def load_db(db_path, list_db, spatial_size, objective_flag):
db_name = list_db[0]
db_home = db_path + db_name + "/"
db_images = db_path + db_name + "/" + db_name + "/"
cross_db_flag = 0
print(db_name)
if db_name == 'CASME2_TIM':
table = loading_casme_table(db_home + 'CASME2_label_Ver_2.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
db_path, list_db)
r = w = spatial_size
subjects = 26
samples = 246
n_exp = 5
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
channel = 1
if os.path.isdir(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'CASME2_Optical':
print("arrived")
table = loading_casme_table(db_home + 'CASME2_label_Ver_2.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
db_path, list_db)
r = w = spatial_size
subjects = 26
samples = 246
n_exp = 5
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
channel = 3
if os.path.isdir(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SMIC_TIM10':
table = loading_smic_table(db_path, db_name)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
r = w = spatial_size
subjects = 16
samples = 164
n_exp = 3
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
channel = 3
if os.path.isdir(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SAMM_Optical':
table, table_objective = loading_samm_table(db_path, db_name,
objective_flag)
# print(table)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
r = w = spatial_size
subjects = 29
samples = 159
n_exp = 8
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
channel = 3
if os.path.isdir(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SAMM_TIM10':
table, table_objective = loading_samm_table(db_path, db_name,
objective_flag)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
samples = 159
n_exp = 8
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
channel = 3
#########################################################
if os.path.isdir(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SAMM_Strain':
table, table_objective = loading_samm_table(db_path, db_name,
objective_flag)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
samples = 159
n_exp = 8
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
channel = 3
#########################################################
if os.path.isdir(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SAMM_CASME_Optical':
# total amount of videos 253
table, table_objective = loading_samm_table(db_path, db_name)
table = table_objective
table2 = loading_casme_objective_table(db_path, db_name)
# merge samm and casme tables
table = np.concatenate((table, table2), axis=1)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
sub_items = np.empty([0])
list_samples = filter_objective_samples(table)
r = w = spatial_size
subjects = 47 # some subjects were removed because of objective classes and ignore samples: 47
n_exp = 5
samples = 253
VidPerSubject, list_samples = get_subfolders_num_crossdb(
db_images, IgnoredSamples_index, sub_items, table, list_samples)
timesteps_TIM = 9
data_dim = r * w
channel = 3
if os.path.isdir(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
cross_db_flag = 1
return r, w, subjects, samples, n_exp, VidPerSubject, timesteps_TIM, data_dim, channel, table, list_samples, db_home, db_images, cross_db_flag
return r, w, subjects, samples, n_exp, VidPerSubject, timesteps_TIM, data_dim, channel, table, listOfIgnoredSamples, db_home, db_images, cross_db_flag
def test_samm(batch_size, spatial_epochs, temporal_epochs, train_id, dB,
spatial_size, flag, tensorboard):
############## Path Preparation ######################
root_db_path = "/media/viprlab/01D31FFEF66D5170/"
workplace = root_db_path + dB + "/"
inputDir = root_db_path + dB + "/" + dB + "/"
######################################################
classes = 5
if dB == 'CASME2_TIM':
table = loading_casme_table(workplace + 'CASME2-ObjectiveClasses.xlsx')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
inputDir)
############## Variables ###################
r = w = spatial_size
subjects = 2
n_exp = 5
# VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
listOfIgnoredSamples = []
VidPerSubject = [2, 1]
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
############################################
os.remove(workplace + "Classification/CASME2_TIM_label.txt")
elif dB == 'CASME2_Optical':
table = loading_casme_table(workplace + 'CASME2-ObjectiveClasses.xlsx')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
inputDir)
############## Variables ###################
r = w = spatial_size
subjects = 26
n_exp = 5
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
pad_sequence = 9
channel = 3
############################################
# os.remove(workplace + "Classification/CASME2_TIM_label.txt")
elif dB == 'SAMM_TIM10':
table, table_objective = loading_samm_table(root_db_path, dB)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
n_exp = 8
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
classes = 8
#########################################################
elif dB == 'SAMM_CASME_Optical':
# total amount of videos 253
table, table_objective = loading_samm_table(root_db_path, dB)
table = table_objective
table2 = loading_casme_objective_table(root_db_path, dB)
# merge samm and casme tables
table = np.concatenate((table, table2), axis=1)
# print(table.shape)
# listOfIgnoredSamples, IgnoredSamples_index, sub_items = ignore_casme_samples(inputDir)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
sub_items = np.empty([0])
list_samples = filter_objective_samples(table)
r = w = spatial_size
subjects = 47 # some subjects were removed because of objective classes and ignore samples: 47
n_exp = 5
# TODO:
# 1) Further decrease the video amount, the one with objective classes >= 6
# list samples: samples with wanted objective class
VidPerSubject, list_samples = get_subfolders_num_crossdb(
inputDir, IgnoredSamples_index, sub_items, table, list_samples)
# print(VidPerSubject)
# print(len(VidPerSubject))
# print(sum(VidPerSubject))
timesteps_TIM = 9
data_dim = r * w
channel = 3
classes = 5
if os.path.isfile(workplace +
"Classification/SAMM_CASME_Optical_label.txt"):
os.remove(workplace +
"Classification/SAMM_CASME_Optical_label.txt")
##################### Variables ######################
######################################################
############## Flags ####################
tensorboard_flag = tensorboard
resizedFlag = 1
train_spatial_flag = 0
train_temporal_flag = 0
svm_flag = 0
finetuning_flag = 0
cam_visualizer_flag = 0
channel_flag = 0
if flag == 'st':
train_spatial_flag = 1
train_temporal_flag = 1
finetuning_flag = 1
elif flag == 's':
train_spatial_flag = 1
finetuning_flag = 1
elif flag == 't':
train_temporal_flag = 1
elif flag == 'nofine':
svm_flag = 1
elif flag == 'scratch':
train_spatial_flag = 1
train_temporal_flag = 1
elif flag == 'st4':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 1
elif flag == 'st7':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 2
#########################################
############ Reading Images and Labels ################
SubperdB = Read_Input_Images_SAMM_CASME(inputDir, list_samples,
listOfIgnoredSamples, dB,
resizedFlag, table, workplace,
spatial_size, channel)
print("Loaded Images into the tray...")
labelperSub = label_matching(workplace, dB, subjects, VidPerSubject)
print("Loaded Labels into the tray...")
if channel_flag == 1:
SubperdB_strain = Read_Input_Images_SAMM_CASME(
inputDir, list_samples, listOfIgnoredSamples, 'SAMM_CASME_Strain',
resizedFlag, table, workplace, spatial_size, 1)
elif channel_flag == 2:
SubperdB_strain = Read_Input_Images(inputDir, listOfIgnoredSamples,
'CASME2_Strain_TIM10', resizedFlag,
table, workplace, spatial_size, 1)
SubperdB_gray = Read_Input_Images(inputDir, listOfIgnoredSamples,
'CASME2_TIM', resizedFlag, table,
workplace, spatial_size, 3)
#######################################################
########### Model Configurations #######################
sgd = optimizers.SGD(lr=0.0001, decay=1e-7, momentum=0.9, nesterov=True)
adam = optimizers.Adam(lr=0.00001, decay=0.000001)
adam2 = optimizers.Adam(lr=0.00075, decay=0.0001)
# Different Conditions for Temporal Learning ONLY
if train_spatial_flag == 0 and train_temporal_flag == 1 and dB != 'CASME2_Optical':
data_dim = spatial_size * spatial_size
elif train_spatial_flag == 0 and train_temporal_flag == 1 and dB == 'CASME2_Optical':
data_dim = spatial_size * spatial_size * 3
else:
data_dim = 4096
########################################################
########### Training Process ############
# total confusion matrix to be used in the computation of f1 score
tot_mat = np.zeros((n_exp, n_exp))
# model checkpoint
spatial_weights_name = 'vgg_spatial_' + str(train_id) + '_casme2_'
temporal_weights_name = 'temporal_ID_' + str(train_id) + '_casme2_'
history = LossHistory()
stopping = EarlyStopping(monitor='loss',
min_delta=0,
mode='min',
patience=5)
# model checkpoint
if os.path.isdir('/media/viprlab/01D31FFEF66D5170/Weights/' +
str(train_id)) == False:
os.mkdir('/media/viprlab/01D31FFEF66D5170/Weights/' + str(train_id))
for sub in range(subjects):
# sub = sub + 25
# if sub > subjects:
# break
############### Reinitialization & weights reset of models ########################
spatial_weights_name = '/media/viprlab/01D31FFEF66D5170/Weights/' + str(
train_id) + '/vgg_spatial_' + str(train_id) + '_' + str(
dB) + '_' + str(sub) + '.h5'
spatial_weights_name_strain = '/media/viprlab/01D31FFEF66D5170/Weights/' + str(
train_id) + '/vgg_spatial_strain_' + str(train_id) + '_' + str(
dB) + '_' + str(sub) + '.h5'
temporal_weights_name = '/media/viprlab/01D31FFEF66D5170/Weights/' + str(
train_id) + '/temporal_ID_' + str(train_id) + '_' + str(
dB) + '_' + str(sub) + '.h5'
ae_weights_name = '/media/viprlab/01D31FFEF66D5170/Weights/' + str(
train_id) + '/autoencoder_' + str(train_id) + '_' + str(
