diff --git a/model_zoo/official/cv/yolov3_darknet53/eval.py b/model_zoo/official/cv/yolov3_darknet53/eval.py
index 6680b104762..f04ed2447c9 100644
--- a/model_zoo/official/cv/yolov3_darknet53/eval.py
+++ b/model_zoo/official/cv/yolov3_darknet53/eval.py
@@ -90,35 +90,35 @@ class DetectionEngine:
                             for i in keep_index]
                 self.det_boxes.extend(keep_box)
 
-    def _nms(self, dets, thresh):
+    def _nms(self, predicts, threshold):
         """Calculate NMS."""
         # conver xywh -> xmin ymin xmax ymax
-        x1 = dets[:, 0]
-        y1 = dets[:, 1]
-        x2 = x1 + dets[:, 2]
-        y2 = y1 + dets[:, 3]
-        scores = dets[:, 4]
+        x1 = predicts[:, 0]
+        y1 = predicts[:, 1]
+        x2 = x1 + predicts[:, 2]
+        y2 = y1 + predicts[:, 3]
+        scores = predicts[:, 4]
 
         areas = (x2 - x1 + 1) * (y2 - y1 + 1)
         order = scores.argsort()[::-1]
 
-        keep = []
+        reserved_boxes = []
         while order.size > 0:
             i = order[0]
-            keep.append(i)
-            xx1 = np.maximum(x1[i], x1[order[1:]])
-            yy1 = np.maximum(y1[i], y1[order[1:]])
-            xx2 = np.minimum(x2[i], x2[order[1:]])
-            yy2 = np.minimum(y2[i], y2[order[1:]])
+            reserved_boxes.append(i)
+            max_x1 = np.maximum(x1[i], x1[order[1:]])
+            max_y1 = np.maximum(y1[i], y1[order[1:]])
+            min_x2 = np.minimum(x2[i], x2[order[1:]])
+            min_y2 = np.minimum(y2[i], y2[order[1:]])
 
-            w = np.maximum(0.0, xx2 - xx1 + 1)
-            h = np.maximum(0.0, yy2 - yy1 + 1)
-            inter = w * h
-            ovr = inter / (areas[i] + areas[order[1:]] - inter)
+            intersect_w = np.maximum(0.0, min_x2 - max_x1 + 1)
+            intersect_h = np.maximum(0.0, min_y2 - max_y1 + 1)
+            intersect_area = intersect_w * intersect_h
+            ovr = intersect_area / (areas[i] + areas[order[1:]] - intersect_area)
 
-            inds = np.where(ovr <= thresh)[0]
-            order = order[inds + 1]
-        return keep
+            indexs = np.where(ovr <= threshold)[0]
+            order = order[indexs + 1]
+        return reserved_boxes
 
     def write_result(self):
         """Save result to file."""
diff --git a/model_zoo/official/cv/yolov3_darknet53/src/transforms.py b/model_zoo/official/cv/yolov3_darknet53/src/transforms.py
index 837d1a25ead..4756a141f04 100644
--- a/model_zoo/official/cv/yolov3_darknet53/src/transforms.py
+++ b/model_zoo/official/cv/yolov3_darknet53/src/transforms.py
@@ -73,42 +73,35 @@ def statistic_normalize_img(img, statistic_norm):
 
 
 def get_interp_method(interp, sizes=()):
-    """Get the interpolation method for resize functions.
+    """
+    Get the interpolation method for resize functions.
     The major purpose of this function is to wrap a random interp method selection
     and a auto-estimation method.
 
-    Parameters
-    ----------
-    interp : int
-        interpolation method for all resizing operations
-
-        Possible values:
-        0: Nearest Neighbors Interpolation.
-        1: Bilinear interpolation.
-        2: Bicubic interpolation over 4x4 pixel neighborhood.
-        3: Nearest Neighbors. [Originally it should be Area-based,
-        as we cannot find Area-based, so we use NN instead.
-        Area-based (resampling using pixel area relation). It may be a
-        preferred method for image decimation, as it gives moire-free
-        results. But when the image is zoomed, it is similar to the Nearest
-        Neighbors method. (used by default).
-        4: Lanczos interpolation over 8x8 pixel neighborhood.
-        9: Cubic for enlarge, area for shrink, bilinear for others
-        10: Random select from interpolation method metioned above.
-        Note:
+    Note:
         When shrinking an image, it will generally look best with AREA-based
         interpolation, whereas, when enlarging an image, it will generally look best
-        with Bicubic (slow) or Bilinear (faster but still looks OK).
-        More details can be found in the documentation of OpenCV, please refer to
-        http://docs.opencv.org/master/da/d54/group__imgproc__transform.html.
-    sizes : tuple of int
-        (old_height, old_width, new_height, new_width), if None provided, auto(9)
-        will return Area(2) anyway.
+        with Bicubic or Bilinear.
 
-    Returns
-    -------
-    int
-        interp method from 0 to 4
+    Args:
+        interp (int): Interpolation method for all resizing operations.
+
+            - 0: Nearest Neighbors Interpolation.
+            - 1: Bilinear interpolation.
+            - 2: Bicubic interpolation over 4x4 pixel neighborhood.
+            - 3: Nearest Neighbors. Originally it should be Area-based, as we cannot find Area-based,
+              so we use NN instead. Area-based (resampling using pixel area relation).
+              It may be a preferred method for image decimation, as it gives moire-free results.
+              But when the image is zoomed, it is similar to the Nearest Neighbors method. (used by default).
+            - 4: Lanczos interpolation over 8x8 pixel neighborhood.
+            - 9: Cubic for enlarge, area for shrink, bilinear for others.
+            - 10: Random select from interpolation method mentioned above.
+
+        sizes (tuple): Format should like (old_height, old_width, new_height, new_width),
+            if None provided, auto(9) will return Area(2) anyway. Default: ()
+
+    Returns:
+        int, interp method from 0 to 4.
     """
     if interp == 9:
         if sizes: