本文整理汇总了Python中official.resnet.resnet_model.DEFAULT_DTYPE属性的典型用法代码示例。如果您正苦于以下问题：Python resnet_model.DEFAULT_DTYPE属性的具体用法？Python resnet_model.DEFAULT_DTYPE怎么用？Python resnet_model.DEFAULT_DTYPE使用的例子？那么恭喜您, 这里精选的属性代码示例或许可以为您提供帮助。您也可以进一步了解该属性所在类official.resnet.resnet_model的用法示例。
在下文中一共展示了resnet_model.DEFAULT_DTYPE属性的7个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: __init__
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import DEFAULT_DTYPE [as 别名]
def __init__(self, resnet_size, data_format=None, num_classes=_NUM_CLASSES,
version=resnet_model.DEFAULT_VERSION,
dtype=resnet_model.DEFAULT_DTYPE):
"""These are the parameters that work for CIFAR-10 data.
Args:
resnet_size: The number of convolutional layers needed in the model.
data_format: Either 'channels_first' or 'channels_last', specifying which
data format to use when setting up the model.
num_classes: The number of output classes needed from the model. This
enables users to extend the same model to their own datasets.
version: Integer representing which version of the ResNet network to use.
See README for details. Valid values: [1, 2]
dtype: The TensorFlow dtype to use for calculations.
Raises:
ValueError: if invalid resnet_size is chosen
"""
if resnet_size % 6 != 2:
raise ValueError('resnet_size must be 6n + 2:', resnet_size)
num_blocks = (resnet_size - 2) // 6
super(Cifar10Model, self).__init__(
resnet_size=resnet_size,
bottleneck=False,
num_classes=num_classes,
num_filters=16,
kernel_size=3,
conv_stride=1,
first_pool_size=None,
first_pool_stride=None,
block_sizes=[num_blocks] * 3,
block_strides=[1, 2, 2],
final_size=64,
version=version,
data_format=data_format,
dtype=dtype
) 示例2: __init__
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import DEFAULT_DTYPE [as 别名]
def __init__(self, resnet_size, data_format=None, num_classes=_NUM_CLASSES,
version=resnet_model.DEFAULT_VERSION,
dtype=resnet_model.DEFAULT_DTYPE):
"""These are the parameters that work for Imagenet data.
Args:
resnet_size: The number of convolutional layers needed in the model.
data_format: Either 'channels_first' or 'channels_last', specifying which
data format to use when setting up the model.
num_classes: The number of output classes needed from the model. This
enables users to extend the same model to their own datasets.
version: Integer representing which version of the ResNet network to use.
See README for details. Valid values: [1, 2]
dtype: The TensorFlow dtype to use for calculations.
"""
# For bigger models, we want to use "bottleneck" layers
if resnet_size < 50:
bottleneck = False
final_size = 512
else:
bottleneck = True
final_size = 2048
super(ImagenetModel, self).__init__(
resnet_size=resnet_size,
bottleneck=bottleneck,
num_classes=num_classes,
num_filters=64,
kernel_size=7,
conv_stride=2,
first_pool_size=3,
first_pool_stride=2,
block_sizes=_get_block_sizes(resnet_size),
block_strides=[1, 2, 2, 2],
final_size=final_size,
version=version,
data_format=data_format,
dtype=dtype
) 示例3: __init__
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import DEFAULT_DTYPE [as 别名]
def __init__(self, resnet_size, data_format=None, num_classes=NUM_CLASSES,
resnet_version=resnet_model.DEFAULT_VERSION,
dtype=resnet_model.DEFAULT_DTYPE):
"""These are the parameters that work for CIFAR-10 data.
Args:
resnet_size: The number of convolutional layers needed in the model.
data_format: Either 'channels_first' or 'channels_last', specifying which
data format to use when setting up the model.
num_classes: The number of output classes needed from the model. This
enables users to extend the same model to their own datasets.
resnet_version: Integer representing which version of the ResNet network
to use. See README for details. Valid values: [1, 2]
dtype: The TensorFlow dtype to use for calculations.
Raises:
ValueError: if invalid resnet_size is chosen
"""
if resnet_size % 6 != 2:
raise ValueError('resnet_size must be 6n + 2:', resnet_size)
num_blocks = (resnet_size - 2) // 6
super(Cifar10Model, self).__init__(
resnet_size=resnet_size,
bottleneck=False,
num_classes=num_classes,
num_filters=16,
kernel_size=3,
conv_stride=1,
first_pool_size=None,
first_pool_stride=None,
block_sizes=[num_blocks] * 3,
block_strides=[1, 2, 2],
resnet_version=resnet_version,
data_format=data_format,
dtype=dtype
) 示例4: __init__
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import DEFAULT_DTYPE [as 别名]
def __init__(self, resnet_size, data_format=None, num_classes=NUM_CLASSES,
resnet_version=resnet_model.DEFAULT_VERSION,
dtype=resnet_model.DEFAULT_DTYPE):
"""These are the parameters that work for Imagenet data.
