-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathtest.py
More file actions
187 lines (153 loc) · 7.48 KB
/
test.py
File metadata and controls
187 lines (153 loc) · 7.48 KB
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
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import joblib
from scipy.optimize import linear_sum_assignment
from sklearn.cluster import KMeans
from train import *
from utilities import get_dataset, block_warnings, to_float32
def bipartite_match(preds, labels, n_classes=None, presence=None):
"""Does maximum biprartite matching between `pred` and `gt`."""
if n_classes is not None:
n_gt_labels, n_pred_labels = n_classes, n_classes
else:
n_gt_labels = np.unique(labels).shape[0]
n_pred_labels = np.unique(preds).shape[0]
cost_matrix = np.zeros([n_gt_labels, n_pred_labels], dtype=np.int32)
for label in range(n_gt_labels):
label_idx = (labels == label)
for new_label in range(n_pred_labels):
errors = np.equal(preds[label_idx], new_label).astype(np.float32)
if presence is not None:
errors *= presence[label_idx]
num_errors = errors.sum()
cost_matrix[label, new_label] = -num_errors
labels_idx, preds_idx = linear_sum_assignment(cost_matrix)
preds_2_labels = [0 for _ in range(n_classes)]
for i in range(n_classes):
preds_2_labels[preds_idx[i]] = labels_idx[i]
return preds_2_labels
if __name__ == '__main__':
block_warnings()
config = config_fashion_mnist
batch_size = 100
max_train_steps = 300
learning_rate = 3e-5
snapshot = './checkpoints/{}/model.ckpt'.format(config['dataset'])
model = SCAE(
input_size=[batch_size, config['canvas_size'], config['canvas_size'], config['n_channels']],
num_classes=config['num_classes'],
n_part_caps=config['n_part_caps'],
n_obj_caps=config['n_obj_caps'],
n_channels=config['n_channels'],
colorize_templates=config['colorize_templates'],
use_alpha_channel=config['use_alpha_channel'],
prior_within_example_sparsity_weight=config['prior_within_example_sparsity_weight'],
prior_between_example_sparsity_weight=config['prior_between_example_sparsity_weight'],
posterior_within_example_sparsity_weight=config['posterior_within_example_sparsity_weight'],
posterior_between_example_sparsity_weight=config['posterior_between_example_sparsity_weight'],
template_size=config['template_size'],
template_nonlin=config['template_nonlin'],
color_nonlin=config['color_nonlin'],
part_encoder_noise_scale=0.,
obj_decoder_noise_type=None,
obj_decoder_noise_scale=0.,
set_transformer_n_layers=config['set_transformer_n_layers'],
set_transformer_n_heads=config['set_transformer_n_heads'],
set_transformer_n_dims=config['set_transformer_n_dims'],
set_transformer_n_output_dims=config['set_transformer_n_output_dims'],
part_cnn_strides=config['part_cnn_strides'],
prep=config['prep'],
is_training=False,
scope='SCAE',
snapshot=snapshot
)
if config['dataset'] == 'gtsrb':
trainset = get_gtsrb('train', shape=[config['canvas_size'], config['canvas_size']], file_path='./datasets',
save_only=False, gtsrb_raw_file_path=GTSRB_DATASET_PATH, gtsrb_classes=config['classes'])
testset = get_gtsrb('test', shape=[config['canvas_size'], config['canvas_size']], file_path='./datasets',
save_only=False, gtsrb_raw_file_path=GTSRB_DATASET_PATH, gtsrb_classes=config['classes'])
else:
trainset = get_dataset(config['dataset'], 'train', shape=[config['canvas_size'], config['canvas_size']],
file_path='./datasets', save_only=False)
testset = get_dataset(config['dataset'], 'test', shape=[config['canvas_size'], config['canvas_size']],
file_path='./datasets', save_only=False)
path = snapshot[:snapshot.rindex('/')]
if not os.path.exists(path):
raise FileExistsError('Cannot access checkpoint files')
len_trainset = len(trainset['image'])
len_testset = len(testset['image'])
train_batches = np.int(np.ceil(np.float(len_trainset) / np.float(batch_size)))
test_batches = np.int(np.ceil(np.float(len_testset) / np.float(batch_size)))
# Supervised Classification
test_acc_prior = 0.
