From 58d1c6786fbcb736b01a2020eabccd6b801fa7aa Mon Sep 17 00:00:00 2001
From: ktro2828 <kotaro.uetake@tier4.jp>
Date: Wed, 10 Jan 2024 22:28:55 +0900
Subject: [PATCH 1/3] docs: update documents for result

Signed-off-by: ktro2828 <kotaro.uetake@tier4.jp>
---
 .../result/perception/__init__.py             |   2 +-
 .../result/perception/frame_config.py         |  40 ++---
 .../result/perception/frame_result.py         |  43 +++--
 .../result/perception/object_result.py        | 170 +++++++++---------
 .../result/perception/pass_fail_result.py     |  48 ++---
 .../result/sensing/__init__.py                |   2 +-
 .../result/sensing/frame_config.py            |  12 +-
 .../result/sensing/frame_result.py            |  17 +-
 .../result/sensing/object_result.py           |  26 +--
 9 files changed, 158 insertions(+), 202 deletions(-)

diff --git a/perception_eval/perception_eval/result/perception/__init__.py b/perception_eval/perception_eval/result/perception/__init__.py
index 799ab305..caaa01dc 100644
--- a/perception_eval/perception_eval/result/perception/__init__.py
+++ b/perception_eval/perception_eval/result/perception/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2023 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
diff --git a/perception_eval/perception_eval/result/perception/frame_config.py b/perception_eval/perception_eval/result/perception/frame_config.py
index 2e310d80..8374d58a 100644
--- a/perception_eval/perception_eval/result/perception/frame_config.py
+++ b/perception_eval/perception_eval/result/perception/frame_config.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -30,26 +30,7 @@
 
 
 class PerceptionFrameConfig:
-    """
-    Config class for critical object filter
-
-    Attributes:
-        self.target_labels (List[str]): Target list
-        self.max_x_position_list (Optional[List[float]]):
-            Maximum x position threshold list for each label. Defaults to None.
-        self.max_y_position_list (Optional[List[float]]):
-            Maximum y position threshold list for each label. Defaults to None.
-        self.max_distance_list (Optional[List[float]]]):
-            Maximum distance threshold list for each label. Defaults to None.
-        self.min_distance_list (Optional[List[float]]):
-            Minimum distance threshold list for object. Defaults to None.
-        self.min_point_numbers (Optional[List[int]]):
-            Minimum number of points to be included in object's box. Defaults to None.
-        self.confidence_threshold_list (Optional[List[float]]):
-            The list of confidence threshold for each label. Defaults to None.
-        self.target_uuids (Optional[List[str]]): The list of target uuid. Defaults to None.
-        self.filtering_params: (Dict[str, Any]): The container of filtering parameters.
-    """
+    """Frame level configuration for perception evaluation."""
 
     def __init__(
         self,
@@ -67,8 +48,9 @@ def __init__(
     ) -> None:
         """
         Args:
+        -----
             evaluator_config (PerceptionEvaluationConfig): Evaluation config
-            target_labels (List[str]): The list of target label.
+            target_labels (List[Union[str, LabelType]): The list of target labels.
             max_x_position_list (Optional[List[float]]):
                 Maximum x position threshold list for each label. Defaults to None.
             max_y_position_list (Optional[List[float]]):
@@ -82,6 +64,8 @@ def __init__(
             confidence_threshold_list (Optional[List[float]]):
                 The list of confidence threshold for each label. Defaults to None.
             target_uuids (Optional[List[str]]): The list of target uuid. Defaults to None.
+            ignore_attributes (Optional[List[str]]): The list of attribute names to ignore. Defaults to None.
+            success_thresholds (Optional[List[float]]): The list of thresholds to be success. Defaults to None.
         """
         self.evaluation_task = evaluator_config.evaluation_task
         if all([isinstance(label, str) for label in target_labels]):
@@ -110,6 +94,18 @@ def __init__(
 
     @classmethod
     def from_eval_cfg(cls, eval_cfg: PerceptionEvaluationConfig) -> PerceptionFrameConfig:
+        """Constructs from scene level configuration.
+
+        This method should be used if frame level configuration was not specified by user.
+
+        Args:
+        -----
+            eval_cfg (PerceptionEvaluationConfig): Scene level configuration.
+
+        Returns:
+        --------
+            PerceptionFrameConfig: Frame level configuration.
+        """
         if eval_cfg.evaluation_task.is_3d():
             success_thresholds = eval_cfg.metrics_param.plane_distance_thresholds
         else:
diff --git a/perception_eval/perception_eval/result/perception/frame_result.py b/perception_eval/perception_eval/result/perception/frame_result.py
index 5b432d29..a1085c83 100644
--- a/perception_eval/perception_eval/result/perception/frame_result.py
+++ b/perception_eval/perception_eval/result/perception/frame_result.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -37,25 +37,15 @@
 
 
 class PerceptionFrameResult:
-    """The result for 1 frame (the pair of estimated objects and ground truth objects)
-
-    Attributes:
-        object_results (List[PerceptionObjectResult]): Filtered object results to each estimated object.
-        frame_ground_truth (FrameGroundTruth): Filtered ground truth of frame.
-        frame_number (int): The file name of frame in the datasets.
-        unix_time (int): The unix time for frame [us].
-        target_labels (List[AutowareLabel]): The list of target label.
-        metrics_score (MetricsScore): Metrics score results.
-        pass_fail_result (PassFailResult): Pass fail results.
+    """Frame level result for perception evaluation.
 
     Args:
-        object_results (List[PerceptionObjectResult]): The list of object result.
-        frame_ground_truth (FrameGroundTruth): FrameGroundTruth instance.
-        metrics_config (MetricsScoreConfig): Metrics config class.
-        critical_object_filter_config (CriticalObjectFilterConfig): Critical object filter config.
-        frame_pass_fail_config (PerceptionPassFailConfig): Frame pass fail config.
-        unix_time (int): The unix time for frame [us]
-        target_labels (List[AutowareLabel]): The list of target label.
+    -----
+        unix_time (int): Unix timestamp.
+        frame_config (PerceptionFrameConfig): Frame level configuration.
+        metrics_config (MetricsScoreConfig): Configuration for metrics score.
+        object_results (List[PerceptionObjectResult]): List of object results.
+        frame_ground_truth (FrameGroundTruth): Ground truth at one frame.
     """
 
     def __init__(
@@ -99,10 +89,17 @@ def evaluate_frame(
         previous_result: Optional[PerceptionFrameResult] = None,
     ) -> None:
         """
-        Evaluate a frame from the pair of estimated objects and ground truth objects
+        Evaluate one frame.
+
+        Internally, this method runs the following.
+        1. Calculate metrics score if corresponding metrics config exists.
+        2. Evaluate pass fail result.
+
         Args:
-            ros_critical_ground_truth_objects (List[ObjectType]): The list of Ground truth objects filtered by ROS node.
-            previous_result (Optional[PerceptionFrameResult]): The previous frame result. If None, set it as empty list []. Defaults to None.
+        -----
+            critical_ground_truth_objects (List[ObjectType]): List of ground truth objects.
+            previous_result (Optional[PerceptionFrameResult]): If the previous result exist
+                it should be input to perform time series evaluation. Defaults to None.
         """
         # Divide objects by label to dict
         object_results_dict = objects_filter.divide_objects(self.object_results, self.target_labels)
@@ -140,10 +137,12 @@ def get_object_status(frame_results: List[PerceptionFrameResult]) -> List[Ground
     """Returns the number of TP/FP/TN/FN ratios per frame as tuple.
 
     Args:
+    -----
         frame_results (List[PerceptionFrameResult]): List of frame results.
 
     Returns:
-        List[GroundTruthStatus]: Sequence of matching status ratios for each GT.
+    --------
+        List[GroundTruthStatus]: List of matching status ratios for each GT.
     """
     status_infos: List[GroundTruthStatus] = []
     for frame_result in frame_results:
diff --git a/perception_eval/perception_eval/result/perception/object_result.py b/perception_eval/perception_eval/result/perception/object_result.py
index 72f8870e..0e61199b 100644
--- a/perception_eval/perception_eval/result/perception/object_result.py
+++ b/perception_eval/perception_eval/result/perception/object_result.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -42,17 +42,15 @@
 
 
 class PerceptionObjectResult:
-    """Object result class for perception evaluation.
+    """Object level result for perception evaluation.
 
-    Attributes:
+    This class consists of the pair of estimated and ground truth objects.
+    In case of ground truth object is `None`, it means the estimated object is FP regardless of the matching score.
+
+    Args:
+    -----
         estimated_object (ObjectType): Estimated object.
-        ground_truth_object (Optional[ObjectType]): Ground truth object.
-        is_label_correct (bool): Whether the label both of `estimated_object` and `ground_truth_object` are same.
-        center_distance (Optional[CenterDistanceMatching]): CenterDistanceMatching instance.
-        plane_distance (Optional[PlaneDistanceMatching]): PlaneDistanceMatching instance.
-            In 2D evaluation, this is None.
-        iou_2d (IOU2dMatching): IOU2dMatching instance.
-        iou_3d (IOU3dMatching): IOU3dMatching instance. In 2D evaluation, this is None.
+        ground_truth_object (Optional[ObjectType]): Ground truth object. `None` is allowed.
     """
 
     def __init__(
@@ -60,13 +58,6 @@ def __init__(
         estimated_object: ObjectType,
         ground_truth_object: Optional[ObjectType],
     ) -> None:
-        """[summary]
-        Evaluation result for an object estimated object.
-
-        Args:
-            estimated_object (ObjectType): The estimated object by inference like CenterPoint
-            ground_truth_objects (Optional[ObjectType]): The list of Ground truth objects
-        """
         if ground_truth_object is not None:
             assert type(estimated_object) is type(
                 ground_truth_object
@@ -108,11 +99,22 @@ def get_status(
     ) -> Tuple[MatchingStatus, Optional[MatchingStatus]]:
         """Returns matching status both of estimation and GT as `tuple`.
 