dB) + '_' + str(sub) + '.h5'
ae_weights_name_strain = '/media/viprlab/01D31FFEF66D5170/Weights/' + str(
train_id) + '/autoencoder_strain_' + str(train_id) + '_' + str(
dB) + '_' + str(sub) + '.h5'
temporal_model = temporal_module(data_dim=data_dim,
timesteps_TIM=timesteps_TIM,
weights_path=temporal_weights_name)
temporal_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
if channel_flag == 1 or channel_flag == 2:
vgg_model = VGG_16_4_channels(spatial_size=spatial_size,
weights_path=spatial_weights_name)
vgg_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
else:
vgg_model = VGG_16(spatial_size=spatial_size,
weights_path='VGG_Face_Deep_16.h5')
vgg_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
svm_classifier = SVC(kernel='linear', C=1)
####################################################################################
Train_X, Train_Y, Test_X, Test_Y, Test_Y_gt = data_loader_with_LOSO(
sub, SubperdB, labelperSub, subjects)
# Rearrange Training labels into a vector of images, breaking sequence
Train_X_spatial = Train_X.reshape(Train_X.shape[0] * timesteps_TIM, r,
w, channel)
Test_X_spatial = Test_X.reshape(Test_X.shape[0] * timesteps_TIM, r, w,
channel)
# Special Loading for 4-Channel
if channel_flag == 1:
Train_X_Strain, _, Test_X_Strain, _, _ = data_loader_with_LOSO(
sub, SubperdB_strain, labelperSub, subjects)
Train_X_Strain = Train_X_Strain.reshape(
Train_X_Strain.shape[0] * timesteps_TIM, r, w, 1)
Test_X_Strain = Test_X_Strain.reshape(
Test_X.shape[0] * timesteps_TIM, r, w, 1)
# Concatenate Train X & Train_X_Strain
Train_X_spatial = np.concatenate((Train_X_spatial, Train_X_Strain),
axis=3)
Test_X_spatial = np.concatenate((Test_X_spatial, Test_X_Strain),
axis=3)
total_channel = 4
elif channel_flag == 2:
Train_X_Strain, _, Test_X_Strain, _, _ = data_loader_with_LOSO(
sub, SubperdB_strain, labelperSub, subjects, classes)
Train_X_gray, _, Test_X_gray, _, _ = data_loader_with_LOSO(
sub, SubperdB_gray, labelperSub, subjects)
Train_X_Strain = Train_X_Strain.reshape(
Train_X_Strain.shape[0] * timesteps_TIM, r, w, 1)
Test_X_Strain = Test_X_Strain.reshape(
Test_X_Strain.shape[0] * timesteps_TIM, r, w, 1)
Train_X_gray = Train_X_gray.reshape(
Train_X_gray.shape[0] * timesteps_TIM, r, w, 3)
Test_X_gray = Test_X_gray.reshape(
Test_X_gray.shape[0] * timesteps_TIM, r, w, 3)
# Concatenate Train_X_Strain & Train_X & Train_X_gray
Train_X_spatial = np.concatenate(
(Train_X_spatial, Train_X_Strain, Train_X_gray), axis=3)
Test_X_spatial = np.concatenate(
(Test_X_spatial, Test_X_Strain, Test_X_gray), axis=3)
total_channel = 7
if channel == 1:
# Duplicate channel of input image
Train_X_spatial = duplicate_channel(Train_X_spatial)
Test_X_spatial = duplicate_channel(Test_X_spatial)
# Extend Y labels 10 fold, so that all images have labels
Train_Y_spatial = np.repeat(Train_Y, timesteps_TIM, axis=0)
Test_Y_spatial = np.repeat(Test_Y, timesteps_TIM, axis=0)
print("Train_X_shape: " + str(np.shape(Train_X_spatial)))
print("Train_Y_shape: " + str(np.shape(Train_Y_spatial)))
print("Test_X_shape: " + str(np.shape(Test_X_spatial)))
print("Test_Y_shape: " + str(np.shape(Test_Y_spatial)))
# print(Train_X_spatial)
##################### Training & Testing #########################
X = Train_X_spatial.reshape(Train_X_spatial.shape[0], total_channel, r,
w)
y = Train_Y_spatial.reshape(Train_Y_spatial.shape[0], classes)
normalized_X = X.astype('float32') / 255.
test_X = Test_X_spatial.reshape(Test_X_spatial.shape[0], total_channel,
r, w)
test_y = Test_Y_spatial.reshape(Test_Y_spatial.shape[0], classes)
normalized_test_X = test_X.astype('float32') / 255.
print(X.shape)
###### conv weights must be freezed for transfer learning ######
if finetuning_flag == 1:
for layer in vgg_model.layers[:33]:
layer.trainable = False
for layer in vgg_model_cam.layers[:31]:
layer.trainable = False
if train_spatial_flag == 1 and train_temporal_flag == 1:
# record f1 and loss
file_loss = open(
workplace + 'Classification/' + 'Result/' + dB + '/loss_' +
str(train_id) + '.txt', 'a')
file_loss.write(str(history.losses) + "\n")
file_loss.close()
file_loss = open(
workplace + 'Classification/' + 'Result/' + dB + '/accuracy_' +
str(train_id) + '.txt', 'a')
file_loss.write(str(history.accuracy) + "\n")
file_loss.close()
model = Model(inputs=vgg_model.input,
outputs=vgg_model.layers[35].output)
plot_model(model,
to_file="spatial_module_FULL_TRAINING.png",
show_shapes=True)
# Testing
output = model.predict(test_X, batch_size=batch_size)
features = output.reshape(Test_X.shape[0], timesteps_TIM,
output.shape[1])
predict = temporal_model.predict_classes(features,
batch_size=batch_size)
##############################################################
#################### Confusion Matrix Construction #############
print(predict)
print(Test_Y_gt)
ct = confusion_matrix(Test_Y_gt, predict)
# check the order of the CT
order = np.unique(np.concatenate((predict, Test_Y_gt)))
# create an array to hold the CT for each CV
mat = np.zeros((n_exp, n_exp))
# put the order accordingly, in order to form the overall ConfusionMat
for m in range(len(order)):
for n in range(len(order)):
mat[int(order[m]), int(order[n])] = ct[m, n]
tot_mat = mat + tot_mat
################################################################
#################### cumulative f1 plotting ######################
microAcc = np.trace(tot_mat) / np.sum(tot_mat)
[f1, precision, recall] = fpr(tot_mat, n_exp)
file = open(
workplace + 'Classification/' + 'Result/' + dB + '/f1_' +
str(train_id) + '.txt', 'a')
file.write(str(f1) + "\n")
file.close()
##################################################################
################# write each CT of each CV into .txt file #####################
record_scores(workplace, dB, ct, sub, order, tot_mat, n_exp, subjects)
###############################################################################
def train_samm_cross(batch_size, spatial_epochs, temporal_epochs, train_id, dB, spatial_size, flag, tensorboard):
############## Path Preparation ######################
root_db_path = "/media/ice/OS/Datasets/"
workplace = root_db_path + dB + "/"
inputDir = root_db_path + dB + "/" + dB + "/"
######################################################
classes = 5
if dB == 'CASME2_TIM':
table = loading_casme_table(workplace + 'CASME2-ObjectiveClasses.xlsx')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(inputDir)
############## Variables ###################
r = w = spatial_size
subjects=2
n_exp = 5
# VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
listOfIgnoredSamples = []
VidPerSubject = [2,1]
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
############################################
os.remove(workplace + "Classification/CASME2_TIM_label.txt")
elif dB == 'CASME2_Optical':
table = loading_casme_table(workplace + 'CASME2-ObjectiveClasses.xlsx')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(inputDir)
############## Variables ###################
r = w = spatial_size
subjects=26
n_exp = 5
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
pad_sequence = 9
channel = 3
############################################
# os.remove(workplace + "Classification/CASME2_TIM_label.txt")
elif dB == 'SAMM_TIM10':
table, table_objective = loading_samm_table(root_db_path, dB)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
n_exp = 8
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
classes = 8
#########################################################
elif dB == 'SAMM_CASME_Optical':
# total amount of videos 253
# table, table_objective = loading_samm_table(root_db_path, dB)
# table = table_objective
table = loading_casme_objective_table(root_db_path, dB)
# merge samm and casme tables
# table = np.concatenate((table, table2), axis=1)
# print(table.shape)
# listOfIgnoredSamples, IgnoredSamples_index, sub_items = ignore_casme_samples(inputDir)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
sub_items = np.empty([0])
list_samples = filter_objective_samples(table)
r = w = spatial_size
subjects = 26 # some subjects were removed because of objective classes and ignore samples: 47
n_exp = 5
# TODO:
# 1) Further decrease the video amount, the one with objective classes >= 6
# list samples: samples with wanted objective class
VidPerSubject, list_samples = get_subfolders_num_crossdb(inputDir, IgnoredSamples_index, sub_items, table, list_samples)
# print(VidPerSubject)
# print(len(VidPerSubject))
# print(sum(VidPerSubject))
timesteps_TIM = 9
data_dim = r * w
channel = 3
classes = 5
if os.path.isfile(workplace + "Classification/SAMM_CASME_Optical_label.txt"):
os.remove(workplace + "Classification/SAMM_CASME_Optical_label.txt")
##################### Variables ######################
######################################################
############## Flags ####################
tensorboard_flag = tensorboard
resizedFlag = 1
train_spatial_flag = 0
train_temporal_flag = 0
svm_flag = 0
finetuning_flag = 0
cam_visualizer_flag = 0
channel_flag = 0
if flag == 'st':
train_spatial_flag = 1
train_temporal_flag = 1
finetuning_flag = 1
elif flag == 's':
train_spatial_flag = 1
finetuning_flag = 1
elif flag == 't':
train_temporal_flag = 1
elif flag == 'nofine':
svm_flag = 1
elif flag == 'scratch':
train_spatial_flag = 1
train_temporal_flag = 1
elif flag == 'st4':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 1
elif flag == 'st7':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 2
#########################################
############ Reading Images and Labels ################
SubperdB = Read_Input_Images_SAMM_CASME(inputDir, list_samples, listOfIgnoredSamples, dB, resizedFlag, table, workplace, spatial_size, channel)
print("Loaded Images into the tray...")
labelperSub = label_matching(workplace, dB, subjects, VidPerSubject)
print("Loaded Labels into the tray...")