test_acc_posterior = 0.
prior_pres_list = []
posterior_pres_list = []
for i_batch in trange(train_batches, desc='Testing trainset'):
i_end = min((i_batch + 1) * batch_size, len_trainset)
i_start = min(i_batch * batch_size, i_end - batch_size)
images = to_float32(trainset['image'][i_start:i_end])
labels = trainset['label'][i_start:i_end]
test_pred_prior, test_pred_posterior, prior_pres, posterior_pres = model.sess.run(
[model.res.prior_cls_pred,
model.res.posterior_cls_pred,
model.res.caps_presence_prob,
model.res.posterior_mixing_probs],
feed_dict={model.batch_input: images})
n_samples = i_end - (i_batch * batch_size)
test_acc_prior += (test_pred_prior == labels)[:n_samples].sum()
test_acc_posterior += (test_pred_posterior == labels)[:n_samples].sum()
prior_pres_list.append(prior_pres[:n_samples])
posterior_pres_list.append(posterior_pres[:n_samples])
for i_batch in trange(test_batches, desc='Testing testset'):
i_end = min((i_batch + 1) * batch_size, len_testset)
i_start = min(i_batch * batch_size, i_end - batch_size)
images = to_float32(testset['image'][i_start:i_end])
labels = testset['label'][i_start:i_end]
test_pred_prior, test_pred_posterior, prior_pres, posterior_pres = model.sess.run(
[model.res.prior_cls_pred,
model.res.posterior_cls_pred,
model.res.caps_presence_prob,
model.res.posterior_mixing_probs],
feed_dict={model.batch_input: images})
n_samples = i_end - (i_batch * batch_size)
test_acc_prior += (test_pred_prior == labels)[:n_samples].sum()
test_acc_posterior += (test_pred_posterior == labels)[:n_samples].sum()
prior_pres_list.append(prior_pres[:n_samples])
posterior_pres_list.append(posterior_pres[:n_samples])
print('Supervised acc: prior={:.6f}, posterior={:.6f}'
.format(test_acc_prior / (len_trainset + len_testset), test_acc_posterior / (len_trainset + len_testset)))
# Unsupervised Classification
prior_pres_list = np.concatenate(prior_pres_list)
posterior_pres_list = np.concatenate(posterior_pres_list).sum(1)
kmeans_prior = KMeans(
n_clusters=config['num_classes'],
precompute_distances=True,
n_jobs=-1,
max_iter=1000,
).fit(prior_pres_list)
kmeans_posterior = KMeans(
n_clusters=config['num_classes'],
precompute_distances=True,
n_jobs=-1,
max_iter=1000,
).fit(posterior_pres_list)
kmeans_pred_list_prior = kmeans_prior.predict(prior_pres_list)
kmeans_pred_list_posterior = kmeans_posterior.predict(posterior_pres_list)
ground_truth_list = np.concatenate([trainset['label'], testset['label']])
p2l_prior = bipartite_match(kmeans_pred_list_prior[:len_trainset], trainset['label'], config['num_classes'])
p2l_posterior = bipartite_match(kmeans_pred_list_posterior[:len_trainset], trainset['label'], config['num_classes'])
for i in range(len(kmeans_pred_list_prior)):
# kmeans_label to gt_label
kmeans_pred_list_prior[i] = p2l_prior[kmeans_pred_list_prior[i]]
for i in range(len(kmeans_pred_list_posterior)):
# kmeans_label to gt_label
kmeans_pred_list_posterior[i] = p2l_posterior[kmeans_pred_list_posterior[i]]
print('Unsupervised acc: prior={:.6f}, posterior={:.6f}'
.format((kmeans_pred_list_prior == ground_truth_list).sum() / (len_trainset + len_testset),
(kmeans_pred_list_posterior == ground_truth_list).sum() / (len_trainset + len_testset)))
joblib.dump(kmeans_prior, '{}/kmeans_prior.m'.format(path))
np.savez_compressed('{}/kmeans_labels_prior.npz'.format(path), preds_2_labels=p2l_prior)
joblib.dump(kmeans_posterior, '{}/kmeans_posterior.m'.format(path))
np.savez_compressed('{}/kmeans_labels_posterior.npz'.format(path), preds_2_labels=p2l_posterior)