View source on GitHub |
Creates a table initializer from a tf.data.Dataset
.
tf.lookup.experimental.DatasetInitializer(
dataset
)
Sample usage:
keys = tf.data.Dataset.range(100)
values = tf.data.Dataset.range(100).map(
lambda x: string_ops.as_string(x * 2))
ds = tf.data.Dataset.zip((keys, values))
init = tf.lookup.experimental.DatasetInitializer(ds)
table = tf.lookup.StaticHashTable(init, "")
table.lookup(tf.constant([0, 1, 2], dtype=tf.int64)).numpy()
array([b'0', b'2', b'4'], dtype=object)
Raises: ValueError if dataset
doesn't conform to specifications.
Args | |
---|---|
dataset
|
A tf.data.Dataset object that produces tuples of scalars. The
first scalar is treated as a key and the second as value.
|
Attributes | |
---|---|
dataset
|
A tf.data.Dataset object that produces tuples of scalars. The
first scalar is treated as a key and the second as value.
|
key_dtype
|
The expected table key dtype. |
value_dtype
|
The expected table value dtype. |
Methods
initialize
initialize(
table
)
Returns the table initialization op.