Update app.py

app.py

@@ -17,7 +17,7 @@ from PIL import Image
 
 warnings.filterwarnings("ignore")
 
-def scale_image(input_image):
+def scale_input_image(input_image):
     original_shape = input_image.shape
     im = Image.fromarray(input_image)
     newsize = (config.IMAGE_SIZE, config.IMAGE_SIZE)
@@ -25,7 +25,7 @@ def scale_image(input_image):
     input_image_resized = np.array(im_resized)
     return input_image_resized, original_shape
 
-def convert_to_transformed_tensor(input_image):
+def get_transformed_image(input_image):
     test_transforms = A.Compose(
     [
         #A.LongestMaxSize(max_size=config.IMAGE_SIZE),
@@ -45,7 +45,7 @@ def convert_to_transformed_tensor(input_image):
 
 
 def get_detection_output(input_image, cam, targets, conf_threshold, iou_threshold, renormalize_cam, image_weight):
-    input_tensor = convert_to_transformed_tensor(input_image)
+    input_tensor = get_transformed_image(input_image)
     boxes_list, classes_list, labels_list, confidences_list = predict_single_image(input_tensor, 
                                                                                    yolo_model_trained, 
                                                                                    scaled_anchors, 
@@ -58,7 +58,7 @@ def get_detection_output(input_image, cam, targets, conf_threshold, iou_threshol
     print(confidences_list)
     grayscale_cam_input_image = cam(input_tensor, targets=targets)
 
-    input_image_np, original_shape = scale_image(input_image)
+    input_image_np, original_shape = scale_input_image(input_image)
     input_image_np = input_image_np/255.
     original_width, original_height, _ = original_shape
     
@@ -94,7 +94,7 @@ def get_detection_output(input_image, cam, targets, conf_threshold, iou_threshol
     return image_w_bb_resized, composite_img_w_bb_resized
 
 
-def get_detection_on_samples():
+def detect_samples():
     """
     Run inference on one batch of test images
     """
@@ -134,7 +134,7 @@ def get_detection_on_samples():
     return sample_detections_list, sample_grad_cam_maps_list
 
 
-def yolo_v3_interface(input_image, 
+def yolo_predictor(input_image, 
                       conf_threshold=0.6, 
                       iou_threshold=0.3,
                       renormalize_cam=False,
@@ -213,25 +213,21 @@ cam = EigenCAM(yolo_model_trained,
                target_layers,
                use_cuda=torch.cuda.is_available(),
                reshape_transform=None)
-sample_detections_list, sample_grad_cam_maps_list = get_detection_on_samples()
+sample_detections_list, sample_grad_cam_maps_list = detect_samples()
 
 # Define Interface
 description = 'Perform object detection using YOLOv3 algorithm, trained from scratch on PASCAL VOC dataset. Display the bounding boxes of the detected objects, and show the corresponding EigenCAM outputs'
 #title = 'Interface to perform Object detection using YOLOv3 algorithm'
 title = 'YOLOv3 object detection trained from scratch'
 
-yolo_examples = [['bicycle.webp', 0.6, 0.3, False, 0.8, True],
-                ['bus.jpg', 0.6, 0.3, False, 0.8, True],
-                ['dining_table.webp', 0.6, 0.3, False, 0.8, True], 
-                ['potted_plant.jpeg', 0.6, 0.3, False, 0.8, True],  
-                ['traffic.jpg', 0.6, 0.3, False, 0.8, True],
-                ['boat.webp', 0.6, 0.3, False, 0.8, True],     
-                ['car.jpg', 0.6, 0.3, False, 0.8, True],   
-                ['motorcycle.jpg', 0.6, 0.3, False, 0.8, True],   
-                ['sheep_and_horse', 0.6, 0.3, False, 0.8, True], 
-                ['train.jpeg', 0.6, 0.3, False, 0.8, True]]
-
-demo = gr.Interface(yolo_v3_interface,
+yolo_examples = [['images/cycle.jpg', 0.6, 0.3, False, 0.8, True],
+                ['images/bus.webp', 0.6, 0.3, False, 0.8, True],
+                ['images/train.jpg', 0.6, 0.3, False, 0.8, True], 
+                ['images/motorcycle.jpg', 0.6, 0.3, False, 0.8, True],  
+                ['images/boat.jpg', 0.6, 0.3, False, 0.8, True],
+                ['images/car.jpg', 0.6, 0.3, False, 0.8, True]]
+
+demo = gr.Interface(yolo_predictor,
                     inputs = [#gr.Image(shape=(config.IMAGE_SIZE, config.IMAGE_SIZE)),
                               gr.Image(),
                               gr.Slider(0,1.,0.6, label="Confidence Threshold"),