+        In case of the ground truth is `None`, the status of GT is `None` and estimation is FP.
+        Otherwise, the status is determined by the following rules.
+        1. The matching score is better than the threshold:
+            a. Evaluation task is FP validation:    (FP, TN)
+            b. Otherwise:                           (TP, TP)
+        2. Otherwise:
+            a. Evaluation task is FP validation:    (FP, FP)
+            b. Otherwise:                           (FP, FN)
+
         Args:
-            matching_mode (MatchingMode): Matching policy.
+        -----
+            matching_mode (MatchingMode): Matching mode.
             matching_threshold (float): Matching threshold.
 
         Returns:
+        --------
             Tuple[MatchingStatus, Optional[MatchingStatus]]: Matching status of estimation and GT.
         """
         if self.ground_truth_object is None:
@@ -131,24 +133,18 @@ def get_status(
                 else (MatchingStatus.FP, MatchingStatus.FN)
             )
 
-    def is_result_correct(
-        self,
-        matching_mode: MatchingMode,
-        matching_threshold: Optional[float],
-    ) -> bool:
-        """The function judging whether the result is target or not.
-        Return `False`, if label of GT is "FP" and matching.
+    def is_result_correct(self, matching_mode: MatchingMode, matching_threshold: Optional[float]) -> bool:
+        """The method judging whether the result is target or not.
+
+        Returns `False`, if label of GT is "FP" and matching.
 
         Args:
-            matching_mode (MatchingMode):
-                    The matching mode to evaluate.
-            matching_threshold (float):
-                    The matching threshold to evaluate.
-                    For example, if matching_mode = IOU3d and matching_threshold = 0.5,
-                    and IoU of the object is higher than "matching_threshold",
-                    this function appends to return objects.
+        -----
+            matching_mode (MatchingMode): The matching mode to evaluate.
+            matching_threshold (float): The matching threshold to evaluate.
 
         Returns:
+        --------
             bool: If label is correct and satisfy matching threshold, return True
         """
         if self.ground_truth_object is None:
@@ -167,15 +163,18 @@ def is_result_correct(
         return not is_matching if self.ground_truth_object.semantic_label.is_fp() else is_matching
 
     def get_matching(self, matching_mode: MatchingMode) -> Optional[MatchingMethod]:
-        """Get MatchingMethod instance with corresponding MatchingMode.
+        """Returns the matching module for corresponding matching mode.
 
-        Args:
-            matching_mode (MatchingMode): MatchingMode instance.
+        Note that,
+            - 2D detection/tracking tasks, input `PLANEDISTANCE` and `IOU3D` return `None`.
+            - 2D classification task, all return `None`.
 
-        Raises:
-            NotImplementedError: When unexpected MatchingMode is input.
+        Args:
+        -----
+            matching_mode (MatchingMode): The matching mode.
 
         Returns:
+        --------
             Optional[MatchingMethod]: Corresponding MatchingMethods instance.
         """
         if matching_mode == MatchingMode.CENTERDISTANCE:
@@ -187,16 +186,17 @@ def get_matching(self, matching_mode: MatchingMode) -> Optional[MatchingMethod]:
         elif matching_mode == MatchingMode.IOU3D:
             return self.iou_3d
         else:
-            raise NotImplementedError
+            raise NotImplementedError(f"Unexpected matching mode: {matching_mode}")
 
     @property
     def distance_error_bev(self) -> Optional[float]:
-        """Get error center distance between ground truth and estimated object in BEV space.
+        """Returns the center distance error in BEV coords between estimation and GT.
 
-        If `self.ground_truth_object=None`, returns None.
+        If GT is `None` returns `None`.
 
         Returns:
-            Optional[float]: error center distance between ground truth and estimated object.
+        --------
+            Optional[float]: Calculated error [m].
         """
         if self.ground_truth_object is None:
             return None
@@ -204,12 +204,13 @@ def distance_error_bev(self) -> Optional[float]:
 
     @property
     def distance_error(self) -> Optional[float]:
-        """Get error center distance between ground truth and estimated object.
+        """Returns the center distance error in BEV coords between estimation and GT.
 
-        If `self.ground_truth_object=None`, returns None.
+        If GT is `None` returns `None`.
 
         Returns:
-            Optional[float]: error center distance between ground truth and estimated object.
+        --------
+            Optional[float]: Calculated error [m].
         """
         if self.ground_truth_object is None:
             return None
@@ -217,41 +218,36 @@ def distance_error(self) -> Optional[float]:
 
     @property
     def position_error(self) -> Optional[Tuple[float, float, float]]:
-        """Get the position error vector from estimated to ground truth object.
+        """Returns the position error vector from estimated to ground truth object.
 
-        If `self.ground_truth_object=None`, returns None.
+        If GT is `None` returns `None`.
 
         Returns:
-            float: x-axis position error[m].
-            float: y-axis position error[m].
-            float: z-axis position error[m].
+        --------
+            Optional[Tuple[float, float, float]]: Errors ordering (x, y, z) [m].
         """
         return self.estimated_object.get_position_error(self.ground_truth_object)
 
     @property
     def heading_error(self) -> Optional[Tuple[float, float, float]]:
-        """Get the heading error vector from estimated to ground truth object.
+        """Returns the heading error vector from estimated to ground truth object.
 
-        If `self.ground_truth_object=None`, returns None.
+        If GT is `None` returns `None`.
 
         Returns:
-            float: Roll error, in [-pi, pi].
-            float: Pitch error, in [-pi, pi].
-            float: Yaw error, in [-pi, pi].
+            Optional[Tuple[float, float, float]]: Errors ordering (roll, pitch, yaw) [rad] in `[-pi, pi]`.
         """
         return self.estimated_object.get_heading_error(self.ground_truth_object)
 
     @property
     def velocity_error(self) -> Optional[Tuple[float, float, float]]:
-        """Get the velocity error vector from estimated to ground truth object.
+        """Returns the velocity error vector from estimated to ground truth object.
 
-        If `self.ground_truth_object=None`, returns None.
+        If GT is `None` returns `None`.
         Also, velocity of estimated or ground truth object is None, returns None too.
 
         Returns:
-            float: x-axis velocity error[m/s].
-            float: y-axis velocity error[m/s].
-            float: z-axis velocity error[m/s].
+            Optional[Tuple[float, float, float]]: Errors ordering (x, y, z) [m/s].
         """
         return self.estimated_object.get_velocity_error(self.ground_truth_object)
 
@@ -272,7 +268,7 @@ def get_object_results(
     target_labels: Optional[List[LabelType]] = None,
     matching_policy: MatchingPolicy = MatchingPolicy(),
 ) -> List[PerceptionObjectResult]:
-    """Returns list of DynamicObjectWithPerceptionResult.
+    """Returns list of object results.
 
     For classification, matching objects their uuid.
     Otherwise, matching them depending on their center distance by default.
@@ -281,14 +277,16 @@ def get_object_results(
     Otherwise, they all are FP.
 
     Args:
+    -----
         evaluation_task (EvaluationTask): Evaluation task.
-        estimated_objects (List[ObjectType]): Estimated objects list.
-        ground_truth_objects (List[ObjectType]): Ground truth objects list.
-        matching_mode (MatchingMode): MatchingMode instance.
-        matchable_thresholds (Optional[List[float]]): Thresholds to be
+        estimated_objects (List[ObjectType]): List of estimated objects.
+        ground_truth_objects (List[ObjectType]): List of ground truth objects.
+        target_labels (Optional[List[LabelType]]): List of target labels to match. Defaults to None.
+        matching_policy (MatchingPolicy): Policy of matching.
 
     Returns:
-        object_results (List[DynamicObjectWithPerceptionResult]): Object results list.
+    --------
+        object_results (List[PerceptionObjectResult]): List of object results.
     """
     # There is no estimated object (= all FN)
     if not estimated_objects:
@@ -346,16 +344,18 @@ def _get_object_results_with_id(
     estimated_objects: List[DynamicObject2D],
     ground_truth_objects: List[DynamicObject2D],
 ) -> List[PerceptionObjectResult]:
-    """Returns the list of DynamicObjectWithPerceptionResult considering their uuids.
+    """Returns the list of object results considering their uuids.
 
     This function is used in 2D classification evaluation.
 
     Args:
-        estimated_objects (List[DynamicObject2D]): Estimated objects list.
-        ground_truth_objects (List[DynamicObject2D]): Ground truth objects list.
+    -----
+        estimated_objects (List[DynamicObject2D]): List of estimated objects.
+        ground_truth_objects (List[DynamicObject2D]): List of ground truth objects.
 
     Returns:
-        object_results (List[PerceptionObjectResult]): Object results list.
+    --------
+        object_results (List[PerceptionObjectResult]): List of object results.
     """
     object_results: List[PerceptionObjectResult] = []
     estimated_objects_ = estimated_objects.copy()
@@ -387,26 +387,18 @@ def _get_object_results_with_id(
     return object_results
 
 
-def _get_fp_object_results(
-    estimated_objects: List[ObjectType],
-) -> List[PerceptionObjectResult]:
-    """Returns the list of DynamicObjectWithPerceptionResult that have no ground truth.
+def _get_fp_object_results(estimated_objects: List[ObjectType]) -> List[PerceptionObjectResult]:
+    """Returns the list of FP object results that have no ground truth.
 
     Args:
-        estimated_objects (List[ObjectType]): Estimated objects list.
+    -----
+        estimated_objects (List[ObjectType]): List of object results.
 
     Returns:
-        object_results (List[DynamicObjectWithPerceptionResult]): FP object results list.
+    --------
+        List[PerceptionObjectResult]: List of FP object results.
     """
-    object_results: List[PerceptionObjectResult] = []
-    for est_obj_ in estimated_objects:
-        object_result_ = PerceptionObjectResult(
-            estimated_object=est_obj_,
-            ground_truth_object=None,
-        )
-        object_results.append(object_result_)
-
-    return object_results
+    return [PerceptionObjectResult(est, None) for est in estimated_objects]
 
 
 def _get_score_table(
@@ -418,17 +410,15 @@ def _get_score_table(
     """Returns score table, in shape (num_estimation, num_ground_truth).
 
