Update app.py

app.py

@@ -4,12 +4,38 @@ import torch
 from torchvision import transforms
 from transformers import AutoModelForImageClassification
 
-# Load the model (no AutoImageProcessor since it is unsupported)
-model_name = "anismizi/skin-type-classifier"
-model = AutoModelForImageClassification.from_pretrained(model_name)
-model.eval()  # Set model to eval mode
+# Define model names and corresponding labels
+MODEL_CONFIGS = [
+    {
+        "name": "anismizi/skin-type-classifier",
+        "labels": ["dry", "oily"],
+        "key": "oil_vs_dry"
+    },
+    {
+        "name": "naamalia23/acne-severity-classification",      
+        "labels": ["no_acne", "acne"],
+        "key": "acne"
+    },
+    {
+        "name": "Siraja704/DermaAI",   
+        "labels": ["no_redness", "redness"],
+        "key": "redness"
+    },
+    {
+        "name": "imfarzanansari/skintelligent-wrinkles",
+        "labels": ["no_wrinkles", "wrinkles"],
+        "key": "wrinkles"
+    },
+]
 
-# Define manual preprocessing transforms similar to ResNet50 training
+# Load all models at startup
+MODELS = []
+for config in MODEL_CONFIGS:
+    model = AutoModelForImageClassification.from_pretrained(config["name"])
+    model.eval()
+    MODELS.append(model)
+
+# Common preprocessing (adjust if any model requires different input specs)
 preprocess = transforms.Compose([
     transforms.Resize(256),
     transforms.CenterCrop(224),
@@ -18,37 +44,33 @@ preprocess = transforms.Compose([
                          std=[0.229, 0.224, 0.225]),
 ])
 
-# Labels according to model info
-labels = ["dry", "oily"]
-
 def analyze_skin(image: Image.Image):
-    # Convert input image to RGB
     image = image.convert("RGB")
-    # Preprocess image
     input_tensor = preprocess(image)
-    # Add batch dimension
-    input_batch = input_tensor.unsqueeze(0)
+    input_batch = input_tensor.unsqueeze(0)  # add batch dimension
 
-    # Run inference
+    results = {}
     with torch.no_grad():
-        outputs = model(input_batch)
-        logits = outputs.logits
-        probabilities = torch.nn.functional.softmax(logits, dim=1)
-        confidence, predicted_idx = torch.max(probabilities, dim=1)
-    
-    predicted_label = labels[predicted_idx.item()]
-    confidence_score = confidence.item()
-
-    # Format results for display and API output
-    return {predicted_label: confidence_score}
+        for idx, config in enumerate(MODEL_CONFIGS):
+            model, labels, key = MODELS[idx], config["labels"], config["key"]
+            outputs = model(input_batch)
+            logits = outputs.logits
+            probs = torch.softmax(logits, dim=1)
+            confidence, pred_idx = torch.max(probs, dim=1)
+            predicted_label = labels[pred_idx.item()]
+            confidence_score = confidence.item()
+            results[key] = {
+                "label": predicted_label,
+                "confidence": f"{confidence_score:.2%}"
+            }
+    return results
 
-# Create Gradio interface
 iface = gr.Interface(
     fn=analyze_skin,
     inputs=gr.Image(type="pil"),
-    outputs=gr.Label(num_top_classes=2),  
-    title="Skin Condition Analyzer",
-    description="Classify skin as dry or oily from an image."
+    outputs=gr.JSON(label="Skin Analysis Results"),
+    title="Comprehensive Skin Condition Analyzer",
+    description="Classifies skin image for oily/dry, acne, redness, wrinkles using multiple models."
 )
 
 if __name__ == "__main__":