test_algorithms.py 6.56 KB
Newer Older
1 2 3 4
#!/usr/bin/env python
# vim: set fileencoding=utf-8 :
#

Pavel KORSHUNOV's avatar
Pavel KORSHUNOV committed
5 6
from __future__ import print_function

7
import numpy as np
Pavel KORSHUNOV's avatar
Pavel KORSHUNOV committed
8

9 10
import bob.io.image  # for image loading functionality
import bob.bio.video
Pavel KORSHUNOV's avatar
Pavel KORSHUNOV committed
11 12
import bob.pad.base

13
from bob.pad.base.algorithm import SVM
14
from bob.pad.base.algorithm import OneClassGMM
15
from bob.pad.base.algorithm import MLP
16 17 18

import random

19 20 21
from bob.pad.base.utils import (
    convert_array_to_list_of_frame_cont,
    convert_list_of_frame_cont_to_array,
22
    convert_frame_cont_to_array
23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
)

from bob.pad.base.database import PadFile
from bob.pad.base.algorithm import Predictions
from bob.pad.base import padfile_to_label


def test_prediction():
    alg = Predictions()
    sample = [0, 1]
    assert alg.score(sample)[0] == sample[1]


def test_padfile_to_label():
    f = PadFile(client_id='', path='', attack_type=None, file_id=1)
    assert padfile_to_label(f) is True, padfile_to_label(f)
    f = PadFile(client_id='', path='', attack_type='print', file_id=1)
    assert padfile_to_label(f) is False, padfile_to_label(f)
41

42 43 44

def test_video_svm_pad_algorithm():
    """
45
    Test the SVM PAD algorithm.
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
    """

    random.seed(7)

    N = 20000
    mu = 1
    sigma = 1
    real_array = np.transpose(
        np.vstack([[random.gauss(mu, sigma) for _ in range(N)],
                   [random.gauss(mu, sigma) for _ in range(N)]]))

    mu = 5
    sigma = 1
    attack_array = np.transpose(
        np.vstack([[random.gauss(mu, sigma) for _ in range(N)],
                   [random.gauss(mu, sigma) for _ in range(N)]]))

    real = convert_array_to_list_of_frame_cont(real_array)
    attack = convert_array_to_list_of_frame_cont(attack_array)

    training_features = [real, attack]

    MACHINE_TYPE = 'C_SVC'
    KERNEL_TYPE = 'RBF'
    N_SAMPLES = 1000
    TRAINER_GRID_SEARCH_PARAMS = {'cost': [1], 'gamma': [0.5, 1]}
    MEAN_STD_NORM_FLAG = True  # enable mean-std normalization
    FRAME_LEVEL_SCORES_FLAG = True  # one score per frame(!) in this case

    algorithm = SVM(
        machine_type=MACHINE_TYPE,
        kernel_type=KERNEL_TYPE,
        n_samples=N_SAMPLES,
        trainer_grid_search_params=TRAINER_GRID_SEARCH_PARAMS,
        mean_std_norm_flag=MEAN_STD_NORM_FLAG,
        frame_level_scores_flag=FRAME_LEVEL_SCORES_FLAG)

    machine = algorithm.train_svm(
        training_features=training_features,
        n_samples=algorithm.n_samples,
        machine_type=algorithm.machine_type,
        kernel_type=algorithm.kernel_type,
        trainer_grid_search_params=algorithm.trainer_grid_search_params,
        mean_std_norm_flag=algorithm.mean_std_norm_flag,
        projector_file="",
        save_debug_data_flag=False)

    assert machine.n_support_vectors == [148, 150]
    assert machine.gamma == 0.5

    real_sample = convert_frame_cont_to_array(real[0])

    prob = machine.predict_class_and_probabilities(real_sample)[1]

    assert prob[0, 0] > prob[0, 1]

    precision = algorithm.comp_prediction_precision(machine, real_array,
                                                    attack_array)

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
    assert precision > 0.99


def test_video_gmm_pad_algorithm():
    """
    Test the OneClassGMM PAD algorithm.
    """

    random.seed(7)