     Args:
+    -----
         estimated_objects (List[ObjectType]): Estimated objects list.
         ground_truth_objects (List[ObjectType]): Ground truth objects list.
-        allow_matching_unknown (bool): Indicates whether allow to match with unknown label.
-        matching_method_module (Callable): MatchingMethod instance.
-        target_labels (Optional[List[LabelType]]): Target labels to be evaluated.
-        matching_thresholds (Optional[List[float]]): List of thresholds
+        target_labels (List[LabelType]): Target labels to be evaluated.
+        matching_policy (MatchingPolicy): Policy of matching.
 
     Returns:
-        score_table (numpy.ndarray): in shape (num_estimation, num_ground_truth).
+        np.ndarray: Array in shape (num_estimation, num_ground_truth).
     """
-    # fill matching score table, in shape (NumEst, NumGT)
     num_row: int = len(estimated_objects)
     num_col: int = len(ground_truth_objects)
     score_table = np.full((num_row, num_col), np.nan)
diff --git a/perception_eval/perception_eval/result/perception/pass_fail_result.py b/perception_eval/perception_eval/result/perception/pass_fail_result.py
index 4e0b68f4..f1712be6 100644
--- a/perception_eval/perception_eval/result/perception/pass_fail_result.py
+++ b/perception_eval/perception_eval/result/perception/pass_fail_result.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -18,7 +18,6 @@
 from typing import Optional
 from typing import Tuple
 from typing import TYPE_CHECKING
-import warnings
 
 import numpy as np
 from perception_eval.matching import MatchingMode
@@ -32,22 +31,14 @@
 
 
 class PassFailResult:
-    """Class for keeping TP/FP/TN/FP object results and GT objects for critical GT objects.
-
-    Attributes:
-        frame_config (PerceptionFrameConfig): Critical object filter config.
-        tn_objects (List[ObjectType]): TN ground truth objects list.
-        fn_objects (List[ObjectType]): FN ground truth objects list.
-        fp_object_results (List[PerceptionObjectResult]): FP object results list.
-        tp_object_results (List[PerceptionObjectResult]): TP object results list.
+    """Class to determine pass fail.
 
     Args:
+    -----
         unix_time (int): UNIX timestamp.
         frame_number (int): The Number of frame.
-        critical_object_filter_config (CriticalObjectFilterConfig): Critical object filter config.
-        frame_id (str): `base_link` or `map`.
-        ego2map (Optional[numpy.ndarray]): Array of 4x4 matrix to transform coordinates from ego to map.
-            Defaults to None.
+        frame_config (PerceptionFrameConfig): Frame level configuration.
+        ego2map (Optional[np.ndarray]): 4x4 matrix to transform coordinates from ego to map. Defaults to None.
     """
 
     def __init__(
@@ -76,8 +67,9 @@ def evaluate(
         """Evaluate object results' pass fail.
 
         Args:
-            object_results (List[PerceptionObjectResult]): Object results list.
-            ros_critical_ground_truth_objects (List[ObjectType]): Critical ground truth objects
+        -----
+            object_results (List[PerceptionObjectResult]): List of object results.
+            critical_ground_truth_objects (List[ObjectType]): Critical ground truth objects
                 must be evaluated at current frame.
         """
         critical_ground_truth_objects = objects_filter.filter_objects(
@@ -105,6 +97,7 @@ def get_num_success(self) -> int:
         """Returns the number of success.
 
         Returns:
+        --------
             int: Number of success.
         """
         return len(self.tp_object_results) + len(self.tn_objects)
@@ -113,28 +106,17 @@ def get_num_fail(self) -> int:
         """Returns the number of fail.
 
         Returns:
+        --------
             int: Number of fail.
         """
         return len(self.fp_object_results) + len(self.fn_objects)
 
-    def get_fail_object_num(self) -> int:
-        """Get the number of fail objects.
-
-        Returns:
-            int: Number of fail objects.
-        """
-        warnings.warn(
-            "`get_fail_object_num()` is removed in next minor update, please use `get_num_fail()`",
-            DeprecationWarning,
-        )
-        return len(self.fn_objects) + len(self.fp_object_results)
-
     def __get_positive_object_results(
         self,
         object_results: List[PerceptionObjectResult],
         critical_ground_truth_objects: List[ObjectType],
     ) -> Tuple[List[PerceptionObjectResult], List[PerceptionObjectResult]]:
-        """Get TP and FP object results list from `object_results`.
+        """Returns list of TP and FP results from object results.
 
         Args:
             object_results (List[PerceptionObjectResult]): Object results list.
@@ -142,8 +124,8 @@ def __get_positive_object_results(
                 must be evaluated at current frame.
 
         Returns:
-            List[PerceptionObjectResult]: TP object results.
-            List[PerceptionObjectResult]: FP object results.
+        --------
+            Tuple[List[PerceptionObjectResult], List[PerceptionObjectResult]]: TP and FP object results.
         """
         tp_object_results, fp_object_results = objects_filter.get_positive_objects(
             object_results=object_results,
@@ -155,12 +137,12 @@ def __get_positive_object_results(
         )
 
         # filter by critical_ground_truth_objects
-        tp_critical_results: List[PerceptionObjectResult] = [
+        tp_critical_results = [
             tp_result
             for tp_result in tp_object_results
             if tp_result.ground_truth_object in critical_ground_truth_objects
         ]
-        fp_critical_results: List[PerceptionObjectResult] = [
+        fp_critical_results = [
             fp_result
             for fp_result in fp_object_results
             if fp_result.ground_truth_object in critical_ground_truth_objects
diff --git a/perception_eval/perception_eval/result/sensing/__init__.py b/perception_eval/perception_eval/result/sensing/__init__.py
index 025f5eb8..5bc65556 100644
--- a/perception_eval/perception_eval/result/sensing/__init__.py
+++ b/perception_eval/perception_eval/result/sensing/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
diff --git a/perception_eval/perception_eval/result/sensing/frame_config.py b/perception_eval/perception_eval/result/sensing/frame_config.py
index 454a49e1..f150d093 100644
--- a/perception_eval/perception_eval/result/sensing/frame_config.py
+++ b/perception_eval/perception_eval/result/sensing/frame_config.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -17,14 +17,10 @@
 
 
 class SensingFrameConfig:
-    """Config class for sensing detection evaluation per frame.
-
-    Attributes:
-        box_scale_0m (float): Scale factor for bounding box at 0m.
-        box_scale_100m (float): Scale factor for bounding box at 100m.
-        min_points_threshold (int): The minimum number of points should be detected in bounding box.
+    """Frame level configuration for sensing evaluation.
 
     Args:
+    -----
         target_uuids (Optional[List[str]]): List of target uuids to be filtered.
         box_scale_0m (float): Scale factor for bounding box at 0m.
         box_scale_100m (float): Scale factor for bounding box at 100m.
@@ -52,9 +48,11 @@ def get_scale_factor(self, distance: float) -> float:
             scale = ((box_scale_100m - box_scale_0m) / (100 - 0)) * (distance - 0) + box_scale_0m
 
         Args:
+        -----
             distance (float): The distance from vehicle to target bounding box.
 
         Returns:
+        --------
             float: Calculated scale factor.
         """
         return self.scale_slope_ * distance + self.box_scale_0m
diff --git a/perception_eval/perception_eval/result/sensing/frame_result.py b/perception_eval/perception_eval/result/sensing/frame_result.py
index 1557af01..fc72e48b 100644
--- a/perception_eval/perception_eval/result/sensing/frame_result.py
+++ b/perception_eval/perception_eval/result/sensing/frame_result.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -44,17 +44,13 @@ class SensingFrameResult:
             in non-detection area.
 
     Args:
-        sensing_frame_config (SensingFrameConfig): The configuration of sensing evaluation.
+    -----
+        sensing_frame_config (SensingFrameConfig): Frame level configuration for sensing evaluation.
         unix_time (int): Unix time [us].
-        frame_name (str): Frame name in string.
+        frame_number (int): Number of frame.
     """
 
-    def __init__(
-        self,
-        sensing_frame_config: SensingFrameConfig,
-        unix_time: int,
-        frame_number: int,
-    ) -> None:
+    def __init__(self, sensing_frame_config: SensingFrameConfig, unix_time: int, frame_number: int) -> None:
         # Config
         self.sensing_frame_config = sensing_frame_config
 
@@ -77,6 +73,7 @@ def evaluate_frame(
         """Evaluate each object at current frame.
 
         Args:
+        -----
             ground_truth_objects (list[DynamicObject]): Ground truth objects list.
             pointcloud_for_detection (numpy.ndarray): Array of pointcloud in detection area.
             pointcloud_for_non_detection (List[numpy.ndarray]): List of pointcloud array in non-detection area.
@@ -101,6 +98,7 @@ def _evaluate_pointcloud_for_detection(
         If the object is occluded, the result is appended to `self.detection_warning_results`.
 
         Args:
+        -----
             ground_truth_objects (list[DynamicObject]): Ground truth objects list.
             pointcloud_for_detection (numpy.ndarray): Array of pointcloud in detection area.
         """
@@ -136,6 +134,7 @@ def _evaluate_pointcloud_for_non_detection(
         `self.pointcloud_failed_non_detection`.
 
         Args:
+        -----
             ground_truth_objects (list[DynamicObject]): Ground truth objects list.
             pointcloud_for_non_detection (list[numpy.ndarray]): List of pointcloud array in non-detection area.
         """
diff --git a/perception_eval/perception_eval/result/sensing/object_result.py b/perception_eval/perception_eval/result/sensing/object_result.py
index 86be57a8..de451383 100644
--- a/perception_eval/perception_eval/result/sensing/object_result.py
+++ b/perception_eval/perception_eval/result/sensing/object_result.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -20,15 +20,13 @@
 
 
 class SensingObjectResult:
-    """The class to evaluate sensing result for dynamic object.
+    """Object level result for sensing evaluation.
 