if channel_flag == 1:
SubperdB_strain = Read_Input_Images(inputDir, listOfIgnoredSamples, 'CASME2_Strain_TIM10', resizedFlag, table, workplace, spatial_size, 1)
elif channel_flag == 2:
SubperdB_strain = Read_Input_Images(inputDir, listOfIgnoredSamples, 'CASME2_Strain_TIM10', resizedFlag, table, workplace, spatial_size, 1)
SubperdB_gray = Read_Input_Images(inputDir, listOfIgnoredSamples, 'CASME2_TIM', resizedFlag, table, workplace, spatial_size, 3)
#######################################################
########### Model Configurations #######################
sgd = optimizers.SGD(lr=0.0001, decay=1e-7, momentum=0.9, nesterov=True)
adam = optimizers.Adam(lr=0.00001, decay=0.000001)
adam2 = optimizers.Adam(lr= 0.00075, decay= 0.0001)
# Different Conditions for Temporal Learning ONLY
if train_spatial_flag == 0 and train_temporal_flag == 1 and dB != 'CASME2_Optical':
data_dim = spatial_size * spatial_size
elif train_spatial_flag == 0 and train_temporal_flag == 1 and dB == 'CASME2_Optical':
data_dim = spatial_size * spatial_size * 3
else:
data_dim = 4096
########################################################
########### Training Process ############
# total confusion matrix to be used in the computation of f1 score
tot_mat = np.zeros((n_exp,n_exp))
# model checkpoint
spatial_weights_name = 'vgg_spatial_'+ str(train_id) + '_casme2_'
temporal_weights_name = 'temporal_ID_' + str(train_id) + '_casme2_'
history = LossHistory()
stopping = EarlyStopping(monitor='loss', min_delta = 0, mode = 'min')
############### Reinitialization & weights reset of models ########################
vgg_model_cam = VGG_16(spatial_size=spatial_size, classes=classes, weights_path='VGG_Face_Deep_16.h5')
temporal_model = temporal_module(data_dim=data_dim, classes=classes, timesteps_TIM=timesteps_TIM)
temporal_model.compile(loss='categorical_crossentropy', optimizer=adam, metrics=[metrics.categorical_accuracy])
conv_ae = convolutional_autoencoder(spatial_size = spatial_size, classes = classes)
conv_ae.compile(loss='binary_crossentropy', optimizer=adam)
if channel_flag == 1 or channel_flag == 2:
vgg_model = VGG_16_4_channels(classes=classes, spatial_size = spatial_size)
vgg_model.compile(loss='categorical_crossentropy', optimizer=adam, metrics=[metrics.categorical_accuracy])
else:
vgg_model = VGG_16(spatial_size = spatial_size, classes=classes, weights_path='VGG_Face_Deep_16.h5')
vgg_model.compile(loss='categorical_crossentropy', optimizer=adam, metrics=[metrics.categorical_accuracy])
svm_classifier = SVC(kernel='linear', C=1)
####################################################################################
############ for tensorboard ###############
if tensorboard_flag == 1:
cat_path = tensorboard_path + str(sub) + "/"
os.mkdir(cat_path)
tbCallBack = keras.callbacks.TensorBoard(log_dir=cat_path, write_graph=True)
cat_path2 = tensorboard_path + str(sub) + "spat/"
os.mkdir(cat_path2)
tbCallBack2 = keras.callbacks.TensorBoard(log_dir=cat_path2, write_graph=True)
#############################################
image_label_mapping = np.empty([0])
Train_X, Train_Y= standard_data_loader(SubperdB, labelperSub, subjects, classes)
# Rearrange Training labels into a vector of images, breaking sequence
Train_X_spatial = Train_X.reshape(Train_X.shape[0]*timesteps_TIM, r, w, channel)
# Test_X_spatial = Test_X.reshape(Test_X.shape[0]* timesteps_TIM, r, w, channel)
# Special Loading for 4-Channel
if channel_flag == 1:
Train_X_Strain, _, Test_X_Strain, _, _ = data_loader_with_LOSO(sub, SubperdB_strain, labelperSub, subjects, classes)
Train_X_Strain = Train_X_Strain.reshape(Train_X_Strain.shape[0]*timesteps_TIM, r, w, 1)
Test_X_Strain = Test_X_Strain.reshape(Test_X.shape[0]*timesteps_TIM, r, w, 1)
# Concatenate Train X & Train_X_Strain
Train_X_spatial = np.concatenate((Train_X_spatial, Train_X_Strain), axis=3)
Test_X_spatial = np.concatenate((Test_X_spatial, Test_X_Strain), axis=3)
channel = 4
elif channel_flag == 2:
Train_X_Strain, _, Test_X_Strain, _, _ = data_loader_with_LOSO(sub, SubperdB_strain, labelperSub, subjects, classes)
Train_X_gray, _, Test_X_gray, _, _ = data_loader_with_LOSO(sub, SubperdB_gray, labelperSub, subjects)
Train_X_Strain = Train_X_Strain.reshape(Train_X_Strain.shape[0]*timesteps_TIM, r, w, 1)
Test_X_Strain = Test_X_Strain.reshape(Test_X_Strain.shape[0]*timesteps_TIM, r, w, 1)
Train_X_gray = Train_X_gray.reshape(Train_X_gray.shape[0]*timesteps_TIM, r, w, 3)
Test_X_gray = Test_X_gray.reshape(Test_X_gray.shape[0]*timesteps_TIM, r, w, 3)
# Concatenate Train_X_Strain & Train_X & Train_X_gray
Train_X_spatial = np.concatenate((Train_X_spatial, Train_X_Strain, Train_X_gray), axis=3)
Test_X_spatial = np.concatenate((Test_X_spatial, Test_X_Strain, Test_X_gray), axis=3)
channel = 7
if channel == 1:
# Duplicate channel of input image
Train_X_spatial = duplicate_channel(Train_X_spatial)
# Test_X_spatial = duplicate_channel(Test_X_spatial)
# Extend Y labels 10 fold, so that all images have labels
Train_Y_spatial = np.repeat(Train_Y, timesteps_TIM, axis=0)
# Test_Y_spatial = np.repeat(Test_Y, timesteps_TIM, axis=0)
# print ("Train_X_shape: " + str(np.shape(Train_X_spatial)))
# print ("Train_Y_shape: " + str(np.shape(Train_Y_spatial)))
# print ("Test_X_shape: " + str(np.shape(Test_X_spatial)))
# print ("Test_Y_shape: " + str(np.shape(Test_Y_spatial)))
# print(Train_X_spatial)
##################### Training & Testing #########################
X = Train_X_spatial.reshape(Train_X_spatial.shape[0], channel, r, w)
y = Train_Y_spatial.reshape(Train_Y_spatial.shape[0], classes)
normalized_X = X.astype('float32') / 255.
# test_X = Test_X_spatial.reshape(Test_X_spatial.shape[0], channel, r, w)
# test_y = Test_Y_spatial.reshape(Test_Y_spatial.shape[0], classes)
# normalized_test_X = test_X.astype('float32') / 255.
print(X.shape)
###### conv weights must be freezed for transfer learning ######
if finetuning_flag == 1:
for layer in vgg_model.layers[:33]:
layer.trainable = False
for layer in vgg_model_cam.layers[:31]:
layer.trainable = False
if train_spatial_flag == 1 and train_temporal_flag == 1:
# Autoencoder features
# conv_ae.fit(normalized_X, normalized_X, batch_size=batch_size, epochs=spatial_epochs, shuffle=True)
# Spatial Training
if tensorboard_flag == 1:
vgg_model.fit(X, y, batch_size=batch_size, epochs=spatial_epochs, shuffle=True, callbacks=[tbCallBack2])
else:
vgg_model.fit(X, y, batch_size=batch_size, epochs=spatial_epochs, shuffle=True, callbacks=[history, stopping])
# record f1 and loss
file_loss = open(workplace+'Classification/'+ 'Result/'+dB+'/loss_' + str(train_id) + '.txt', 'a')
file_loss.write(str(history.losses) + "\n")
file_loss.close()
file_loss = open(workplace+'Classification/'+ 'Result/'+dB+'/accuracy_' + str(train_id) + '.txt', 'a')
file_loss.write(str(history.accuracy) + "\n")
file_loss.close()
vgg_model.save_weights(spatial_weights_name + 'HDE'+ ".h5")
model = Model(inputs=vgg_model.input, outputs=vgg_model.layers[35].output)
plot_model(model, to_file="spatial_module_FULL_TRAINING.png", show_shapes=True)
model_ae = Model(inputs=conv_ae.input, outputs=conv_ae.output)
plot_model(model_ae, to_file='autoencoders.png', show_shapes=True)
# Autoencoding
output_ae = model_ae.predict(normalized_X, batch_size = batch_size)
for i in range(batch_size):
visual_ae = output_ae[i].reshape(224,224,channel)
# de-normalize
visual_ae = ( ( visual_ae - min(visual_ae) ) / ( max(visual_ae) - min(visual_ae) ) ) * 255
fname = '{prefix}_{index}_{hash}.{format}'.format(prefix='AE_output', index=str(sub),
hash=np.random.randint(1e7), format='png')
cv2.imwrite(workplace+'Classification/Result/ae_train/'+fname, visual_ae)
output_ae = model.predict(output_ae, batch_size = batch_size)
# Spatial Encoding
output = model.predict(X, batch_size = batch_size)
# features = output.reshape(int(Train_X.shape[0]), timesteps_TIM, output.shape[1])
# merging autoencoded features and spatial features
output = np.concatenate((output, output_ae), axis=1)
# print(output.shape)
features = output.reshape(int(Train_X.shape[0]), timesteps_TIM, output.shape[1])
# Temporal Training
if tensorboard_flag == 1:
temporal_model.fit(features, Train_Y, batch_size=batch_size, epochs=temporal_epochs, callbacks=[tbCallBack])
else:
temporal_model.fit(features, Train_Y, batch_size=batch_size, epochs=temporal_epochs)
temporal_model.save_weights(temporal_weights_name + 'HDE' + ".h5")
# # Testing
# output = model.predict(test_X, batch_size = batch_size)
# output_ae = model_ae.predict(normalized_test_X, batch_size = batch_size)
# for i in range(batch_size):
# visual_ae = output_ae[i].reshape(224,224,channel)
# # de-normalize
# visual_ae = ( ( visual_ae - min(visual_ae) ) / ( max(visual_ae) - min(visual_ae) ) ) * 255
# fname = '{prefix}_{index}_{hash}.{format}'.format(prefix='AE_output', index=str(sub),
# hash=np.random.randint(1e7), format='png')
# cv2.imwrite(workplace+'Classification/Result/ae_train/'+fname, visual_ae)
# output_ae = model.predict(output_ae, batch_size = batch_size)
# output = np.concatenate((output, output_ae), axis=1)
# features = output.reshape(Test_X.shape[0], timesteps_TIM, output.shape[1])
# predict = temporal_model.predict_classes(features, batch_size=batch_size)
# ##############################################################
# #################### Confusion Matrix Construction #############
# print (predict)
# print (Test_Y_gt)
# ct = confusion_matrix(Test_Y_gt,predict)
# # check the order of the CT
# order = np.unique(np.concatenate((predict,Test_Y_gt)))
# # create an array to hold the CT for each CV
# mat = np.zeros((n_exp,n_exp))
# # put the order accordingly, in order to form the overall ConfusionMat
# for m in range(len(order)):
# for n in range(len(order)):
# mat[int(order[m]),int(order[n])]=ct[m,n]
# tot_mat = mat + tot_mat
# ################################################################
# #################### cumulative f1 plotting ######################
# microAcc = np.trace(tot_mat) / np.sum(tot_mat)
# [f1,precision,recall] = fpr(tot_mat,n_exp)
# file = open(workplace+'Classification/'+ 'Result/'+dB+'/f1_' + str(train_id) + '.txt', 'a')
# file.write(str(f1) + "\n")
# file.close()
##################################################################
################# write each CT of each CV into .txt file #####################
# record_scores(workplace, dB, ct, sub, order, tot_mat, n_exp, subjects)
###############################################################################
IgnoredSamples_index = np.empty([0])
for item in IgnoredSamples:
item = item.split('sub', 1)[1]
item = int(item.split('/', 1)[0]) - 1
IgnoredSamples_index = np.append(IgnoredSamples_index, item)
#######################################################################
############## Variables ###################
spatial_size = 224
r = w = spatial_size
resizedFlag = 1
subjects = 26
samples = 246
n_exp = 5
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
############################################
################## Clearing labels.txt ################
os.remove(workplace + "Classification/CASME2_TIM_label.txt")
#######################################################
############ Reading Images and Labels ################
SubperdB = Read_Input_Images(inputDir, listOfIgnoredSamples, dB, resizedFlag,
table, workplace, spatial_size)
print("Loaded Images into the tray...")
