tf.RaggedTensor

TensorFlow 1 version View source on GitHub

Represents a ragged tensor.

A RaggedTensor is a tensor with one or more ragged dimensions, which are dimensions whose slices may have different lengths. For example, the inner (column) dimension of rt=[[3, 1, 4, 1], [], [5, 9, 2], [6], []] is ragged, since the column slices (rt[0, :], ..., rt[4, :]) have different lengths. Dimensions whose slices all have the same length are called uniform dimensions. The outermost dimension of a RaggedTensor is always uniform, since it consists of a single slice (and so there is no possibility for differing slice lengths).

The total number of dimensions in a RaggedTensor is called its rank, and the number of ragged dimensions in a RaggedTensor is called its ragged-rank. A RaggedTensor's ragged-rank is fixed at graph creation time: it can't depend on the runtime values of Tensors, and can't vary dynamically for different session runs.

Potentially Ragged Tensors

Many ops support both Tensors and RaggedTensors. The term "potentially ragged tensor" may be used to refer to a tensor that might be either a Tensor or a RaggedTensor. The ragged-rank of a Tensor is zero.

Documenting RaggedTensor Shapes

When documenting the shape of a RaggedTensor, ragged dimensions can be indicated by enclosing them in parentheses. For example, the shape of a 3-D RaggedTensor that stores the fixed-size word embedding for each word in a sentence, for each sentence in a batch, could be written as [num_sentences, (num_words), embedding_size]. The parentheses around (num_words) indicate that dimension is ragged, and that the length of each element list in that dimension may vary for each item.

Component Tensors

Internally, a RaggedTensor consists of a concatenated list of values that are partitioned into variable-length rows. In particular, each RaggedTensor consists of:

  • A values tensor, which concatenates the variable-length rows into a flattened list. For example, the values tensor for [[3, 1, 4, 1], [], [5, 9, 2], [6], []] is [3, 1, 4, 1, 5, 9, 2, 6].

  • A row_splits vector, which indicates how those flattened values are divided into rows. In particular, the values for row rt[i] are stored in the slice rt.values[rt.row_splits[i]:rt.row_splits[i+1]].

Example:

print(tf.RaggedTensor.from_row_splits(
      values=[3, 1, 4, 1, 5, 9, 2, 6],
      row_splits=[0, 4, 4, 7, 8, 8]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>

Alternative Row-Partitioning Schemes

In addition to row_splits, ragged tensors provide support for four other row-partitioning schemes:

  • row_lengths: a vector with shape [nrows], which specifies the length of each row.

  • value_rowids and nrows: value_rowids is a vector with shape [nvals], corresponding one-to-one with values, which specifies each value's row index. In particular, the row rt[row] consists of the values rt.values[j] where value_rowids[j]==row. nrows is an integer scalar that specifies the number of rows in the RaggedTensor. (nrows is used to indicate trailing empty rows.)

  • row_starts: a vector with shape [nrows], which specifies the start offset of each row. Equivalent to row_splits[:-1].

  • row_limits: a vector with shape [nrows], which specifies the stop offset of each row. Equivalent to row_splits[1:].

  • uniform_row_length: A scalar tensor, specifying the length of every row. This row-partitioning scheme may only be used if all rows have the same length.

Example: The following ragged tensors are equivalent, and all represent the nested list [[3, 1, 4, 1], [], [5, 9, 2], [6], []].

values = [3, 1, 4, 1, 5, 9, 2, 6]
rt1 = RaggedTensor.from_row_splits(values, row_splits=[0, 4, 4, 7, 8, 8])
rt2 = RaggedTensor.from_row_lengths(values, row_lengths=[4, 0, 3, 1, 0])
rt3 = RaggedTensor.from_value_rowids(
    values, value_rowids=[0, 0, 0, 0, 2, 2, 2, 3], nrows=5)
rt4 = RaggedTensor.from_row_starts(values, row_starts=[0, 4, 4, 7, 8])
rt5 = RaggedTensor.from_row_limits(values, row_limits=[4, 4, 7, 8, 8])

Multiple Ragged Dimensions

RaggedTensors with multiple ragged dimensions can be defined by using a nested RaggedTensor for the values tensor. Each nested RaggedTensor adds a single ragged dimension.

inner_rt = RaggedTensor.from_row_splits(  # =rt1 from above
    values=[3, 1, 4, 1, 5, 9, 2, 6], row_splits=[0, 4, 4, 7, 8, 8])
outer_rt = RaggedTensor.from_row_splits(
    values=inner_rt, row_splits=[0, 3, 3, 5])
print(outer_rt.to_list())
[[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]]
print(outer_rt.ragged_rank)
2

The factory function RaggedTensor.from_nested_row_splits may be used to construct a RaggedTensor with multiple ragged dimensions directly, by providing a list of row_splits tensors:

RaggedTensor.from_nested_row_splits(
    flat_values=[3, 1, 4, 1, 5, 9, 2, 6],
    nested_row_splits=([0, 3, 3, 5], [0, 4, 4, 7, 8, 8])).to_list()
[[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]]

Uniform Inner Dimensions

RaggedTensors with uniform inner dimensions can be defined by using a multidimensional Tensor for values.

rt = RaggedTensor.from_row_splits(values=tf.ones([5, 3], tf.int32),
                                  row_splits=[0, 2, 5])
print(rt.to_list())
[[[1, 1, 1], [1, 1, 1]],
 [[1, 1, 1], [1, 1, 1], [1, 1, 1]]]
print(rt.shape)
(2, None, 3)