-    Attributes:
-        self.ground_truth_object (DynamicObject): The target DynamicObject.
-        self.inside_pointcloud (numpy.ndarray): The array of pointcloud in bounding box.
-        self.inside_pointcloud_num (int): The number of pointcloud in bounding box.
-        self.is_detected (bool): The boolean flag indicates whether pointcloud is in bounding box.
-        self.nearest_point (np.ndarray): The nearest point from base_link.
-        self.is_occluded (bool): Whether the object is occluded.
+    Args:
+        ground_truth_object (DynamicObject): Ground truth object.
+        pointcloud (numpy.ndarray): Array of pointcloud after removing ground.
+        scale_factor (float): Scale factor for bounding box.
+        min_points_threshold (int): The minimum number of points should be detected in bounding box.
     """
 
     def __init__(
@@ -38,13 +36,6 @@ def __init__(
         scale_factor: float,
         min_points_threshold: int,
     ) -> None:
-        """[summary]
-        Args:
-            ground_truth_object (DynamicObject): Ground truth object.
-            pointcloud (numpy.ndarray): Array of pointcloud after removing ground.
-            scale_factor (float): Scale factor for bounding box.
-            min_points_threshold (int): The minimum number of points should be detected in bounding box.
-        """
         self.ground_truth_object: DynamicObject = ground_truth_object
 
         # Evaluate
@@ -58,10 +49,11 @@ def __init__(
         self.is_occluded: bool = ground_truth_object.visibility == Visibility.NONE
 
     def _get_nearest_point(self) -> Optional[np.ndarray]:
-        """[summary]
+        """
         Returns the nearest point from base_link. The pointcloud must be base_link coords.
 
         Returns:
+        --------
             Optional[np.ndarray]: The nearest point included in the object's bbox, in shape (3,).
                 If there is no point in bbox, returns None.
         """

From 772548c102e5149fb55e636be6262c3c49837df5 Mon Sep 17 00:00:00 2001
From: ktro2828 <kotaro.uetake@tier4.jp>
Date: Wed, 10 Jan 2024 22:38:58 +0900
Subject: [PATCH 2/3] docs: update documents for object

Signed-off-by: ktro2828 <kotaro.uetake@tier4.jp>
---
 .../perception_eval/object/__init__.py        | 28 ++++++++-----
 .../perception_eval/object/object2d.py        | 33 ++++-----------
 .../perception_eval/object/object3d.py        | 41 ++++++++++++++-----
 3 files changed, 56 insertions(+), 46 deletions(-)

diff --git a/perception_eval/perception_eval/object/__init__.py b/perception_eval/perception_eval/object/__init__.py
index 595542c0..641784f2 100644
--- a/perception_eval/perception_eval/object/__init__.py
+++ b/perception_eval/perception_eval/object/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2023 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -36,13 +36,16 @@
 
 
 def distance_objects(object_1: ObjectType, object_2: ObjectType) -> float:
-    """
-    Calculate the 3D/2D center distance between two objects.
+    """Calculates the 3D/2D center distance between two objects.
 
     Args:
-         object_1 (ObjectType): An object
-         object_2 (ObjectType): An object
-    Returns: float: The center distance between object_1 and object_2.
+    -----
+        object_1 (ObjectType): An object.
+        object_2 (ObjectType): An object.
+
+    Returns:
+    --------
+        float: The center distance between object_1 and object_2.
     """
     if type(object_1) is not type(object_2):
         raise TypeError(f"objects' type must be same, but got {type(object_1) and {type(object_2)}}")
@@ -53,13 +56,16 @@ def distance_objects(object_1: ObjectType, object_2: ObjectType) -> float:
 
 
 def distance_objects_bev(object_1: DynamicObject, object_2: DynamicObject) -> float:
-    """
-    Calculate the BEV 2d center distance between two objects.
+    """Calculates the BEV 2d center distance between two objects.
 
     Args:
-         object_1 (DynamicObject): An object
-         object_2 (DynamicObject): An object
-    Returns: float: The 2d center distance from object_1 to object_2.
+    -----
+        object_1 (DynamicObject): An object.
+        object_2 (DynamicObject): An object.
+
+    Returns:
+    --------
+        float: The 2d center distance from object_1 to object_2.
     """
     assert isinstance(object_1, DynamicObject) and isinstance(object_2, DynamicObject)
     return distance_points_bev(object_1.state.position, object_2.state.position)
diff --git a/perception_eval/perception_eval/object/object2d.py b/perception_eval/perception_eval/object/object2d.py
index c211c7b2..3fe08619 100644
--- a/perception_eval/perception_eval/object/object2d.py
+++ b/perception_eval/perception_eval/object/object2d.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -34,22 +34,12 @@ class Roi:
     TODO:
     - Support multi roi input for Blinker and BrakeLamp objects.
 
-    Attributes:
-        offset (Tuple[int, int]): Top-left pixels from (0, 0), (x, y) order.
-        size (Tuple[int, int]): Size of ROI, (width, height) order.
-        center (Tuple[int, int]): Center of ROI, (x, y) order.
-        height (int): Height of ROI.
-        width (int): Width  of ROI.
-        area (int): Area of ROI.
-
     Args:
+    -----
         roi (Tuple[int, int, int, int]): (xmin, ymin, width, height) of ROI.
     """
 
-    def __init__(
-        self,
-        roi: Tuple[int, int, int, int],
-    ) -> None:
+    def __init__(self, roi: Tuple[int, int, int, int]) -> None:
         if len(roi) != 4:
             raise ValueError("`roi` must be 4 length int array.")
 
@@ -109,17 +99,9 @@ def corners(self) -> np.ndarray:
 class DynamicObject2D:
     """Dynamic object class for 2D object.
 
-    Attributes:
-        unix_time (int): Unix time[us].
-        frame_id (FrameID): FrameID instance, where 2D objects are with respect, related to CAM_**.
-        semantic_score (float): Object's confidence [0, 1].
-        semantic_label (Label): Object's Label.
-        roi (Optional[Roi]): ROI in image. For classification, None is OK. Defaults to None.
-        uuid (Optional[str]): Unique ID. For traffic light objects, set lane ID. Defaults to None.
-        visibility (Optional[Visibility]): Visibility status. Defaults to None.
-
     Args:
-        unix_time (int): Unix time[us].
+    -----
+        unix_time (int): Unix time [us].
         frame_id (FrameID): FrameID instance, where 2D objects are with respect, related to CAM_**.
         semantic_score (float): Object's confidence [0, 1].
         semantic_label (Label): Object's Label.
@@ -152,7 +134,8 @@ def get_corners(self) -> np.ndarray:
         """Returns the corners of bounding box in pixel.
 
         Returns:
-            numpy.ndarray: (top_left, top_right, bottom_right, bottom_left), in shape (4, 2).
+        --------
+            np.ndarray: (top_left, top_right, bottom_right, bottom_left), in shape (4, 2).
         """
         assert self.roi is not None, "self.roi is None."
         return self.roi.corners
@@ -161,6 +144,7 @@ def get_area(self) -> int:
         """Returns the area of bounding box in pixel.
 
         Returns:
+        --------
             int: Area of bounding box[px].
         """
         assert self.roi is not None, "self.roi is None."
@@ -170,6 +154,7 @@ def get_polygon(self) -> Polygon:
         """Returns the corners as polygon.
 
         Returns:
+        --------
             Polygon: Corners as Polygon. ((x0, y0), ..., (x0, y0))
         """
         assert self.roi is not None, "self.roi is None."
diff --git a/perception_eval/perception_eval/object/object3d.py b/perception_eval/perception_eval/object/object3d.py
index 5f43c2db..1f541151 100644
--- a/perception_eval/perception_eval/object/object3d.py
+++ b/perception_eval/perception_eval/object/object3d.py
@@ -38,17 +38,12 @@
 class ObjectState:
     """Object state class.
 
-    Attributes:
-        position (Tuple[float, float, float]): (center_x, center_y, center_z)[m].
-        orientation (Quaternion) : Quaternion instance.
-        size (Tuple[float, float, float]): Bounding box size, (wx, wy, wz)[m].
-        velocity (Optional[Tuple[float, float, float]]): Velocity, (vx, vy, vz)[m/s].
-
     Args:
-        position (Tuple[float, float, float]): (center_x, center_y, center_z)[m].
+    -----
+        position (Tuple[float, float, float]): (center_x, center_y, center_z) [m].
         orientation (Quaternion) : Quaternion instance.
         shape (Shape): Shape instance.
-        velocity (Optional[Tuple[float, float, float]]): Velocity, (vx, vy, vz)[m/s].
+        velocity (Optional[Tuple[float, float, float]]): Velocity, (vx, vy, vz) [m/s].
     """
 
     def __init__(
@@ -80,6 +75,7 @@ class DynamicObject:
     """Dynamic object class for 3D object.
 
     Attributes:
+    -----------
         unix_time (int) : Unix time [us].
         frame_id (FrameID): FrameID instance, where 3D objects are with respect, BASE_LINK or MAP.
 
@@ -102,6 +98,7 @@ class DynamicObject:
         visibility (Optional[Visibility]): Visibility status. Defaults to None.
 
     Args:
+    -----
         unix_time (int): Unix time [us].
         frame_id (FrameID): FrameID instance, where 3D objects are with respect, BASE_LINK or MAP.
         position (Tuple[float, float, float]): (center_x, center_y, center_z)[m].
@@ -198,9 +195,11 @@ def __eq__(self, other: Optional[DynamicObject]) -> bool:
         it causes error.
 
         Args:
+        -----
             other (Optional[DynamicObject]): Another object.
 
         Returns:
+        --------
             bool
         """
         if other is None:
@@ -217,10 +216,12 @@ def get_distance(self, ego2map: Optional[np.ndarray] = None) -> float:
         """Get the 3d distance to the object from ego vehicle in bird eye view.
 
