classifying image:', error);
  }
}
// Example usage
classifyImage('https://example.com/path/to/image.jpg');
// Predicted class: sfw