Uniform Outer Dimensions

RaggedTensors with uniform outer dimensions can be defined by using one or more RaggedTensor with a uniform_row_length row-partitioning tensor. For example, a RaggedTensor with shape [2, 2, None] can be constructed with this method from a RaggedTensor values with shape [4, None]:

values = tf.ragged.constant([[1, 2, 3], [4], [5, 6], [7, 8, 9, 10]])
print(values.shape)
(4, None)
rt6 = tf.RaggedTensor.from_uniform_row_length(values, 2)
print(rt6)
<tf.RaggedTensor [[[1, 2, 3], [4]], [[5, 6], [7, 8, 9, 10]]]>
print(rt6.shape)
(2, 2, None)

Note that rt6 only contains one ragged dimension (the innermost dimension). In contrast, if from_row_splits is used to construct a similar RaggedTensor, then that RaggedTensor will have two ragged dimensions:

rt7 = tf.RaggedTensor.from_row_splits(values, [0, 2, 4])
print(rt7.shape)
(2, None, None)

Uniform and ragged outer dimensions may be interleaved, meaning that a tensor with any combination of ragged and uniform dimensions may be created. For example, a RaggedTensor t4 with shape [3, None, 4, 8, None, 2] could be constructed as follows:

t0 = tf.zeros([1000, 2])                           # Shape:         [1000, 2]
t1 = RaggedTensor.from_row_lengths(t0, [...])      #           [160, None, 2]
t2 = RaggedTensor.from_uniform_row_length(t1, 8)   #         [20, 8, None, 2]
t3 = RaggedTensor.from_uniform_row_length(t2, 4)   #       [5, 4, 8, None, 2]
t4 = RaggedTensor.from_row_lengths(t3, [...])      # [3, None, 4, 8, None, 2]

values A potentially ragged tensor of any dtype and shape [nvals, ...].
row_splits A 1-D integer tensor with shape [nrows+1].
cached_row_lengths A 1-D integer tensor with shape [nrows]
cached_value_rowids A 1-D integer tensor with shape [nvals].
cached_nrows A 1-D integer scalar tensor.
internal True if the constructor is being called by one of the factory methods. If false, an exception will be raised.
uniform_row_length A scalar tensor.

TypeError If a row partitioning tensor has an inappropriate dtype.
TypeError If exactly one row partitioning argument was not specified.
ValueError If a row partitioning tensor has an inappropriate shape.
ValueError If multiple partitioning arguments are specified.
ValueError If nrows is specified but value_rowids is not None.

dtype The DType of values in this tensor.
flat_values The innermost values tensor for this ragged tensor.

Concretely, if rt.values is a Tensor, then rt.flat_values is rt.values; otherwise, rt.flat_values is rt.values.flat_values.

Conceptually, flat_values is the tensor formed by flattening the outermost dimension and all of the ragged dimensions into a single dimension.

rt.flat_values.shape = [nvals] + rt.shape[rt.ragged_rank + 1:] (where nvals is the number of items in the flattened dimensions).

Example:

rt = tf.ragged.constant([[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]])
print(rt.flat_values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)

nested_row_splits A tuple containing the row_splits for all ragged dimensions.

rt.nested_row_splits is a tuple containing the row_splits tensors for all ragged dimensions in rt, ordered from outermost to innermost. In particular, rt.nested_row_splits = (rt.row_splits,) + value_splits where:

  • value_splits = () if rt.values is a Tensor.
  • value_splits = rt.values.nested_row_splits otherwise.

Example:

rt = tf.ragged.constant(
    [[[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]]])
for i, splits in enumerate(rt.nested_row_splits):
  print('Splits for dimension %d: %s' % (i+1, splits.numpy()))
Splits for dimension 1: [0 3]
Splits for dimension 2: [0 3 3 5]
Splits for dimension 3: [0 4 4 7 8 8]

ragged_rank The number of ragged dimensions in this ragged tensor.
row_splits The row-split indices for this ragged tensor's values.

rt.row_splits specifies where the values for each row begin and end in rt.values. In particular, the values for row rt[i] are stored in the slice rt.values[rt.row_splits[i]:rt.row_splits[i+1]].

Example:

rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
print(rt.row_splits)  # indices of row splits in rt.values
tf.Tensor([0 4 4 7 8 8], shape=(6,), dtype=int64)

shape The statically known shape of this ragged tensor.

tf.ragged.constant([[0], [1, 2]]).shape
TensorShape([2, None])
tf.ragged.constant([[[0, 1]], [[1, 2], [3, 4]]], ragged_rank=1).shape
TensorShape([2, None, 2])

values The concatenated rows for this ragged tensor.

rt.values is a potentially ragged tensor formed by flattening the two outermost dimensions of rt into a single dimension.

rt.values.shape = [nvals] + rt.shape[2:] (where nvals is the number of items in the outer two dimensions of rt).

rt.ragged_rank = self.ragged_rank - 1

Example:

rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
print(rt.values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)

Methods

bounding_shape

View source

Returns the tight bounding box shape for this RaggedTensor.

Args
axis An integer scalar or vector indicating which axes to return the bounding box for. If not specified, then the full bounding box is returned.
name A name prefix for the returned tensor (optional).
out_type dtype for the returned tensor. Defaults to self.row_splits.dtype.

Returns
An integer Tensor (dtype=self.row_splits.dtype). If axis is not specified, then output is a vector with output.shape=[self.shape.ndims]. If axis is a scalar, then the output is a scalar. If axis is a vector, then output is a vector, where output[i] is the bounding size for dimension axis[i].

Example:

rt = tf.ragged.constant([[1, 2, 3, 4], [5], [], [6, 7, 8, 9], [10]])
rt.bounding_shape().numpy()
array([5, 4])

consumers

View source

from_nested_row_lengths

View source

Creates a RaggedTensor from a nested list of row_lengths tensors.

Equivalent to:

result = flat_values
for row_lengths in reversed(nested_row_lengths):
  result = from_row_lengths(result, row_lengths)

Args
flat_values A potentially ragged tensor.
nested_row_lengths A list of 1-D integer tensors. The ith tensor is used as the row_lengths for the ith ragged dimension.
name A name prefix for the RaggedTensor (optional).
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.