         Args:
-            ego2map (Optional[numpy.ndarray]):4x4 Transform matrix
+        -----
+            ego2map (Optional[numpy.ndarray]): 4x4 Transform matrix
                 from base_link coordinate system to map coordinate system.
 
         Returns:
+        --------
             float: The 3d distance to the object from ego vehicle in bird eye view.
         """
         if self.frame_id == FrameID.BASE_LINK:
@@ -236,10 +237,12 @@ def get_distance_bev(self, ego2map: Optional[np.ndarray] = None) -> float:
         """Get the 2d distance to the object from ego vehicle in bird eye view.
 
         Args:
+        -----
             ego2map (Optional[numpy.ndarray]):4x4 Transform matrix
                 from base_link coordinate system to map coordinate system.
 
         Returns:
+        --------
             float: The 2d distance to the object from ego vehicle in bird eye view.
         """
         if self.frame_id == FrameID.BASE_LINK:
@@ -255,10 +258,12 @@ def get_heading_bev(self, ego2map: Optional[np.ndarray] = None) -> float:
         """Get the object heading from ego vehicle in bird eye view.
 
         Args:
+        -----
             ego2map (Optional[numpy.ndarray]):4x4 Transform matrix.
                 from base_link coordinate system to map coordinate system.
 
         Returns:
+        --------
             float: The heading (radian)
         """
         if self.frame_id == FrameID.MAP:
@@ -280,9 +285,11 @@ def get_corners(self, scale: float = 1.0) -> np.ndarray:
         """Get the bounding box corners.
 
         Args:
+        -----
             scale (float): Scale factor to scale the corners. Defaults to 1.0.
 
         Returns:
+        --------
             corners (numpy.ndarray): Objects corners array.
         """
         # NOTE: This is with respect to base_link or map coordinate system.
@@ -299,17 +306,18 @@ def get_corners(self, scale: float = 1.0) -> np.ndarray:
         return corners
 
     def get_footprint(self, scale: float = 1.0) -> Polygon:
-        """[summary]
-        Get footprint polygon from an object with respect ot base_link or map coordinate system.
+        """Get footprint polygon from an object with respect ot base_link or map coordinate system.
 
         Args:
             scale (float): Scale factor for footprint. Defaults to 1.0.
 
         Returns:
+        --------
             Polygon: The footprint polygon of object with respect to base_link or map coordinate system.
                 It consists of 4 corner 2d position of the object and  start and end points are same point.
                 ((x0, y0, 0), (x1, y1, 0), (x2, y2, 0), (x3, y3, 0), (x0, y0, 0))
         Notes:
+        ------
             center_position: (xc, yc)
             vector_center_to_corners[0]: (x0 - xc, y0 - yc)
         """
@@ -333,6 +341,7 @@ def get_position_error(
         """Get the position error between myself and other. If other is None, returns None.
 
         Returns:
+        --------
             err_x (float): x-axis position error[m].
             err_y (float): y-axis position error[m].
             err_z (float): z-axis position error[m].
@@ -351,6 +360,7 @@ def get_heading_error(
         """Get the heading error between myself and other. If other is None, returns None.
 
         Returns:
+        --------
             err_x (float): Roll error[rad], in [-pi, pi].
             err_y (float): Pitch error[rad], in [-pi, pi].
             err_z (float): Yaw error[rad], in [-pi, pi].
@@ -383,6 +393,7 @@ def get_velocity_error(
         If other is None, returns None. Also, velocity of myself or other is None, returns None too.
 
         Returns:
+        --------
             err_vx (float): x-axis velocity error[m/s].
             err_vy (float): y-axis velocity error[m/s].
             err_vz (float): z-axis velocity error[m/s].
@@ -403,6 +414,7 @@ def get_area_bev(self) -> float:
         """Get area of object BEV.
 
         Returns:
+        --------
             float: Area of footprint.
         """
         return self.state.footprint.area
@@ -411,6 +423,7 @@ def get_volume(self) -> float:
         """Get volume of bounding box.
 
         Returns:
+        --------
             float: Box volume.
         """
         return self.get_area_bev() * self.state.size[2]
@@ -427,11 +440,13 @@ def crop_pointcloud(
         Otherwise, returns pointcloud array outside of box.
 
         Args:
+        -----
             pointcloud (np.ndarray): The Array of pointcloud, in shape (N, 3).
             bbox_scale (float): Scale factor for bounding box. Defaults to 1.0.
             inside (bool): Whether crop inside pointcloud or outside. Defaults to True.
 
         Returns:
+        --------
             numpy.ndarray: Pointcloud array inside or outside box.
         """
         corners: np.ndarray = self.get_corners(scale=bbox_scale)
@@ -445,10 +460,12 @@ def get_inside_pointcloud_num(
         """Calculate the number of pointcloud inside of own bounding box.
 
         Args:
+        -----
             pointcloud (np.ndarray): The Array of pointcloud, in shape (N, 3).
             bbox_scale (float): Scale factor for bounding box. Defaults to 1.0.
 
         Returns:
+        --------
             int: Number of points inside of the own box.
         """
         inside_pointcloud: np.ndarray = self.crop_pointcloud(pointcloud, bbox_scale)
@@ -477,12 +494,14 @@ def _set_states(
         """Set object state from positions, orientations, sizes, and twists.
 
         Args:
+        -----
             positions (Optional[List[Tuple[float, float, float]]]): Sequence of positions. Defaults to None.
             orientations (Optional[List[Quaternion]], optional): Sequence of orientations. Defaults to None.
             sizes (Optional[List[Tuple[float, float, float]]]): Sequence of boxes sizes. Defaults to None.
             twists (Optional[List[Tuple[float, float, float]]]): Sequence of velocities. Defaults to None.
 
         Returns:
+        --------
             Optional[List[ObjectState]]: The list of ObjectState
         """
         if positions is None or orientations is None:

From 3a6033fc790140c7691cf648f6f343c4ec38926f Mon Sep 17 00:00:00 2001
From: ktro2828 <kotaro.uetake@tier4.jp>
Date: Wed, 10 Jan 2024 23:32:22 +0900
Subject: [PATCH 3/3] docs: update documents for metrics

Signed-off-by: ktro2828 <kotaro.uetake@tier4.jp>
---
 .../perception_eval/metrics/__init__.py       |  2 +-
 .../metrics/classification/__init__.py        |  2 +-
 .../metrics/classification/accuracy.py        | 26 +++---
 .../classification_metrics_score.py           |  5 +-
 .../metrics/config/__init__.py                |  2 +-
 .../metrics/config/_metrics_config_base.py    |  3 +-
 .../config/classification_metrics_config.py   | 11 +--
 .../config/detection_metrics_config.py        | 10 +--
 .../config/prediction_metrics_config.py       | 11 +--
 .../metrics/config/tracking_metrics_config.py | 11 +--
 .../metrics/detection/__init__.py             |  2 +-
 .../perception_eval/metrics/detection/ap.py   | 83 ++++++++-----------
 .../perception_eval/metrics/detection/map.py  |  9 +-
 .../metrics/detection/tp_metrics.py           | 28 ++-----
 .../perception_eval/metrics/metrics.py        | 34 ++++----
 .../metrics/metrics_score_config.py           | 19 ++---
 .../metrics/tracking/__init__.py              |  2 +-
 .../metrics/tracking/_metrics_base.py         | 17 ++--
 .../perception_eval/metrics/tracking/clear.py | 24 ++----
 .../tracking/tracking_metrics_score.py        |  9 +-
 20 files changed, 115 insertions(+), 195 deletions(-)

diff --git a/perception_eval/perception_eval/metrics/__init__.py b/perception_eval/perception_eval/metrics/__init__.py
index 0310fa43..7b7fad70 100644
--- a/perception_eval/perception_eval/metrics/__init__.py
+++ b/perception_eval/perception_eval/metrics/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
diff --git a/perception_eval/perception_eval/metrics/classification/__init__.py b/perception_eval/perception_eval/metrics/classification/__init__.py
index a529cf2f..c3139590 100644
--- a/perception_eval/perception_eval/metrics/classification/__init__.py
+++ b/perception_eval/perception_eval/metrics/classification/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
diff --git a/perception_eval/perception_eval/metrics/classification/accuracy.py b/perception_eval/perception_eval/metrics/classification/accuracy.py
index e8846437..80905793 100644
--- a/perception_eval/perception_eval/metrics/classification/accuracy.py
+++ b/perception_eval/perception_eval/metrics/classification/accuracy.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -26,22 +26,10 @@
 
 
 class ClassificationAccuracy:
-    """[summary]
-    Class to calculate classification accuracy.
-
-    Attributes:
-        target_labels (List[LabelType]): Target labels list.
-        num_ground_truth (int): Number of ground truths.
-        objects_results_num (int): Number of object results.
-        num_tp (int): Number of TP results.
-        num_fp (int): Number of FP results.
-        accuracy (float): Accuracy score. When `num_ground_truth+objects_results_num-num_tp=0`, this is float('inf').
-        precision (float): Precision score. When `num_tp+num_fp=0`, this is float('inf').
-        recall (float): Recall score. When `num_ground_truth=0`, this is float('inf').
-        f1score (float): F1 score. When `precision+recall=0`, this is float('inf').
-        results (Dict[str, float]): Dict that items are scores mapped by score names.
+    """Class to calculate classification accuracy.
 
     Args:
+    -----
         object_results (List[PerceptionObjectResult]): Object results list.
         num_ground_truth (int): Number of ground truths.
         target_labels (List[LabelType]): Target labels list.
@@ -83,9 +71,11 @@ def calculate_tp_fp(self, object_results: List[PerceptionObjectResult]) -> Tuple
         """Calculate accuracy score.
 
         Args:
+        -----
             object_results (List[PerceptionObjectResult]): Object results list.
 
