-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathb_analyze_encoding_results.py
207 lines (172 loc) · 10.2 KB
/
b_analyze_encoding_results.py
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
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
import sqlite3
import numpy as np
import ntpath
import matplotlib.pyplot as plt
from collections import defaultdict
from utils.u_utils_common import easy_logging, get_mean_metric_value_file_size_bytes, get_metric_value_file_size_bytes, \
get_print_string
from utils.u_mysql_execute import get_unique_sorted, query_for_codec, apply_checks_before_analyzing, apply_size_check, \
get_unique_sorted_with_sub_sampling
from Config.config_analyse import args_analyze_config
blanket = 20
codec_len = 15
def eve_line(logger, all_codec_results_terse_all, source_image_all, CR):
num_size = len(all_codec_results_terse_all)
max_arr = np.ones(num_size) * np.inf * -1
min_arr = np.ones(num_size) * np.inf
for i in range(len(all_codec_results_terse_all[0])):
consolidated_results = ""
for k, a in enumerate(all_codec_results_terse_all):
consolidated_results += '{:.2f}%'.format(a[i]).rjust(codec_len) if CR == False else '{:.2f}'.format(
a[i]).rjust(codec_len)
max_arr[k] = max(max_arr[k], a[i])
min_arr[k] = min(min_arr[k], a[i])
prefix = source_image_all[i]
if CR:
prefix += " CR"
if args_analyze_config().every_images == 1:
logger.info('{}:{}'.format(str(prefix).ljust(blanket), consolidated_results))
return (max_arr, min_arr)
def avg_line(logger, codec_results_terse, target, all_codec_results_terse_all, prefix):
logger.info('-' * (blanket + 1 + codec_len * len(all_codec_results_terse_all)))
consolidated_results = ""
for a in codec_results_terse:
consolidated_results += a.ljust(codec_len) if prefix == 'Terse' else '{:.2f}'.format(a).rjust(codec_len)
num_size = len(all_codec_results_terse_all[0])
t_prefix = str(
'T {}({})'.format(target, num_size) if not args_analyze_config().lossless else 'avg {}({})'.format(prefix,
num_size)).ljust(
blanket)
logger.info('{}:{}'.format(t_prefix, consolidated_results))
def max_min_len(logger, array_max_min, prefix):
max_line, min_line = array_max_min
tt = ""
if prefix == 'Terse':
max_line, min_line = min_line, max_line
tt = '%'
def tmp_line(data_line, tmp_prefix):
consolidated_results = ""
for a in data_line:
consolidated_results += '{:.2f}{}'.format(a, tt).rjust(codec_len)
prefix_ = '{} {}'.format(tmp_prefix, prefix).ljust(blanket)
logger.info('{}:{}'.format(prefix_, consolidated_results))
tmp_line(max_line, 'max')
tmp_line(min_line, 'min')
def show_result(logger, codecs, sub_sampling, unique_sorted_metric_values, source_image_all, results_dict):
print('\n')
logger.info('=' * (blanket + 1 + codec_len * len(codecs)))
sub_sampling_report = '{} subsampling'.format(sub_sampling)
logger.info(sub_sampling_report)
logger.info('-' * len(sub_sampling_report))
codecs_string = ' ' * (blanket + 1)
for codec in codecs:
codecs_string += codec.rjust(codec_len)
logger.info(codecs_string)
for target in unique_sorted_metric_values:
results_list_compress_rate, results_list_compress_rate_all, codec_results_terse, all_codec_results_terse_all = \
results_dict[target]
# ======================== EVE ======================== #
terse_max_min = eve_line(logger, all_codec_results_terse_all, source_image_all, False)
cr_max_min = eve_line(logger, results_list_compress_rate_all, source_image_all, True)
# ======================== AVG ======================== #
avg_line(logger, codec_results_terse, target, all_codec_results_terse_all, 'Terse')
max_min_len(logger, terse_max_min, 'Terse')
avg_line(logger, results_list_compress_rate, target, results_list_compress_rate_all, 'CR')
max_min_len(logger, cr_max_min, 'CR')
logger.info('=' * (blanket + 1 + codec_len * len(codecs)))
logger.info("\n\n")
def main():
# =================================== BASIC CONFIGS =================================== #
metric_name = args_analyze_config().metric
db_file_name = args_analyze_config().db_file_name
verbose = args_analyze_config().quiet
logger = easy_logging(file_prefix="bitrate_savings", db_file_name=db_file_name) # logging
connection = sqlite3.connect(db_file_name) # connect database
baseline_codec = args_analyze_config().baseline_codec # baseline
sub_sampling_arr = get_unique_sorted(connection, "SUB_SAMPLING") # subsampling
color_list = ['tab:blue', 'tab:orange', 'tab:green', 'tab:red', 'tab:purple', 'tab:brown', 'tab:pink', 'tab:gray',
