本文整理汇总了Python中dragnn.python.dragnn_ops.bulk_fixed_embeddings方法的典型用法代码示例。如果您正苦于以下问题：Python dragnn_ops.bulk_fixed_embeddings方法的具体用法？Python dragnn_ops.bulk_fixed_embeddings怎么用？Python dragnn_ops.bulk_fixed_embeddings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.dragnn_ops的用法示例。
在下文中一共展示了dragnn_ops.bulk_fixed_embeddings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: fetch_fast_fixed_embeddings
# 需要导入模块: from dragnn.python import dragnn_ops [as 别名]
# 或者: from dragnn.python.dragnn_ops import bulk_fixed_embeddings [as 别名]
def fetch_fast_fixed_embeddings(comp, state):
"""Looks up fixed features with fast, non-differentiable, op.
Since BulkFixedEmbeddings is non-differentiable with respect to the
embeddings, the idea is to call this function only when the graph is
not being used for training.
Args:
comp: Component whose fixed features we wish to look up.
state: live MasterState object for the component.
Returns:
state handle: updated state handle to be used after this call
fixed_embeddings: list of NamedTensor objects
"""
_validate_embedded_fixed_features(comp)
num_channels = len(comp.spec.fixed_feature)
if not num_channels:
return state.handle, []
tf.logging.info('[%s] Adding %d fast fixed features', comp.name, num_channels)
state.handle, bulk_embeddings, _ = dragnn_ops.bulk_fixed_embeddings(
state.handle, [
comp.get_variable(network_units.fixed_embeddings_name(c))
for c in range(num_channels)
],
component=comp.name)
bulk_embeddings = network_units.NamedTensor(bulk_embeddings,
'bulk-%s-fixed-features' %
comp.name)
return state.handle, [bulk_embeddings] 示例2: fetch_fast_fixed_embeddings
# 需要导入模块: from dragnn.python import dragnn_ops [as 别名]
# 或者: from dragnn.python.dragnn_ops import bulk_fixed_embeddings [as 别名]
def fetch_fast_fixed_embeddings(comp,
state,
pad_to_batch=None,
pad_to_steps=None):
"""Looks up fixed features with fast, non-differentiable, op.
Since BulkFixedEmbeddings is non-differentiable with respect to the
embeddings, the idea is to call this function only when the graph is
not being used for training. If the function is being called with fixed step
and batch sizes, it will use the most efficient possible extractor.
Args:
comp: Component whose fixed features we wish to look up.
state: live MasterState object for the component.
pad_to_batch: Optional; the number of batch elements to pad to.
pad_to_steps: Optional; the number of steps to pad to.
Returns:
state handle: updated state handle to be used after this call
fixed_embeddings: list of NamedTensor objects
"""
_validate_embedded_fixed_features(comp)
num_channels = len(comp.spec.fixed_feature)
if not num_channels:
return state.handle, []
tf.logging.info('[%s] Adding %d fast fixed features', comp.name, num_channels)
features = [
comp.get_variable(network_units.fixed_embeddings_name(c))
for c in range(num_channels)
]
if pad_to_batch is not None and pad_to_steps is not None:
# If we have fixed padding numbers, we can use 'bulk_embed_fixed_features',
# which is the fastest embedding extractor.
state.handle, bulk_embeddings, _ = dragnn_ops.bulk_embed_fixed_features(
state.handle,
features,
component=comp.name,
pad_to_batch=pad_to_batch,
pad_to_steps=pad_to_steps)
else:
state.handle, bulk_embeddings, _ = dragnn_ops.bulk_fixed_embeddings(
state.handle, features, component=comp.name)
bulk_embeddings = network_units.NamedTensor(
bulk_embeddings, 'bulk-%s-fixed-features' % comp.name)
return state.handle, [bulk_embeddings]