-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathexperiment_sanity_runner.py
160 lines (139 loc) · 9.3 KB
/
experiment_sanity_runner.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
"""
Runs an active learning simulation for experiment 2.
Experiment Sanity Checker: randomly scramble task labels and check that no label leakage occurs; the selected compounds should not change.
Usage:
python experiment_sanity_runner.py \
--pipeline_params_json_file=../param_configs/exp_sanity_pipeline_config.json \
--hyperparams_json_file=../param_configs/first_pass_hyperparams/top/batch_size_1536/ClusterBasedWCSelector_201.json \
--initial_dataset_file=../datasets/aid624173_cv_96_scrambled/unlabeled_1179.csv
--iter_max=1 \
--batch_size_index=0 \
--no-precompute_dissimilarity_matrix
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import argparse
import json
import pathlib
import numpy as np
import pandas as pd
import csv
import time
import os
import shutil
from active_learning_dd.active_learning_dd import get_next_batch
from active_learning_dd.database_loaders.prepare_loader import prepare_loader
from active_learning_dd.utils.data_utils import get_duplicate_smiles_in1d
from active_learning_dd.utils.generate_dissimilarity_matrix import compute_dissimilarity_matrix
from simulation_utils import *
if __name__ == '__main__':
# read args
parser = argparse.ArgumentParser()
parser.add_argument('--pipeline_params_json_file', action="store", dest="pipeline_params_json_file", required=True)
parser.add_argument('--hyperparams_json_file', action="store", dest="hyperparams_json_file", required=True)
parser.add_argument('--initial_dataset_file', action="store", dest="initial_dataset_file", required=True)
parser.add_argument('--iter_max', type=int, default=10, action="store", dest="iter_max", required=False)
parser.add_argument('--batch_size_index', type=int, default=0, action="store", dest="batch_size_index", required=False)
parser.add_argument('--precompute_dissimilarity_matrix', dest='precompute_dissimilarity_matrix', action='store_true')
parser.add_argument('--no-precompute_dissimilarity_matrix', dest='precompute_dissimilarity_matrix', action='store_false')
given_args = parser.parse_args()
pipeline_params_json_file = given_args.pipeline_params_json_file
hyperparams_json_file = given_args.hyperparams_json_file
initial_dataset_file = given_args.initial_dataset_file
iter_max = given_args.iter_max
batch_size_index = given_args.batch_size_index
precompute_dissimilarity_matrix = given_args.precompute_dissimilarity_matrix
start_iter = 0
# load param json configs
with open(pipeline_params_json_file) as f:
pipeline_config = json.load(f)
with open(hyperparams_json_file) as f:
next_batch_selector_params = json.load(f)
# setup initial training plate
[os.remove(f) for f in glob.glob(pipeline_config['training_data_params']['data_path_format'].format('*'))]
shutil.copyfile(initial_dataset_file, pipeline_config['training_data_params']['data_path_format'].format(0))
print('Set {} as starting initial set.'.format(initial_dataset_file))
assert pd.read_csv(initial_dataset_file).equals(pd.read_csv(pipeline_config['training_data_params']['data_path_format'].format(0)))
rnd_seed = next_batch_selector_params['rnd_seed']
params_set_results_dir = pipeline_config['common']['params_set_results_dir'].format(next_batch_selector_params['hyperparameter_group'],
next_batch_selector_params['hyperparameter_id'],
initial_dataset_file.split('_')[-1].replace('.csv',''))
params_set_config_csv = params_set_results_dir+'/'+pipeline_config['common']['params_set_config_csv']
pathlib.Path(params_set_config_csv).parent.mkdir(parents=True, exist_ok=True)
with open(params_set_config_csv,'w') as f:
csv_w = csv.writer(f)
csv_w.writerow(list(next_batch_selector_params.keys()) + ['rnd_seed'])
csv_w.writerow(list(next_batch_selector_params.values()) + [rnd_seed])
# run this param set for each batch size
batch_size_list = next_batch_selector_params["batch_size"]