    N = 1000
    mu = 1
    sigma = 1
    real_array = np.transpose(
        np.vstack([[random.gauss(mu, sigma) for _ in range(N)],
                   [random.gauss(mu, sigma) for _ in range(N)]]))

    mu = 5
    sigma = 1
    attack_array = np.transpose(
        np.vstack([[random.gauss(mu, sigma) for _ in range(N)],
                   [random.gauss(mu, sigma) for _ in range(N)]]))

    real = convert_array_to_list_of_frame_cont(real_array)

    N_COMPONENTS = 1
    RANDOM_STATE = 3
    FRAME_LEVEL_SCORES_FLAG = True

    algorithm = OneClassGMM(
        n_components=N_COMPONENTS,
        random_state=RANDOM_STATE,
        frame_level_scores_flag=FRAME_LEVEL_SCORES_FLAG)

    # training_features[0] - training features for the REAL class.
    real_array_converted = convert_list_of_frame_cont_to_array(real)  # output is array

    assert (real_array == real_array_converted).all()

    # Train the OneClassGMM machine and get normalizers:
    machine, features_mean, features_std = algorithm.train_gmm(
        real=real_array_converted,
        n_components=algorithm.n_components,
        random_state=algorithm.random_state)

    algorithm.machine = machine

    algorithm.features_mean = features_mean

    algorithm.features_std = features_std

    scores_real = algorithm.project(real_array_converted)

    scores_attack = algorithm.project(attack_array)

    assert (np.min(scores_real) + 7.9423798970985917) < 0.000001
    assert (np.max(scores_real) + 1.8380480068281055) < 0.000001
    assert (np.min(scores_attack) + 38.831260843070098) < 0.000001
    assert (np.max(scores_attack) + 5.3633030621521272) < 0.000001
164

165

166
def test_convert_list_of_frame_cont_to_array():
167

168 169 170 171 172 173
  N = 1000
  mu = 1
  sigma = 1
  real_array = np.transpose(np.vstack([[random.gauss(mu, sigma) for _ in range(N)], [random.gauss(mu, sigma) for _ in range(N)]]))

  features_array = convert_list_of_frame_cont_to_array(real_array)
174
  assert isinstance(features_array[0], np.ndarray)
175 176
  features_fm = convert_array_to_list_of_frame_cont(real_array)
  assert isinstance(features_fm[0], bob.bio.video.FrameContainer)
177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209


def test_MLP():
    """
    Test the MLP PAD algorithm.
    """

    random.seed(7)

    N = 20000
    mu = 1
    sigma = 1
    real_array = np.transpose(
        np.vstack([[random.gauss(mu, sigma) for _ in range(N)],
                   [random.gauss(mu, sigma) for _ in range(N)]]))

    mu = 5
    sigma = 1
    attack_array = np.transpose(
        np.vstack([[random.gauss(mu, sigma) for _ in range(N)],
                   [random.gauss(mu, sigma) for _ in range(N)]]))


    training_features = [real_array, attack_array]

    mlp = MLP(max_iter=100)
    mlp.train_projector(training_features, '/tmp/mlp.hdf5')

    real_sample = real_array[0]
    prob = mlp.project(real_sample)
    assert prob[0] > prob[1]


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
def test_LDA():
    """
    Test the LDA PAD algorithm.
    """

    random.seed(7)

    N = 20000
    mu = 1
    sigma = 1
    real_array = np.transpose(
        np.vstack([[random.gauss(mu, sigma) for _ in range(N)],
                   [random.gauss(mu, sigma) for _ in range(N)]]))

    mu = 5
    sigma = 1
    attack_array = np.transpose(
        np.vstack([[random.gauss(mu, sigma) for _ in range(N)],
                   [random.gauss(mu, sigma) for _ in range(N)]]))

    training_features = [real_array, attack_array]

    lda = PadLDA()
    lda.train_projector(training_features, '/tmp/lda.hdf5')

    real_sample = real_array[0]
    prob = lda.project(real_sample)
    assert prob[0] > prob[1]