tf.one_hot(indices, depth, on_value=None, off_value=None, axis=None)

 Tensor Transformations > Slicing and Joining

tf.one_hot(indices, depth, on_value=None, off_value=None, axis=None, dtype=None, name=None)

tf.one_hot(indices, depth, on_value=None, off_value=None, axis=None, dtype=None, name=None)

See the guide: Tensor Transformations > Slicing and Joining

Returns a one-hot tensor.

The locations represented by indices in 

indices

 take value 

on_value

,
while all other locations take value 

off_value

.

on_value

 and 

off_value

 must
have matching data types. If 

dtype

 is also provided, they must be the same data
type as specified by 

dtype

.

If 

on_value

 is not provided, it will default to the value 

1

 with
type 

dtype

If 

off_value

 is not provided, it will default to the value 

0

 with
type 

dtype

If the input 

indices

 is rank 

N

,
the output will have rank 

N+1

. The new axis is created at dimension 

axis

 (default:
the new axis is appended at the end).

If 

indices

 is a scalar the output shape will be a vector of length 

depth

If 

indices

 is a vector of length 

features

,
the output shape will be:

  features x depth if axis == -1
  depth x features if axis == 0

If 

indices

 is a matrix (batch) with shape 

[batch,
features]

, the output shape will be:

  batch x features x depth if axis == -1
  batch x depth x features if axis == 1
  depth x batch x features if axis == 0

If 

dtype

 is not provided, it will attempt to assume the data type of 

on_value

 or 

off_value

,
if one or both are passed in. If none of 

on_value

, 

off_value

,
or 

dtype

 are provided, 

dtype

 will
default to the value 

tf.float32

.
Note: If a non-numeric data type output is desired (

tf.string

, 

tf.bool

,
etc.), both 

on_value

 and 

off_value

 must be
provided to 

one_hot

.

Examples

Suppose that

  indices = [0, 2, -1, 1]
  depth = 3
  on_value = 5.0
  off_value = 0.0
  axis = -1

Then output is 

[4 x 3]

:

  output =
  [5.0 0.0 0.0]  // one_hot(0)
  [0.0 0.0 5.0]  // one_hot(2)
  [0.0 0.0 0.0]  // one_hot(-1)
  [0.0 5.0 0.0]  // one_hot(1)

Suppose that

  indices = [[0, 2], [1, -1]]
  depth = 3
  on_value = 1.0
  off_value = 0.0
  axis = -1

Then output is 

[2 x 2 x 3]

:

  output =
  [
    [1.0, 0.0, 0.0]  // one_hot(0)
    [0.0, 0.0, 1.0]  // one_hot(2)
  ][
    [0.0, 1.0, 0.0]  // one_hot(1)
    [0.0, 0.0, 0.0]  // one_hot(-1)
  ]

Using default values for 

on_value

 and 

off_value

:

  indices = [0, 1, 2]
  depth = 3

The output will be

  output =
  [[1., 0., 0.],
   [0., 1., 0.],
   [0., 0., 1.]]

Args:

indices

: A 

Tensor

 of
indices.

depth

: A scalar
defining the depth of the one hot dimension.

on_value

: A scalar
defining the value to fill in output when 

indices[j] = i

. (default: 1)

off_value

: A scalar
defining the value to fill in output when 

indices[j] != i

. (default: 0)

axis

: The axis
to fill (default: -1, a new inner-most axis).

dtype

: The data
type of the output tensor.

Returns:

output

: The one-hot
tensor.

Raises:

TypeError

: If dtype
of either 

on_value

 or 

off_value

 don't
match 

dtype

TypeError

: If dtype
of 

on_value

 and 

off_value

 don't
match one another

Defined in 

tensorflow/python/ops/array_ops.py

.