labelperSub = label_matching(workplace, dB, subjects, VidPerSubject)
print("Loaded Labels into the tray...")
def test_casme(batch_size, spatial_epochs, temporal_epochs, train_id, dB,
spatial_size, flag, tensorboard):
############## Path Preparation ######################
root_db_path = "/media/ice/OS/Datasets/"
workplace = root_db_path + dB + "/"
inputDir = root_db_path + dB + "/" + dB + "/"
######################################################
classes = 5
if dB == 'CASME2_TIM':
table = loading_casme_table(workplace + 'CASME2_label_Ver_2.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
inputDir)
############## Variables ###################
r = w = spatial_size
subjects = 2
samples = 246
n_exp = 5
# VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
listOfIgnoredSamples = []
VidPerSubject = [2, 1]
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
############################################
os.remove(workplace + "Classification/CASME2_TIM_label.txt")
elif dB == 'CASME2_Optical':
table = loading_casme_table(workplace + 'CASME2_label_Ver_2.xls')
listOfIgnoredSamples, IgnoredSamples_index, _ = ignore_casme_samples(
inputDir)
############## Variables ###################
r = w = spatial_size
subjects = 26
samples = 246
n_exp = 5
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
pad_sequence = 9
channel = 3
############################################
# os.remove(workplace + "Classification/CASME2_TIM_label.txt")
elif dB == 'CASME2_RGB':
# print(inputDir)
table = loading_casme_table(workplace +
'CASME2_RGB/CASME2_label_Ver_2.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casmergb_samples(
inputDir)
############## Variables ###################
r = w = spatial_size
subjects = 26
samples = 245 # not used, delete it later
n_exp = 5
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
############################################
elif dB == 'SMIC_TIM10':
table = loading_smic_table(root_db_path, dB)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 16
samples = 164
n_exp = 3
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 1
classes = 3
#########################################################
elif dB == 'SAMM_Optical':
table, table_objective = loading_samm_table(root_db_path, dB)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
samples = 159
n_exp = 8
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
pad_sequence = 10
channel = 3
classes = 8
#########################################################
elif dB == 'SAMM_TIM10':
table, table_objective = loading_samm_table(root_db_path, dB)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
samples = 159
n_exp = 8
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
classes = 8
#########################################################
# print(VidPerSubject)
############## Flags ####################
tensorboard_flag = tensorboard
resizedFlag = 1
train_spatial_flag = 0
train_temporal_flag = 0
svm_flag = 0
finetuning_flag = 0
cam_visualizer_flag = 0
channel_flag = 0
if flag == 'st':
train_spatial_flag = 1
train_temporal_flag = 1
finetuning_flag = 1
elif flag == 's':
train_spatial_flag = 1
finetuning_flag = 1
elif flag == 't':
train_temporal_flag = 1
elif flag == 'nofine':
svm_flag = 1
elif flag == 'scratch':
train_spatial_flag = 1
train_temporal_flag = 1
elif flag == 'st4':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 1
elif flag == 'st7':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 2
elif flag == 'st4vis':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 3
#########################################
############ Reading Images and Labels ################
SubperdB = Read_Input_Images(inputDir, listOfIgnoredSamples, dB,
resizedFlag, table, workplace, spatial_size,
channel)
print("Loaded Images into the tray...")
labelperSub = label_matching(workplace, dB, subjects, VidPerSubject)
print("Loaded Labels into the tray...")
if channel_flag == 1:
inputDir = root_db_path + dB + "/" + dB + "/"
SubperdB_strain = Read_Input_Images(
root_db_path + 'CASME2_Strain_TIM10' + '/' +
'CASME2_Strain_TIM10' + '/', listOfIgnoredSamples,
'CASME2_Strain_TIM10', resizedFlag, table, workplace, spatial_size,
3)
SubperdB_gray = Read_Input_Images(
root_db_path + 'CASME2_TIM' + '/' + 'CASME2_TIM' + '/',
listOfIgnoredSamples, 'CASME2_TIM', resizedFlag, table, workplace,
spatial_size, 3)
elif channel_flag == 3:
inputDir_strain = '/media/ice/OS/Datasets/CASME2_Strain_TIM10/CASME2_Strain_TIM10/'
SubperdB_strain = Read_Input_Images(inputDir_strain,
listOfIgnoredSamples,
'CASME2_Strain_TIM10', resizedFlag,
table, workplace, spatial_size, 3)
inputDir_gray = '/media/ice/OS/Datasets/CASME2_TIM/CASME2_TIM/'
SubperdB_gray = Read_Input_Images(inputDir_gray, listOfIgnoredSamples,
'CASME2_TIM', resizedFlag, table,
workplace, spatial_size, 3)
elif channel_flag == 2:
SubperdB_strain = Read_Input_Images(inputDir, listOfIgnoredSamples,
'CASME2_Strain_TIM10', resizedFlag,
table, workplace, spatial_size, 1)
SubperdB_gray = Read_Input_Images(inputDir, listOfIgnoredSamples,
'CASME2_TIM', resizedFlag, table,
workplace, spatial_size, 3)
#######################################################
########### Model Configurations #######################
sgd = optimizers.SGD(lr=0.0001, decay=1e-7, momentum=0.9, nesterov=True)
adam = optimizers.Adam(lr=0.00001, decay=0.000001)
# Different Conditions for Temporal Learning ONLY
if train_spatial_flag == 0 and train_temporal_flag == 1 and dB != 'CASME2_Optical':
data_dim = spatial_size * spatial_size
elif train_spatial_flag == 0 and train_temporal_flag == 1 and dB == 'CASME2_Optical':
data_dim = spatial_size * spatial_size * 3
else:
data_dim = 8192
########################################################
########### Image Data Generator ##############
image_generator = ImageDataGenerator(zca_whitening=True,
rotation_range=0.2,
width_shift_range=0.2,
height_shift_range=0.2,
zoom_range=0.2,
horizontal_flip=True,
rescale=1.5)
###############################################
########### Training Process ############
# Todo:
# 1) LOSO (done)
# 2) call model (done)
# 3) saving model architecture
# 4) Saving Checkpoint (done)
# 5) make prediction (done)
if tensorboard_flag == 1:
tensorboard_path = "/home/ice/Documents/Micro-Expression/tensorboard/"
# total confusion matrix to be used in the computation of f1 score
tot_mat = np.zeros((n_exp, n_exp))
weights_dir = '/media/ice/OS/Datasets/Weights/53/'
image_path = '/home/ice/Documents/Micro-Expression/image/'
table_count = 0
for sub in range(subjects):
############### Reinitialization & weights reset of models ########################
temporal_model_weights = weights_dir + 'temporal_enrichment_ID_' + str(
train_id) + '_' + str(dB) + '_' + str(sub) + '.h5'
vgg_model_weights = weights_dir + 'vgg_spatial_' + str(
train_id) + '_' + str(dB) + '_' + str(sub) + '.h5'
vgg_model_strain_weights = weights_dir + 'vgg_spatial_strain_' + str(
train_id) + '_' + str(dB) + '_' + str(sub) + '.h5'
conv_ae_weights = weights_dir + 'autoencoder_' + str(
train_id) + '_' + str(dB) + '_' + str(sub) + '.h5'
conv_ae_strain_weights = weights_dir + 'autoencoder_strain_' + str(
train_id) + '_' + str(dB) + '_' + str(sub) + '.h5'
temporal_model = temporal_module(data_dim=data_dim,
timesteps_TIM=timesteps_TIM,
weights_path=temporal_model_weights)
temporal_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
conv_ae = convolutional_autoencoder(spatial_size=spatial_size,
weights_path=conv_ae_weights)
conv_ae.compile(loss='binary_crossentropy', optimizer=adam)
conv_ae_strain = convolutional_autoencoder(
spatial_size=spatial_size, weights_path=conv_ae_strain_weights)
conv_ae_strain.compile(loss='binary_crossentropy', optimizer=adam)
vgg_model = VGG_16(spatial_size=spatial_size,
classes=classes,
weights_path=vgg_model_weights)
vgg_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
vgg_model_strain = VGG_16(spatial_size=spatial_size,
classes=classes,
weights_path=vgg_model_strain_weights)
vgg_model_strain.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
svm_classifier = SVC(kernel='linear', C=1)
####################################################################################
Train_X, Train_Y, Test_X, Test_Y, Test_Y_gt = data_loader_with_LOSO(
sub, SubperdB, labelperSub, subjects, classes)
# Rearrange Training labels into a vector of images, breaking sequence
Train_X_spatial = Train_X.reshape(Train_X.shape[0] * timesteps_TIM, r,
w, channel)
Test_X_spatial = Test_X.reshape(Test_X.shape[0] * timesteps_TIM, r, w,
channel)
# Special Loading for 4-Channel
if channel_flag == 1 or channel_flag == 3:
Train_X_Strain, _, Test_X_Strain, _, _ = data_loader_with_LOSO(
sub, SubperdB_strain, labelperSub, subjects, classes)
