本文整理汇总了Python中dragnn.python.network_units.StoredActivations方法的典型用法代码示例。如果您正苦于以下问题：Python network_units.StoredActivations方法的具体用法？Python network_units.StoredActivations怎么用？Python network_units.StoredActivations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dragnn.python.network_units的用法示例。
在下文中一共展示了network_units.StoredActivations方法的2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: update_network_states
# 需要导入模块: from dragnn.python import network_units [as 别名]
# 或者: from dragnn.python.network_units import StoredActivations [as 别名]
def update_network_states(comp, tensors, network_states, stride):
"""Stores Tensor objects corresponding to layer outputs.
For use in subsequent tasks.
Args:
comp: Component for which the tensor handles are being stored.
tensors: list of Tensors to store
network_states: dictionary of component NetworkState objects
stride: stride of the stored tensor.
"""
network_state = network_states[comp.name]
with tf.name_scope(comp.name + '/stored_act'):
for index, network_tensor in enumerate(tensors):
network_state.activations[comp.network.layers[index].name] = (
network_units.StoredActivations(tensor=network_tensor, stride=stride,
dim=comp.network.layers[index].dim)) 示例2: update_network_states
# 需要导入模块: from dragnn.python import network_units [as 别名]
# 或者: from dragnn.python.network_units import StoredActivations [as 别名]
def update_network_states(comp, tensors, network_states, stride):
"""Stores Tensor objects corresponding to layer outputs.
For use in subsequent tasks.
Args:
comp: Component for which the tensor handles are being stored.
tensors: list of Tensors to store
network_states: dictionary of component NetworkState objects
stride: stride of the stored tensor.
"""
network_state = network_states[comp.name]
with tf.name_scope(comp.name + '/stored_act'):
for index, network_tensor in enumerate(tensors):
network_state.activations[comp.network.layers[index].name] = (
network_units.StoredActivations(
tensor=network_tensor,
stride=stride,
dim=comp.network.layers[index].dim))