Returns
A RaggedTensor (or flat_values if nested_row_lengths is empty).

from_nested_row_splits

View source

Creates a RaggedTensor from a nested list of row_splits tensors.

Equivalent to:

result = flat_values
for row_splits in reversed(nested_row_splits):
  result = from_row_splits(result, row_splits)

Args
flat_values A potentially ragged tensor.
nested_row_splits A list of 1-D integer tensors. The ith tensor is used as the row_splits for the ith ragged dimension.
name A name prefix for the RaggedTensor (optional).
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.

Returns
A RaggedTensor (or flat_values if nested_row_splits is empty).

from_nested_value_rowids

View source

Creates a RaggedTensor from a nested list of value_rowids tensors.

Equivalent to:

result = flat_values
for (rowids, nrows) in reversed(zip(nested_value_rowids, nested_nrows)):
  result = from_value_rowids(result, rowids, nrows)

Args
flat_values A potentially ragged tensor.
nested_value_rowids A list of 1-D integer tensors. The ith tensor is used as the value_rowids for the ith ragged dimension.
nested_nrows A list of integer scalars. The ith scalar is used as the nrows for the ith ragged dimension.
name A name prefix for the RaggedTensor (optional).
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.

Returns
A RaggedTensor (or flat_values if nested_value_rowids is empty).

Raises
ValueError If len(nested_values_rowids) != len(nested_nrows).

from_row_lengths

View source

Creates a RaggedTensor with rows partitioned by row_lengths.

The returned RaggedTensor corresponds with the python list defined by:

result = [[values.pop(0) for i in range(length)]
          for length in row_lengths]

Args
values A potentially ragged tensor with shape [nvals, ...].
row_lengths A 1-D integer tensor with shape [nrows]. Must be nonnegative. sum(row_lengths) must be nvals.
name A name prefix for the RaggedTensor (optional).
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.

Returns
A RaggedTensor. result.rank = values.rank + 1. result.ragged_rank = values.ragged_rank + 1.

Example:

print(tf.RaggedTensor.from_row_lengths(
    values=[3, 1, 4, 1, 5, 9, 2, 6],
    row_lengths=[4, 0, 3, 1, 0]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>

from_row_limits

View source

Creates a RaggedTensor with rows partitioned by row_limits.

Equivalent to: from_row_splits(values, concat([0, row_limits])).

Args
values A potentially ragged tensor with shape [nvals, ...].
row_limits A 1-D integer tensor with shape [nrows]. Must be sorted in ascending order. If nrows>0, then row_limits[-1] must be nvals.
name A name prefix for the RaggedTensor (optional).
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.

Returns
A RaggedTensor. result.rank = values.rank + 1. result.ragged_rank = values.ragged_rank + 1.

Example:

print(tf.RaggedTensor.from_row_limits(
    values=[3, 1, 4, 1, 5, 9, 2, 6],
    row_limits=[4, 4, 7, 8, 8]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>

from_row_splits

View source

Creates a RaggedTensor with rows partitioned by row_splits.

The returned RaggedTensor corresponds with the python list defined by:

result = [values[row_splits[i]:row_splits[i + 1]]
          for i in range(len(row_splits) - 1)]

Args
values A potentially ragged tensor with shape [nvals, ...].
row_splits A 1-D integer tensor with shape [nrows+1]. Must not be empty, and must be sorted in ascending order. row_splits[0] must be zero and row_splits[-1] must be nvals.
name A name prefix for the RaggedTensor (optional).
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.

Returns
A RaggedTensor. result.rank = values.rank + 1. result.ragged_rank = values.ragged_rank + 1.

Raises
ValueError If row_splits is an empty list.

Example:

print(tf.RaggedTensor.from_row_splits(
    values=[3, 1, 4, 1, 5, 9, 2, 6],
    row_splits=[0, 4, 4, 7, 8, 8]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>

from_row_starts

View source

Creates a RaggedTensor with rows partitioned by row_starts.

Equivalent to: from_row_splits(values, concat([row_starts, nvals])).

Args
values A potentially ragged tensor with shape [nvals, ...].
row_starts A 1-D integer tensor with shape [nrows]. Must be nonnegative and sorted in ascending order. If nrows>0, then row_starts[0] must be zero.
name A name prefix for the RaggedTensor (optional).
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.

Returns
A RaggedTensor. result.rank = values.rank + 1. result.ragged_rank = values.ragged_rank + 1.

Example:

print(tf.RaggedTensor.from_row_starts(
    values=[3, 1, 4, 1, 5, 9, 2, 6],
    row_starts=[0, 4, 4, 7, 8]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>

from_sparse

View source

Converts a 2D tf.SparseTensor to a RaggedTensor.

Each row of the output RaggedTensor will contain the explicit values from the same row in st_input. st_input must be ragged-right. If not it is not ragged-right, then an error will be generated.

Example:

st = tf.SparseTensor(indices=[[0, 0], [0, 1], [0, 2], [1, 0], [3, 0]],
                     values=[1, 2, 3, 4, 5],
                     dense_shape=[4, 3])
tf.RaggedTensor.from_sparse(st).to_list()
[[1, 2, 3], [4], [], [5]]

Currently, only two-dimensional SparseTensors are supported.

Args
st_input The sparse tensor to convert. Must have rank 2.
name A name prefix for the returned tensors (optional).
row_splits_dtype dtype for the returned RaggedTensor's row_splits tensor. One of tf.int32 or tf.int64.

Returns
A RaggedTensor with the same values as st_input. output.ragged_rank = rank(st_input) - 1. output.shape = [st_input.dense_shape[0], None].

Raises
ValueError If the number of dimensions in st_input is not known statically, or is not two.

from_tensor

View source

Converts a tf.Tensor into a RaggedTensor.

