File size: 1,611 Bytes
9da5b91
758c978
 
59a3399
70a751e
758c978
 
 
 
 
 
70a751e
59a3399
a135d91
c5f5082
a135d91
c5f5082
 
452a079
 
59a3399
 
 
 
 
 
 
 
 
 
 
452a079
59a3399
01d090f
4700b60
70a751e
406e1b2
a135d91
63dbfda
758c978
 
70a751e
 
758c978
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
import gradio as gr
from transformers import ViTFeatureExtractor, ViTForImageClassification
from hugsvision.inference.VisionClassifierInference import VisionClassifierInference
from PIL import Image, ImageDraw, ImageFont

# Load the pre-trained ViT model
path = "mrm8488/vit-base-patch16-224_finetuned-kvasirv2-colonoscopy"
classifier = VisionClassifierInference(
    feature_extractor=ViTFeatureExtractor.from_pretrained(path),
    model=ViTForImageClassification.from_pretrained(path),
)


def classify_image(image_file):
    """Classify an image using a pre-trained ViT model."""
    label = classifier.predict(img_path=image_file.name)
    # Add a confidence score to the output
    confidence = classifier.predict_proba(img_path=image_file.name)[0][label]

    # Get the PIL Image object for the uploaded image
    image = Image.open(image_file)

    # Draw the predicted label on the image
    draw = ImageDraw.Draw(image)
    font = ImageFont.truetype("arial.ttf", 20)
    draw.text((10, 10), f"Predicted class: {label} (confidence: {confidence:.2f})", font=font, fill=(255, 255, 255))

    # Save the modified image to a BytesIO object
    output_image = BytesIO()
    image.save(output_image, format="JPEG")
    output_image.seek(0)

    return output_image, f"Predicted class: {label} (confidence: {confidence:.2f})"


iface = gr.Interface(
    fn=classify_image,
    inputs=gr.inputs.Image(type="filepath", label="Upload an image"),
    outputs=[gr.outputs.Image(type="numpy"), "text"],
    title="Image Classifier",
    description="Classify images using a pre-trained ViT model",
)

iface.launch()