Train_X_Strain = Train_X_Strain.reshape(
Train_X_Strain.shape[0] * timesteps_TIM, r, w, 3)
Test_X_Strain = Test_X_Strain.reshape(
Test_X.shape[0] * timesteps_TIM, r, w, 3)
Train_X_Gray, _, Test_X_Gray, _, _ = data_loader_with_LOSO(
sub, SubperdB_gray, labelperSub, subjects, classes)
Test_X_Gray = Test_X_Gray.reshape(Test_X_Gray.shape[0] * 10, r, w,
3)
# print(Train_X_Strain.shape)
# Train_X_Strain = Train_X_Strain[0]
# Train_X_Strain = Train_X_Strain.reshape((224, 224, 3, 1))
# Train_X_Strain = Train_X_Strain.reshape((224, 224, 3))
# cv2.imwrite('steveharvey.png', Train_X_Strain)
# Concatenate Train X & Train_X_Strain
# Train_X_spatial = np.concatenate((Train_X_spatial, Train_X_Strain), axis=3)
# Test_X_spatial = np.concatenate((Test_X_spatial, Test_X_Strain), axis=3)
total_channel = 4
# Extend Y labels 10 fold, so that all images have labels
Train_Y_spatial = np.repeat(Train_Y, timesteps_TIM, axis=0)
Test_Y_spatial = np.repeat(Test_Y, timesteps_TIM, axis=0)
##################### Training & Testing #########################
# print(Train_X_spatial.shape)
test_X = Test_X_spatial.reshape(Test_X_spatial.shape[0], channel, r, w)
test_y = Test_Y_spatial.reshape(Test_Y_spatial.shape[0], classes)
normalized_test_X = test_X.astype('float32') / 255.
Test_X_Strain = Test_X_Strain.reshape(Test_X_Strain.shape[0], channel,
r, w)
Test_X_Gray = Test_X_Gray.reshape(Test_X_Gray.shape[0], channel, r, w)
# test_y = Test_Y_spatial.reshape(Test_Y_spatial.shape[0], classes)
normalized_test_X_strain = test_X.astype('float32') / 255.
# print(X.shape)
###### conv weights must be freezed for transfer learning ######
if finetuning_flag == 1:
for layer in vgg_model.layers[:33]:
layer.trainable = False
if train_spatial_flag == 1 and train_temporal_flag == 1:
# vgg
model = Model(inputs=vgg_model.input,
outputs=vgg_model.layers[35].output)
plot_model(model,
to_file="spatial_module_FULL_TRAINING.png",
show_shapes=True)
output = model.predict(test_X)
# vgg strain
model_strain = Model(inputs=vgg_model_strain.input,
outputs=vgg_model_strain.layers[35].output)
plot_model(model_strain,
to_file="spatial_module_FULL_TRAINING_strain.png",
show_shapes=True)
output_strain = model_strain.predict(Test_X_Strain)
# ae
# model_ae = Model(inputs=conv_ae.input, outputs=conv_ae.output)
# plot_model(model_ae, to_file='autoencoders.png', show_shapes=True)
# output_ae = model_ae.predict(normalized_test_X)
# output_ae = model.predict(output_ae)
# ae strain
# model_ae_strain = Model(inputs=conv_ae_strain.input, outputs=conv_ae_strain.output)
# plot_model(model_ae, to_file='autoencoders.png', show_shapes=True)
# output_ae_strain = model_ae_strain.predict(normalized_test_X_strain)
# output_ae_strain = model_ae_strain.predict(output_ae_strain)
# concatenate features
output = np.concatenate((output, output_strain), axis=1)
features = output.reshape(int(Test_X.shape[0]), timesteps_TIM,
output.shape[1])
# temporal
predict = temporal_model.predict_classes(features,
batch_size=batch_size)
# visualize cam
countcam = 0
file = open(
workplace + 'Classification/' + 'Result/' + dB + '/log_hde' +
str(train_id) + '.txt', 'a')
file.write(str(sub + 1) + "\n")
for item_idx in range(len(predict)):
test_strain = Test_X_Gray[item_idx + countcam]
test_strain = test_strain.reshape((224, 224, 3))
item = test_strain
cam_output = visualize_cam(model, 29, 0, item)
cam_output2 = visualize_cam(model, 29, 1, item)
cam_output3 = visualize_cam(model, 29, 2, item)
cam_output4 = visualize_cam(model, 29, 3, item)
cam_output5 = visualize_cam(model, 29, 4, item)
overlaying_cam = overlay(item, cam_output)
overlaying_cam2 = overlay(item, cam_output2)
overlaying_cam3 = overlay(item, cam_output3)
overlaying_cam4 = overlay(item, cam_output4)
overlaying_cam5 = overlay(item, cam_output5)
cv2.imwrite(
image_path + '_' + str(sub) + '_' + str(item_idx) + '_' +
str(predict[item_idx]) + '_' + str(Test_Y_gt[item_idx]) +
'_coverlayingcam0.png', overlaying_cam)
cv2.imwrite(
image_path + '_' + str(sub) + '_' + str(item_idx) + '_' +
str(predict[item_idx]) + '_' + str(Test_Y_gt[item_idx]) +
'_coverlayingcam1.png', overlaying_cam2)
cv2.imwrite(
image_path + '_' + str(sub) + '_' + str(item_idx) + '_' +
str(predict[item_idx]) + '_' + str(Test_Y_gt[item_idx]) +
'_coverlayingcam2.png', overlaying_cam3)
cv2.imwrite(
image_path + '_' + str(sub) + '_' + str(item_idx) + '_' +
str(predict[item_idx]) + '_' + str(Test_Y_gt[item_idx]) +
'_coverlayingcam3.png', overlaying_cam4)
cv2.imwrite(
image_path + '_' + str(sub) + '_' + str(item_idx) + '_' +
str(predict[item_idx]) + '_' + str(Test_Y_gt[item_idx]) +
'_coverlayingcam4.png', overlaying_cam5)
countcam += 9
######## write the log file for megc 2018 ############
result_string = table[table_count, 1] + ' ' + str(
int(Test_Y_gt[item_idx])) + ' ' + str(
predict[item_idx]) + '\n'
file.write(result_string)
######################################################
table_count += 1
##############################################################
#################### Confusion Matrix Construction #############
print(predict)
print(Test_Y_gt)
ct = confusion_matrix(Test_Y_gt, predict)
# print(type(ct))a
# check the order of the CT
order = np.unique(np.concatenate((predict, Test_Y_gt)))
# create an array to hold the CT for each CV
mat = np.zeros((n_exp, n_exp))
# put the order accordingly, in order to form the overall ConfusionMat
for m in range(len(order)):
for n in range(len(order)):
mat[int(order[m]), int(order[n])] = ct[m, n]
tot_mat = mat + tot_mat
################################################################
#################### cumulative f1 plotting ######################
microAcc = np.trace(tot_mat) / np.sum(tot_mat)
[f1, precision, recall] = fpr(tot_mat, n_exp)
file = open(
workplace + 'Classification/' + 'Result/' + dB + '/f1_' +
str(train_id) + '.txt', 'a')
file.write(str(f1) + "\n")
file.close()
##################################################################
################# write each CT of each CV into .txt file #####################
record_scores(workplace, dB, ct, sub, order, tot_mat, n_exp, subjects)
###############################################################################
tot_mat_cm = np.asarray(tot_mat, dtype=int)
plt.figure()
classes_test = [0, 1, 2, 3, 4]
plot_confusion_matrix(tot_mat_cm,
classes_test,
normalize=True,
title='Confusion matrix_single_db')
plt.show()
ws=wb.sheet_by_index(0)
colm=ws.col_slice(colx=0,start_rowx=1,end_rowx=None)
iD=[str(x.value) for x in colm]
colm=ws.col_slice(colx=1,start_rowx=1,end_rowx=None)
vidName=[str(x.value) for x in colm]
colm=ws.col_slice(colx=6,start_rowx=1,end_rowx=None)
expression=[str(x.value) for x in colm]
table=np.transpose(np.array([np.array(iD),np.array(vidName),np.array(expression)],dtype=str))
# print(table)
r=50; w=50
resizedFlag=1;
subjects=26
samples=246
n_exp=5
VidPerSubject = get_subfolders_num(inputDir)
# VidPerSubject = [9,13,7,5,19,5,9,3,13,13,10,12,8,4,3,4,34,3,15,11,2,2,12,7,7,16]
IgnoredSamples=['sub09/EP13_02/','sub09/EP02_02f/','sub10/EP13_01/','sub17/EP15_01/',
'sub17/EP15_03/','sub19/EP19_04/','sub24/EP10_03/','sub24/EP07_01/',
'sub24/EP07_04f/','sub24/EP02_07/','sub26/EP15_01/']
listOfIgnoredSamples=[]
for s in range(len(IgnoredSamples)):
if s==0:
listOfIgnoredSamples=[inputDir+IgnoredSamples[s]]
else:
listOfIgnoredSamples.append(inputDir+IgnoredSamples[s])
elif dB == "SMIC":
inputDir="/srv/oyh/DataBase/SMIC/HS_naming_modified/"
inputDir="/media/ice/OS/Datasets/SMIC"
wb=xlrd.open_workbook('/srv/oyh/DataBase/SMIC_label.xlsx');
def load_db(db_path, list_db, spatial_size, objective_flag):
db_name = list_db[0]
db_home = db_path + db_name + "/"
db_images = db_path + db_name + "/" + db_name + "/"
cross_db_flag = 0
print("retrieving data for:" + db_name)
if db_name == 'CASME2_TIM10':
table = loading_casme_table(db_home + 'CASME2_label_Ver_3.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
db_path, list_db)
r = w = spatial_size
subjects = 26
samples = 257
n_exp = 4
#VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
VidPerSubject, vidList = get_vid_per_subject_augmented(db_images)
timesteps_TIM = 9
data_dim = r * w
channel = 1
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'CASME2_Optical' or db_name == 'CASME2_Optical2' or db_name == 'CASME2_Color_TIM10' or db_name == 'CASME2_Color_TIM20':
print("arrived")
table = loading_casme_table(db_home + 'CASME2_label_Ver_3.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
db_path, list_db)
r = w = spatial_size
subjects = 26
if db_name != 'CASME2_Color_TIM10' and db_name != 'CASME2_Color_TIM20' and db_name != 'CASME2_Optical2':
samples = 257
else:
samples = 255
n_exp = 4
#VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
#VidPerSubject = get_vid_per_subject(table, ['fear', 'sadness'])
