Newer
Older
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
# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
#
# SPDX-License-Identifier: GPL-3.0-or-later
"""Tests for our CLI applications."""
import contextlib
import os
import re
import tempfile
import tomli_w
from click.testing import CliRunner
@contextlib.contextmanager
def rc_context(**new_config):
with tempfile.TemporaryDirectory() as tmpdir:
config_filename = "ptbench.toml"
with open(os.path.join(tmpdir, config_filename), "wb") as f:
tomli_w.dump(new_config, f)
f.flush()
old_config_home = os.environ.get("XDG_CONFIG_HOME")
os.environ["XDG_CONFIG_HOME"] = tmpdir
yield
if old_config_home is None:
del os.environ["XDG_CONFIG_HOME"]
else:
os.environ["XDG_CONFIG_HOME"] = old_config_home
@contextlib.contextmanager
def stdout_logging():
# copy logging messages to std out
import io
import logging
buf = io.StringIO()
ch = logging.StreamHandler(buf)
ch.setFormatter(logging.Formatter("%(message)s"))
ch.setLevel(logging.INFO)
logger = logging.getLogger("ptbench")
logger.addHandler(ch)
yield buf
logger.removeHandler(ch)
def _assert_exit_0(result):
assert (
result.exit_code == 0
), f"Exit code {result.exit_code} != 0 -- Output:\n{result.output}"
def _check_help(entry_point):
runner = CliRunner()
result = runner.invoke(entry_point, ["--help"])
_assert_exit_0(result)
assert result.output.startswith("Usage:")
def test_config_help():
from ptbench.scripts.config import config
_check_help(config)
def test_config_list_help():
from ptbench.scripts.config import list
_check_help(list)
def test_config_list():
from ptbench.scripts.config import list
runner = CliRunner()
result = runner.invoke(list)
_assert_exit_0(result)
assert "module: ptbench.configs.datasets" in result.output
assert "module: ptbench.configs.models" in result.output
def test_config_list_v():
from ptbench.scripts.config import list
result = CliRunner().invoke(list, ["--verbose"])
_assert_exit_0(result)
assert "module: ptbench.configs.datasets" in result.output
assert "module: ptbench.configs.models" in result.output
def test_config_describe_help():
from ptbench.scripts.config import describe
_check_help(describe)
def test_config_describe_montgomery():
from ptbench.scripts.config import describe
runner = CliRunner()
result = runner.invoke(describe, ["montgomery"])
_assert_exit_0(result)
assert "Montgomery dataset for TB detection" in result.output
def test_dataset_help():
from ptbench.scripts.dataset import dataset
_check_help(dataset)
def test_dataset_list_help():
from ptbench.scripts.dataset import list
_check_help(list)
def test_dataset_list():
from ptbench.scripts.dataset import list
runner = CliRunner()
result = runner.invoke(list)
_assert_exit_0(result)
assert result.output.startswith("Supported datasets:")
def test_dataset_check_help():
from ptbench.scripts.dataset import check
_check_help(check)
def test_dataset_check():
from ptbench.scripts.dataset import check
runner = CliRunner()
result = runner.invoke(check, ["--verbose", "--limit=2"])
_assert_exit_0(result)
def test_main_help():
from ptbench.scripts.cli import cli
_check_help(cli)
def test_train_help():
from ptbench.scripts.train import train
_check_help(train)
def _str_counter(substr, s):
return sum(1 for _ in re.finditer(substr, s, re.MULTILINE))
def test_predict_help():
from ptbench.scripts.predict import predict
_check_help(predict)
def test_predtojson_help():
from ptbench.scripts.predtojson import predtojson
_check_help(predtojson)
def test_aggregpred_help():
from ptbench.scripts.aggregpred import aggregpred
_check_help(aggregpred)
def test_evaluate_help():
from ptbench.scripts.evaluate import evaluate
_check_help(evaluate)
def test_compare_help():
from ptbench.scripts.compare import compare
_check_help(compare)
def test_train_pasa_montgomery(temporary_basedir, montgomery_datadir):
# Temporarily modifies Montgomery datadir if need be
new_value = {"datadir.montgomery": str(montgomery_datadir)}
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
with rc_context(**new_value):
from ptbench.scripts.train import train
runner = CliRunner()
with stdout_logging() as buf:
output_folder = str(temporary_basedir / "results")
result = runner.invoke(
train,
[
"pasa",
"montgomery",
"-vv",
"--epochs=1",
"--batch-size=1",
"--normalization=current",
f"--output-folder={output_folder}",
],
)
_assert_exit_0(result)
assert os.path.exists(
os.path.join(output_folder, "model_final_epoch.pth")