batch_size = batch_size_list[batch_size_index]
print('---------------------------------------------------------------')
print('Starting AL pipeline with batch_size: {}'.format(batch_size))
next_batch_selector_params["batch_size"] = batch_size
batch_size_results_dir = params_set_results_dir + pipeline_config['common']['batch_size_results_dir'].format(batch_size)
if not os.path.exists(batch_size_results_dir):
pathlib.Path(batch_size_results_dir+'/'+pipeline_config['common']['params_set_config_csv']).parent.mkdir(parents=True, exist_ok=True)
start_iter = len(glob.glob(batch_size_results_dir + '/training_data/iter_*')) - 1
# modify location of training data to be able to continue jobs
if not os.path.exists(batch_size_results_dir + '/training_data/'):
import shutil
shutil.copytree(pathlib.Path(pipeline_config['training_data_params']['data_path_format']).parent,
batch_size_results_dir + '/training_data')
start_iter = 0
pipeline_config['training_data_params']['data_path_format'] = batch_size_results_dir + '/training_data/iter_{}.csv'
with open(batch_size_results_dir+'/'+pipeline_config['common']['params_set_config_csv'],'w') as f:
csv_w = csv.writer(f)
csv_w.writerow(list(next_batch_selector_params.keys()) + ['rnd_seed'])
csv_w.writerow(list(next_batch_selector_params.values()) + [rnd_seed])
try:
pipeline_config['common']['dissimilarity_memmap_filename']
except:
pipeline_config['common']['dissimilarity_memmap_filename'] = None
if precompute_dissimilarity_matrix:
if pipeline_config['common']['dissimilarity_memmap_filename'] is None:
pipeline_config['common']['dissimilarity_memmap_filename'] = '../datasets/dissimilarity_matrix.dat'
compute_dissimilarity_matrix(csv_file_or_dir=pipeline_config['unlabeled_data_params']['data_path_format'],
output_dir=pipeline_config['common']['dissimilarity_memmap_filename'])
# run iterations for this simulation
for iter_num in range(start_iter, iter_max):
iter_start_time = time.time()
print('---------------------------------------------------------------')
print('Processing iteration number: {}...'.format(iter_num))
#### Run single iteration of active learning pipeline ####
selection_start_time = time.time()
exploitation_df, exploration_df, exploitation_array, exploration_array = get_next_batch(training_loader_params=pipeline_config['training_data_params'],
unlabeled_loader_params=pipeline_config['unlabeled_data_params'],
model_params=pipeline_config['model'],
task_names=pipeline_config['common']['task_names'],
next_batch_selector_params=next_batch_selector_params,
dissimilarity_memmap_filename=pipeline_config['common']['dissimilarity_memmap_filename'])
selection_end_time = time.time()
total_selection_time = selection_end_time - selection_start_time
#### Evaluation ####
# save results
print('Evaluating selected batch...')
eval_start_time = time.time()
evaluate_selected_batch(exploitation_df, exploration_df,
exploitation_array, exploration_array,
batch_size_results_dir,
pipeline_config,
iter_num,
batch_size,
total_selection_time,
add_mean_medians=False)
eval_end_time = time.time()
print('Time it took to evaluate batch {} seconds.'.format(eval_end_time-eval_start_time))
if exploitation_df is not None or exploration_df is not None:
# finally save the exploitation, exploration dataframes to training data directory for next iteration
pd.concat([exploitation_df, exploration_df]).to_csv(pipeline_config['training_data_params']['data_path_format'].format(iter_num+1),
index=False)
iter_end_time = time.time()
print('Finished processing iteration {}. Took {} seconds.'.format(iter_num, iter_end_time-iter_start_time))
# terminate if both exploitation and exploration df are None
if exploitation_df is None and exploration_df is None:
print('Both exploitation and exploration selections are empty. Terminating program.')
break
# summarize the evaluation results into a single csv file
summarize_simulation(batch_size_results_dir,
pipeline_config)