The set of absent/default values may be specified using a vector of lengths or a padding value (but not both). If lengths is specified, then the output tensor will satisfy output[row] = tensor[row][:lengths[row]]. If 'lengths' is a list of lists or tuple of lists, those lists will be used as nested row lengths. If padding is specified, then any row suffix consisting entirely of padding will be excluded from the returned RaggedTensor. If neither lengths nor padding is specified, then the returned RaggedTensor will have no absent/default values.

Examples:

dt = tf.constant([[5, 7, 0], [0, 3, 0], [6, 0, 0]])
tf.RaggedTensor.from_tensor(dt)
<tf.RaggedTensor [[5, 7, 0], [0, 3, 0], [6, 0, 0]]>
tf.RaggedTensor.from_tensor(dt, lengths=[1, 0, 3])
<tf.RaggedTensor [[5], [], [6, 0, 0]]>
tf.RaggedTensor.from_tensor(dt, padding=0)
<tf.RaggedTensor [[5, 7], [0, 3], [6]]>
dt = tf.constant([[[5, 0], [7, 0], [0, 0]],
                  [[0, 0], [3, 0], [0, 0]],
                  [[6, 0], [0, 0], [0, 0]]])
tf.RaggedTensor.from_tensor(dt, lengths=([2, 0, 3], [1, 1, 2, 0, 1]))
<tf.RaggedTensor [[[5], [7]], [], [[6, 0], [], [0]]]>

Args
tensor The Tensor to convert. Must have rank ragged_rank + 1 or higher.
lengths An optional set of row lengths, specified using a 1-D integer Tensor whose length is equal to tensor.shape[0] (the number of rows in tensor). If specified, then output[row] will contain tensor[row][:lengths[row]]. Negative lengths are treated as zero. You may optionally pass a list or tuple of lengths to this argument, which will be used as nested row lengths to construct a ragged tensor with multiple ragged dimensions.
padding An optional padding value. If specified, then any row suffix consisting entirely of padding will be excluded from the returned RaggedTensor. padding is a Tensor with the same dtype as tensor and with shape=tensor.shape[ragged_rank + 1:].
ragged_rank Integer specifying the ragged rank for the returned RaggedTensor. Must be greater than zero.
name A name prefix for the returned tensors (optional).
row_splits_dtype dtype for the returned RaggedTensor's row_splits tensor. One of tf.int32 or tf.int64.

Returns
A RaggedTensor with the specified ragged_rank. The shape of the returned ragged tensor is compatible with the shape of tensor.

Raises
ValueError If both lengths and padding are specified.

from_uniform_row_length

View source

Creates a RaggedTensor with rows partitioned by uniform_row_length.

This method can be used to create RaggedTensors with multiple uniform outer dimensions. For example, a RaggedTensor with shape [2, 2, None] can be constructed with this method from a RaggedTensor values with shape [4, None]:

values = tf.ragged.constant([[1, 2, 3], [4], [5, 6], [7, 8, 9, 10]])
print(values.shape)
(4, None)
rt1 = tf.RaggedTensor.from_uniform_row_length(values, 2)
print(rt1)
<tf.RaggedTensor [[[1, 2, 3], [4]], [[5, 6], [7, 8, 9, 10]]]>
print(rt1.shape)
(2, 2, None)

Note that rt1 only contains one ragged dimension (the innermost dimension). In contrast, if from_row_splits is used to construct a similar RaggedTensor, then that RaggedTensor will have two ragged dimensions:

rt2 = tf.RaggedTensor.from_row_splits(values, [0, 2, 4])
print(rt2.shape)
(2, None, None)

Args
values A potentially ragged tensor with shape [nvals, ...].
uniform_row_length A scalar integer tensor. Must be nonnegative. The size of the outer axis of values must be evenly divisible by uniform_row_length.
nrows The number of rows in the constructed RaggedTensor. If not specified, then it defaults to nvals/uniform_row_length (or 0 if uniform_row_length==0). nrows only needs to be specified if uniform_row_length might be zero. uniform_row_length*nrows must be nvals.
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.
name A name prefix for the RaggedTensor (optional).

Returns
A RaggedTensor that corresponds with the python list defined by:

result = [[values.pop(0) for i in range(uniform_row_length)]
for _ in range(nrows)]

result.rank = values.rank + 1. result.ragged_rank = values.ragged_rank + 1.

from_value_rowids

View source

Creates a RaggedTensor with rows partitioned by value_rowids.

The returned RaggedTensor corresponds with the python list defined by:

result = [[values[i] for i in range(len(values)) if value_rowids[i] == row]
          for row in range(nrows)]

Args
values A potentially ragged tensor with shape [nvals, ...].
value_rowids A 1-D integer tensor with shape [nvals], which corresponds one-to-one with values, and specifies each value's row index. Must be nonnegative, and must be sorted in ascending order.
nrows An integer scalar specifying the number of rows. This should be specified if the RaggedTensor may containing empty training rows. Must be greater than value_rowids[-1] (or zero if value_rowids is empty). Defaults to value_rowids[-1] (or zero if value_rowids is empty).
name A name prefix for the RaggedTensor (optional).
validate If true, then use assertions to check that the arguments form a valid RaggedTensor.

Returns
A RaggedTensor. result.rank = values.rank + 1. result.ragged_rank = values.ragged_rank + 1.

Raises
ValueError If nrows is incompatible with value_rowids.

Example:

print(tf.RaggedTensor.from_value_rowids(
    values=[3, 1, 4, 1, 5, 9, 2, 6],
    value_rowids=[0, 0, 0, 0, 2, 2, 2, 3],
    nrows=5))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>

merge_dims

View source

Merges outer_axis...inner_axis into a single dimension.

Returns a copy of this RaggedTensor with the specified range of dimensions flattened into a single dimension, with elements in row-major order.

