util.py 7.51 KB
Newer Older
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
1
2
3
4
5
6
7
8
9
#!/usr/bin/env python
# vim: set fileencoding=utf-8 :
# @author: Tiago de Freitas Pereira <tiago.pereira@idiap.ch>
# @date: Wed 11 May 2016 09:39:36 CEST 

import numpy
import tensorflow as tf
numpy.random.seed(10)

10

Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
11
12
13
14
15
16
def compute_euclidean_distance(x, y):
    """
    Computes the euclidean distance between two tensorflow variables
    """

    with tf.name_scope('euclidean_distance') as scope:
17
        d = tf.sqrt(tf.reduce_sum(tf.square(tf.subtract(x, y)), 1))
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
18
19
        return d

20
21

def load_mnist(perc_train=0.9):
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
22
23

    import bob.db.mnist
24
    db = bob.db.mnist.Database()
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
25
26
27
28
29
30
    raw_data = db.data()

    # data  = raw_data[0].astype(numpy.float64)
    data = raw_data[0]
    labels = raw_data[1]

31
32
33
34
35
36
37
38
39
    # Shuffling
    total_samples = data.shape[0]
    indexes = numpy.array(range(total_samples))
    numpy.random.shuffle(indexes)

    # Spliting train and validation
    n_train = int(perc_train*indexes.shape[0])
    n_validation = total_samples - n_train

40
    train_data = data[0:n_train, :].astype("float32") * 0.00390625
41
42
    train_labels = labels[0:n_train]

43
    validation_data = data[n_train:n_train+n_validation, :].astype("float32") * 0.00390625
44
45
46
    validation_labels = labels[n_train:n_train+n_validation]

    return train_data, train_labels, validation_data, validation_labels
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
47
48


49
def create_mnist_tfrecord(tfrecords_filename, data, labels, n_samples=6000):
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
50

51
52
    def _bytes_feature(value):
        return tf.train.Feature(bytes_list=tf.train.BytesList(value=[value]))
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
53

54
55
    def _int64_feature(value):
        return tf.train.Feature(int64_list=tf.train.Int64List(value=[value]))
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
56

57
    writer = tf.python_io.TFRecordWriter(tfrecords_filename)
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
58

59
60
61
62
63
64
65
66
67
68
69
    for i in range(n_samples):
        img = data[i]
        img_raw = img.tostring()
        
        feature = {'train/data': _bytes_feature(img_raw),
                   'train/label': _int64_feature(labels[i])
                  }
        
        example = tf.train.Example(features=tf.train.Features(feature=feature))
        writer.write(example.SerializeToString())
    writer.close()
Tiago de Freitas Pereira's avatar
Scratch  
Tiago de Freitas Pereira committed
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


def compute_eer(data_train, labels_train, data_validation, labels_validation, n_classes):
    import bob.measure
    from scipy.spatial.distance import cosine

    # Creating client models
    models = []
    for i in range(n_classes):
        indexes = labels_train == i
        models.append(numpy.mean(data_train[indexes, :], axis=0))

    # Probing
    positive_scores = numpy.zeros(shape=0)
    negative_scores = numpy.zeros(shape=0)

    for i in range(n_classes):
        # Positive scoring
        indexes = labels_validation == i
        positive_data = data_validation[indexes, :]
        p = [cosine(models[i], positive_data[j]) for j in range(positive_data.shape[0])]
        positive_scores = numpy.hstack((positive_scores, p))

        # negative scoring
        indexes = labels_validation != i
        negative_data = data_validation[indexes, :]
        n = [cosine(models[i], negative_data[j]) for j in range(negative_data.shape[0])]
        negative_scores = numpy.hstack((negative_scores, n))

    # Computing performance based on EER
    negative_scores = (-1) * negative_scores
    positive_scores = (-1) * positive_scores

    threshold = bob.measure.eer_threshold(negative_scores, positive_scores)
    far, frr = bob.measure.farfrr(negative_scores, positive_scores, threshold)
    eer = (far + frr) / 2.

    return eer


def compute_accuracy(data_train, labels_train, data_validation, labels_validation, n_classes):
    from scipy.spatial.distance import cosine

    # Creating client models
    models = []
    for i in range(n_classes):
        indexes = labels_train == i
        models.append(numpy.mean(data_train[indexes, :], axis=0))

