def init_resnet(stack_fn, preact, use_bias, model_name, input_shape,
block_name_to_hyperparameters_dict, preprocess_input_mode):
# Downloads the weights
file_name = model_name + "_weights_tf_dim_ordering_tf_kernels_notop.h5"
file_hash = WEIGHTS_HASHES[model_name][1]
weights_path = data_utils.get_file(file_name,
BASE_WEIGHTS_PATH + file_name,
cache_subdir="models",
file_hash=file_hash)
# Define and initialize the first block
first_block = ResNet(stack_fn=lambda x: x,
preact=preact,
use_bias=use_bias,
include_top=False,
weights=None,
input_shape=input_shape)
first_block.load_weights(weights_path, by_name=True)
submodel_list = [first_block]
# Define and initialize each block
for block_name, (filters, blocks,
stride1) in block_name_to_hyperparameters_dict.items():
input_tensor = Input(shape=K.int_shape(submodel_list[-1].output)[1:])
output_tensor = stack_fn(input_tensor,
filters=filters,
blocks=blocks,
stride1=stride1,
name=block_name)
submodel = Model(inputs=input_tensor,
outputs=output_tensor,
name="{}_block".format(block_name))
submodel.load_weights(weights_path, by_name=True)
submodel_list.append(submodel)
return submodel_list, lambda x: preprocess_input(x,
mode=preprocess_input_mode)
def ResNeXt50(include_top=True,
weights='imagenet',
input_tensor=None,
input_shape=None,
pooling=None,
classes=1000,
**kwargs):
def stack_fn(x):
x = stack3(x, 128, 3, stride1=1, name='conv2')
x = stack3(x, 256, 4, name='conv3')
x = stack3(x, 512, 6, name='conv4')
x = stack3(x, 1024, 3, name='conv5')
return x
return ResNet(stack_fn, False, False, 'resnext50', include_top, weights,
input_tensor, input_shape, pooling, classes, **kwargs)
def ResNet34(include_top=True,
weights='imagenet',
input_tensor=None,
input_shape=None,
pooling=None,
classes=1000,
**kwargs):
"""Instantiates the ResNet34 architecture."""
def stack_fn(x):
x = basicstack1(x, 64, 3, stride1=1, name='conv2')
x = basicstack1(x, 128, 4, name='conv3')
x = basicstack1(x, 256, 6, name='conv4')
return basicstack1(x, 512, 3, name='conv5')
return ResNet(stack_fn, False, True, 'resnet34', include_top, weights,
input_tensor, input_shape, pooling, classes, **kwargs)
def ResNet34Bottleneck(include_top=True,
weights='imagenet',
input_tensor=None,
input_shape=None,
pooling=None,
classes=1000,
**kwargs):
"""Instantiates the ResNet34Bottleneck architecture.
Warning: this is actually the same with ResNet50.
"""
def stack_fn(x):
x = stack1(x, 64, 3, stride1=1, name='conv2')
x = stack1(x, 128, 4, name='conv3')
x = stack1(x, 256, 6, name='conv4')
return stack1(x, 512, 3, name='conv5')
return ResNet(stack_fn, False, True, 'resnet34_bottleneck', include_top, weights,
input_tensor, input_shape, pooling, classes, **kwargs)
def ResNet18Bottleneck(include_top=True,
weights='imagenet',
input_tensor=None,
input_shape=None,
pooling=None,
classes=1000,
**kwargs):
"""Instantiates the ResNet18Bottleneck architecture.
Warning: this would be inexistent ResNet26.
"""
def stack_fn(x):
x = stack1(x, 64, 2, stride1=1, name='conv2')
x = stack1(x, 128, 2, name='conv3')
x = stack1(x, 256, 2, name='conv4')
return stack1(x, 512, 2, name='conv5')
return ResNet(stack_fn, False, True, 'resnet18_bottleneck', include_top, weights,
input_tensor, input_shape, pooling, classes, **kwargs)
def stacks(x):
x = resnet.stack1(x, 64, 3, stride1=1, name='conv2')
x = resnet.stack1(x, 128, 4, name='conv3')
x = resnet.stack1(x, 256, 6, name='conv4')
x = resnet.stack1(x, 512, 3, name='conv5')
return x
input_name = 'my_input'
inp = tf.keras.layers.Input(shape=(img_dim, img_dim, 3), name=input_name)
model = ResNet(stacks,
preact=False,
use_bias=True,
model_name='resnet50',
weights='imagenet',
input_tensor=inp,
pooling='avg',
input_shape=(img_dim, img_dim, 3))
y = model(inp)
print(tf.shape(y))
print(len(model.layers))
names = {}
def show_layers():
for i, layer in enumerate(model.layers):
print(i, layer.name)