Examples:

rt = tf.ragged.constant([[[1, 2], [3]], [[4, 5, 6]]])
print(rt.merge_dims(0, 1))
<tf.RaggedTensor [[1, 2], [3], [4, 5, 6]]>
print(rt.merge_dims(1, 2))
<tf.RaggedTensor [[1, 2, 3], [4, 5, 6]]>
print(rt.merge_dims(0, 2))
tf.Tensor([1 2 3 4 5 6], shape=(6,), dtype=int32)

To mimic the behavior of np.flatten (which flattens all dimensions), use rt.merge_dims(0, -1). To mimic the behavior oftf.layers.Flatten(which flattens all dimensions except the outermost batch dimension), usert.merge_dims(1, -1)`.

Args
outer_axis int: The first dimension in the range of dimensions to merge. May be negative if self.shape.rank is statically known.
inner_axis int: The last dimension in the range of dimensions to merge. May be negative if self.shape.rank is statically known.

Returns
A copy of this tensor, with the specified dimensions merged into a single dimension. The shape of the returned tensor will be self.shape[:outer_axis] + [N] + self.shape[inner_axis + 1:], where N is the total number of slices in the merged dimensions.

nested_row_lengths

View source

Returns a tuple containing the row_lengths for all ragged dimensions.

rt.nested_row_lengths() is a tuple containing the row_lengths tensors for all ragged dimensions in rt, ordered from outermost to innermost.

Args
name A name prefix for the returned tensors (optional).

Returns
A tuple of 1-D integer Tensors. The length of the tuple is equal to self.ragged_rank.

nested_value_rowids

View source

Returns a tuple containing the value_rowids for all ragged dimensions.

rt.nested_value_rowids is a tuple containing the value_rowids tensors for all ragged dimensions in rt, ordered from outermost to innermost. In particular, rt.nested_value_rowids = (rt.value_rowids(),) + value_ids where:

* `value_ids = ()` if `rt.values` is a `Tensor`.
* `value_ids = rt.values.nested_value_rowids` otherwise.

Args
name A name prefix for the returned tensors (optional).

Returns
A tuple of 1-D integer Tensors.

Example:

rt = tf.ragged.constant(
    [[[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]]])
for i, ids in enumerate(rt.nested_value_rowids()):
  print('row ids for dimension %d: %s' % (i+1, ids.numpy()))
row ids for dimension 1: [0 0 0]
row ids for dimension 2: [0 0 0 2 2]
row ids for dimension 3: [0 0 0 0 2 2 2 3]

nrows

View source

Returns the number of rows in this ragged tensor.

I.e., the size of the outermost dimension of the tensor.

Args
out_type dtype for the returned tensor. Defaults to self.row_splits.dtype.
name A name prefix for the returned tensor (optional).

Returns
A scalar Tensor with dtype out_type.

Example:

rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
print(rt.nrows())  # rt has 5 rows.
tf.Tensor(5, shape=(), dtype=int64)

row_lengths

View source

Returns the lengths of the rows in this ragged tensor.

rt.row_lengths()[i] indicates the number of values in the ith row of rt.

Args
axis An integer constant indicating the axis whose row lengths should be returned.
name A name prefix for the returned tensor (optional).

Returns
A potentially ragged integer Tensor with shape self.shape[:axis].

Raises
ValueError If axis is out of bounds.

Example:

rt = tf.ragged.constant(
    [[[3, 1, 4], [1]], [], [[5, 9], [2]], [[6]], []])
print(rt.row_lengths())  # lengths of rows in rt
tf.Tensor([2 0 2 1 0], shape=(5,), dtype=int64)
print(rt.row_lengths(axis=2))  # lengths of axis=2 rows.
<tf.RaggedTensor [[3, 1], [], [2, 1], [1], []]>

row_limits

View source

Returns the limit indices for rows in this ragged tensor.

These indices specify where the values for each row end in self.values. rt.row_limits(self) is equal to rt.row_splits[:-1].

Args
name A name prefix for the returned tensor (optional).

Returns
A 1-D integer Tensor with shape [nrows]. The returned tensor is nonnegative, and is sorted in ascending order.

Example:

rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
print(rt.values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)
print(rt.row_limits())  # indices of row limits in rt.values
tf.Tensor([4 4 7 8 8], shape=(5,), dtype=int64)

row_starts

View source

Returns the start indices for rows in this ragged tensor.

These indices specify where the values for each row begin in self.values. rt.row_starts() is equal to rt.row_splits[:-1].

Args
name A name prefix for the returned tensor (optional).

Returns
A 1-D integer Tensor with shape [nrows]. The returned tensor is nonnegative, and is sorted in ascending order.

Example:

rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
print(rt.values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)
print(rt.row_starts())  # indices of row starts in rt.values
tf.Tensor([0 4 4 7 8], shape=(5,), dtype=int64)

to_list

View source

Returns a nested Python list with the values for this RaggedTensor.

Requires that rt was constructed in eager execution mode.

Returns
A nested Python list.

to_sparse

View source

Converts this RaggedTensor into a tf.SparseTensor.

Example:

rt = tf.ragged.constant([[1, 2, 3], [4], [], [5, 6]])
print(rt.to_sparse())
SparseTensor(indices=tf.Tensor(
                 [[0 0] [0 1] [0 2] [1 0] [3 0] [3 1]],
                 shape=(6, 2), dtype=int64),
             values=tf.Tensor([1 2 3 4 5 6], shape=(6,), dtype=int32),
             dense_shape=tf.Tensor([4 3], shape=(2,), dtype=int64))

Args
name A name prefix for the returned tensors (optional).

Returns
A SparseTensor with the same values as self.

to_tensor

View source

Converts this RaggedTensor into a tf.Tensor.

If shape is specified, then the result is padded and/or truncated to the specified shape.

