-
Notifications
You must be signed in to change notification settings - Fork 2
/
reproduce.py
58 lines (45 loc) · 3.06 KB
/
reproduce.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
import datetime
import logging
from clize import run
import steps
from helper.utils import generate_run_folder
def reproduce(*, force_fresh_data=True, debug=False, parallel=False, profiling=False, name=None, sampling_rate=80, resample_sr=80):
logging_level = logging.DEBUG if debug else logging.INFO
scheduler = 'processes' if parallel else 'sync'
logging.basicConfig(
level=logging_level, format='[%(levelname)s] %(asctime)-15s - %(message)s', datefmt='%Y-%m-%d %H:%M:%S')
logging.info('Downloading data...')
data_folderpath = steps.download_data(force=force_fresh_data)
logging.info(
'Data is downloaded and unzipped to {}'.format(data_folderpath))
output_folder = generate_run_folder(
data_folderpath, debug=debug, name=name)
logging.info('Preparing class set...')
classset_path = steps.prepare_class_set(data_folderpath, output_folder=output_folder,
debug=debug, scheduler=scheduler, profiling=profiling, force=force_fresh_data)
logging.info('Class set is generated to {}'.format(classset_path))
logging.info('Preparing feature set...')
feature_set_path = steps.prepare_feature_set(data_folderpath, output_folder=output_folder, debug=debug, scheduler=scheduler,
profiling=profiling, force=force_fresh_data, sampling_rate=sampling_rate, resample_sr=resample_sr)
logging.info('Feature set is generated to {}'.format(feature_set_path))
logging.info('Preparing validation sets...')
dataset_path = steps.prepare_validation_set(data_folderpath, output_folder=output_folder, debug=debug,
scheduler=scheduler, profiling=profiling, force=force_fresh_data, include_nonwear=False)
logging.info('Validation sets are saved to {}'.format(dataset_path))
logging.info('Running validation experiments...')
prediction_path = steps.run_validation_experiments(data_folderpath, output_folder=output_folder, debug=debug,
scheduler=scheduler, profiling=profiling, force=force_fresh_data, include_nonwear=False, model_type='svm')
logging.info('Prediction results are saved to {}'.format(prediction_path))
logging.info('Computing metrics...')
metric_path, cm_path = steps.compute_metrics(
data_folderpath, output_folder=output_folder,
debug=debug, scheduler=scheduler, profiling=profiling, force=force_fresh_data)
logging.info('Metrics results are saved to {}'.format(metric_path))
logging.info('Confusion matrices are saved to {}'.format(cm_path))
logging.info('Generating publication figures and tables...')
figure_path = steps.get_figures(
data_folderpath, output_folder=output_folder, debug=debug, force=force_fresh_data)
logging.info('Figures and tables are saved to {}'.format(figure_path))
if __name__ == "__main__":
run(reproduce)
# reproduce(force_fresh_data=False, debug=True, parallel=True, profiling=False, run_ts='2019-07-15-16-42-19', sampling_rate=80, resample_sr=80)