Update app.py

app.py

@@ -53,13 +53,13 @@ def dice_coef(y_true, y_pred, smooth=1):
 seg_model = tf.keras.models.load_model('seg_unet_model.h5', custom_objects={'Combo_loss': Combo_loss, 'dice_coef': dice_coef})
 
 inputs = gr.inputs.Image(type="pil", label="Upload an image", source="upload")
-image_output = gr.outputs.Image(type="pil", label="Output Image")
+image_output = gr.outputs.Image(type="numpy", label="Output Image")
 # outputs = gr.outputs.HTML() #uncomment for single class output 
 
 def gen_pred(img=inputs, model=seg_model):
     # rgb_path = os.path.join(test_image_dir,img)
     # img = cv2.imread(rgb_path)
-    # img = cv2.imread("./003e2c95d.jpg")
+    img = cv2.imread("./003e2c95d.jpg")
     # pil_image = Image.open('./003b50a15.jpg')
     # img = cv2.cvtColor(np.array(pil_image), cv2.COLOR_RGB2BGR)
     img = img[::IMG_SCALING[0], ::IMG_SCALING[1]]
@@ -72,8 +72,8 @@ def gen_pred(img=inputs, model=seg_model):
     # color_coverted = cv2.cvtColor(pred, cv2.COLOR_BGR2RGB)
     # pil_image = Image.fromarray(pred)
     PIL_image = Image.fromarray(pred.astype('uint8'), 'RGB')
-    return "UI in developing process ..."
-
+    # return "UI in developing process ..."
+    return PIL_image
 
 title = "<h1 style='text-align: center;'>Semantic Segmentation</h1>"
 description = "Upload an image and get prediction mask"
@@ -81,7 +81,7 @@ description = "Upload an image and get prediction mask"
 
 gr.Interface(fn=gen_pred, 
              inputs=inputs, 
-             outputs="text", 
+             outputs="image", 
              title=title, 
              examples=[["003e2c95d.jpg"], ["003b50a15.jpg"], ["003b48a9e.jpg"], ["0038cbe45.jpg"], ["00371aa92.jpg"]],
              # css=css_code,