Examples:

rt = tf.ragged.constant([[9, 8, 7], [], [6, 5], [4]])
print(rt.to_tensor())
tf.Tensor(
    [[9 8 7] [0 0 0] [6 5 0] [4 0 0]], shape=(4, 3), dtype=int32)
print(rt.to_tensor(shape=[5, 2]))
tf.Tensor(
    [[9 8] [0 0] [6 5] [4 0] [0 0]], shape=(5, 2), dtype=int32)

Args
default_value Value to set for indices not specified in self. Defaults to zero. default_value must be broadcastable to self.shape[self.ragged_rank + 1:].
name A name prefix for the returned tensors (optional).
shape The shape of the resulting dense tensor. In particular, result.shape[i] is shape[i] (if shape[i] is not None), or self.bounding_shape(i) (otherwise).shape.rank must be None or equal to self.rank.

Returns
A Tensor with shape ragged.bounding_shape(self) and the values specified by the non-empty values in self. Empty values are assigned default_value.

value_rowids

View source

Returns the row indices for the values in this ragged tensor.

rt.value_rowids() corresponds one-to-one with the outermost dimension of rt.values, and specifies the row containing each value. In particular, the row rt[row] consists of the values rt.values[j] where rt.value_rowids()[j] == row.

Args
name A name prefix for the returned tensor (optional).

Returns
A 1-D integer Tensor with shape self.values.shape[:1]. The returned tensor is nonnegative, and is sorted in ascending order.

Example:

rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
print(rt.values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)
print(rt.value_rowids())  # corresponds 1:1 with rt.values
tf.Tensor([0 0 0 0 2 2 2 3], shape=(8,), dtype=int64)

with_flat_values

View source

Returns a copy of self with flat_values replaced by new_value.

Preserves cached row-partitioning tensors such as self.cached_nrows and self.cached_value_rowids if they have values.

Args
new_values Potentially ragged tensor that should replace self.flat_values. Must have rank > 0, and must have the same number of rows as self.flat_values.

Returns
A RaggedTensor. result.rank = self.ragged_rank + new_values.rank. result.ragged_rank = self.ragged_rank + new_values.ragged_rank.

with_row_splits_dtype

View source

Returns a copy of this RaggedTensor with the given row_splits dtype.

For RaggedTensors with multiple ragged dimensions, the row_splits for all nested RaggedTensor objects are cast to the given dtype.

Args
dtype The dtype for row_splits. One of tf.int32 or tf.int64.

Returns
A copy of this RaggedTensor, with the row_splits cast to the given type.

with_values

View source

Returns a copy of self with values replaced by new_value.

Preserves cached row-partitioning tensors such as self.cached_nrows and self.cached_value_rowids if they have values.

Args
new_values Potentially ragged tensor to use as the values for the returned RaggedTensor. Must have rank > 0, and must have the same number of rows as self.values.

Returns
A RaggedTensor. result.rank = 1 + new_values.rank. result.ragged_rank = 1 + new_values.ragged_rank

__abs__

View source

Computes the absolute value of a tensor.

Given a tensor of integer or floating-point values, this operation returns a tensor of the same type, where each element contains the absolute value of the corresponding element in the input.

Given a tensor x of complex numbers, this operation returns a tensor of type float32 or float64 that is the absolute value of each element in x. All elements in x must be complex numbers of the form \(a + bj\). The absolute value is computed as \( \sqrt{a^2 + b^2}\). For example:

x = tf.constant([[-2.25 + 4.75j], [-3.25 + 5.75j]])
tf.abs(x)  # [5.25594902, 6.60492229]

Args
x A Tensor or SparseTensor of type float16, float32, float64, int32, int64, complex64 or complex128.
name A name for the operation (optional).

Returns
A Tensor or SparseTensor the same size, type, and sparsity as x with absolute values. Note, for complex64 or complex128 input, the returned Tensor will be of type float32 or float64, respectively.

If x is a SparseTensor, returns SparseTensor(x.indices, tf.math.abs(x.values, ...), x.dense_shape)

__add__

Returns x + y element-wise.

Args
x A Tensor. Must be one of the following types: bfloat16, half, float32, float64, uint8, int8, int16, int32, int64, complex64, complex128, string.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

__and__

Returns the truth value of x AND y element-wise.

Args
x A Tensor of type bool.
y A Tensor of type bool.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__bool__

View source

Dummy method to prevent a RaggedTensor from being used as a Python bool.

__div__

View source

Divides x / y elementwise (using Python 2 division operator semantics). (deprecated)

This function divides x and y, forcing Python 2 semantics. That is, if x and y are both integers then the result will be an integer. This is in contrast to Python 3, where division with / is always a float while division with // is always an integer.

Args
x Tensor numerator of real numeric type.
y Tensor denominator of real numeric type.
name A name for the operation (optional).

Returns
x / y returns the quotient of x and y.

__floordiv__

View source

Divides x / y elementwise, rounding toward the most negative integer.

The same as tf.compat.v1.div(x,y) for integers, but uses tf.floor(tf.compat.v1.div(x,y)) for floating point arguments so that the result is always an integer (though possibly an integer represented as floating point). This op is generated by x // y floor division in Python 3 and in Python 2.7 with from __future__ import division.

x and y must have the same type, and the result will have the same type as well.

Args
x Tensor numerator of real numeric type.
y Tensor denominator of real numeric type.
name A name for the operation (optional).

Returns
x / y rounded down.

Raises
TypeError If the inputs are complex.

__ge__

Returns the truth value of (x >= y) element-wise.

Example:

x = tf.constant([5, 4, 6, 7])
y = tf.constant([5, 2, 5, 10])
tf.math.greater_equal(x, y) ==> [True, True, True, False]

x = tf.constant([5, 4, 6, 7])
y = tf.constant([5])
tf.math.greater_equal(x, y) ==> [True, False, True, True]

Args
x A Tensor. Must be one of the following types: float32, float64, int32, uint8, int16, int8, int64, bfloat16, uint16, half, uint32, uint64.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__getitem__

View source

Returns the specified piece of this RaggedTensor.

