Update app.py

app.py

@@ -1,23 +1,45 @@
 import gradio as gr
-from transformers import pipeline
+from transformers import AutoImageProcessor
+from transformers import SiglipForImageClassification
+from transformers.image_utils import load_image
 from PIL import Image
+import torch
 
-# Load the pipeline for age classification
-pipe = pipeline("image-classification", model="prithivMLmods/Age-Classification-SigLIP2")
+# Load model and processor
+model_name = "prithivMLmods/Age-Classification-SigLIP2"
+model = SiglipForImageClassification.from_pretrained(model_name)
+processor = AutoImageProcessor.from_pretrained(model_name)
 
-# Define the prediction function
-def predict(input_img):
-    # Get the predictions
-    predictions = pipe(input_img)
-    # Format the predictions into a human-readable string
-    result_str = "\n".join([f"{p['label']}: {p['score']:.4f}" for p in predictions])
-    return result_str
+def age_classification(image):
+    """Predicts the age group of a person from an image."""
+    image = Image.fromarray(image).convert("RGB")
+    inputs = processor(images=image, return_tensors="pt")
+    
+    with torch.no_grad():
+        outputs = model(**inputs)
+        logits = outputs.logits
+        probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist()
+    
+    labels = {
+        "0": "Child 0-12", 
+        "1": "Teenager 13-20", 
+        "2": "Adult 21-44", 
+        "3": "Middle Age 45-64", 
+        "4": "Aged 65+"
+    }
+    predictions = {labels[str(i)]: round(probs[i], 3) for i in range(len(probs))}
+    
+    return predictions
 
-# Create a Gradio interface
-iface = gr.Interface(fn=predict,
-                     inputs=gr.Image(type="pil"),  # Define input type as an image
-                     outputs=gr.Textbox(label="Class Confidence Scores", interactive=False),  # Output as plain text
-                     )  # Set live=True to update results as soon as the image is uploaded
+# Create Gradio interface
+iface = gr.Interface(
+    fn=age_classification,
+    inputs=gr.Image(type="numpy"),
+    outputs=gr.Label(label="Prediction Scores"),
+    title="Age Group Classification",
+    description="Upload an image to predict the person's age group."
+)
 
-# Launch the Gradio app
-iface.launch()
+# Launch the app
+if __name__ == "__main__":
+    iface.launch()