FrameDrop-RL/make_data.py at main · webb-c/FrameDrop-RL · 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
from collections import Counter
import csv
import os
from utils.yolov5.detect import inference
from utils.cal_F1 import get_F1
from torch.utils.tensorboard import SummaryWriter

root_detection = "./data/detect/test/"
video_dir = "../DATASET/test/"


def f1_test(video_name, idx_list, writer):
    video_path = os.path.join(video_dir, video_name + ".mp4")
    origin_detection_path = os.path.join(root_detection, video_name + "/labels")
    if not os.path.exists(origin_detection_path):
        command = ["--weights", "models/yolov5s6.pt", "--source", video_path, "--project", root_detection, "--name", video_name, "--save-txt", "--save-conf", "--nosave"]
        inference(command)

    origin_list = os.listdir(origin_detection_path)
    num_file = min(len(origin_list), len(idx_list))
    print(len(origin_list))
    print(len(idx_list))

    total_F1 = 0
    for i in range(num_file):
        skip_idx = idx_list[i]
        if skip_idx == -1: skip_idx = 0

        #print("idx: ", i, "skip_idx: ", skip_idx)
        origin, skip = origin_list[i], origin_list[skip_idx]
        origin_file = os.path.join(origin_detection_path, origin)
        skip_file = os.path.join(origin_detection_path, skip)
        F1_score = get_F1(origin_file, skip_file)

        if writer is not None:
            writer.add_scalar("F1_score/changes", F1_score, i)
        total_F1 += F1_score

    if writer is not None:
        writer.add_scalar("F1_score/total", total_F1, 1)
        writer.add_scalar("F1_score/average", total_F1/num_file, 1)

    F1_score = total_F1/num_file

    return F1_score


def make_idx_to_fraction_change_reducto(idx_list):
    segment = 120


def make_idx_to_fraction_change(idx_list, writer, interval=30):

    send_frame = 0
    for i in range(len(idx_list)):
        skip_idx = idx_list[i]
        if skip_idx == -1: skip_idx = 0

        #print("idx: ", i, "skip_idx: ", skip_idx)
        if i == skip_idx:
            send_frame += 1


    Fraction = send_frame/len(idx_list)

    return Fraction


def save_parameters_to_csv(data, prefix, f1_score, fraction):
    csv_file_path = 'test_exp_result.csv'

    existing_data = []
    if os.path.exists(csv_file_path):
        with open(csv_file_path, 'r') as csvfile:
            csv_reader = csv.reader(csvfile)
            for row in csv_reader:
                existing_data.append(row)

    new_row = [data, prefix, str(f1_score), str(fraction)]
    existing_data.append(new_row)

    with open(csv_file_path, 'w', newline='') as csvfile:
        csv_writer = csv.writer(csvfile)
        csv_writer.writerows(existing_data)


if __name__ == "__main__":
    matching = {
        'JK': 'JK-1',
        'SD': 'SD-1',
        'JN': 'JN',
    }

    exp_matching = {}
    with open('matching.csv', 'r', newline='') as file:
        reader = csv.reader(file)
        for row in reader:
            if len(row) >= 2:
                key = row[0]
                value = row[1]
                exp_matching[key] = value

    #path_list = ['config/test_config_LRLO.csv', 'config/test_config_V_effect.csv']
    path_list = ['config/test_config_LRLO_re_re.csv', 'config/test_config_LRLO_V_effect.csv', 'config/test_config_448_LRLO.csv']
    video_name = ['JK', 'SD']


    for config_path in path_list:
        print("In test config", config_path)

        with open(config_path, newline='') as csvfile:
            reader = csv.reader(csvfile)

            for row in reader:
                value = list(row)
                prefix = value[0]
                if value[2] != 'LRLO':
                    continue
                    idx_list_2d = eval(value[-2])
                    idx_list = [item for sublist in idx_list_2d for item in sublist]
                    video = value[3].split("/")[3]
                    if video == 'JN-1':
                        video = 'JN'
                    video_path = os.path.join(video_dir, video + ".mp4")
                    origin_detection_path = os.path.join(root_detection, video + "/labels")

                    if not os.path.exists(origin_detection_path):
                        command = ["--weights", "models/yolov5s6.pt", "--source", video_path, "--project", root_detection, "--name", video, "--save-txt", "--save-conf", "--nosave"]
                        inference(command)
                    origin_list = os.listdir(origin_detection_path)
                    fraction = len(idx_list) / len(origin_list)

                else:
                    idx_list = eval(value[-2])
                    print(idx_list[0], type(idx_list[0]))
                    idx_list = list(map(int, value[-2].strip('[]').split(',')))
                    video = value[3].split("/")[-1].split(".")[0]
                    f1_score = f1_test(video, idx_list, None)
                    fraction = make_idx_to_fraction_change(idx_list, None)

                fraction_jetson = value[-1]

                print("prefix:", prefix, "\t(fraction:", fraction, "*jetson score:", fraction_jetson, ")\n")
                save_parameters_to_csv(video, prefix, f1_score, fraction)