Update app.py

app.py

@@ -21,16 +21,7 @@ labels = categories.readline().split(";")
 radio = gr.inputs.Radio(models_name, default="DenseNet", type="value")
 
 def predict_image(image, model_name):
-
-
-    print("======================")
-    print(type(image))
-    print(type(model_name))
-    print("==========")
-    print(image)
-    print(model_name)
-    print("======================")
-
+    # model create by keras 
     if model_name == "DenseNet":
         image = np.array(image) / 255
         image = np.expand_dims(image, axis=0)
@@ -39,8 +30,9 @@ def predict_image(image, model_name):
         pred = model.predict(image)
 
         pred = dict((labels[i], "%.2f" % pred[0][i]) for i in range(len(labels)))
-    else:
         
+    # model create by HugsVision
+    else:
         image = Image.fromarray(np.uint8(image)).convert('RGB')
         classifier = TorchVisionClassifierInference(
             model_path = "./models/" + model_name
@@ -58,7 +50,7 @@ def predict_image(image, model_name):
 image = gr.inputs.Image(shape=(300, 300), label="Upload Your Image Here")
 label = gr.outputs.Label(num_top_classes=len(labels))
 
-samples = [["samples/" + p + ".jpg"] for p in labels]
+samples = [["samples/" + p + ".jpg", p] for p in labels]
         
 interface = gr.Interface(
     fn=predict_image,