count=n,
stride=(2, 2))(conv3) # "Stage 3 (spatial size: 8x8)"
net = tflearn.batch_normalization(conv4)
net = tflearn.activation(net, 'twins_relu')
net = tflearn.avg_pool_2d(net, 8)
#net = tflearn.avg_pool_2d(net, kernel_size=8, strides=1, padding='same')
net = tflearn.fully_connected(net, 10, activation='softmax')
return net
if __name__ == '__main__':
net = create_model()
mom = tflearn.Momentum(0.0001,
lr_decay=0.1,
decay_step=1000000,
staircase=True)
net = tflearn.regression(net,
optimizer=mom,
loss='categorical_crossentropy')
model = tflearn.DNN(net, tensorboard_verbose=1)
model.load(
"/home/lfwin/my_tmp/tflearn_logs/cifar10_WRN2_tReLU_130/model.tfl")
model.fit(X,
Y,
n_epoch=10,
shuffle=True,
validation_set=(testX, testY),
show_metric=True,
img_prep.add_custom_preprocessing(my_func)
# Building Residual Network
net = tflearn.input_data(shape=[None, 32, 32, 3], data_preprocessing=img_prep)
net = tflearn.conv_2d(net, 16, 3, regularizer='L2', weight_decay=0.0001)
net = tflearn.resnext_block(net, n, 16, 32)
net = tflearn.resnext_block(net, 1, 32, 32, downsample=True)
net = tflearn.resnext_block(net, n - 1, 32, 32)
net = tflearn.resnext_block(net, 1, 64, 32, downsample=True)
net = tflearn.resnext_block(net, n - 1, 64, 32)
net = tflearn.batch_normalization(net)
net = tflearn.activation(net, 'relu')
net = tflearn.global_avg_pool(net)
# Regression
loss = tflearn.fully_connected(net, 3, activation='softmax')
opt = tflearn.Momentum(0.1, lr_decay=0.1, decay_step=32000, staircase=True)
network = tflearn.regression(loss,
optimizer=opt,
learning_rate=0.01,
loss='categorical_crossentropy')
# Training
model = tflearn.DNN(
network,
checkpoint_path='C:/Users/Administrator/Desktop/Resnet_test/model_resnext',
max_checkpoints=1,
tensorboard_verbose=3,
tensorboard_dir='C:/Users/Administrator/Desktop/Resnet_test',
best_checkpoint_path=
'C:/Users/Administrator/Desktop/Resnet_test/model_resnet')
model.fit(X,
img_aug = tflearn.ImageAugmentation()
img_aug.add_random_flip_leftright()
img_aug.add_random_crop([48, 48], padding=4)
# Building Residual Network
net = tflearn.input_data(shape=[None, 48, 48, 1],
data_preprocessing=img_prep,
data_augmentation=img_aug)
net = tflearn.conv_2d(net, 16, 3, regularizer='L2', weight_decay=0.0001)
net = tflearn.resnext_block(net, n, 16, 32)
net = tflearn.resnext_block(net, 1, 32, 32, downsample=True)
net = tflearn.resnext_block(net, n-1, 32, 32)
net = tflearn.resnext_block(net, 1, 64, 32, downsample=True)
net = tflearn.resnext_block(net, n-1, 64, 32)
net = tflearn.batch_normalization(net)
net = tflearn.activation(net, 'relu')
net = tflearn.global_avg_pool(net)
# Regression
net = tflearn.fully_connected(net, 7, activation='softmax')
opt = tflearn.Momentum(learning_rate=0.1, lr_decay=0.1, decay_step=32000, staircase=True , momentum=0.9)
net = tflearn.regression(net, optimizer=opt,
loss='categorical_crossentropy')
# Training
model = tflearn.DNN(net, checkpoint_path='models/model_resnet_emotion',
max_checkpoints=20, tensorboard_verbose=0,
clip_gradients=0.)
