fix docstring & typing

graphic_element_v1.py

@@ -8,7 +8,7 @@ class Exp:
     Configuration class for the graphic element model.
 
     This class contains all configuration parameters for the YOLOX-based
-    page element detection model, including architecture settings, inference
+    graphic element detection model, including architecture settings, inference
     parameters, and class-specific thresholds.
     """
 

utils.py

@@ -30,17 +30,20 @@ def reformat_for_plotting(
 ) -> Tuple[List[npt.NDArray[np.int_]], List[npt.NDArray[np.float64]]]:
     """
     Reformat YOLOX predictions for plotting.
+    - Unnormalizes boxes to original image size.
+    - Reformats boxes to [xmin, ymin, width, height].
+    - Converts to list of boxes and scores per class.
 
     Args:
-        boxes (np.ndarray): Array of bounding boxes.
-        labels (np.ndarray): Array of labels.
-        scores (np.ndarray): Array of confidence scores.
-        shape (tuple): Shape of the image.
+        boxes (np.ndarray [N, 4]): Array of bounding boxes in format [xmin, ymin, xmax, ymax].
+        labels (np.ndarray [N]): Array of labels.
+        scores (np.ndarray [N]): Array of confidence scores.
+        shape (tuple [2]): Shape of the image (height, width).
         num_classes (int): Number of classes.
 
     Returns:
-        list[np.ndarray]: List of box bounding boxes per class.
-        list[np.ndarray]: List of confidence scores per class.
+        list[np.ndarray[N]]: List of box bounding boxes per class.
+        list[np.ndarray[N]]: List of confidence scores per class.
     """
     boxes_plot = boxes.copy()
     boxes_plot[:, [0, 2]] *= shape[1]
@@ -58,6 +61,7 @@ def plot_sample(
     boxes_list: List[npt.NDArray[np.int_]],
     confs_list: List[npt.NDArray[np.float64]],
     labels: List[str],
+    show_text: bool = True,
 ) -> None:
     """
     Plots an image with bounding boxes.
@@ -68,6 +72,7 @@ def plot_sample(
         boxes_list (list[np.ndarray]): List of box bounding boxes per class.
         confs_list (list[np.ndarray]): List of confidence scores per class.
         labels (list): List of class labels.
+        show_text (bool, optional): Whether to show the text. Defaults to True.
     """
     plt.imshow(img, cmap="gray")
     plt.axis(False)
@@ -92,15 +97,16 @@ def plot_sample(
             plt.gca().add_patch(rect)
 
             # Add class and index label with proper alignment
-            plt.text(
-                box[0], box[1],
-                f"{l}_{box_idx}   conf={confs[box_idx]:.3f}",
-                color='white',
-                fontsize=8,
-                bbox=dict(facecolor=col, alpha=1, edgecolor=col, pad=0, linewidth=2),
-                verticalalignment='bottom',
-                horizontalalignment='left'
-            )
+            if show_text:
+                plt.text(
+                    box[0], box[1],
+                    f"{l}_{box_idx}   conf={confs[box_idx]:.3f}",
+                    color='white',
+                    fontsize=8,
+                    bbox=dict(facecolor=col, alpha=1, edgecolor=col, pad=0, linewidth=2),
+                    verticalalignment='bottom',
+                    horizontalalignment='left'
+                )
 
 
 def reorder_boxes(
@@ -117,13 +123,13 @@ def reorder_boxes(
     Ordering depends on the class.
 
     Args:
-        boxes (np.ndarray): Array of bounding boxes of shape (N, 4) in format [x1, y1, x2, y2].
-        labels (np.ndarray): Array of labels of shape (N,).
+        boxes (np.ndarray [N, 4]): Array of bounding boxes in format [xmin, ymin, xmax, ymax].
+        labels (np.ndarray [N]): Array of labels.
         classes (list, optional): List of class labels. Defaults to None.
-        scores (np.ndarray, optional): Array of confidences of shape (N,). Defaults to None.
+        scores (np.ndarray [N], optional): Array of confidence scores. Defaults to None.
 
     Returns:
-        np.ndarray [N, 4]: Ordered boxes.
+        np.ndarray [N, 4]: Ordered boxes in format [xmin, ymin, xmax, ymax].
         np.ndarray [N]: Ordered labels.
         np.ndarray [N]: Ordered scores if scores is not None.
     """