)
assert os.path.exists(
os.path.join(output_folder, "model_lowest_valid_loss.pth")
)
assert os.path.exists(
os.path.join(output_folder, "last_checkpoint")
)
assert os.path.exists(os.path.join(output_folder, "constants.csv"))
assert os.path.exists(os.path.join(output_folder, "trainlog.csv"))
assert os.path.exists(
os.path.join(output_folder, "model_summary.txt")
)
keywords = {
r"^Found \(dedicated\) '__train__' set for training$": 1,
r"^Found \(dedicated\) '__valid__' set for validation$": 1,
r"^Continuing from epoch 0$": 1,
r"^Saving model summary at.*$": 1,
r"^Model has.*$": 1,
r"^Saving checkpoint": 2,
r"^Total training time:": 1,
r"^Z-normalization with mean": 1,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_predict_pasa_montgomery(
temporary_basedir, montgomery_datadir, datadir
):
# Temporarily modifies Montgomery datadir if need be
new_value = {"datadir.montgomery": str(montgomery_datadir)}
with rc_context(**new_value):
from ptbench.scripts.predict import predict
runner = CliRunner()
with stdout_logging() as buf:
output_folder = str(temporary_basedir / "predictions")
result = runner.invoke(
predict,
[
"pasa",
"montgomery",
"-vv",
"--batch-size=1",
"--relevance-analysis",
f"--weight={str(datadir / 'lfs' / 'models' / 'pasa.pth')}",
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
f"--output-folder={output_folder}",
],
)
_assert_exit_0(result)
# check predictions are there
predictions_file1 = os.path.join(
output_folder, "train/predictions.csv"
)
predictions_file2 = os.path.join(
output_folder, "validation/predictions.csv"
)
predictions_file3 = os.path.join(
output_folder, "test/predictions.csv"
)
assert os.path.exists(predictions_file1)
assert os.path.exists(predictions_file2)
assert os.path.exists(predictions_file3)
keywords = {
r"^Loading checkpoint from.*$": 1,
r"^Total time:.*$": 3,
r"^Relevance analysis.*$": 3,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_predtojson(datadir, temporary_basedir, montgomery_datadir):
# Temporarily modify Montgomery datadir if need be
new_value = {"datadir.montgomery": str(montgomery_datadir)}
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
with rc_context(**new_value):
from ptbench.scripts.predtojson import predtojson
runner = CliRunner()
with stdout_logging() as buf:
predictions = str(datadir / "test_predictions.csv")
output_folder = str(temporary_basedir / "pred_to_json")
result = runner.invoke(
predtojson,
[
"-vv",
"train",
f"{predictions}",
"test",
f"{predictions}",
f"--output-folder={output_folder}",
],
)
_assert_exit_0(result)
# check json file is there
assert os.path.exists(os.path.join(output_folder, "dataset.json"))
keywords = {
f"Output folder: {output_folder}": 1,
r"Saving JSON file...": 1,
r"^Loading predictions from.*$": 2,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_evaluate_pasa_montgomery(temporary_basedir, montgomery_datadir):
# Temporarily modify Montgomery datadir if need be
new_value = {"datadir.montgomery": str(montgomery_datadir)}
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
with rc_context(**new_value):
from ptbench.scripts.evaluate import evaluate
runner = CliRunner()
with stdout_logging() as buf:
prediction_folder = str(temporary_basedir / "predictions")
output_folder = str(temporary_basedir / "evaluations")
result = runner.invoke(
evaluate,
[
"-vv",
"montgomery",
f"--predictions-folder={prediction_folder}",
f"--output-folder={output_folder}",
"--threshold=train",
"--steps=2000",
],
)
_assert_exit_0(result)
# check evaluations are there
assert os.path.exists(os.path.join(output_folder, "test.csv"))
assert os.path.exists(os.path.join(output_folder, "train.csv"))
assert os.path.exists(
os.path.join(output_folder, "test_score_table.pdf")
)
assert os.path.exists(
os.path.join(output_folder, "train_score_table.pdf")
)
keywords = {
r"^Skipping dataset '__train__'": 1,
r"^Evaluating threshold on.*$": 1,
r"^Maximum F1-score of.*$": 4,
r"^Set --f1_threshold=.*$": 1,
r"^Set --eer_threshold=.*$": 1,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_compare_pasa_montgomery(temporary_basedir, montgomery_datadir):
# Temporarily modify Montgomery datadir if need be
new_value = {"datadir.montgomery": str(montgomery_datadir)}
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
with rc_context(**new_value):
from ptbench.scripts.compare import compare
runner = CliRunner()
with stdout_logging() as buf:
predictions_folder = str(temporary_basedir / "predictions")
