def __init__(self, weights=None):
super(Backbone, self).__init__()
model = resnet.resnet_18()
model.build(input_shape=(None, 256, 256, 3))
if weights is not None:
model.load_weights(weights)
self.conv1 = model.conv1
self.bn1 = model.bn1
self.pool1 = model.pool1
self.layer1 = model.layer1
self.layer2 = model.layer2
self.layer3 = model.layer3
self.layer4 = model.layer4
if False:
self.conv1.trainable = False
self.bn1.trainable = False
self.pool1.trainable = False
for layer in self.layer1.layers:
layer.trainable = False
for layer in self.layer2.layers:
layer.trainable = False
for layer in self.layer3.layers:
layer.trainable = False
for layer in self.layer4.layers:
layer.trainable = False
def get_model():
model = resnet_50()
if config.model == "resnet18":
model = resnet_18()
if config.model == "resnet34":
model = resnet_34()
if config.model == "resnet101":
model = resnet_101()
if config.model == "resnet152":
model = resnet_152()
model.build(input_shape=(None, config.image_height, config.image_width,
config.channels))
model.summary()
return model
def get_model():
if model_index == 0:
return mobilenet_v1.MobileNetV1()
elif model_index == 1:
return mobilenet_v2.MobileNetV2()
elif model_index == 2:
return mobilenet_v3_large.MobileNetV3Large()
elif model_index == 3:
return mobilenet_v3_small.MobileNetV3Small()
elif model_index == 4:
return efficientnet.efficient_net_b0()
elif model_index == 5:
return efficientnet.efficient_net_b1()
elif model_index == 6:
return efficientnet.efficient_net_b2()
elif model_index == 7:
return efficientnet.efficient_net_b3()
elif model_index == 8:
return efficientnet.efficient_net_b4()
elif model_index == 9:
return efficientnet.efficient_net_b5()
elif model_index == 10:
return efficientnet.efficient_net_b6()
elif model_index == 11:
return efficientnet.efficient_net_b7()
elif model_index == 12:
return resnext.ResNeXt50()
elif model_index == 13:
return resnext.ResNeXt101()
elif model_index == 14:
return inception_v4.InceptionV4()
elif model_index == 15:
return inception_resnet_v1.InceptionResNetV1()
elif model_index == 16:
return inception_resnet_v2.InceptionResNetV2()
elif model_index == 17:
return se_resnet.se_resnet_50()
elif model_index == 18:
return se_resnet.se_resnet_101()
elif model_index == 19:
return se_resnet.se_resnet_152()
elif model_index == 20:
return squeezenet.SqueezeNet()
elif model_index == 21:
return densenet.densenet_121()
elif model_index == 22:
return densenet.densenet_169()
elif model_index == 23:
return densenet.densenet_201()
elif model_index == 24:
return densenet.densenet_264()
elif model_index == 25:
return shufflenet_v2.shufflenet_0_5x()
elif model_index == 26:
return shufflenet_v2.shufflenet_1_0x()
elif model_index == 27:
return shufflenet_v2.shufflenet_1_5x()
elif model_index == 28:
return shufflenet_v2.shufflenet_2_0x()
elif model_index == 29:
return resnet.resnet_18()
elif model_index == 30:
return resnet.resnet_34()
elif model_index == 31:
return resnet.resnet_50()
elif model_index == 32:
return resnet.resnet_101()
elif model_index == 33:
return resnet.resnet_152()
elif model_index == 34:
return se_resnext.SEResNeXt50()
elif model_index == 35:
return se_resnext.SEResNeXt101()
else:
raise ValueError("The model_index does not exist.")
wavegan = wrap_frozen_graph(graph_def, inputs='z:0', outputs='G_z:0')
string_to_label = {
"Zero": 0,
"One": 1,
"Two": 2,
"Three": 3,
"Four": 4,
"Five": 5,
"Six": 6,
"Seven": 7,
"Eight": 8,
"Nine": 9
}
inverse_mapping_model = resnet.resnet_18()
inverse_mapping_model.load_weights(
"inverse_mapping_model_checkpoint/generated_and_real_training.ckpt")
classifier = resnet.resnet_18(num_classes=10,
activation=tf.keras.activations.softmax)
classifier.load_weights("classifier_checkpoint/model.ckpt")
for l in classifier.layers:
l.trainable = False
classifier_middle_layer = tf.keras.Sequential(classifier.layers[0:5])
for l in inverse_mapping_model.layers:
l.trainable = False
string_to_label = {
"Zero": 0,
"One": 1,
"Two": 2,
"Three": 3,
"Four": 4,
"Five": 5,
"Six": 6,
"Seven": 7,
"Eight": 8,
"Nine": 9
}
#inverse_mapping_model = tf.keras.applications.ResNet50(classes=100, weights=None, input_shape=(256, 256, 3))
classifier = resnet.resnet_18(num_classes=10,
activation=tf.keras.activations.softmax)
#inverse_mapping_model.load_weights("inverse_mapping_model_checkpoint/inverse_mapping_weights_1d_mse.ckpt")
loss_object = tf.keras.losses.CategoricalCrossentropy()
optimizer = tf.keras.optimizers.Adam()
train_loss = tf.keras.metrics.Mean(name='train_loss')
train_acc = tf.keras.metrics.Accuracy(name='train_acc')
test_loss = tf.keras.metrics.Mean(name='test_loss')
test_acc = tf.keras.metrics.Accuracy(name='test_acc')
@tf.function
def train_step(audio, labels):
audio = tf.signal.stft(audio, 256, 128, pad_end=True)
INPUT_SHAPE = (256, 256, 3)
EPOCHS = args.epochs
BATCH_SIZE = args.batch_size
PREFETCH_SIZE = args.prefetch_size
NUM_CLASSES = 1000
NUM_IMAGES = args.max_images
VAL_SPLIT = 0.3
LOG_DIR = './logs_resnet'
CHECKPOINT_DIR = './checkpoints_resnet'
model = resnet_18()
model.build(input_shape=(None, *INPUT_SHAPE))
model.summary()
model.compile(optimizer='adam',
loss='categorical_crossentropy',
metrics=['categorical_accuracy'])
print('Preparing imagenet data')
imagenet = tfds.load('imagenet2012',
data_dir='/media/mldata/imagenet',
as_supervised=True)
def resize_image(image, label):
image = tf.image.convert_image_dtype(image, tf.float32) / 255
image = tf.image.resize(image, (INPUT_SHAPE[0], INPUT_SHAPE[1]))