         Returns:
+        --------
             num_tp (int): Number of TP results.
             num_fp (int): Number of FP results.
         """
@@ -104,9 +94,11 @@ def calculate_accuracy(self, num_tp: int) -> float:
         Accuracy = (TP + TN) / (TP + TN + FP + FN)
 
         Args:
+        -----
             num_tp (int): Number of TP results.
 
         Returns:
+        --------
             float: Accuracy score. When `objects_results_num+num_ground_truth-num_tp=0`, returns float('inf').
         """
         return (
@@ -122,9 +114,11 @@ def calculate_precision_recall(self, num_tp: int) -> Tuple[float, float]:
         Recall = TP / (TP + FN)
 
         Args:
+        -----
             num_tp (int): Number of TP results.
 
         Returns:
+        --------
             precision (float): Precision score. When `self.object_results_num=0`, returns float('inf').
             recall (float): Recall score. When `self.num_ground_truth=0`, returns float('inf').
         """
@@ -138,11 +132,13 @@ def calculate_f1score(self, precision: float, recall: float, beta: float = 1.0)
         F score = (1 + beta**2) * precision * recall / (beta**2 * precision + recall)
 
         Args:
+        -----
             precision (float): Precision score.
             recall (float): Recall score.
             beta (float): Defaults 1.0.
 
         Returns:
+        --------
             f1score (float): F1 score. When `precision+recall=0`, returns float('inf').
         """
         return (
diff --git a/perception_eval/perception_eval/metrics/classification/classification_metrics_score.py b/perception_eval/perception_eval/metrics/classification/classification_metrics_score.py
index 99f4297e..7310b7ec 100644
--- a/perception_eval/perception_eval/metrics/classification/classification_metrics_score.py
+++ b/perception_eval/perception_eval/metrics/classification/classification_metrics_score.py
@@ -29,10 +29,8 @@
 class ClassificationMetricsScore:
     """Metrics score class for classification evaluation.
 
-    Attributes:
-        self.accuracies (List[ClassificationAccuracy]): List of ClassificationAccuracy instances.
-
     Args:
+    -----
         object_results_dict (Dict[LabelType, List[List[PerceptionObjectResult]]]):
             Dict that are list of PerceptionObjectResult mapped by their labels.
         num_ground_truth_dict (Dict[LabelType, int]): Dict that are number of PerceptionObjectResult
@@ -62,6 +60,7 @@ def _summarize(self) -> Tuple[float, float, float, float]:
         """Summarize all ClassificationAccuracy.
 
         Returns:
+        --------
             accuracy (float): Accuracy score. When `num_est+num_gt-num_tp=0`, this is float('inf').
             precision (float): Precision score. When `num_gt+num_fp=0`, this is float('inf').
             recall (float): Recall score. When `num_gt=0`, this is float('inf').
diff --git a/perception_eval/perception_eval/metrics/config/__init__.py b/perception_eval/perception_eval/metrics/config/__init__.py
index 9be33190..e0c0beaf 100644
--- a/perception_eval/perception_eval/metrics/config/__init__.py
+++ b/perception_eval/perception_eval/metrics/config/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
diff --git a/perception_eval/perception_eval/metrics/config/_metrics_config_base.py b/perception_eval/perception_eval/metrics/config/_metrics_config_base.py
index 7085b6b3..011bd509 100644
--- a/perception_eval/perception_eval/metrics/config/_metrics_config_base.py
+++ b/perception_eval/perception_eval/metrics/config/_metrics_config_base.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -35,6 +35,7 @@ class _MetricsConfigBase(ABC):
     Then, threshold for car objects is 1.0. threshold for bike objects is 0.5. threshold for pedestrian object is 0.5.
 
     Args:
+    -----
         target_labels (List[LabelType]): Target labels list.
         center_distance_thresholds (List[List[float]]):
                 Thresholds List of center distance. Defaults to None.
diff --git a/perception_eval/perception_eval/metrics/config/classification_metrics_config.py b/perception_eval/perception_eval/metrics/config/classification_metrics_config.py
index 606adb9a..e9f399bd 100644
--- a/perception_eval/perception_eval/metrics/config/classification_metrics_config.py
+++ b/perception_eval/perception_eval/metrics/config/classification_metrics_config.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -29,15 +29,8 @@
 class ClassificationMetricsConfig(_MetricsConfigBase):
     """Configuration class for classification evaluation metrics.
 
-    Attributes:
-        evaluation_task (EvaluationTask.CLASSIFICATION)
-        target_labels (List[LabelType]): Target labels list.
-        center_distance_thresholds (List[float]): Thresholds list of center distance matching.
-        plane_distance_thresholds (List[float]): Threshold list of plane distance matching.
-        iou_2d_thresholds (List[float]): Thresholds list of 2d iou matching.
-        iou_3d_thresholds (List[float]): Thresholds list of 3d iou matching.
-
     Args:
+    -----
         target_labels (List[LabelType]): Target labels list.
         center_distance_thresholds (List[float]): Thresholds list of center distance matching.
         plane_distance_thresholds (List[float]): Threshold list of plane distance matching.
diff --git a/perception_eval/perception_eval/metrics/config/detection_metrics_config.py b/perception_eval/perception_eval/metrics/config/detection_metrics_config.py
index 50f123e9..71df912b 100644
--- a/perception_eval/perception_eval/metrics/config/detection_metrics_config.py
+++ b/perception_eval/perception_eval/metrics/config/detection_metrics_config.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -29,14 +29,6 @@
 class DetectionMetricsConfig(_MetricsConfigBase):
     """Configuration class for detection evaluation metrics.
 
-    Attributes:
-        evaluation_task (EvaluationTask.DETECTION)
-        target_labels (List[LabelType]): Target labels list.
-        center_distance_thresholds (List[float]): Thresholds list of center distance matching.
-        plane_distance_thresholds (List[float]): Threshold list of plane distance matching.
-        iou_2d_thresholds (List[float]): Thresholds list of 2d iou matching.
-        iou_3d_thresholds (List[float]): Thresholds list of 3d iou matching.
-
     Args:
         target_labels (List[LabelType]): Target labels list.
         center_distance_thresholds (List[float]): Thresholds list of center distance matching.
diff --git a/perception_eval/perception_eval/metrics/config/prediction_metrics_config.py b/perception_eval/perception_eval/metrics/config/prediction_metrics_config.py
index 5af96f20..4702921f 100644
--- a/perception_eval/perception_eval/metrics/config/prediction_metrics_config.py
+++ b/perception_eval/perception_eval/metrics/config/prediction_metrics_config.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -28,15 +28,8 @@
 class PredictionMetricsConfig(_MetricsConfigBase):
     """Configuration class for prediction evaluation metrics.
 
-    Attributes:
-        evaluation_task (EvaluationTask.PREDICTION)
-        target_labels (List[LabelType]): Target labels list.
-        center_distance_thresholds (List[float]): Thresholds list of center distance matching.
-        plane_distance_thresholds (List[float]): Threshold list of plane distance matching.
-        iou_2d_thresholds (List[float]): Thresholds list of 2d iou matching.
-        iou_3d_thresholds (List[float]): Thresholds list of 3d iou matching.
-
     Args:
+    -----
         target_labels (List[LabelType]): Target labels list.
         center_distance_thresholds (List[float]): Thresholds list of center distance matching.
         plane_distance_thresholds (List[float]): Threshold list of plane distance matching.
diff --git a/perception_eval/perception_eval/metrics/config/tracking_metrics_config.py b/perception_eval/perception_eval/metrics/config/tracking_metrics_config.py
index e2e9bcea..5ced4ae2 100644
--- a/perception_eval/perception_eval/metrics/config/tracking_metrics_config.py
+++ b/perception_eval/perception_eval/metrics/config/tracking_metrics_config.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -28,15 +28,8 @@
 class TrackingMetricsConfig(_MetricsConfigBase):
     """Configuration class for tracking evaluation metrics.
 
-    Attributes:
-        evaluation_task (EvaluationTask.TRACKING)
-        target_labels (List[LabelType]): Target labels list.
-        center_distance_thresholds (List[float]): Thresholds list of center distance matching.
-        plane_distance_thresholds (List[float]): Threshold list of plane distance matching.
-        iou_2d_thresholds (List[float]): Thresholds list of 2d iou matching.
-        iou_3d_thresholds (List[float]): Thresholds list of 3d iou matching.
-
     Args:
+    -----
         target_labels (List[LabelType]): Target labels list.
         center_distance_thresholds (List[float]): Thresholds list of center distance matching.
         plane_distance_thresholds (List[float]): Threshold list of plane distance matching.
diff --git a/perception_eval/perception_eval/metrics/detection/__init__.py b/perception_eval/perception_eval/metrics/detection/__init__.py
index acaaa366..31acc2a8 100644
--- a/perception_eval/perception_eval/metrics/detection/__init__.py
+++ b/perception_eval/perception_eval/metrics/detection/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
diff --git a/perception_eval/perception_eval/metrics/detection/ap.py b/perception_eval/perception_eval/metrics/detection/ap.py
index 076d45b1..c6afb7cc 100644
--- a/perception_eval/perception_eval/metrics/detection/ap.py
+++ b/perception_eval/perception_eval/metrics/detection/ap.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -39,21 +39,8 @@
 class Ap:
     """AP class.
 
-    Attributes:
-        ap (float): AP (Average Precision) score.
-        matching_average (Optional[float]): Average of matching score.
-            If there are no object results, this variable is None.
-        matching_mode (MatchingMode): MatchingMode instance.
-        matching_threshold (List[float]): Thresholds list for matching.
-        matching_standard_deviation (Optional[float]): Standard deviation of matching score.
-            If there are no object results, this variable is None.
-        target_labels (List[LabelType]): Target labels list.
-        tp_metrics (TPMetrics): Mode of TP metrics.
-        ground_truth_objects_num (int): Number ground truths.
-        tp_list (List[float]): List of the number of TP objects ordered by their confidences.
-        fp_list (List[float]): List of the number of FP objects ordered by their confidences.
-
     Args:
+    -----
         tp_metrics (TPMetrics): Mode of TP (True positive) metrics.
         object_results (List[List[PerceptionObjectResult]]): Object results list.
         num_ground_truth (int): Number of ground truths.
@@ -115,16 +102,13 @@ def __init__(
             matching_mode=self.matching_mode,
         )
 
-    def save_precision_recall_graph(
-        self,
-        result_directory: str,
-        frame_name: str,
-    ) -> None:
-        """[summary]
-        Save visualization image of precision and recall curve.
+    def save_precision_recall_graph(self, result_directory: str, frame_name: str) -> None:
+        """Save visualization image of precision and recall curve.
+
         The circle points represent original values and the square points represent interpolated ones.
 