Supports multidimensional indexing and slicing, with one restriction: indexing into a ragged inner dimension is not allowed. This case is problematic because the indicated value may exist in some rows but not others. In such cases, it's not obvious whether we should (1) report an IndexError; (2) use a default value; or (3) skip that value and return a tensor with fewer rows than we started with. Following the guiding principles of Python ("In the face of ambiguity, refuse the temptation to guess"), we simply disallow this operation.

Any dimensions added by array_ops.newaxis will be ragged if the following dimension is ragged.

Args
self The RaggedTensor to slice.
key Indicates which piece of the RaggedTensor to return, using standard Python semantics (e.g., negative values index from the end). key may have any of the following types:

  • int constant
  • Scalar integer Tensor
  • slice containing integer constants and/or scalar integer Tensors
  • Ellipsis
  • tf.newaxis
  • tuple containing any of the above (for multidimentional indexing)

Returns
A Tensor or RaggedTensor object. Values that include at least one ragged dimension are returned as RaggedTensor. Values that include no ragged dimensions are returned as Tensor. See above for examples of expressions that return Tensors vs RaggedTensors.

Raises
ValueError If key is out of bounds.
ValueError If key is not supported.
TypeError If the indices in key have an unsupported type.

Examples:

# A 2-D ragged tensor with 1 ragged dimension.
rt = tf.ragged.constant([['a', 'b', 'c'], ['d', 'e'], ['f'], ['g']])
rt[0].numpy()                 # First row (1-D `Tensor`)
array([b'a', b'b', b'c'], dtype=object)
rt[:3].to_list()              # First three rows (2-D RaggedTensor)
[[b'a', b'b', b'c'], [b'd', b'e'], [b'f']]
rt[3, 0].numpy()              # 1st element of 4th row (scalar)
b'g'
# A 3-D ragged tensor with 2 ragged dimensions.
rt = tf.ragged.constant([[[1, 2, 3], [4]],
                         [[5], [], [6]],
                         [[7]],
                         [[8, 9], [10]]])
rt[1].to_list()               # Second row (2-D RaggedTensor)
[[5], [], [6]]
rt[3, 0].numpy()              # First element of fourth row (1-D Tensor)
array([8, 9], dtype=int32)
rt[:, 1:3].to_list()          # Items 1-3 of each row (3-D RaggedTensor)
[[[4]], [[], [6]], [], [[10]]]
rt[:, -1:].to_list()          # Last item of each row (3-D RaggedTensor)
[[[4]], [[6]], [[7]], [[10]]]

__gt__

Returns the truth value of (x > y) element-wise.

Example:

x = tf.constant([5, 4, 6])
y = tf.constant([5, 2, 5])
tf.math.greater(x, y) ==> [False, True, True]

x = tf.constant([5, 4, 6])
y = tf.constant([5])
tf.math.greater(x, y) ==> [False, False, True]

Args
x A Tensor. Must be one of the following types: float32, float64, int32, uint8, int16, int8, int64, bfloat16, uint16, half, uint32, uint64.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__invert__

Returns the truth value of NOT x element-wise.

Args
x A Tensor of type bool.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__le__

Returns the truth value of (x <= y) element-wise.

Example:

x = tf.constant([5, 4, 6])
y = tf.constant([5])
tf.math.less_equal(x, y) ==> [True, True, False]

x = tf.constant([5, 4, 6])
y = tf.constant([5, 6, 6])
tf.math.less_equal(x, y) ==> [True, True, True]

Args
x A Tensor. Must be one of the following types: float32, float64, int32, uint8, int16, int8, int64, bfloat16, uint16, half, uint32, uint64.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__lt__

Returns the truth value of (x < y) element-wise.

Example:

x = tf.constant([5, 4, 6])
y = tf.constant([5])
tf.math.less(x, y) ==> [False, True, False]

x = tf.constant([5, 4, 6])
y = tf.constant([5, 6, 7])
tf.math.less(x, y) ==> [False, True, True]

Args
x A Tensor. Must be one of the following types: float32, float64, int32, uint8, int16, int8, int64, bfloat16, uint16, half, uint32, uint64.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__mod__

Returns element-wise remainder of division. When x < 0 xor y < 0 is

true, this follows Python semantics in that the result here is consistent with a flooring divide. E.g. floor(x / y) * y + mod(x, y) = x.

Args
x A Tensor. Must be one of the following types: int32, int64, bfloat16, half, float32, float64.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

__mul__

View source

Returns x * y element-wise.

Args
x A Tensor. Must be one of the following types: bfloat16, half, float32, float64, uint8, int8, uint16, int16, int32, int64, complex64, complex128.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

__neg__

Computes numerical negative value element-wise.

I.e., \(y = -x\).

Args
x A Tensor. Must be one of the following types: bfloat16, half, float32, float64, int32, int64, complex64, complex128.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

If x is a SparseTensor, returns SparseTensor(x.indices, tf.math.negative(x.values, ...), x.dense_shape)

__nonzero__

View source

Dummy method to prevent a RaggedTensor from being used as a Python bool.

__or__

Returns the truth value of x OR y element-wise.

Args
x A Tensor of type bool.
y A Tensor of type bool.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__pow__

View source

Computes the power of one value to another.

Given a tensor x and a tensor y, this operation computes \(x^y\) for corresponding elements in x and y. For example:

x = tf.constant([[2, 2], [3, 3]])
y = tf.constant([[8, 16], [2, 3]])
tf.pow(x, y)  # [[256, 65536], [9, 27]]

Args
x A Tensor of type float16, float32, float64, int32, int64, complex64, or complex128.
y A Tensor of type float16, float32, float64, int32, int64, complex64, or complex128.
name A name for the operation (optional).

