KPI_analysis20220106/data_matching.py at main · WeixianFu/KPI_analysis20220106 · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
@author: Weixian Fu
"""

import numpy as np
from iou_calculation import iou_cv
import pandas as pd

# old data matching function
# def data_matching(df_true: pd.core.frame.DataFrame, df_pred: pd.core.frame.DataFrame,
#                   dimensions: int = 2, iou_threshold: float = 0.5) -> (list, list):
#     """
#     old version data matching function, have be replaced by new data matching function
#     data matching function
#     :param iou_threshold: threshold of iou matching
#     :param dimension: 1 or 2, (2D or 3D), default = 2
#     :param df_true: dataframe of true data
#     :param df_pred: dataframe of pred data
#     :return: two list, pred_list = ['Car', 'Car', ..., ], pred_list = ['Car', 'Van', 'UnKonwn', ... ,]
#     """
#     pred_list = []
#     true_list = []
#
#     for num in set(df_pred['msg_number']):
#         pred_msg_frame = df_pred[(df_pred['msg_number'] == num)]
#         true_msg_frame = df_true[(df_true['msg_number'] == num)]
#         true_class = list(true_msg_frame['class_label_true'])
#         true_class_set = set(true_class)
#         dict_true = np.zeros(true_msg_frame.shape[0])
#
#         for index, msg_bbox_pred in pred_msg_frame.iterrows():
#             pred_matched_or_not = False
#             if msg_bbox_pred['class_label_pred'] not in true_class_set:
#                 pred_list.append(msg_bbox_pred['class_label_pred'])
#                 true_list.append('UnKnown')
#                 pred_matched_or_not = True
#             else:
#                 bbox_counter = -1
#                 for index_true, msg_bbox_true in true_msg_frame.iterrows():
#                     bbox_counter += 1
#                     if iou_result(msg_bbox_true, msg_bbox_pred, dimensions, iou_threshold):
#                         pred_list.append(msg_bbox_pred['class_label_pred'])
#                         true_list.append(msg_bbox_true['class_label_true'])
#                         dict_true[bbox_counter] = 1
#                         pred_matched_or_not = True
#                         break
#             if not pred_matched_or_not:
#                 pred_list.append(msg_bbox_pred['class_label_pred'])
#                 true_list.append('UnKnown')
#
#         for num1 in range(len(dict_true)):
#             if dict_true[num1] == 0:
#                 true_list.append(true_class[num1])
#                 pred_list.append('UnKnown')
#
#     for num in (set(df_true['msg_number']) - set(df_pred['msg_number'])):
#         for index, msg_bbox_true in df_true[(df_true['msg_number'] == num)].iterrows():
#             true_list.append(msg_bbox_true['class_label_true'])
#             pred_list.append('UnKnown')
#
#     return true_list, pred_list


def data_matching(df_true: pd.core.frame.DataFrame, df_pred: pd.core.frame.DataFrame,
                  dimensions: int = 2, iou_threshold: float = 0.5) -> (list, list):
    """
    data matching function
    :param dimensions: 1 or 2, (2D or 3D), default = 2
    :param iou_threshold: threshold of iou matching
    :param df_true: dataframe of true data
    :param df_pred: dataframe of pred data
    :return: two list, pred_list = ['Car', 'Car', ..., ], pred_list = ['Car', 'Van', 'UnKonwn', ... ,]
    """
    pred_list = []
    true_list = []

    for num in set(df_true['msg_number']) - set(df_pred['msg_number']):
        for index, msg_object_true in df_true[(df_true['msg_number'] == num)].iterrows():
            true_list.append(msg_object_true['class_label_true'])
            pred_list.append('UnKnown')
    for num in set(df_pred['msg_number']) - set(df_true['msg_number']):
        for index, msg_object_pred in df_pred[(df_pred['msg_number'] == num)].iterrows():
            pred_list.append(msg_object_pred['class_label_pred'])
            true_list.append('UnKnown')
    for num in set(df_true['msg_number']) & set(df_pred['msg_number']):
        pred_frame_df = df_pred[(df_pred['msg_number'] == num)]
        true_frame_df = df_true[(df_true['msg_number'] == num)]
        set_class_pred = set(pred_frame_df['class_label_pred'])
        set_class_true = set(true_frame_df['class_label_true'])
        for classes in set_class_true - set_class_pred:
            for index, msg_object_true in true_frame_df[(true_frame_df['class_label_true'] == classes)].iterrows():
                true_list.append(classes)
                pred_list.append('UnKnown')
        for classes in set_class_pred - set_class_true:
            for index, msg_object_pred in pred_frame_df[(pred_frame_df['class_label_true'] == classes)].iterrows():
                true_list.append('Unknown')
                pred_list.append(classes)
        for classes in set_class_true & set_class_pred:
            pred_frame_class_df = pred_frame_df[(pred_frame_df['class_label_pred'] == classes)]
            true_frame_class_df = true_frame_df[(true_frame_df['class_label_true'] == classes)]
            matching_results = iou_cv(true_frame_class_df, pred_frame_class_df, dimensions, iou_threshold)
            for i in range(matching_results[0]):
                true_list.append(classes)
                pred_list.append(classes)
            for i in range(matching_results[1]):
                true_list.append(classes)
                pred_list.append('UnKnown')
            for i in range(matching_results[2]):
                true_list.append('UnKnown')
                pred_list.append(classes)
    return true_list, pred_list


if __name__ == '__main__':
    import time
    import math
    from data_initialization import rosbag_ouput_to_dataframe
    from distance_filter import distance_filter_remove

    start = time.clock()  # start time
    list_args = [
        ["/home/ubuntu18/liangdao/Data/19-merge.bag", "/home/ubuntu18/liangdao/Data/19_ALT_OD_CP2.bag"],
        ["/home/ubuntu18/liangdao/Data/20-merge.bag", "/home/ubuntu18/liangdao/Data/20_ALT_OD_CP2.bag"],
        ["/home/ubuntu18/liangdao/Data/21-merge.bag", "/home/ubuntu18/liangdao/Data/21_ALT_OD_CP2.bag"],
        ["/home/ubuntu18/liangdao/Data/40-merge.bag", "/home/ubuntu18liangdao/Data/40_ALT_OD_CP2.bag"],
        ["/home/ubuntu18/liangdao/Data/42-merge.bag", "/home/ubuntu18/liangdao/Data/42_ALT_OD_CP2.bag"],
        ["/home/ubuntu18/liangdao/Data/48-merge.bag", "/home/ubuntu18/liangdao/Data/48_ALT_OD_CP2.bag"]
    ]
    # obtain data from rosbag document

    msg_df_pred, msg_df_pred_number = rosbag_ouput_to_dataframe("/home/ubuntu18/liangdao/Data/19-merge.bag",
                                                                topic=['/ld_object_lists'])
    msg_df_true, msg_df_true_number = rosbag_ouput_to_dataframe("/home/ubuntu18/liangdao/Data/19_ALT_OD_CP2.bag",
                                                                topic=['/ld_object_lists'])
    # distance filter
    distance_range = [[0, 30], [-10, 10], [-math.inf, math.inf]]
    # msg_df_true_filtered = distance_filter(msg_df_true, distance_range)
    # msg_df_pred_filtered = distance_filter(msg_df_pred, distance_range)

    # distance filter remove
    msg_df_true_filtered_removed = distance_filter_remove(msg_df_true, distance_range)
    msg_df_pred_filtered_removed = distance_filter_remove(msg_df_pred, distance_range)

    end = time.clock()  # end time
    print('run time:', str(end - start))