tf.contrib.layers.safe_embedding_lookup_sparse
Stay organized with collections
Save and categorize content based on your preferences.
Lookup embedding results, accounting for invalid IDs and empty features.
tf.contrib.layers.safe_embedding_lookup_sparse(
embedding_weights, sparse_ids, sparse_weights=None, combiner=None,
default_id=None, name=None, partition_strategy='div', max_norm=None
)
The partitioned embedding in embedding_weights must all be the same shape
except for the first dimension. The first dimension is allowed to vary as the
vocabulary size is not necessarily a multiple of P. embedding_weights
may be a PartitionedVariable as returned by using
tf.compat.v1.get_variable() with a
partitioner.
Invalid IDs (< 0) are pruned from input IDs and weights, as well as any IDs
with non-positive weight. For an entry with no features, the embedding vector
for default_id is returned, or the 0-vector if default_id is not supplied.
The ids and weights may be multi-dimensional. Embeddings are always aggregated
along the last dimension.
Args |
|---|
embedding_weights
|
A list of P float tensors or values representing
partitioned embedding tensors. Alternatively, a PartitionedVariable,
created by partitioning along dimension 0. The total unpartitioned shape
should be [e_0, e_1, ..., e_m], where e_0 represents the vocab size
and e_1, ..., e_m are the embedding dimensions.
|
sparse_ids
|
SparseTensor of shape [d_0, d_1, ..., d_n] containing the
ids. d_0 is typically batch size.
|
sparse_weights
|
SparseTensor of same shape as sparse_ids, containing
float weights corresponding to sparse_ids, or None if all weights are
be assumed to be 1.0.
|
combiner
|
A string specifying how to combine embedding results for each
entry. Currently "mean", "sqrtn" and "sum" are supported, with "mean" the
default.
|
default_id
|
The id to use for an entry with no features.
|
name
|
A name for this operation (optional).
|
partition_strategy
|
A string specifying the partitioning strategy. Currently
"div" and "mod" are supported. Default is "div".
|
max_norm
|
If not None, all embeddings are l2-normalized to max_norm before
combining.
|
Returns |
|---|
Dense tensor of shape [d_0, d_1, ..., d_{n-1}, e_1, ..., e_m].
|
Raises |
|---|
ValueError
|
if embedding_weights is empty.
|
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2020-10-01 UTC.
[null,null,["Last updated 2020-10-01 UTC."],[],[],null,["# tf.contrib.layers.safe_embedding_lookup_sparse\n\n\u003cbr /\u003e\n\n|--------------------------------------------------------------------------------------------------------------------------------------------------|\n| [View source on GitHub](https://github.com/tensorflow/tensorflow/blob/v1.15.0/tensorflow/contrib/layers/python/layers/embedding_ops.py#L48-L184) |\n\nLookup embedding results, accounting for invalid IDs and empty features. \n\n tf.contrib.layers.safe_embedding_lookup_sparse(\n embedding_weights, sparse_ids, sparse_weights=None, combiner=None,\n default_id=None, name=None, partition_strategy='div', max_norm=None\n )\n\nThe partitioned embedding in `embedding_weights` must all be the same shape\nexcept for the first dimension. The first dimension is allowed to vary as the\nvocabulary size is not necessarily a multiple of `P`. `embedding_weights`\nmay be a `PartitionedVariable` as returned by using\n[`tf.compat.v1.get_variable()`](../../../tf/get_variable) with a\npartitioner.\n\nInvalid IDs (\\\u003c 0) are pruned from input IDs and weights, as well as any IDs\nwith non-positive weight. For an entry with no features, the embedding vector\nfor `default_id` is returned, or the 0-vector if `default_id` is not supplied.\n\nThe ids and weights may be multi-dimensional. Embeddings are always aggregated\nalong the last dimension.\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Args ---- ||\n|----------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| `embedding_weights` | A list of `P` float tensors or values representing partitioned embedding tensors. Alternatively, a `PartitionedVariable`, created by partitioning along dimension 0. The total unpartitioned shape should be `[e_0, e_1, ..., e_m]`, where `e_0` represents the vocab size and `e_1, ..., e_m` are the embedding dimensions. |\n| `sparse_ids` | `SparseTensor` of shape `[d_0, d_1, ..., d_n]` containing the ids. `d_0` is typically batch size. |\n| `sparse_weights` | `SparseTensor` of same shape as `sparse_ids`, containing float weights corresponding to `sparse_ids`, or `None` if all weights are be assumed to be 1.0. |\n| `combiner` | A string specifying how to combine embedding results for each entry. Currently \"mean\", \"sqrtn\" and \"sum\" are supported, with \"mean\" the default. |\n| `default_id` | The id to use for an entry with no features. |\n| `name` | A name for this operation (optional). |\n| `partition_strategy` | A string specifying the partitioning strategy. Currently `\"div\"` and `\"mod\"` are supported. Default is `\"div\"`. |\n| `max_norm` | If not None, all embeddings are l2-normalized to max_norm before combining. |\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Returns ------- ||\n|---|---|\n| Dense tensor of shape `[d_0, d_1, ..., d_{n-1}, e_1, ..., e_m]`. ||\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Raises ------ ||\n|--------------|----------------------------------|\n| `ValueError` | if `embedding_weights` is empty. |\n\n\u003cbr /\u003e"]]
View source on GitHub