model.fit(X, Y, n_epoch=150, validation_set=(X_test, Y_test),
snapshot_epoch=False, snapshot_step=500,
show_metric=True, batch_size=128, shuffle=True,
run_id='resnext_emotion')
net = tflearn.residual_block(net,1,256,downsample=True)
# [12,12,256]
net = tflearn.residual_block(net,5,256)
# [12,12,256]
net = tflearn.residual_block(net,1,512,downsample=True)
# [6,6,512]
net = tflearn.residual_block(net,2,512)
# [6,6,512]
net = tflearn.global_avg_pool(net)
# [512]
fully_connect = tflearn.fully_connected(net,HIDDEN,activation='relu')
# [1000]
net = tflearn.fully_connected(fully_connect,LEN_OUT,activation='softmax')
# [7211]
#mom = tflearn.SGD(0.1,lr_decay=0.1,decay_step=3086 * 20)
mom = tflearn.Momentum(0.01,lr_decay=0.1,decay_step=int(395000 / BATCH_SIZE) * 10,staircase=True)
reg = tflearn.regression(net,optimizer=mom,loss='categorical_crossentropy')
model = tflearn.DNN(reg,checkpoint_path='models/{}'.format(MODEL_FILE),max_checkpoints=100,session=sess)
elif NET_TYPE == 'resnet50' or NET_TYPE == 'resnet101' or NET_TYPE == 'resnet152':
with tf.device("/gpu:{}".format(GPU_CORE)):
net = tflearn.input_data(shape=[None,IMG_SIZE,IMG_SIZE,3])
net = tflearn.conv_2d(net,64,7,2, regularizer='L2', weight_decay=0.0001)
# [64,64,32]
net = tflearn.max_pool_2d(net,3,2)
# [32,32,32]
net = tflearn.residual_bottleneck(net,3,64,256)
# [32,32,64]
net = tflearn.residual_bottleneck(net,1,128,512,downsample=True)
# [16,16,128]
net = tflearn.residual_bottleneck(net,3,128,512)
# [16,16,128]
def return2img():
n = 5
# Real-time data preprocessing
img_prep = tflearn.ImagePreprocessing()
img_prep.add_featurewise_zero_center(per_channel=True)
# Real-time data augmentation
img_aug = tflearn.ImageAugmentation()
img_aug.add_random_flip_leftright()
img_aug.add_random_crop([32, 32], padding=4)
inp = tflearn.input_data(shape=[None, 32, 32, 1],
data_preprocessing=img_prep,
data_augmentation=img_aug,
name='input')
conv1_1 = tflearn.conv_2d(inp, 64, 3, activation='relu', name="conv1_1")
conv1_2 = tflearn.conv_2d(conv1_1,
64,
3,
activation='relu',
name="conv1_2")
pool1 = tflearn.max_pool_2d(conv1_2, 2, strides=2)
conv2_1 = tflearn.conv_2d(pool1, 128, 3, activation='relu', name="conv2_1")
conv2_2 = tflearn.conv_2d(conv2_1,
128,
3,
activation='relu',
name="conv2_2")
pool2 = tflearn.max_pool_2d(conv2_2, 2, strides=2)
conv3_1 = tflearn.conv_2d(pool2, 256, 3, activation='relu', name="conv3_1")
conv3_2 = tflearn.conv_2d(conv3_1,
256,
3,
activation='relu',
name="conv3_2")
conv3_3 = tflearn.conv_2d(conv3_2,
256,
3,
activation='relu',
name="conv3_3")
pool3 = tflearn.max_pool_2d(conv3_3, 2, strides=2)
conv4_1 = tflearn.conv_2d(pool3, 512, 3, activation='relu', name="conv4_1")
conv4_2 = tflearn.conv_2d(conv4_1,
512,
3,
activation='relu',
name="conv4_2")
conv4_3 = tflearn.conv_2d(conv4_2,
512,
3,
activation='relu',
name="conv4_3")
pool4 = tflearn.max_pool_2d(conv4_3, 2, strides=2)
conv5_1 = tflearn.conv_2d(pool4, 512, 3, activation='relu', name="conv5_1")
conv5_2 = tflearn.conv_2d(conv5_1,
512,
3,
activation='relu',
name="conv5_2")
conv5_3 = tflearn.conv_2d(conv5_2,
512,
3,
activation='relu',
name="conv5_3")
pool5 = tflearn.max_pool_2d(conv5_3, 2, strides=2)
fc6 = tflearn.fully_connected(pool5, 4096, activation='relu', name="fc6")
fc6_dropout = tflearn.dropout(fc6, 0.5)
fc7 = tflearn.fully_connected(fc6_dropout,
4096,
activation='relu',
name="fc7")
fc7_droptout = tflearn.dropout(fc7, 0.5)