output_folder = str(temporary_basedir / "comparisons")
result = runner.invoke(
compare,
[
"-vv",
"train",
f"{predictions_folder}/train/predictions.csv",
"test",
f"{predictions_folder}/test/predictions.csv",
f"--output-figure={output_folder}/compare.pdf",
f"--output-table={output_folder}/table.txt",
"--threshold=0.5",
],
)
_assert_exit_0(result)
# check comparisons are there
assert os.path.exists(os.path.join(output_folder, "compare.pdf"))
assert os.path.exists(os.path.join(output_folder, "table.txt"))
keywords = {
r"^Dataset '\*': threshold =.*$": 1,
r"^Loading predictions from.*$": 2,
r"^Tabulating performance summary...": 1,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_train_signstotb_montgomery_rs(temporary_basedir, datadir):
from ptbench.scripts.train import train
runner = CliRunner()
with stdout_logging() as buf:
output_folder = str(temporary_basedir / "results")
result = runner.invoke(
train,
[
"signs_to_tb",
"montgomery_rs",
"-vv",
"--epochs=1",
"--batch-size=1",
f"--weight={str(datadir / 'lfs' / 'models' / 'signstotb.pth')}",
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
f"--output-folder={output_folder}",
],
)
_assert_exit_0(result)
assert os.path.exists(
os.path.join(output_folder, "model_final_epoch.pth")
)
assert os.path.exists(
os.path.join(output_folder, "model_lowest_valid_loss.pth")
)
assert os.path.exists(os.path.join(output_folder, "last_checkpoint"))
assert os.path.exists(os.path.join(output_folder, "constants.csv"))
assert os.path.exists(os.path.join(output_folder, "trainlog.csv"))
assert os.path.exists(os.path.join(output_folder, "model_summary.txt"))
keywords = {
r"^Found \(dedicated\) '__train__' set for training$": 1,
r"^Found \(dedicated\) '__valid__' set for validation$": 1,
r"^Continuing from epoch 0$": 1,
r"^Saving model summary at.*$": 1,
r"^Model has.*$": 1,
r"^Saving checkpoint": 2,
r"^Total training time:": 1,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_predict_signstotb_montgomery_rs(temporary_basedir, datadir):
from ptbench.scripts.predict import predict
runner = CliRunner()
with stdout_logging() as buf:
output_folder = str(temporary_basedir / "predictions")
result = runner.invoke(
predict,
[
"signs_to_tb",
"montgomery_rs",
"-vv",
"--batch-size=1",
"--relevance-analysis",
f"--weight={str(datadir / 'lfs' / 'models' / 'signstotb.pth')}",
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
f"--output-folder={output_folder}",
],
)
_assert_exit_0(result)
# check predictions are there
predictions_file = os.path.join(output_folder, "train/predictions.csv")
RA1 = os.path.join(output_folder, "train_RA.pdf")
RA2 = os.path.join(output_folder, "validation_RA.pdf")
RA3 = os.path.join(output_folder, "test_RA.pdf")
assert os.path.exists(predictions_file)
assert os.path.exists(RA1)
assert os.path.exists(RA2)
assert os.path.exists(RA3)
keywords = {
r"^Loading checkpoint from.*$": 1,
r"^Total time:.*$": 3 * 15,
r"^Starting relevance analysis for subset.*$": 3,
r"^Creating and saving plot at.*$": 3,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_train_logreg_montgomery_rs(temporary_basedir, datadir):
from ptbench.scripts.train import train
runner = CliRunner()
with stdout_logging() as buf:
output_folder = str(temporary_basedir / "results")
result = runner.invoke(
train,
[
"logistic_regression",
"montgomery_rs",
"-vv",
"--epochs=1",
"--batch-size=1",
f"--weight={str(datadir / 'lfs' / 'models' / 'logreg.pth')}",
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
f"--output-folder={output_folder}",
],
)
_assert_exit_0(result)
assert os.path.exists(
os.path.join(output_folder, "model_final_epoch.pth")
)
assert os.path.exists(
os.path.join(output_folder, "model_lowest_valid_loss.pth")
)
assert os.path.exists(os.path.join(output_folder, "last_checkpoint"))
assert os.path.exists(os.path.join(output_folder, "constants.csv"))
assert os.path.exists(os.path.join(output_folder, "trainlog.csv"))
assert os.path.exists(os.path.join(output_folder, "model_summary.txt"))
keywords = {
r"^Found \(dedicated\) '__train__' set for training$": 1,
r"^Found \(dedicated\) '__valid__' set for validation$": 1,
r"^Continuing from epoch 0$": 1,
r"^Saving model summary at.*$": 1,
r"^Model has.*$": 1,
r"^Saving checkpoint": 2,
r"^Total training time:": 1,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_predict_logreg_montgomery_rs(temporary_basedir, datadir):
from ptbench.scripts.predict import predict