Returns
A Tensor.

__radd__

Returns x + y element-wise.

Args
x A Tensor. Must be one of the following types: bfloat16, half, float32, float64, uint8, int8, int16, int32, int64, complex64, complex128, string.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

__rand__

Returns the truth value of x AND y element-wise.

Args
x A Tensor of type bool.
y A Tensor of type bool.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__rdiv__

View source

Divides x / y elementwise (using Python 2 division operator semantics). (deprecated)

This function divides x and y, forcing Python 2 semantics. That is, if x and y are both integers then the result will be an integer. This is in contrast to Python 3, where division with / is always a float while division with // is always an integer.

Args
x Tensor numerator of real numeric type.
y Tensor denominator of real numeric type.
name A name for the operation (optional).

Returns
x / y returns the quotient of x and y.

__rfloordiv__

View source

Divides x / y elementwise, rounding toward the most negative integer.

The same as tf.compat.v1.div(x,y) for integers, but uses tf.floor(tf.compat.v1.div(x,y)) for floating point arguments so that the result is always an integer (though possibly an integer represented as floating point). This op is generated by x // y floor division in Python 3 and in Python 2.7 with from __future__ import division.

x and y must have the same type, and the result will have the same type as well.

Args
x Tensor numerator of real numeric type.
y Tensor denominator of real numeric type.
name A name for the operation (optional).

Returns
x / y rounded down.

Raises
TypeError If the inputs are complex.

__rmod__

Returns element-wise remainder of division. When x < 0 xor y < 0 is

true, this follows Python semantics in that the result here is consistent with a flooring divide. E.g. floor(x / y) * y + mod(x, y) = x.

Args
x A Tensor. Must be one of the following types: int32, int64, bfloat16, half, float32, float64.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

__rmul__

View source

Returns x * y element-wise.

Args
x A Tensor. Must be one of the following types: bfloat16, half, float32, float64, uint8, int8, uint16, int16, int32, int64, complex64, complex128.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

__ror__

Returns the truth value of x OR y element-wise.

Args
x A Tensor of type bool.
y A Tensor of type bool.
name A name for the operation (optional).

Returns
A Tensor of type bool.

__rpow__

View source

Computes the power of one value to another.

Given a tensor x and a tensor y, this operation computes \(x^y\) for corresponding elements in x and y. For example:

x = tf.constant([[2, 2], [3, 3]])
y = tf.constant([[8, 16], [2, 3]])
tf.pow(x, y)  # [[256, 65536], [9, 27]]

Args
x A Tensor of type float16, float32, float64, int32, int64, complex64, or complex128.
y A Tensor of type float16, float32, float64, int32, int64, complex64, or complex128.
name A name for the operation (optional).

Returns
A Tensor.

__rsub__

View source

Returns x - y element-wise.

Args
x A Tensor. Must be one of the following types: bfloat16, half, float32, float64, uint8, int8, uint16, int16, int32, int64, complex64, complex128.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

__rtruediv__

View source

Divides x / y elementwise (using Python 3 division operator semantics).

This function forces Python 3 division operator semantics where all integer arguments are cast to floating types first. This op is generated by normal x / y division in Python 3 and in Python 2.7 with from __future__ import division. If you want integer division that rounds down, use x // y or tf.math.floordiv.

x and y must have the same numeric type. If the inputs are floating point, the output will have the same type. If the inputs are integral, the inputs are cast to float32 for int8 and int16 and float64 for int32 and int64 (matching the behavior of Numpy).

Args
x Tensor numerator of numeric type.
y Tensor denominator of numeric type.
name A name for the operation (optional).

Returns
x / y evaluated in floating point.

Raises
TypeError If x and y have different dtypes.

__rxor__

View source

Logical XOR function.

x ^ y = (x | y) & ~(x & y)

Inputs are tensor and if the tensors contains more than one element, an element-wise logical XOR is computed.

Usage:

x = tf.constant([False, False, True, True], dtype = tf.bool)
y = tf.constant([False, True, False, True], dtype = tf.bool)
z = tf.logical_xor(x, y, name="LogicalXor")
#  here z = [False  True  True False]

Args
x A Tensor type bool.
y A Tensor of type bool.

Returns
A Tensor of type bool with the same size as that of x or y.

__sub__

View source

Returns x - y element-wise.

Args
x A Tensor. Must be one of the following types: bfloat16, half, float32, float64, uint8, int8, uint16, int16, int32, int64, complex64, complex128.
y A Tensor. Must have the same type as x.
name A name for the operation (optional).

Returns
A Tensor. Has the same type as x.

__truediv__

View source

Divides x / y elementwise (using Python 3 division operator semantics).

This function forces Python 3 division operator semantics where all integer arguments are cast to floating types first. This op is generated by normal x / y division in Python 3 and in Python 2.7 with from __future__ import division. If you want integer division that rounds down, use x // y or tf.math.floordiv.

x and y must have the same numeric type. If the inputs are floating point, the output will have the same type. If the inputs are integral, the inputs are cast to float32 for int8 and int16 and float64 for int32 and int64 (matching the behavior of Numpy).

Args
x Tensor numerator of numeric type.
y Tensor denominator of numeric type.
name A name for the operation (optional).

Returns
x / y evaluated in floating point.

Raises
TypeError If x and y have different dtypes.

__xor__

View source

Logical XOR function.

x ^ y = (x | y) & ~(x & y)

Inputs are tensor and if the tensors contains more than one element, an element-wise logical XOR is computed.

Usage:

x = tf.constant([False, False, True, True], dtype = tf.bool)
y = tf.constant([False, True, False, True], dtype = tf.bool)
z = tf.logical_xor(x, y, name="LogicalXor")
#  here z = [False  True  True False]

Args
x A Tensor type bool.
y A Tensor of type bool.

Returns
A Tensor of type bool with the same size as that of x or y.