Args:
resnet_size: The number of convolutional layers needed in the model.
data_format: Either 'channels_first' or 'channels_last', specifying which
data format to use when setting up the model.
num_classes: The number of output classes needed from the model. This
enables users to extend the same model to their own datasets.
resnet_version: Integer representing which version of the ResNet network
to use. See README for details. Valid values: [1, 2]
dtype: The TensorFlow dtype to use for calculations.
"""
# For bigger models, we want to use "bottleneck" layers
if resnet_size < 50:
bottleneck = False
else:
bottleneck = True
super(ImagenetModel, self).__init__(
resnet_size=resnet_size,
bottleneck=bottleneck,
num_classes=num_classes,
num_filters=64,
kernel_size=7,
conv_stride=2,
first_pool_size=3,
first_pool_stride=2,
block_sizes=_get_block_sizes(resnet_size),
block_strides=[1, 2, 2, 2],
resnet_version=resnet_version,
data_format=data_format,
dtype=dtype
) 示例5: __init__
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import DEFAULT_DTYPE [as 别名]
def __init__(self, resnet_size, data_format=None, num_classes=_NUM_CLASSES,
resnet_version=resnet_model.DEFAULT_VERSION,
dtype=resnet_model.DEFAULT_DTYPE):
"""These are the parameters that work for CIFAR-10 data.
Args:
resnet_size: The number of convolutional layers needed in the model.
data_format: Either 'channels_first' or 'channels_last', specifying which
data format to use when setting up the model.
num_classes: The number of output classes needed from the model. This
enables users to extend the same model to their own datasets.
resnet_version: Integer representing which version of the ResNet network
to use. See README for details. Valid values: [1, 2]
dtype: The TensorFlow dtype to use for calculations.
Raises:
ValueError: if invalid resnet_size is chosen
"""
if resnet_size % 6 != 2:
raise ValueError('resnet_size must be 6n + 2:', resnet_size)
num_blocks = (resnet_size - 2) // 6
super(Cifar10Model, self).__init__(
resnet_size=resnet_size,
bottleneck=False,
num_classes=num_classes,
num_filters=16,
kernel_size=3,
conv_stride=1,
first_pool_size=None,
first_pool_stride=None,
block_sizes=[num_blocks] * 3,
block_strides=[1, 2, 2],
resnet_version=resnet_version,
data_format=data_format,
dtype=dtype
) 示例6: __init__
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import DEFAULT_DTYPE [as 别名]
def __init__(self, resnet_size, data_format=None, num_classes=_NUM_CLASSES,
resnet_version=resnet_model.DEFAULT_VERSION,
dtype=resnet_model.DEFAULT_DTYPE):
"""These are the parameters that work for Imagenet data.
Args:
resnet_size: The number of convolutional layers needed in the model.
data_format: Either 'channels_first' or 'channels_last', specifying which
data format to use when setting up the model.
num_classes: The number of output classes needed from the model. This
enables users to extend the same model to their own datasets.
resnet_version: Integer representing which version of the ResNet network
to use. See README for details. Valid values: [1, 2]
dtype: The TensorFlow dtype to use for calculations.
"""
# For bigger models, we want to use "bottleneck" layers
if resnet_size < 50:
bottleneck = False
else:
bottleneck = True
super(ImagenetModel, self).__init__(
resnet_size=resnet_size,
bottleneck=bottleneck,
num_classes=num_classes,
num_filters=64,
kernel_size=7,
conv_stride=2,
first_pool_size=3,
first_pool_stride=2,
block_sizes=_get_block_sizes(resnet_size),
block_strides=[1, 2, 2, 2],
resnet_version=resnet_version,
data_format=data_format,
dtype=dtype
) 示例7: __init__
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import DEFAULT_DTYPE [as 别名]
def __init__(self, resnet_size, data_format=None, num_classes=_NUM_CLASSES,
version=resnet_model.DEFAULT_VERSION,
dtype=resnet_model.DEFAULT_DTYPE):
"""These are the parameters that work for Imagenet data.
Args:
resnet_size: The number of convolutional layers needed in the model.
data_format: Either 'channels_first' or 'channels_last', specifying which
data format to use when setting up the model.
num_classes: The number of output classes needed from the model. This
enables users to extend the same model to their own datasets.
version: Integer representing which version of the ResNet network to use.
See README for details. Valid values: [1, 2]
dtype: The TensorFlow dtype to use for calculations.
"""
# For bigger models, we want to use "bottleneck" layers
if resnet_size < 50:
bottleneck = False
final_size = 512
else:
bottleneck = True
final_size = 2048
super(ImagenetModel, self).__init__(
resnet_size=resnet_size,
bottleneck=bottleneck,
num_classes=num_classes,
num_filters=64,
kernel_size=7,
conv_stride=2,
first_pool_size=3,
first_pool_stride=2,
second_pool_size=7,
second_pool_stride=1,
block_sizes=_get_block_sizes(resnet_size),
block_strides=[1, 2, 2, 2],
final_size=final_size,
version=version,
data_format=data_format,
dtype=dtype
)