VidPerSubject, vidList = get_vid_per_subject_augmented(db_images)
timesteps_TIM = 9
data_dim = r * w
channel = 3
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'CASME1_Color_TIM20':
print("arrived")
table = loading_casme1_table(db_home + 'CASME1.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
db_path, list_db)
r = w = spatial_size
subjects = 19
samples = 189
n_exp = 4
#VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
#VidPerSubject = get_vid_per_subject(table, ['fear', 'sadness'])
VidPerSubject, vidList = get_vid_per_subject_augmented(db_images)
timesteps_TIM = 19
data_dim = r * w
channel = 3
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SMIC_TIM10':
table = loading_smic_table(db_path, db_name)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
r = w = spatial_size
subjects = 16
samples = 164
n_exp = 3
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
channel = 3
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SAMM_Optical':
table, table_objective = loading_samm_table(db_path, db_name,
objective_flag)
# print(table)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
r = w = spatial_size
subjects = 29
samples = 159
n_exp = 8
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
channel = 3
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SAMM_TIM10':
table, table_objective = loading_samm_table(db_path, db_name,
objective_flag)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
samples = 159
n_exp = 8
VidPerSubject, vidList = get_vid_per_subject_augmented(db_images)
timesteps_TIM = 9
data_dim = r * w
channel = 1
#########################################################
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SAMM_Strain':
table, table_objective = loading_samm_table(db_path, db_name,
objective_flag)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
samples = 159
n_exp = 8
VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
channel = 3
#########################################################
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'SAMM_CASME_Optical':
# total amount of videos 253
table, table_objective = loading_samm_table(db_path, db_name)
table = table_objective
table2 = loading_casme_objective_table(db_path, db_name)
# merge samm and casme tables
table = np.concatenate((table, table2), axis=1)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
sub_items = np.empty([0])
list_samples = filter_objective_samples(table)
r = w = spatial_size
subjects = 47 # some subjects were removed because of objective classes and ignore samples: 47
n_exp = 5
samples = 253
VidPerSubject, list_samples = get_subfolders_num_crossdb(
db_images, IgnoredSamples_index, sub_items, table, list_samples)
timesteps_TIM = 9
data_dim = r * w
channel = 3
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
cross_db_flag = 1
return r, w, subjects, samples, n_exp, VidPerSubject, timesteps_TIM, data_dim, channel, table, list_samples, db_home, db_images, cross_db_flag
elif db_name == 'CASME2_Optical2_Aug' or db_name == 'CASME2_Optical_Aug' or db_name == 'CASME2_flowNet_Aug':
print("arrived")
table = loading_casme_table(db_home + 'CASME2_label_Ver_3.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
db_path, list_db)
r = w = spatial_size
subjects = 26
samples = 200 * 4
n_exp = 4
#VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
VidPerSubject, vidList = get_vid_per_subject_augmented(db_images)
timesteps_TIM = 9
data_dim = r * w
channel = 3
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'CASME2_Cropped':
print("arrived casme2 cropped")
table = loading_casme_table(db_home + 'CASME2_label_Ver_3.xls',
full=True)
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
db_path, list_db)
r = w = spatial_size
subjects = 26
samples = 255
n_exp = 4
VidPerSubject, vidList = get_vid_per_subject_augmented(db_images)
timesteps_TIM = 1
data_dim = r * w
channel = 2
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
elif db_name == 'CASME12_Color_TIM10' or db_name == 'CASME12_flowNet':
print("arrived")
table = loading_casme1_table(db_home + 'CASME12.xls')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
db_path, list_db)
r = w = spatial_size
subjects = 45
samples = 500 * 4
n_exp = 4
#VidPerSubject = get_subfolders_num(db_images, IgnoredSamples_index)
VidPerSubject, vidList = get_vid_per_subject_augmented(db_images)
timesteps_TIM = 9
data_dim = r * w
channel = 1
if os.path.exists(db_home + "Classification/" + db_name +
"_label.txt") == True:
os.remove(db_home + "Classification/" + db_name + "_label.txt")
return r, w, subjects, samples, n_exp, VidPerSubject, vidList, timesteps_TIM, data_dim, channel, table, listOfIgnoredSamples, db_home, db_images, cross_db_flag
def train_samm(batch_size, spatial_epochs, temporal_epochs, train_id, dB,
spatial_size, flag, tensorboard):
############## Path Preparation ######################
root_db_path = "/media/ice/OS/Datasets/"
workplace = root_db_path + dB + "/"
inputDir = root_db_path + dB + "/" + dB + "/"
######################################################
classes = 5
if dB == 'CASME2_TIM':
table = loading_casme_table(workplace + 'CASME2-ObjectiveClasses.xlsx')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
inputDir)
############## Variables ###################
r = w = spatial_size
subjects = 2
n_exp = 5
# VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
listOfIgnoredSamples = []
VidPerSubject = [2, 1]
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
############################################
os.remove(workplace + "Classification/CASME2_TIM_label.txt")
elif dB == 'CASME2_Optical':
table = loading_casme_table(workplace + 'CASME2-ObjectiveClasses.xlsx')
listOfIgnoredSamples, IgnoredSamples_index = ignore_casme_samples(
inputDir)
############## Variables ###################
r = w = spatial_size
subjects = 26
n_exp = 5
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 9
data_dim = r * w
pad_sequence = 9
channel = 3
############################################
# os.remove(workplace + "Classification/CASME2_TIM_label.txt")
elif dB == 'SAMM_TIM10':
table, table_objective = loading_samm_table(root_db_path, dB)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
################# Variables #############################
r = w = spatial_size
subjects = 29
n_exp = 8
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
channel = 3
classes = 8
#########################################################
elif dB == 'SAMM_CASME_Strain':
# total amount of videos 253
table, table_objective = loading_samm_table(root_db_path, dB)
table = table_objective
table2 = loading_casme_objective_table(root_db_path, dB)
# merge samm and casme tables
table = np.concatenate((table, table2), axis=1)
# print(table.shape)
# listOfIgnoredSamples, IgnoredSamples_index, sub_items = ignore_casme_samples(inputDir)
listOfIgnoredSamples = []
IgnoredSamples_index = np.empty([0])
sub_items = np.empty([0])
list_samples = filter_objective_samples(table)
r = w = spatial_size
subjects = 47 # some subjects were removed because of objective classes and ignore samples: 47
n_exp = 5
# TODO:
# 1) Further decrease the video amount, the one with objective classes >= 6
# list samples: samples with wanted objective class
VidPerSubject, list_samples = get_subfolders_num_crossdb(
inputDir, IgnoredSamples_index, sub_items, table, list_samples)
# print(VidPerSubject)
# print(len(VidPerSubject))
# print(sum(VidPerSubject))
timesteps_TIM = 9
data_dim = r * w
channel = 3
classes = 5
if os.path.isfile(workplace +
"Classification/SAMM_CASME_Optical_label.txt"):
os.remove(workplace +
"Classification/SAMM_CASME_Optical_label.txt")
##################### Variables ######################
######################################################
############## Flags ####################
tensorboard_flag = tensorboard
resizedFlag = 1
train_spatial_flag = 0
train_temporal_flag = 0
svm_flag = 0
finetuning_flag = 0
cam_visualizer_flag = 0
channel_flag = 0
if flag == 'st':
train_spatial_flag = 1
train_temporal_flag = 1
finetuning_flag = 1
elif flag == 's':
train_spatial_flag = 1
finetuning_flag = 1
elif flag == 't':
train_temporal_flag = 1
elif flag == 'nofine':
svm_flag = 1
elif flag == 'scratch':
train_spatial_flag = 1
train_temporal_flag = 1
elif flag == 'st4':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 1
elif flag == 'st7':
train_spatial_flag = 1
train_temporal_flag = 1
channel_flag = 2
#########################################
############ Reading Images and Labels ################
SubperdB = Read_Input_Images_SAMM_CASME(inputDir, list_samples,
listOfIgnoredSamples, dB,
resizedFlag, table, workplace,
spatial_size, channel)
print("Loaded Images into the tray...")
