Update app.py

app.py

@@ -15,23 +15,25 @@ def classify_image(image):
     # Convert the PIL Image to a format compatible with the feature extractor
     image = np.array(image)
     # Preprocess the image and prepare it for the model
-    inputs = feature_extractor(images=image, return_tensors="pt")
+    inputs_pneumonia = feature_extractor(images=image, return_tensors="pt")
+    inputs_tuberculosis = feature_extractor(images=image, return_tensors="pt")
     # Make prediction
     with torch.no_grad():
-        outputs_pneumonia = model_pneumonia(**inputs)
+        outputs_pneumonia = model_pneumonia(**inputs_pneumonia)
         logits_pneumonia = outputs_pneumonia.logits
+        outputs_tuberculosis = model_tuberculosis(**inputs_tuberculosis)
+        logits_tuberculosis = outputs_tuberculosis.logits
     # Retrieve the highest probability class label index
     predicted_class_idx_pneumonia = logits_pneumonia.argmax(-1).item()
+    predicted_class_idx_tuberculosis = logits_tuberculosis.argmax(-1).item()
     # Define a manual mapping of label indices to human-readable labels
-    index_to_label = {
-        0: "NORMAL",
-        1: "PNEUMONIA"
-    }
-
+    index_to_label_pneumonia = {0: "NORMAL",1: "PNEUMONIA"}
+    index_to_label_tuberculosis = {0: "NORMAL",1: "TUBERCULOSIS"}
     # Convert the index to the model's class label
-    label_pneumonia = index_to_label.get(predicted_class_idx_pneumonia, "Unknown Label")
+    label_pneumonia = index_to_label_pneumonia.get(predicted_class_idx_pneumonia, "Unknown Label")
+    label_tuberculosis = index_to_label_tuberculosis.get(predicted_class_idx_tuberculosis, "Unknown Label")
 
-    return label_pneumonia
+    return label_pneumonia, label_tuberculosis
 
 # Create title, description and article strings
 title = "Classification Demo"