tf.compat.v1.ragged.RaggedTensorValue

Represents the value of a RaggedTensor.

tf.compat.v1.ragged.RaggedTensorValue(
    values, row_splits
)

See tf.RaggedTensor for a description of ragged tensors.

Args
`values`	A numpy array of any type and shape; or a RaggedTensorValue.
`row_splits`	A 1-D int32 or int64 numpy array.

Attributes
`dtype`	The numpy dtype of values in this tensor.
`flat_values`	The innermost `values` array for this ragged tensor value.
`nested_row_splits`	The row_splits for all ragged dimensions in this ragged tensor value.
`ragged_rank`	The number of ragged dimensions in this ragged tensor value.
`row_splits`	The split indices for the ragged tensor value.
`shape`	A tuple indicating the shape of this RaggedTensorValue.
`values`	The concatenated values for all rows in this tensor.

Methods

to_list()

Returns this ragged tensor value as a nested Python list.

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.