         Args:
+        -----
             result_directory (str): The directory path to save images.
             frame_name (str): The frame name.
         """
@@ -161,16 +145,15 @@ def save_precision_recall_graph(
         plt.ylabel("Precision")
         plt.savefig(file_path)
 
-    def get_precision_recall_list(
-        self,
-    ) -> Tuple[List[float], List[float]]:
-        """[summary]
-        Calculate precision recall.
+    def get_precision_recall_list(self) -> Tuple[List[float], List[float]]:
+        """Calculate precision recall.
 
         Returns:
+        --------
             Tuple[List[float], List[float]]: tp_list and fp_list
 
         Example:
+        --------
             state
                 self.tp_list = [1, 1, 2, 3]
                 self.fp_list = [0, 1, 1, 1]
@@ -194,13 +177,15 @@ def interpolate_precision_recall_list(
         self,
         precision_list: List[float],
         recall_list: List[float],
-    ):
-        """[summary]
-        Interpolate precision and recall with maximum precision value per recall bins.
+    ) -> Tuple[List[float], List[float]]:
+        """Interpolates precision and recall with maximum precision value per recall bins.
 
         Args:
-            precision_list (List[float])
-            recall_list (List[float])
+            precision_list (List[float]): List of precisions.
+            recall_list (List[float]): List of recalls.
+
+        Returns:
+            Tuple[List[float], List[float]]: List of max precisions and recalls.
         """
         max_precision_list: List[float] = [precision_list[-1]]
         max_precision_recall_list: List[float] = [recall_list[-1]]
@@ -221,20 +206,22 @@ def _calculate_tp_fp(
         tp_metrics: Union[TPMetricsAp, TPMetricsAph],
         object_results: List[PerceptionObjectResult],
     ) -> Tuple[List[float], List[float]]:
-        """
-        Calculate TP (true positive) and FP (false positive).
+        """Calculates TP (true positive) and FP (false positive).
 
         Args:
+        -----
             tp_metrics (TPMetrics): The mode of TP (True positive) metrics
             object_results (List[PerceptionObjectResult]): the list of objects with result
 
         Return:
+        -------
             Tuple[tp_list, fp_list]
 
             tp_list (List[float]): the list of TP ordered by object confidence
             fp_list (List[float]): the list of FP ordered by object confidence
 
         Example:
+        --------
             whether object label is correct [True, False, True, True]
             return
                 tp_list = [1, 1, 2, 3]
@@ -282,22 +269,20 @@ def _calculate_tp_fp(
 
         return tp_list, fp_list
 
-    def _calculate_ap(
-        self,
-        precision_list: List[float],
-        recall_list: List[float],
-    ) -> float:
-        """[summary]
-        Calculate AP (average precision)
+    def _calculate_ap(self, precision_list: List[float], recall_list: List[float]) -> float:
+        """Calculates AP (average precision).
 
         Args:
+        -----
             precision_list (List[float]): The list of precision
             recall_list (List[float]): The list of recall
 
         Returns:
+        --------
             float: AP
 
         Example:
+        --------
             precision_list = [1.0, 0.5, 0.67, 0.75]
             recall_list = [0.25, 0.25, 0.5, 0.75]
 
@@ -329,14 +314,15 @@ def _calculate_average_sd(
         object_results: List[PerceptionObjectResult],
         matching_mode: MatchingMode,
     ) -> Tuple[Optional[float], Optional[float]]:
-        """[summary]
-        Calculate average and standard deviation.
+        """Calculates average and standard deviation.
 
         Args:
+        -----
             object_results (List[PerceptionObjectResult]): The object results
             matching_mode (MatchingMode): [description]
 
         Returns:
+        --------
             Tuple[float, float]: [description]
         """
 
@@ -352,10 +338,13 @@ def _calculate_average_sd(
 
     @staticmethod
     def _get_flat_str(str_list: List[str]) -> str:
-        """
-        Example:
-            a = _get_flat_str([aaa, bbb, ccc])
-            print(a) # aaa_bbb_ccc
+        """Flattens list of strings.
+
+        Args:
+            str_list (List[str]): List of strings.
+
+        Returns:
+            str: [aa, bb, cc] is flattened to aa_bb_cc.
         """
         output = ""
         for one_str in str_list:
diff --git a/perception_eval/perception_eval/metrics/detection/map.py b/perception_eval/perception_eval/metrics/detection/map.py
index bc29e90e..9a74b91a 100644
--- a/perception_eval/perception_eval/metrics/detection/map.py
+++ b/perception_eval/perception_eval/metrics/detection/map.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -31,12 +31,8 @@
 class Map:
     """mAP metrics score class.
 
-    Attributes:
-        map_config (MapConfig): The config for mAP calculation.
-        aps (List[Ap]): The list of AP (Average Precision) for each label.
-        map (float): mAP value.
-
     Args:
+    -----
         object_results (List[List[PerceptionObjectResult]]): The list of object results
         target_labels (List[LabelType]): Target labels to evaluate mAP
         matching_mode (MatchingMode): Matching mode like distance between the center of
@@ -100,6 +96,7 @@ def __str__(self) -> str:
         """__str__ method
 
         Returns:
+        --------
             str: Formatted string.
         """
 
diff --git a/perception_eval/perception_eval/metrics/detection/tp_metrics.py b/perception_eval/perception_eval/metrics/detection/tp_metrics.py
index 3ae6af1c..e6427d3f 100644
--- a/perception_eval/perception_eval/metrics/detection/tp_metrics.py
+++ b/perception_eval/perception_eval/metrics/detection/tp_metrics.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -27,11 +27,7 @@
 
 
 class TPMetrics(metaclass=ABCMeta):
-    """Tp metrics meta class
-
-    Attributes:
-        mode (str): TP metrics name.
-    """
+    """Tp metrics meta class"""
 
     mode: str
 
@@ -52,12 +48,7 @@ def get_value(self, object_result: PerceptionObjectResult) -> float:
 
 
 class TPMetricsAp(TPMetrics):
-    """
-    Ap metrics class
-
-    Attributes:
-        mode (str): TP metrics name that is TPMetricsAp.
-    """
+    """Ap metrics class."""
 
     mode: str = "TPMetricsAp"
 
@@ -76,12 +67,7 @@ def get_value(self, object_result: PerceptionObjectResult) -> float:
 
 
 class TPMetricsAph(TPMetrics):
-    """
-    Aph metrics class
-
-    Attributes:
-        mode (str): TP metrics name that is TPMetricsAph.
-    """
+    """Aph metrics class."""
 
     mode: str = "TPMetricsAph"
 
@@ -119,11 +105,7 @@ def get_value(self, object_result: PerceptionObjectResult) -> float:
 
 
 class TPMetricsConfidence(TPMetrics):
-    """Confidence TP class.
-
-    Attributes:
-        mode (str): TP metrics name that is TPMetricsConfidence.
-    """
+    """Confidence TP class."""
 
     mode: str = "TPMetricsConfidence"
 
diff --git a/perception_eval/perception_eval/metrics/metrics.py b/perception_eval/perception_eval/metrics/metrics.py
index de787f20..560b0a2e 100644
--- a/perception_eval/perception_eval/metrics/metrics.py
+++ b/perception_eval/perception_eval/metrics/metrics.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -33,19 +33,8 @@
 class MetricsScore:
     """Metrics score class.
 
-    Attributes:
-        detection_config (Optional[DetectionMetricsConfig]): Config for detection evaluation.
-        tracking_config (Optional[DetectionMetricsConfig]): Config for tracking evaluation.
-        prediction_config (Optional[PredictionMetricsConfig]): Config for prediction evaluation.
-        classification_config (Optional[ClassificationMetricsConfig]): Config for classification evaluation.
-        evaluation_task (EvaluationTask): EvaluationTask instance.
-        maps (List[Map]): List of mAP instances. Each mAP is different from threshold
-                               for matching (ex. IoU 0.3).
-        tracking_scores (List[TrackingMetricsScore]): List of TrackingMetricsScore instances.
-        prediction_scores (List[TODO]): TBD
-        classification_score (List[ClassificationMetricsScore]): List of ClassificationMetricsScore instances.
-
     Args:
+    -----
         config (MetricsScoreConfig): MetricsScoreConfig instance.
         used_frame: List[int]: List of frame numbers loaded to evaluate.
     """
@@ -127,10 +116,10 @@ def evaluate_detection(
         object_results: Dict[LabelType, List[PerceptionObjectResult]],
         num_ground_truth: Dict[LabelType, int],
     ) -> None:
-        """[summary]
-        Calculate detection metrics
+        """Calculate detection metrics
 
         Args:
+        -----
             object_results (Dict[LabelType, List[PerceptionObjectResult]]): The dict of object result
         """
         if self.tracking_config is None:
@@ -183,8 +172,7 @@ def evaluate_tracking(
         object_results: Dict[LabelType, List[List[PerceptionObjectResult]]],
         num_ground_truth: Dict[LabelType, int],
     ) -> None:
-        """[summary]
-        Calculate tracking metrics.
+        """Calculate tracking metrics.
 
