tf.compat.v1.ragged.constant_value

Constructs a RaggedTensorValue from a nested Python list.

tf.compat.v1.ragged.constant_value(
    pylist,
    dtype=None,
    ragged_rank=None,
    inner_shape=None,
    row_splits_dtype='int64'
)

Example:

tf.compat.v1.ragged.constant_value([[1, 2], [3], [4, 5, 6]])
tf.RaggedTensorValue(values=array([1, 2, 3, 4, 5, 6]),
                     row_splits=array([0, 2, 3, 6]))

All scalar values in pylist must have the same nesting depth K, and the returned RaggedTensorValue will have rank K. If pylist contains no scalar values, then K is one greater than the maximum depth of empty lists in pylist. All scalar values in pylist must be compatible with dtype.

pylist A nested list, tuple or np.ndarray. Any nested element that is not a list or tuple must be a scalar value compatible with dtype.
dtype numpy.dtype. The type of elements for the returned RaggedTensor. If not specified, then a default is chosen based on the scalar values in pylist.
ragged_rank An integer specifying the ragged rank of the returned RaggedTensorValue. Must be nonnegative and less than K. Defaults to max(0, K - 1) if inner_shape is not specified. Defaults to `max(0, K

  • 1 - len(inner_shape))ifinner_shapeis specified. </td> </tr><tr> <td>inner_shape<a id="inner_shape"></a> </td> <td> A tuple of integers specifying the shape for individual inner values in the returnedRaggedTensorValue. Defaults to()ifragged_rankis not specified. Ifragged_rankis specified, then a default is chosen based on the contents ofpylist. </td> </tr><tr> <td>row_splits_dtype<a id="row_splits_dtype"></a> </td> <td> data type for the constructedRaggedTensorValue's row_splits. One ofnumpy.int32ornumpy.int64`.

A tf.RaggedTensorValue or numpy.array with rank K and the specified ragged_rank, containing the values from pylist.

ValueError If the scalar values in pylist have inconsistent nesting depth; or if ragged_rank or inner_shape are incompatible with pylist.