labelperSub = label_matching(workplace, dB, subjects, VidPerSubject)
print("Loaded Labels into the tray...")
if channel_flag == 1:
SubperdB_strain = Read_Input_Images(inputDir, listOfIgnoredSamples,
'CASME2_Strain_TIM10', resizedFlag,
table, workplace, spatial_size, 1)
elif channel_flag == 2:
SubperdB_strain = Read_Input_Images(inputDir, listOfIgnoredSamples,
'CASME2_Strain_TIM10', resizedFlag,
table, workplace, spatial_size, 1)
SubperdB_gray = Read_Input_Images(inputDir, listOfIgnoredSamples,
'CASME2_TIM', resizedFlag, table,
workplace, spatial_size, 3)
#######################################################
########### Model Configurations #######################
sgd = optimizers.SGD(lr=0.0001, decay=1e-7, momentum=0.9, nesterov=True)
adam = optimizers.Adam(lr=0.00001, decay=0.000001)
adam2 = optimizers.Adam(lr=0.00075, decay=0.0001)
# Different Conditions for Temporal Learning ONLY
if train_spatial_flag == 0 and train_temporal_flag == 1 and dB != 'CASME2_Optical':
data_dim = spatial_size * spatial_size
elif train_spatial_flag == 0 and train_temporal_flag == 1 and dB == 'CASME2_Optical':
data_dim = spatial_size * spatial_size * 3
else:
data_dim = 4096
########################################################
########### Training Process ############
# total confusion matrix to be used in the computation of f1 score
tot_mat = np.zeros((n_exp, n_exp))
# model checkpoint
spatial_weights_name = 'vgg_spatial_' + str(train_id) + '_casme2_'
temporal_weights_name = 'temporal_ID_' + str(train_id) + '_casme2_'
history = LossHistory()
stopping = EarlyStopping(monitor='loss', min_delta=0, mode='min')
for sub in range(subjects):
############### Reinitialization & weights reset of models ########################
vgg_model_cam = VGG_16(spatial_size=spatial_size,
classes=classes,
weights_path='VGG_Face_Deep_16.h5')
temporal_model = temporal_module(data_dim=data_dim,
classes=classes,
timesteps_TIM=timesteps_TIM)
temporal_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
conv_ae = convolutional_autoencoder(spatial_size=spatial_size,
classes=classes)
conv_ae.compile(loss='binary_crossentropy', optimizer=adam)
if channel_flag == 1 or channel_flag == 2:
vgg_model = VGG_16_4_channels(classes=classes,
spatial_size=spatial_size)
vgg_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
else:
vgg_model = VGG_16(spatial_size=spatial_size,
classes=classes,
weights_path='VGG_Face_Deep_16.h5')
vgg_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.categorical_accuracy])
svm_classifier = SVC(kernel='linear', C=1)
####################################################################################
############ for tensorboard ###############
if tensorboard_flag == 1:
cat_path = tensorboard_path + str(sub) + "/"
os.mkdir(cat_path)
tbCallBack = keras.callbacks.TensorBoard(log_dir=cat_path,
write_graph=True)
cat_path2 = tensorboard_path + str(sub) + "spat/"
os.mkdir(cat_path2)
tbCallBack2 = keras.callbacks.TensorBoard(log_dir=cat_path2,
write_graph=True)
#############################################
image_label_mapping = np.empty([0])
Train_X, Train_Y, Test_X, Test_Y, Test_Y_gt = data_loader_with_LOSO(
sub, SubperdB, labelperSub, subjects, classes)
# Rearrange Training labels into a vector of images, breaking sequence
Train_X_spatial = Train_X.reshape(Train_X.shape[0] * timesteps_TIM, r,
w, channel)
Test_X_spatial = Test_X.reshape(Test_X.shape[0] * timesteps_TIM, r, w,
channel)
# Special Loading for 4-Channel
if channel_flag == 1:
Train_X_Strain, _, Test_X_Strain, _, _ = data_loader_with_LOSO(
sub, SubperdB_strain, labelperSub, subjects, classes)
Train_X_Strain = Train_X_Strain.reshape(
Train_X_Strain.shape[0] * timesteps_TIM, r, w, 1)
Test_X_Strain = Test_X_Strain.reshape(
Test_X.shape[0] * timesteps_TIM, r, w, 1)
# Concatenate Train X & Train_X_Strain
Train_X_spatial = np.concatenate((Train_X_spatial, Train_X_Strain),
axis=3)
Test_X_spatial = np.concatenate((Test_X_spatial, Test_X_Strain),
axis=3)
channel = 4
elif channel_flag == 2:
Train_X_Strain, _, Test_X_Strain, _, _ = data_loader_with_LOSO(
sub, SubperdB_strain, labelperSub, subjects, classes)
Train_X_gray, _, Test_X_gray, _, _ = data_loader_with_LOSO(
sub, SubperdB_gray, labelperSub, subjects)
Train_X_Strain = Train_X_Strain.reshape(
Train_X_Strain.shape[0] * timesteps_TIM, r, w, 1)
Test_X_Strain = Test_X_Strain.reshape(
Test_X_Strain.shape[0] * timesteps_TIM, r, w, 1)
Train_X_gray = Train_X_gray.reshape(
Train_X_gray.shape[0] * timesteps_TIM, r, w, 3)
Test_X_gray = Test_X_gray.reshape(
Test_X_gray.shape[0] * timesteps_TIM, r, w, 3)
# Concatenate Train_X_Strain & Train_X & Train_X_gray
Train_X_spatial = np.concatenate(
(Train_X_spatial, Train_X_Strain, Train_X_gray), axis=3)
Test_X_spatial = np.concatenate(
(Test_X_spatial, Test_X_Strain, Test_X_gray), axis=3)
channel = 7
if channel == 1:
# Duplicate channel of input image
Train_X_spatial = duplicate_channel(Train_X_spatial)
Test_X_spatial = duplicate_channel(Test_X_spatial)
# Extend Y labels 10 fold, so that all images have labels
Train_Y_spatial = np.repeat(Train_Y, timesteps_TIM, axis=0)
Test_Y_spatial = np.repeat(Test_Y, timesteps_TIM, axis=0)
# print ("Train_X_shape: " + str(np.shape(Train_X_spatial)))
# print ("Train_Y_shape: " + str(np.shape(Train_Y_spatial)))
# print ("Test_X_shape: " + str(np.shape(Test_X_spatial)))
# print ("Test_Y_shape: " + str(np.shape(Test_Y_spatial)))
# print(Train_X_spatial)
##################### Training & Testing #########################
X = Train_X_spatial.reshape(Train_X_spatial.shape[0], channel, r, w)
y = Train_Y_spatial.reshape(Train_Y_spatial.shape[0], classes)
normalized_X = X.astype('float32') / 255.
test_X = Test_X_spatial.reshape(Test_X_spatial.shape[0], channel, r, w)
test_y = Test_Y_spatial.reshape(Test_Y_spatial.shape[0], classes)
normalized_test_X = test_X.astype('float32') / 255.