fc8 = tflearn.fully_connected(fc7_droptout,
3,
activation='softmax',
name="fc8")
mm = tflearn.Momentum(learning_rate=0.01,
momentum=0.9,
lr_decay=0.1,
decay_step=1000)
network = tflearn.regression(fc8,
optimizer=mm,
loss='categorical_crossentropy',
restore=False)
# Training
model = tflearn.DNN(network)
model.load("model_resnet_+-------cifar10-14000")
img = load_img_label('/home/bai/最新数据/验证数据/经过预处理的原始图像')
livermask = load_mask('/home/bai/最新数据/验证数据/mask')
abc = []
np.array(abc)
for i in range(img.shape[0]):
a = np.zeros((512, 512, 3, 3),
dtype=np.float32) #最后一项表示第几个分割方式，倒数第二项表示预测的三个结果
for j in range(0, 3):
this_mask = livermask[i, :, :, j]
this_mask_num = int(np.max(this_mask))
for k in range(1, this_mask_num + 1):
if os.path.exists('/home/bai/最新数据/验证数据/超像素块/' + str(i) + '_' +
str(j) + '_' + str(k) + '.jpg'):
this_img = io.imread('/home/bai/最新数据/验证数据/超像素块/' + str(i) +
'_' + str(j) + '_' + str(k) + '.jpg')
this_img = np.reshape(this_img, (1, 32, 32, 1))
this_img = this_img.astype(np.float32)
result = model.predict(this_img)
# print(result)
abc.append(result)
thisPartLoc = np.where(this_mask == k)
for num in range(len(thisPartLoc[1])):
a[thisPartLoc[0][num], thisPartLoc[1][num], 0,
j] = result[0, 0]
a[thisPartLoc[0][num], thisPartLoc[1][num], 1,
j] = result[0, 1]
a[thisPartLoc[0][num], thisPartLoc[1][num], 2,
j] = result[0, 2]
b = np.max(a, axis=3)
final = np.argmax(b, axis=2)
misc.imsave(
'/home/bai/PycharmProjects/DeepRESIDUALNETWORKS/result/' + str(i) +
'.jpg', final)
def create(message, keep_existing):
print('creating network...')
number_of_hidden_layers = message['numberOfHiddenLayers']
activation_function = message['activationFunction'].encode(
'ascii', 'ignore')
input_length = int(message['inputLength'])
output_length = int(message['outputLength'])
network_file = message['networkFilePath']
classify = message['classify']
learning_rate = message['learningRate']
bias = message['bias']
use_dropout = message['useDropout']
dropout = message['dropout']
has_custom_shape = message['customShape']
shape = message['shape']
tf.reset_default_graph()
input_layer = tflearn.input_data(shape=[None, input_length],
name='input')
layer_to_connect_to = input_layer
layer_activation = None
if activation_function != 'none':
layer_activation = activation_function
for x in range(0, int(number_of_hidden_layers)):
layer_to_connect_to = tflearn.fully_connected(
layer_to_connect_to,
input_length if not has_custom_shape else shape[x],
weights_init='uniform',
name='dense' + str(x),
activation=layer_activation,
bias=bias,
regularizer='L2')
if use_dropout:
layer_to_connect_to = tflearn.dropout(layer_to_connect_to,
dropout)
finished_network = tflearn.fully_connected(
layer_to_connect_to,
output_length,
regularizer='L2',
activation='softmax' if classify else 'linear',
name='output',
bias=bias)
optimizer = 'adam'
if 'SGD' == message['optimizer']:
optimizer = tflearn.SGD(learning_rate=learning_rate)
elif 'RMSprop' == message['optimizer']:
optimizer = tflearn.RMSProp(learning_rate=learning_rate)
elif 'Momentum' == message['optimizer']:
optimizer = tflearn.Momentum(learning_rate=learning_rate)
network_graph = tflearn.regression(
finished_network,
optimizer=optimizer,
learning_rate=learning_rate,
loss='categorical_crossentropy' if classify else 'mean_square',
metric='default' if classify else 'R2',
)
model = tflearn.DNN(network_graph)
print('network created.')