    # Probing
    tp = 0
    for i in range(data_validation.shape[0]):

        d = data_validation[i,:]
        l = labels_validation[i]

        scores = [cosine(m, d) for m in models]
        predict = numpy.argmax(scores)

        if predict == l:
            tp += 1

    return (float(tp) / data_validation.shape[0]) * 100
Tiago de Freitas Pereira's avatar
Tiago de Freitas Pereira committed
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
    
    
def debug_embbeding(image, architecture, embbeding_dim=2, feature_layer="fc3"):
    """
    """
    import tensorflow as tf
    from bob.learn.tensorflow.utils.session import Session
    
    session = Session.instance(new=False).session    
    inference_graph = architecture.compute_graph(architecture.inference_placeholder, feature_layer=feature_layer, training=False)

    embeddings = numpy.zeros(shape=(image.shape[0], embbeding_dim))
    for i in range(image.shape[0]):
        feed_dict = {architecture.inference_placeholder: image[i:i+1, :,:,:]}
        embedding = session.run([tf.nn.l2_normalize(inference_graph, 1, 1e-10)], feed_dict=feed_dict)[0]
        embedding = numpy.reshape(embedding, numpy.prod(embedding.shape[1:]))
        embeddings[i] = embedding

    return embeddings
152
153
154


def cdist(A):
155
156
157
158
    """
    Compute a pairwise euclidean distance in the same fashion
    as in scipy.spation.distance.cdist
    """
159
    with tf.variable_scope('Pairwisedistance'):
160
161
        #ones_1 = tf.ones(shape=(1, A.shape.as_list()[0]))
        ones_1 = tf.reshape(tf.cast(tf.ones_like(A), tf.float32)[:, 0], [1, -1])
162
163
        p1 = tf.matmul(
            tf.expand_dims(tf.reduce_sum(tf.square(A), 1), 1),
164
            ones_1
165
        )
166
167
168

        #ones_2 = tf.ones(shape=(A.shape.as_list()[0], 1))
        ones_2 = tf.reshape(tf.cast(tf.ones_like(A), tf.float32)[:, 0], [-1, 1])
169
170
        p2 = tf.transpose(tf.matmul(
            tf.reshape(tf.reduce_sum(tf.square(A), 1), shape=[-1, 1]),
171
            ones_2,
172
173
174
175
176
177
            transpose_b=True
        ))

        return tf.sqrt(tf.add(p1, p2) - 2 * tf.matmul(A, A, transpose_b=True))


178
def predict_using_tensors(embedding, labels, num=None):
179
    """
180
181
    Compute the predictions through exhaustive comparisons between
    embeddings using tensors
182
183
184
    """

    # Fitting the main diagonal with infs (removing comparisons with the same sample)
185
    inf = tf.cast(tf.ones_like(labels), tf.float32) * numpy.inf
186
187
188
189

    distances = cdist(embedding)
    distances = tf.matrix_set_diag(distances, inf)
    indexes = tf.argmin(distances, axis=1)
190
191
192
193
194
195
196
197
198
199
200
201
202
    return [labels[i] for i in tf.unstack(indexes, num=num)]


def compute_embedding_accuracy_tensors(embedding, labels, num=None):
    """
    Compute the accuracy through exhaustive comparisons between the embeddings using tensors
    """

    # Fitting the main diagonal with infs (removing comparisons with the same sample)
    predictions = predict_using_tensors(embedding, labels, num=num)
    matching = [tf.equal(p, l) for p, l in zip(tf.unstack(predictions, num=num), tf.unstack(labels, num=num))]

    return tf.reduce_sum(tf.cast(matching, tf.uint8))/len(predictions)
203
204


205
206
207
208
209
210
211
212
213
214
215
def compute_embedding_accuracy(embedding, labels):
    """
    Compute the accuracy through exhaustive comparisons between the embeddings 
    """

    from scipy.spatial.distance import cdist
    
    distances = cdist(embedding, embedding)
    
    n_samples = embedding.shape[0]

216
217
218
    # Fitting the main diagonal with infs (removing comparisons with the same sample)
    numpy.fill_diagonal(distances, numpy.inf)
    
219
    indexes = distances.argmin(axis=1)
220

221
222
    # Computing the argmin excluding comparisons with the same samples
    # Basically, we are excluding the main diagonal
223
224

    #valid_indexes = distances[distances>0].reshape(n_samples, n_samples-1).argmin(axis=1)
225
226

    # Getting the original positions of the indexes in the 1-axis
227
    #corrected_indexes = [ i if i<j else i+1 for i, j in zip(valid_indexes, range(n_samples))]
228

229
    matching = [ labels[i]==labels[j] for i,j in zip(range(n_samples), indexes)]    
230
231
232
    accuracy = sum(matching)/float(n_samples)
    
    return accuracy
Tiago de Freitas Pereira's avatar
Tiago de Freitas Pereira committed
233