print(X.shape)
###### conv weights must be freezed for transfer learning ######
if finetuning_flag == 1:
for layer in vgg_model.layers[:33]:
layer.trainable = False
if train_spatial_flag == 1 and train_temporal_flag == 1:
# Autoencoder features
# conv_ae.fit(normalized_X, normalized_X, batch_size=batch_size, epochs=spatial_epochs, shuffle=True)
# Spatial Training
if tensorboard_flag == 1:
vgg_model.fit(X,
y,
batch_size=batch_size,
epochs=spatial_epochs,
shuffle=True,
callbacks=[tbCallBack2])
else:
vgg_model.fit(X,
y,
batch_size=batch_size,
epochs=spatial_epochs,
shuffle=True,
callbacks=[history, stopping])
# record f1 and loss
file_loss = open(
workplace + 'Classification/' + 'Result/' + dB + '/loss_' +
str(train_id) + '.txt', 'a')
file_loss.write(str(history.losses) + "\n")
file_loss.close()
file_loss = open(
workplace + 'Classification/' + 'Result/' + dB + '/accuracy_' +
str(train_id) + '.txt', 'a')
file_loss.write(str(history.accuracy) + "\n")
file_loss.close()
vgg_model.save_weights(spatial_weights_name + str(sub) + ".h5")
def train_smic(batch_size, spatial_epochs, temporal_epochs, train_id):
############################## Loading Labels & Images ##############################
# /media/ice/OS/Datasets/SMIC_TIM10/SMIC_TIM10
root_db_path = "/media/ice/OS/Datasets/"
dB = "SMIC_TIM10"
inputDir = root_db_path + dB + "/" + dB + "/"
workplace = root_db_path + dB + "/"
subject, filename, label, num_frames = loading_smic_labels(
root_db_path, dB)
filename = filename.as_matrix()
label = label.as_matrix()
table = np.transpose(np.array([filename, label]))
# os.remove(workplace + "Classification/SMIC_label.txt")
################# Variables #############################
spatial_size = 224
r = w = spatial_size
subjects = 16
samples = 164
n_exp = 3
IgnoredSamples_index = np.empty([0])
VidPerSubject = get_subfolders_num(inputDir, IgnoredSamples_index)
listOfIgnoredSamples = []
timesteps_TIM = 10
data_dim = r * w
pad_sequence = 10
#########################################################
############## Flags ####################
resizedFlag = 1
train_spatial_flag = 1
train_temporal_flag = 1
svm_flag = 0
finetuning_flag = 1
tensorboard_flag = 0
cam_visualizer_flag = 1
#########################################
############## Reading Images and Labels ################
# SubperdB = Read_SMIC_Images(inputDir, listOfIgnoredSamples, dB, resizedFlag, table, workplace, spatial_size)
SubperdB = Read_Input_Images(inputDir, listOfIgnoredSamples, dB,
resizedFlag, table, workplace, spatial_size)
labelperSub = label_matching(workplace, dB, subjects, VidPerSubject)
######################################################################################
########### Model #######################
sgd = optimizers.SGD(lr=0.0001, decay=1e-7, momentum=0.9, nesterov=True)
adam = optimizers.Adam(lr=0.00001)
if train_spatial_flag == 0 and train_temporal_flag == 1:
data_dim = spatial_size * spatial_size
else:
data_dim = 4096
temporal_model = temporal_module
#########################################
################# Pretrained Model ###################
vgg_model = VGG_16('VGG_Face_Deep_16.h5')
# keras_vgg = keras_vgg16(weights='imagenet')
# vgg_model = VGG_16('imagenet')
vgg_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.sparse_categorical_accuracy])
plot_model(vgg_model, to_file='model.png', show_shapes=True)
svm_classifier = SVC(kernel='linear', C=1)
######################################################
# model checkpoint
spatial_weights_name = 'vgg_spatial_17a_smic_'
temporal_weights_name = 'temporal_ID_16_smic_'
# model checkpoint
root = "/home/viprlab/Documents/Micro-Expression/" + spatial_weights_name + "weights.{epoch:02d}-{val_loss:.2f}.hdf5"
root_temporal = "/home/viprlab/Documents/Micro-Expression/" + temporal_weights_name + "weights.{epoch:02d}-{val_loss:.2f}.hdf5"
model_checkpoint = keras.callbacks.ModelCheckpoint(root,
monitor='loss',
save_best_only=True,
save_weights_only=True)
model_checkpoint_temporal = keras.callbacks.ModelCheckpoint(
root_temporal,
monitor='loss',
save_best_only=True,
save_weights_only=True)
########### Training Process ############
# Todo:
# 1) LOSO (done)
# 2) call model (done)
# 3) saving model architecture
# 4) Saving Checkpoint
# 5) make prediction (done)
if tensorboard_flag == 1:
tensorboard_path = "/home/ice/Documents/Micro-Expression/tensorboard/"
tot_mat = np.zeros((n_exp, n_exp))
spatial_weights_name = 'vgg_spatial_ID_under_dev_smic.h5'
temporal_weights_name = 'temporal_ID_under_dev_smic.h5'
for sub in range(subjects):
vgg_model = VGG_16('VGG_Face_Deep_16.h5')
vgg_model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=[metrics.sparse_categorical_accuracy])
############ for tensorboard ###############
if tensorboard_flag == 1:
cat_path = tensorboard_path + str(sub) + "/"
os.mkdir(cat_path)
tbCallBack = keras.callbacks.TensorBoard(log_dir=cat_path,
write_graph=True)
cat_path2 = tensorboard_path + str(sub) + "spat/"
os.mkdir(cat_path2)
tbCallBack2 = keras.callbacks.TensorBoard(log_dir=cat_path2,
write_graph=True)
#############################################
image_label_mapping = np.empty([0])
Train_X, Train_Y, Test_X, Test_Y, Test_Y_gt = data_loader_with_LOSO(
sub, SubperdB, labelperSub, subjects)
# Rearrange Training labels into a vector of images, breaking sequence
Train_X_spatial = Train_X.reshape(Train_X.shape[0] * 10, r, w, 1)
Test_X_spatial = Test_X.reshape(Test_X.shape[0] * 10, r, w, 1)
# Extend Y labels 10 fold, so that all images have labels
Train_Y_spatial = np.repeat(Train_Y, 10, axis=0)
Test_Y_spatial = np.repeat(Test_Y, 10, axis=0)
# Duplicate channel of input image
Train_X_spatial = duplicate_channel(Train_X_spatial)
Test_X_spatial = duplicate_channel(Test_X_spatial)
# print ("Train_X_shape: " + str(np.shape(Train_X_spatial)))
# print ("Train_Y_shape: " + str(np.shape(Train_Y_spatial)))
# print ("Test_X_shape: " + str(np.shape(Test_X_spatial)))
# print ("Test_Y_shape: " + str(np.shape(Test_Y_spatial)))
# print(Train_X_spatial)
##################### Training & Testing #########################
X = Train_X_spatial.reshape(Train_X_spatial.shape[0], 3, r, w)
y = Train_Y_spatial.reshape(Train_Y_spatial.shape[0], 5)
test_X = Test_X_spatial.reshape(Test_X_spatial.shape[0], 3, r, w)
test_y = Test_Y_spatial.reshape(Test_Y_spatial.shape[0], 5)
###### conv weights must be freezed for transfer learning ######
if finetuning_flag == 1:
for layer in vgg_model.layers[:33]:
layer.trainable = False
if train_spatial_flag == 1 and train_temporal_flag == 1:
# trains encoder until fc, train temporal
# Spatial Training
if tensorboard_flag == 1:
vgg_model.fit(X,
y,
batch_size=batch_size,
epochs=spatial_epochs,
shuffle=True,
callbacks=[tbCallBack2])
else:
vgg_model.fit(X,
y,
batch_size=batch_size,
epochs=spatial_epochs,
shuffle=True,
callbacks=[model_checkpoint])
# vgg_model.save_weights(spatial_weights_name)
model = Model(inputs=vgg_model.input,
outputs=vgg_model.layers[35].output)
plot_model(model,
to_file="spatial_module_FULL_TRAINING.png",
show_shapes=True)
# Spatial Encoding
output = model.predict(X, batch_size=batch_size)
features = output.reshape(int(output.shape[0] / 10), 10,
output.shape[1])
# Temporal Training
if tensorboard_flag == 1:
temporal_model.fit(features,
Train_Y,
batch_size=batch_size,
epochs=temporal_epochs,
callbacks=[tbCallBack])
else:
temporal_model.fit(features,
Train_Y,
batch_size=batch_size,
epochs=temporal_epochs,
callbacks=[model_checkpoint_temporal])
# temporal_model.save_weights(temporal_weights_name)
# Testing
output = model.predict(test_X, batch_size=batch_size)
features = output.reshape(int(output.shape[0] / 10), 10,
output.shape[1])
predict = temporal_model.predict_classes(features,
batch_size=batch_size)
elif train_spatial_flag == 1 and train_temporal_flag == 0 and cam_visualizer_flag == 0:
# trains spatial module ONLY, no escape
# Spatial Training
if tensorboard_flag == 1:
vgg_model.fit(X,
y,
batch_size=batch_size,
epochs=spatial_epochs,
shuffle=True,
callbacks=[tbCallBack2])
else:
vgg_model.fit(X,
y,
batch_size=batch_size,
epochs=spatial_epochs,
shuffle=True,
callbacks=[model_checkpoint])
# vgg_model.save_weights(spatial_weights_name)
plot_model(vgg_model,
to_file="spatial_module_ONLY.png",
show_shapes=True)
# Testing
predict = vgg_model.predict(test_X, batch_size=batch_size)
Test_Y_gt = np.repeat(Test_Y_gt, 10, axis=0)
elif train_spatial_flag == 0 and train_temporal_flag == 1:
# trains temporal module ONLY.
# Temporal Training
if tensorboard_flag == 1:
temporal_model.fit(Train_X,
Train_Y,
batch_size=batch_size,
epochs=spatial_epochs,
callbacks=[tbCallBack])
else:
temporal_model.fit(Train_X,
Train_Y,
batch_size=batch_size,
epochs=spatial_epochs,
callbacks=[model_checkpoint_temporal])
# temporal_model.save_weights(temporal_weights_name)
# Testing
predict = temporal_model.predict_classes(Test_X,
batch_size=batch_size)
elif svm_flag == 1 and finetuning_flag == 0:
# no finetuning
X = vgg_model.predict(X, batch_size=batch_size)
y_for_svm = np.argmax(y, axis=1)
svm_classifier.fit(X, y_for_svm)
test_X = vgg_model.predict(test_X, batch_size=batch_size)
predict = svm_classifier.predict(test_X)
Test_Y_gt = np.repeat(Test_Y_gt, 10, axis=0)
elif train_spatial_flag == 1 and train_temporal_flag == 0 and cam_visualizer_flag == 1:
# trains spatial module & CAM ONLY
# Spatial Training
if tensorboard_flag == 1:
vgg_model.fit(X,
y,
batch_size=batch_size,
epochs=spatial_epochs,
shuffle=True,
callbacks=[tbCallBack2])
else:
vgg_model.fit(X,
y,
batch_size=batch_size,
epochs=spatial_epochs,
shuffle=True,
callbacks=[model_checkpoint])
# vgg_model.save_weights(spatial_weights_name)
plot_model(vgg_model,
to_file="spatial_module_CAM_ONLY.png",
show_shapes=True)
# Testing
predict = vgg_model.predict(test_X, batch_size=batch_size)
Test_Y_gt = np.repeat(Test_Y_gt, 10, axis=0)
##############################################################
#################### Confusion Matrix Construction #############
print(predict)
print(Test_Y_gt)
ct = confusion_matrix(Test_Y_gt, predict)
# check the order of the CT
order = np.unique(np.concatenate((predict, Test_Y_gt)))
# create an array to hold the CT for each CV
mat = np.zeros((n_exp, n_exp))
# put the order accordingly, in order to form the overall ConfusionMat
for m in range(len(order)):
for n in range(len(order)):
mat[int(order[m]), int(order[n])] = ct[m, n]
tot_mat = mat + tot_mat
################################################################
#################### cumulative f1 plotting ######################
microAcc = np.trace(tot_mat) / np.sum(tot_mat)
[f1, precision, recall] = fpr(tot_mat, n_exp)
file = open(workplace + 'Classification/' + 'Result/' + '/f1.txt', 'a')
file.write(str(f1) + "\n")
file.close()
##################################################################
################# write each CT of each CV into .txt file #####################
record_scores(workplace, dB, ct, sub, order, tot_mat, n_exp, subjects)
###############################################################################