'tab:olive', 'tab:cyan']
marker_list = ['o', 'v', '>', '<', 's', 'p', 'd', '4', 'P', 'X'] # 'D', '|', '_'
assert len(color_list) == len(marker_list)
# =================================== CHECK DATA =================================== #
if not args_analyze_config().lossless:
unique_sorted_metric_values, total_pixels = apply_checks_before_analyzing(connection, metric_name)
else:
total_pixels = apply_size_check(connection)
unique_sorted_metric_values = get_unique_sorted(connection, "TARGET_VALUE")
# =================================== ANALYZE sub_sampling =================================== #
for sub_sampling in sub_sampling_arr:
results_dict = dict()
results_bpp = defaultdict(list)
results_quality = defaultdict(list)
codecs = get_unique_sorted_with_sub_sampling(connection, "CODEC", sub_sampling)
# =================================== ANALYZE target =================================== #
for target in unique_sorted_metric_values:
baseline_results = connection.execute(
query_for_codec(baseline_codec, sub_sampling, metric_name, target)).fetchall()
baseline_metric_value, baseline_file_size, \
baseline_count_nums, baseline_vmaf_value, \
baseline_compress_rate = get_mean_metric_value_file_size_bytes(baseline_results)
baseline_metric_values_all, baseline_file_size_all, \
baselne_vmaf_values_all, baseline_source_image_all, \
baseline_compress_rate_all, baseline_bpp_all = get_metric_value_file_size_bytes(baseline_results)
if verbose == 0:
print('Baseline is ' + get_print_string(baseline_codec, sub_sampling, baseline_count_nums,
baseline_metric_value,
baseline_file_size, metric_name, baseline_vmaf_value))
results_bpp[baseline_codec].append(baseline_file_size * 8.0 / total_pixels)
results_quality[baseline_codec].append(baseline_metric_value)
results_list_compress_rate = list()
results_list_compress_rate_all = list()
results_list_terse = list()
results_list_terse_all = list()
# =================================== ANALYZE codec =================================== #
for codec in codecs:
results = connection.execute(query_for_codec(codec, sub_sampling, metric_name, target)).fetchall()
metric_value, file_size, count, vmaf_value, compress_rate = get_mean_metric_value_file_size_bytes(
results)
if verbose == 0:
print(' Compared codec is ' + get_print_string(codec, sub_sampling, count, metric_value, file_size,
metric_name, vmaf_value))
print(' Average reduction is {:.2f}%'.format(
(file_size - baseline_file_size) / baseline_file_size * 100.0))
# negative is better. Positive means increase in file_size
results_list_compress_rate.append(compress_rate)
results_list_terse.append(
'{:.2f}%'.format((file_size - baseline_file_size) / baseline_file_size * 100.0).rjust(codec_len))
metric_values_all, file_size_all, \
vmaf_values_all, source_image_all, \
compress_rate_all, bpp_all = get_metric_value_file_size_bytes(
results)
assert source_image_all == baseline_source_image_all
results_list_terse_all.append((np.array(file_size_all) - np.array(baseline_file_size_all)) / np.array(
baseline_file_size_all) * 100)
results_list_compress_rate_all.append(compress_rate_all)
results_bpp[codec].append(file_size * 8.0 / total_pixels)
results_quality[codec].append(metric_value)
results_dict[target] = (
results_list_compress_rate, results_list_compress_rate_all, results_list_terse, results_list_terse_all)
print("")
# =================================== SHOW RESULT =================================== #
show_result(logger, codecs, sub_sampling, unique_sorted_metric_values, source_image_all, results_dict)
# =================================== PLOT =================================== #
fig = plt.figure(figsize=(12.8, 7.2))
marker_num = 0
plt.plot(results_bpp[baseline_codec], results_quality[baseline_codec], linewidth=2.0,
color=color_list[marker_num], marker=marker_list[marker_num], label=baseline_codec)
for codec in codecs:
if codec not in baseline_codec and results_quality[codec][0] < float("inf"):
marker_num += 1
plt.plot(results_bpp[codec], results_quality[codec], linewidth=2.0,
# color=color_list[marker_num],
# marker=marker_list[marker_num],
label=codec)
plt.legend(loc='lower right')
plt.grid()
plt.xlabel('bit per pixel [bpp]')
plt.ylabel(metric_name)
plt.title('{} subsampling, using metric {}'.format(sub_sampling, metric_name.upper()))
plt.tight_layout()
fig.savefig('{}_{}_{}.png'.format(sub_sampling, metric_name, ntpath.basename(db_file_name)))
if __name__ == '__main__':
main()