runner = CliRunner()
with stdout_logging() as buf:
output_folder = str(temporary_basedir / "predictions")
result = runner.invoke(
predict,
[
"logistic_regression",
"montgomery_rs",
"-vv",
"--batch-size=1",
f"--weight={str(datadir / 'lfs' / 'models' / 'logreg.pth')}",
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
f"--output-folder={output_folder}",
],
)
_assert_exit_0(result)
# check predictions are there
predictions_file = os.path.join(output_folder, "train/predictions.csv")
wfile = os.path.join(output_folder, "LogReg_Weights.pdf")
assert os.path.exists(predictions_file)
assert os.path.exists(wfile)
keywords = {
r"^Loading checkpoint from.*$": 1,
r"^Total time:.*$": 3,
r"^Logistic regression identified: saving model weights.*$": 1,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
def test_aggregpred(temporary_basedir, montgomery_datadir):
# Temporarily modify Montgomery datadir if need be
new_value = {"datadir.montgomery": str(montgomery_datadir)}
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
with rc_context(**new_value):
from ptbench.scripts.aggregpred import aggregpred
runner = CliRunner()
with stdout_logging() as buf:
predictions = str(
temporary_basedir / "predictions" / "train" / "predictions.csv"
)
output_folder = str(temporary_basedir / "aggregpred")
result = runner.invoke(
aggregpred,
[
"-vv",
f"{predictions}",
f"{predictions}",
f"--output-folder={output_folder}",
],
)
_assert_exit_0(result)
# check csv file is there
assert os.path.exists(os.path.join(output_folder, "aggregpred.csv"))
keywords = {
f"Output folder: {output_folder}": 1,
r"Saving aggregated CSV file...": 1,
r"^Loading predictions from.*$": 2,
}
buf.seek(0)
logging_output = buf.read()
for k, v in keywords.items():
assert _str_counter(k, logging_output) == v, (
f"Count for string '{k}' appeared "
f"({_str_counter(k, logging_output)}) "
f"instead of the expected {v}:\nOutput:\n{logging_output}"
)
# Not enough RAM available to do this test
# def test_predict_densenetrs_montgomery(temporary_basedir, montgomery_datadir, datadir):
# # Temporarily modify Montgomery datadir if need be
# new_value = {"datadir.montgomery": str(montgomery_datadir)}
# with rc_context(**new_value):
# from ptbench.scripts.predict import predict
# runner = CliRunner()
# with stdout_logging() as buf:
# output_folder = str(temporary_basedir / "predictions")
# result = runner.invoke(
# predict,
# [
# "densenet_rs",
# "montgomery_f0_rgb",
# "-vv",
# "--batch-size=1",
# f"--weight={str(datadir / 'lfs' / 'models' / 'densenetrs.pth')}",
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
# f"--output-folder={output_folder}",
# "--grad-cams"
# ],
# )
# _assert_exit_0(result)
# # check predictions are there
# predictions_file1 = os.path.join(output_folder, "train/predictions.csv")
# predictions_file2 = os.path.join(output_folder, "validation/predictions.csv")
# predictions_file3 = os.path.join(output_folder, "test/predictions.csv")
# assert os.path.exists(predictions_file1)
# assert os.path.exists(predictions_file2)
# assert os.path.exists(predictions_file3)
# # check some grad cams are there
# cam1 = os.path.join(output_folder, "train/cams/MCUCXR_0002_0_cam.png")
# cam2 = os.path.join(output_folder, "train/cams/MCUCXR_0126_1_cam.png")
# cam3 = os.path.join(output_folder, "train/cams/MCUCXR_0275_1_cam.png")
# cam4 = os.path.join(output_folder, "validation/cams/MCUCXR_0399_1_cam.png")
# cam5 = os.path.join(output_folder, "validation/cams/MCUCXR_0113_1_cam.png")
# cam6 = os.path.join(output_folder, "validation/cams/MCUCXR_0013_0_cam.png")
# cam7 = os.path.join(output_folder, "test/cams/MCUCXR_0027_0_cam.png")
# cam8 = os.path.join(output_folder, "test/cams/MCUCXR_0094_0_cam.png")
# cam9 = os.path.join(output_folder, "test/cams/MCUCXR_0375_1_cam.png")
# assert os.path.exists(cam1)
# assert os.path.exists(cam2)
# assert os.path.exists(cam3)
# assert os.path.exists(cam4)
# assert os.path.exists(cam5)
# assert os.path.exists(cam6)
# assert os.path.exists(cam7)
# assert os.path.exists(cam8)
# assert os.path.exists(cam9)
# keywords = {
# r"^Loading checkpoint from.*$": 1,
# r"^Total time:.*$": 3,
# r"^Grad cams folder:.*$": 3,
# }
# buf.seek(0)
# logging_output = buf.read()
# for k, v in keywords.items():
# assert _str_counter(k, logging_output) == v, (
# f"Count for string '{k}' appeared "
# f"({_str_counter(k, logging_output)}) "
# f"instead of the expected {v}:\nOutput:\n{logging_output}"
# )