if not keep_existing:
save_network(model, network_file)
return model
# Building deep neural network
input_layer = tflearn.input_data(shape=[None, 11])
dense1 = tflearn.fully_connected(input_layer, 32, activation=activation_func,
regularizer='L2')
dense2 = tflearn.fully_connected(dense1, 32, activation=activation_func,
regularizer='L2')
dense3 = tflearn.fully_connected(dense2, 32, activation=activation_func,
regularizer='L2')
dense4 = tflearn.fully_connected(dense3, 32, activation=activation_func,
regularizer='L2')
dense5 = tflearn.fully_connected(dense4, 32, activation=activation_func,
regularizer='L2')
output_layer = tflearn.fully_connected(dense5, 1, activation='linear')
# Regression using SGD + momentum with learning rate decay
sgd = tflearn.Momentum(learning_rate=0.001, lr_decay=0.96, decay_step=200)
net = tflearn.regression(output_layer, optimizer=sgd, loss='mean_square', metric='R2')
# Training
model = tflearn.DNN(net, tensorboard_verbose=0)
model.fit(X, Y, n_epoch=1000, validation_set=0.2,
show_metric=True, run_id="dense_model")
predictY = model.predict(X)
# Plot the results
steps_in_future = 1
plt.figure()
plt.title('Prediction of Max Temperature (degrees Celsius)')
tflearn.config.init_training_mode()
input_layer = tflearn.input_data([None, IMG_SIZE, IMG_SIZE, 3])
net = tflearn.conv_2d(input_layer, 64, 7, strides=2)
net = tflearn.max_pool_2d(net, 3, strides=2)
net = tflearn.residual_block(net, 3, 64)
net = tflearn.residual_block(net, 1, 128, downsample=True)
net = tflearn.residual_block(net, 3, 128)
net = tflearn.residual_block(net, 1, 256, downsample=True)
net = tflearn.residual_block(net, 5, 256)
net = tflearn.residual_block(net, 1, 512, downsample=True)
net = tflearn.residual_block(net, 2, 512)
net = tflearn.global_avg_pool(net)
fully_connected = tflearn.fully_connected(net, HIDDEN, activation="relu")
result = tflearn.fully_connected(fully_connected, 7211, activation='softmax')
mom = tflearn.Momentum(0.01,
lr_decay=0.1,
decay_step=int(395000 / BATCH_SIZE) * 300098,
staircase=True)
net = tflearn.regression(result,
optimizer=mom,
loss="categorical_crossentropy")
model = tflearn.DNN(net,
checkpoint_path='models/model',
session=sess,
max_checkpoints=100,
tensorboard_verbose=0)
# configure for tf session
config_box = None
# session for tensorflow boxing model
sess_box = None
# mtcnn tool for face detection
nb_blocks=1,
out_channels=32,
cardinality=32,
downsample=True) # 残差虚线
network = tflearn.resnext_block(network,
nb_blocks=n,
out_channels=64,
cardinality=32)
network = tflearn.resnext_block(network,
nb_blocks=1,
out_channels=32,
cardinality=32,
downsample=True) # 残差虚线
network = tflearn.batch_normalization(network) # 标准化
network = tflearn.activation(network, 'relu') # 镶嵌激活函数
network = tflearn.global_avg_pool(network) # 镶嵌平均池化
# 全连接（输出）
network = tflearn.fully_connected(network, 10, activation='softmax')
# 设置优化器的参数
# 动态学习率，初始值为0.1,每decay_step时乘以lr_decay
opt = tflearn.Momentum(learning_rate=0.1, lr_decay=0.1, decay_step=32000)
network = tflearn.regression(network, optimizer=opt)
# 训练
model = tflearn.DNN(network,
checkpoint_path='./model/resnet/model_resnet',
tensorboard_dir='./logs')
model.fit(X, Y, n_epoch=200, validation_set=(X_test, Y_test), batch_size=128)