         NOTE:
             object_results and ground_truth_objects must be nested list.
@@ -192,6 +180,7 @@ def evaluate_tracking(
             In case of evaluating multi frame, [[], [t1], [t2], ..., [tn]]
 
         Args:
+        -----
             object_results (List[List[PerceptionObjectResult]]): The list of object result for each frame.
         """
         self.__num_gt += sum(num_ground_truth.values())
@@ -240,10 +229,10 @@ def evaluate_prediction(
         object_results: Dict[LabelType, List[PerceptionObjectResult]],
         num_ground_truth: Dict[LabelType, int],
     ) -> None:
-        """[summary]
-        Calculate prediction metrics
+        """Calculate prediction metrics.
 
         Args:
+        -----
             object_results (List[PerceptionObjectResult]): The list of object result
         """
         pass
@@ -253,6 +242,13 @@ def evaluate_classification(
         object_results: Dict[LabelType, List[List[PerceptionObjectResult]]],
         num_ground_truth: Dict[LabelType, int],
     ) -> None:
+        """Calculates classification metrics score.
+
+        Args:
+        -----
+            object_results (Dict[LabelType, List[List[PerceptionObjectResult]]]): Object results.
+            num_ground_truth (Dict[LabelType, int]): The number of ground truths.
+        """
         self.__num_gt += sum(num_ground_truth.values())
         classification_score_ = ClassificationMetricsScore(
             object_results_dict=object_results,
diff --git a/perception_eval/perception_eval/metrics/metrics_score_config.py b/perception_eval/perception_eval/metrics/metrics_score_config.py
index 6260cbbc..39e0346b 100644
--- a/perception_eval/perception_eval/metrics/metrics_score_config.py
+++ b/perception_eval/perception_eval/metrics/metrics_score_config.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -36,16 +36,9 @@
 class MetricsScoreConfig:
     """A configuration class for each evaluation task metrics.
 
-    Attributes:
-        detection_config (Optional[DetectionMetricsConfig]): Config for detection evaluation.
-        tracking_config (Optional[DetectionMetricsConfig]): Config for tracking evaluation.
-        prediction_config (Optional[PredictionMetricsConfig]): Config for prediction evaluation.
-        classification_config (Optional[ClassificationMetricsConfig]): Config for classification evaluation.
-        evaluation_task (EvaluationTask): EvaluationTask instance.
-
     Args:
-        evaluation_task (EvaluationTask): EvaluationTask instance.
-        **cfg: Configuration data.
+    -----
+        params (PerceptionMetricsParam): Parameters for metrics score.
     """
 
     def __init__(self, params: PerceptionMetricsParam) -> None:
@@ -81,10 +74,16 @@ def _extract_params(config: _MetricsConfigBase, params: PerceptionMetricsParam)
         """Check if input parameters are valid.
 
         Args:
+        -----
             config (_MetricsConfigBase): Metrics score instance.
             params (PerceptionMetricsParam): Parameters for metrics.
 
+        Returns:
+        --------
+            Dict[str, Any]: Extracted params.
+
         Raises:
+        -------
             KeyError: When got invalid parameter names.
         """
         input_params_dict = params.as_dict()
diff --git a/perception_eval/perception_eval/metrics/tracking/__init__.py b/perception_eval/perception_eval/metrics/tracking/__init__.py
index 9fcce01e..8c2bdae9 100644
--- a/perception_eval/perception_eval/metrics/tracking/__init__.py
+++ b/perception_eval/perception_eval/metrics/tracking/__init__.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
diff --git a/perception_eval/perception_eval/metrics/tracking/_metrics_base.py b/perception_eval/perception_eval/metrics/tracking/_metrics_base.py
index 7f984a78..a38a71f8 100644
--- a/perception_eval/perception_eval/metrics/tracking/_metrics_base.py
+++ b/perception_eval/perception_eval/metrics/tracking/_metrics_base.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -34,15 +34,8 @@
 class _TrackingMetricsBase(ABC):
     """Abstract base class for tracking metrics.
 
-    Attributes:
-        num_ground_truth (int): Number of ground truths.
-        target_labels (List[LabelType]): Target labels list.
-        matching_mode (MatchingMode): MatchingMode instance.
-        matching_threshold_list (List[float]): Thresholds list for matching.
-        metrics_field (Optional[List[str]]): Filed name of target metrics. If it is not specified, default supported metrics are used.
-        support_metrics (List[str]): List of supported metrics names.
-
     Args:
+    -----
         num_ground_truth (int): Number of ground truths.
         target_labels (List[LabelType]): Target labels list.
         matching_mode (MatchingMode): MatchingMode instance.
@@ -95,12 +88,15 @@ def _check_metrics(self, metrics_field: Optional[List[str]]) -> List[str]:
         If `metrics_field=None`, returns default supported metrics.
 
         Args:
+        -----
             metrics_field (Optional[List[str]]): The list of executing metrics field.
 
         Returns:
+        --------
             metrics_field (List[str])
 
         Raises:
+        -------
             ValueError: If input metrics field is unsupported.
         """
         if metrics_field is None:
@@ -143,13 +139,16 @@ def get_precision_recall_list(
         """Calculate precision recall.
 
         Args:
+        -----
             tp_list (List[float]): TP results list.
             ground_truth_objects_num (int): Number of ground truths.
 
         Returns:
+        --------
             Tuple[List[float], List[float]]: tp_list and fp_list
 
         Examples:
+        ---------
             >>> tp_list = [1, 1, 2, 3]
             >>> ground_truth_num = 4
             >>> precision_list, recall_list = self.get_precision_recall(tp_list, ground_truth_num)
diff --git a/perception_eval/perception_eval/metrics/tracking/clear.py b/perception_eval/perception_eval/metrics/tracking/clear.py
index 966def37..ed823675 100644
--- a/perception_eval/perception_eval/metrics/tracking/clear.py
+++ b/perception_eval/perception_eval/metrics/tracking/clear.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -43,21 +43,8 @@ class CLEAR(_TrackingMetricsBase):
 
     NOTE: MT, ML, PT is under construction.
 
-    Attributes:
-        target_labels (List[LabelType]): The list of target label.
-        matching_mode (MatchingMode): The target matching mode.
-        metrics_field (Optional[List[str]]): The list of target metrics name. If not specified, set default supported metrics.
-        ground_truth_objects_num (int): The number of ground truth.
-        support_metrics (List[str]): The list of supported metrics name. (["MOTA", "MOTP"])
-        tp (float): The total value/number of TP.
-        fp (float): The total value/number of FP.
-        id_switch (int): The total number of ID switch.
-        tp_matching_score (float): The total value of matching score in TP.
-        mota (float): MOTA score.
-        motp (float): MOTP score.
-        results (OrderedDict[str, Any]): The dict to keep scores.
-
     Args:
+    -----
         object_results (List[List[PerceptionObjectResult]]): The list of object results for each frames.
         num_ground_truth (int): The number of ground truth.
         target_labels (List[LabelType]): The list of target labels.
@@ -138,6 +125,7 @@ def _calculate_score(self):
             if the number of GT is 0, MOTA returns inf and if the TP score is 0, MOTP returns inf.
 
         Returns:
+        --------
             mota (float): MOTA score.
             motp (float): MOTP score.
         """
@@ -162,10 +150,12 @@ def _calculate_tp_fp(
         """Calculate matching compared with previous object results.
 
         Args:
+        -----
             cur_object_results (List[List[PerceptionObjectResult]]): Object results list at current frame.
             prev_object_results (List[List[PerceptionObjectResult]]): Object results list at previous frame.
 
         Returns:
+        --------
             tp (float): Total value of TP. If matching is True, num_tp += 1.0.
             fp (float): Total value of FP. If matching is False, num_fp += 1.0.
             num_id_switch (int): Total number of ID switch. If matching is switched compared with previous result, num_id_switch += 1.
@@ -241,10 +231,12 @@ def _is_id_switched(cur_object_result: PerceptionObjectResult, prev_object_resul
             GT ID is unique between the different labels.
 
         Args:
+        -----
             cur_object_result (PerceptionObjectResult): Object result at current frame.
             prev_object_result (PerceptionObjectResult): Object result at previous frame.
 
         Returns:
+        --------
             bool: Return True if ID is switched.
         """
         # current GT = None -> FP
@@ -282,10 +274,12 @@ def _is_same_match(
         When previous or current GT is None(=FP), return False regardless the ID of estimated.
 
         Args:
+        -----
             cur_object_result (PerceptionObjectResult): Object result at current frame.
             prev_object_result (PerceptionObjectResult):Object result at previous frame.
 
         Returns:
+        --------
             bool: Return True if both estimated and GT ID are same.
         """
         if cur_object_result.ground_truth_object is None or prev_object_result.ground_truth_object is None:
diff --git a/perception_eval/perception_eval/metrics/tracking/tracking_metrics_score.py b/perception_eval/perception_eval/metrics/tracking/tracking_metrics_score.py
index e09829f6..39689a89 100644
--- a/perception_eval/perception_eval/metrics/tracking/tracking_metrics_score.py
+++ b/perception_eval/perception_eval/metrics/tracking/tracking_metrics_score.py
@@ -1,4 +1,4 @@
-# Copyright 2022 TIER IV, Inc.
+# Copyright 2022-2024 TIER IV, Inc.
 
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -33,12 +33,8 @@ class TrackingMetricsScore:
 
     Length of input `target_labels` and `matching_threshold_list` must be same.
 
-    Attributes:
-        target_labels: (List[LabelType]): Target labels list.
-        matching_mode (MatchingMode): MatchingMode instance.
-        clears (List[CLEAR]): List of CLEAR instances.
-
     Args:
+    -----
         object_results_dict (Dict[LabelType, List[List[PerceptionObjectResult]]):
             Dict that items are object results list mapped by their labels.
         num_ground_truth (int): Number of ground truths.
@@ -78,6 +74,7 @@ def _sum_clear(self) -> Tuple[float, float, int]:
         """Summing up multi CLEAR result.
 
         Returns:
+        --------
             mota (float): MOTA score.
             motp (float): MOTP score.
             num_id_switch (int): The number of ID switched.