本文整理汇总了Python中official.resnet.resnet_model.Model方法的典型用法代码示例。如果您正苦于以下问题：Python resnet_model.Model方法的具体用法？Python resnet_model.Model怎么用？Python resnet_model.Model使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类official.resnet.resnet_model的用法示例。
在下文中一共展示了resnet_model.Model方法的3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: cifar10_model_fn
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import Model [as 别名]
def cifar10_model_fn(features, labels, mode, params):
"""Model function for CIFAR-10."""
features = tf.reshape(features, [-1, _HEIGHT, _WIDTH, _NUM_CHANNELS])
learning_rate_fn = resnet_run_loop.learning_rate_with_decay(
batch_size=params['batch_size'], batch_denom=128,
num_images=_NUM_IMAGES['train'], boundary_epochs=[100, 150, 200],
decay_rates=[1, 0.1, 0.01, 0.001])
# We use a weight decay of 0.0002, which performs better
# than the 0.0001 that was originally suggested.
weight_decay = 2e-4
# Empirical testing showed that including batch_normalization variables
# in the calculation of regularized loss helped validation accuracy
# for the CIFAR-10 dataset, perhaps because the regularization prevents
# overfitting on the small data set. We therefore include all vars when
# regularizing and computing loss during training.
def loss_filter_fn(_):
return True
return resnet_run_loop.resnet_model_fn(
features=features,
labels=labels,
mode=mode,
model_class=Cifar10Model,
resnet_size=params['resnet_size'],
weight_decay=weight_decay,
learning_rate_fn=learning_rate_fn,
momentum=0.9,
data_format=params['data_format'],
version=params['version'],
loss_scale=params['loss_scale'],
loss_filter_fn=loss_filter_fn,
multi_gpu=params['multi_gpu'],
dtype=params['dtype']
) 示例2: cifar10_model_fn
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import Model [as 别名]
def cifar10_model_fn(features, labels, mode, params):
"""Model function for CIFAR-10."""
features = tf.reshape(features, [-1, HEIGHT, WIDTH, NUM_CHANNELS])
# Learning rate schedule follows arXiv:1512.03385 for ResNet-56 and under.
learning_rate_fn = resnet_run_loop.learning_rate_with_decay(
batch_size=params['batch_size'], batch_denom=128,
num_images=NUM_IMAGES['train'], boundary_epochs=[91, 136, 182],
decay_rates=[1, 0.1, 0.01, 0.001])
# Weight decay of 2e-4 diverges from 1e-4 decay used in the ResNet paper
# and seems more stable in testing. The difference was nominal for ResNet-56.
weight_decay = 2e-4
# Empirical testing showed that including batch_normalization variables
# in the calculation of regularized loss helped validation accuracy
# for the CIFAR-10 dataset, perhaps because the regularization prevents
# overfitting on the small data set. We therefore include all vars when
# regularizing and computing loss during training.
def loss_filter_fn(_):
return True
return resnet_run_loop.resnet_model_fn(
features=features,
labels=labels,
mode=mode,
model_class=Cifar10Model,
resnet_size=params['resnet_size'],
weight_decay=weight_decay,
learning_rate_fn=learning_rate_fn,
momentum=0.9,
data_format=params['data_format'],
resnet_version=params['resnet_version'],
loss_scale=params['loss_scale'],
loss_filter_fn=loss_filter_fn,
dtype=params['dtype'],
fine_tune=params['fine_tune']
) 示例3: cifar10_model_fn
# 需要导入模块: from official.resnet import resnet_model [as 别名]
# 或者: from official.resnet.resnet_model import Model [as 别名]
def cifar10_model_fn(features, labels, mode, params):
"""Model function for CIFAR-10."""
features = tf.reshape(features, [-1, _HEIGHT, _WIDTH, _NUM_CHANNELS])
# Learning rate schedule follows arXiv:1512.03385 for ResNet-56 and under.
learning_rate_fn = resnet_run_loop.learning_rate_with_decay(
batch_size=params['batch_size'], batch_denom=128,
num_images=_NUM_IMAGES['train'], boundary_epochs=[91, 136, 182],
decay_rates=[1, 0.1, 0.01, 0.001])
# Weight decay of 2e-4 diverges from 1e-4 decay used in the ResNet paper
# and seems more stable in testing. The difference was nominal for ResNet-56.
weight_decay = 2e-4
# Empirical testing showed that including batch_normalization variables
# in the calculation of regularized loss helped validation accuracy
# for the CIFAR-10 dataset, perhaps because the regularization prevents
# overfitting on the small data set. We therefore include all vars when
# regularizing and computing loss during training.
def loss_filter_fn(_):
return True
return resnet_run_loop.resnet_model_fn(
features=features,
labels=labels,
mode=mode,
model_class=Cifar10Model,
resnet_size=params['resnet_size'],
weight_decay=weight_decay,
learning_rate_fn=learning_rate_fn,
momentum=0.9,
data_format=params['data_format'],
resnet_version=params['resnet_version'],
loss_scale=params['loss_scale'],
loss_filter_fn=loss_filter_fn,
dtype=params['dtype'],
fine_tune=params['fine_tune']
)