# Building Residual Network
net = tl.input_data(shape=[None, 42, 42, 3])
net = tl.conv_2d(net, 32, 3)
net = tl.batch_normalization(net)
net = tl.activation(net, 'relu')
net = tl.shallow_residual_block(net, 4, 32, regularizer='L2')
net = tl.shallow_residual_block(net,
1,
32,
downsample=True,
regularizer='L2')
net = tl.shallow_residual_block(net, 4, 64, regularizer='L2')
net = tl.shallow_residual_block(net,
1,
64,
downsample=True,
regularizer='L2')
net = tl.shallow_residual_block(net, 5, 64, regularizer='L2')
net = tl.global_avg_pool(net)
# Regression
net = tl.fully_connected(net, 9, activation='softmax')
mom = tl.Momentum(0.1, lr_decay=0.1, decay_step=16000, staircase=True)
net = tl.regression(net, optimizer=mom, loss='categorical_crossentropy')
# Training
model = tl.DNN(net)
model.load('resnet_analysis-172500')
pred_y = model.predict(X)
print(pred_y)
CWD_PATH = os.getcwd()
from tflearn.data_utils import build_hdf5_image_dataset
network = tflearn.input_data(shape=[None, 128, 128, 3])
network = tflearn.batch_normalization(network)
network = conv_2d(network, 32, 3, activation='linear')
network = max_pool_2d(network, 2)
network = conv_2d(network, 64, 3, activation='linear')
network = conv_2d(network, 64, 3, activation='linear')
network = max_pool_2d(network, 2)
network = fully_connected(network, 512, activation='linear')
network = dropout(network, 0.7)
network = tflearn.batch_normalization(network)
network = fully_connected(network, 9, activation='sigmoid')
optimizer = tflearn.Momentum(0.005)
network = regression(network, optimizer=optimizer, loss='binary_crossentropy')
model = tflearn.DNN(network, tensorboard_verbose=0, clip_gradients=0.)
model.load('./single_prediction_3.tfl')
def encode(data):
class MyEncoder(json.JSONEncoder):
def default(self, obj):
if isinstance(obj, np.integer):
return int(obj)
elif isinstance(obj, np.floating):
return float(obj)
elif isinstance(obj, np.ndarray):
def train_nn_tflearn(data_handler, num_epochs=50):
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.5)
batch_size = data_handler.mini_batch_size
img_prep = tflearn.ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()
img_aug = tflearn.ImageAugmentation()
img_aug.add_random_flip_leftright()
img_aug.add_random_rotation(max_angle=25)
img_aug.add_random_crop([32, 32], padding=4)
x = tflearn.input_data(shape=[None, 32, 32, 3],
dtype='float',
data_preprocessing=img_prep,
data_augmentation=img_aug)
# x = tf.placeholder('float', [None, 32, 32, 3])
#y = tf.placeholder('float', [None, 10])
# test_data, test_labels = data_handler.get_test_data()
# test_data = test_data.reshape([-1,32,32,3])
ntrain = data_handler.train_size
ntest = data_handler.meta['num_cases_per_batch']
# from tflearn.datasets import cifar10
# (X, Y), (X_test, Y_test) = cifar10.load_data(dirname="/home/hamza/meh/bk_fedora24/Documents/tflearn_example/cifar-10-batches-py")
# X, Y = tflearn.data_utils.shuffle(X, Y)
# Y = tflearn.data_utils.to_categorical(Y, 10)
# Y_test = tflearn.data_utils.to_categorical(Y_test, 10)
X, Y = data_handler.get_all_train_data()
X, Y = tflearn.data_utils.shuffle(X, Y)
X = np.dstack((X[:, :1024], X[:, 1024:2048], X[:, 2048:]))
X = X / 255.0
X = X.reshape([-1, 32, 32, 3])
Y = tflearn.data_utils.to_categorical(Y, 10)
X_test, Y_test = data_handler.get_test_data()
X_test = np.dstack((X_test[:, :1024], X_test[:, 1024:2048], X_test[:,
2048:]))
X_test = X_test / 255.0
X_test = X_test.reshape([-1, 32, 32, 3])
#network = tflearn.regression(net3(x),optimizer='adam',loss='categorical_crossentropy',learning_rate=0.001)
mom = tflearn.Momentum(0.1, lr_decay=0.1, decay_step=32000, staircase=True)
network = tflearn.regression(resnet1(x),
optimizer=mom,
loss='categorical_crossentropy')
print np.shape(X)
print np.shape(Y)
print network
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(X,
Y,
n_epoch=50,
shuffle=True,
validation_set=(X_test, Y_test),
show_metric=True,
batch_size=data_handler.mini_batch_size,
run_id='cifar10_cnn')
net = rb(net, 1, 32, downsample=True,activation=args['act'])
net = rb(net, n-1, 32,activation=args['act'])
net = rb(net, 1, 64, downsample=True,activation=args['act'])
net = rb(net, n-1, 64,activation=args['act'])
if block_mode.upper() == 'FULLPREACT' or block_mode.upper() == 'SUB':
net = tflearn.batch_normalization(net)
if block_mode.upper() == 'PREACT' or block_mode.upper() == 'FULLPREACT' or block_mode.upper() == 'SUB':
net = tflearn.activation(net,activation=args['act'])
net = tflearn.global_avg_pool(net)
# Regression
net = tflearn.fully_connected(net, 10, activation='softmax')
mom = tflearn.Momentum(args['lr'], lr_decay=0.1, decay_step=32000, staircase=True)
net = tflearn.regression(net, optimizer=mom,
loss='categorical_crossentropy')
# Training
model = tflearn.DNN(net, checkpoint_path='./checkpoints/', tensorboard_verbose=0,tensorboard_dir='log',
clip_gradients=0.)
model_name = str(layers)+'_layers'+'_'+block_mode.upper()
if suffix2 != None:
model_name = model_name +'_'+suffix2[0]
model.fit(X, Y, n_epoch=EPOCHS, validation_set=(testX, testY),
snapshot_epoch=True,
show_metric=True, batch_size=128, shuffle=True,
def run(self):
# Real-time pre-processing of the image data
img_prep = ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()
# Real-time data augmentation
img_aug = tflearn.ImageAugmentation()
img_aug.add_random_flip_leftright()
# img_aug.add_random_crop([48, 48], padding=8)
# Building Residual Network
net = tflearn.input_data(shape=[None, 48, 48, 1],
data_preprocessing=img_prep,
data_augmentation=img_aug)
net = tflearn.conv_2d(net,
nb_filter=16,
filter_size=3,
regularizer='L2',
weight_decay=0.0001)
net = tflearn.residual_block(net, self.n, 16)
net = tflearn.residual_block(net, 1, 32, downsample=True)
net = tflearn.residual_block(net, self.n - 1, 32)
net = tflearn.residual_block(net, 1, 64, downsample=True)
net = tflearn.residual_block(net, self.n - 1, 64)
net = tflearn.batch_normalization(net)
net = tflearn.activation(net, 'relu')
net = tflearn.global_avg_pool(net)
# Regression
net = tflearn.fully_connected(net, 7, activation='softmax')
mom = tflearn.Momentum(learning_rate=0.1,
lr_decay=0.0001,
decay_step=32000,
staircase=True,
momentum=0.9)
net = tflearn.regression(net,
optimizer=mom,
loss='categorical_crossentropy')
self.model = tflearn.DNN(net,
checkpoint_path='models/model_resnet_emotion',
max_checkpoints=10,
tensorboard_verbose=0,
clip_gradients=0.)
self.model.load('current_model/model_resnet_emotion-42000')
face_cascade = cv2.CascadeClassifier(
'haarcascade_frontalface_default.xml')
cap = cv2.VideoCapture(0)
while True:
ret, img = cap.read()
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, 1.3, 5)
for (x, y, w, h) in faces:
cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 2)
roi_gray = gray[y:y + h, x:x + w]
roi_color = img[y:y + h, x:x + w]
self.process_image(roi_gray, img)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
cap.release()
cv2.destroyAllWindows()
def __init__(self, learning_rate, decay_rate, momentum, settings): tf.reset_default_graph() self.learning_rate = learning_rate self.decay_rate = decay_rate self.momentum = momentum if settings[0] == "1": self.preprocessing = True else: self.preprocessing = False if settings[1] == "1": self.doRegularizer = True else: self.doRegularizer = False if settings[2] == "1": self.dropout = True else: self.dropout = False if settings[3] == "1": self.activation = "sigmoid" else: self.activation = "relu6" if settings[4] == "1": self.loss = "mean_square" else: self.loss = "softmax_categorical_crossentropy" if settings[5] == "1": self.optimization = "adaDelta" else: self.optimization = "momentum" # Reading in data self.targets = tf.placeholder(tf.float32, [None, self.output_nodes]) input_layer = None if self.preprocessing: data_processing = tflearn.data_preprocessing.DataPreprocessing() data_processing.add_featurewise_stdnorm(std=0.2090549348314474) data_processing.add_featurewise_zero_center(mean=0.3040620249839109) input_layer = tflearn.input_data(shape=[None, self.input_nodes], data_preprocessing=data_processing) else: input_layer = tflearn.input_data(shape=[None, self.input_nodes]) h1 = None h2 = None h3 = None h4 = None output_layer = None if self.doRegularizer: # 1st Hidden Layer h1 = tflearn.fully_connected(input_layer, self.h1_nodes, activation=self.activation, regularizer=self.regularization, weight_decay=1/(8 * self.batch_size * self.input_nodes*self.h1_nodes)) if self.dropout: h1 = tflearn.dropout(h1, self.dropout_value) # 2nd Hidden Layer h2 = tflearn.fully_connected(h1, self.h2_nodes, activation=self.activation, regularizer=self.regularization, weight_decay=1/(8 * self.batch_size * self.h1_nodes*self.h2_nodes)) if self.dropout: h2 = tflearn.dropout(h2, self.dropout_value) ''' h3 = tflearn.fully_connected(h2, self.h3_nodes, activation=self.activation, regularizer=self.regularization, weight_decay=1/(8 * self.batch_size * self.h2_nodes*self.h3_nodes)) if self.dropout: h3 = tflearn.dropout(h3, self.dropout_value) # 2nd Hidden Layer h4 = tflearn.fully_connected(h3, self.h4_nodes, activation=self.activation, regularizer=self.regularization, weight_decay=1/(8 * self.batch_size * self.h3_nodes*self.h4_nodes)) if self.dropout: h4 = tflearn.dropout(h4, self.dropout_value) ''' # Output Layer output_layer = tflearn.fully_connected(h2, self.output_nodes, activation=self.activation, regularizer=self.regularization, weight_decay=1/(8 * self.batch_size * self.h2_nodes*self.output_nodes)) else: # 1st Hidden Layer h1 = tflearn.fully_connected(input_layer, self.h1_nodes, activation=self.activation) if self.dropout: h1 = tflearn.dropout(h1, self.dropout_value) # 2nd Hidden Layer h2 = tflearn.fully_connected(h1, self.h2_nodes, activation=self.activation) ''' if self.dropout: h2 = tflearn.dropout(h2, self.dropout_value) h3 = tflearn.fully_connected(h2, self.h3_nodes, activation=self.activation) if self.dropout: h3 = tflearn.dropout(h3, self.dropout_value) # 2nd Hidden Layer h4 = tflearn.fully_connected(h3, self.h4_nodes, activation=self.activation) if self.dropout: h4 = tflearn.dropout(h4, self.dropout_value) ''' # Output Layer output_layer = tflearn.fully_connected(h2, self.output_nodes, activation=self.activation) # Learning # Optimizer opt = None if self.optimization == "momentum": opt = tflearn.Momentum(learning_rate=self.learning_rate, lr_decay=self.decay_rate, momentum=self.momentum) else: opt = tflearn.AdaDelta(learning_rate=self.learning_rate, rho=self.decay_rate) r = tflearn.regression(output_layer, placeholder=self.targets, optimizer=opt, loss=self.loss) self.model = tflearn.DNN(r)
