mindspore.ops.log_softmax

mindspore.ops.log_softmax(logits, axis=- 1)

Applies the Log Softmax function to the input tensor on the specified axis. Supposes a slice in the given axis, \(x\) for each element \(x_i\), the Log Softmax function is shown as follows:

\[\text{output}(x_i) = \log \left(\frac{\exp(x_i)} {\sum_{j = 0}^{N-1}\exp(x_j)}\right),\]

where \(N\) is the length of the Tensor.

Parameters

• logits (Tensor) – The input Tensor, which is the \(x\) in the formula above, it’s shape is \((N, *)\), where \(*\) means, any number of additional dimensions, with float16 or float32 data type.

• axis (int) – The axis to perform the Log softmax operation. Default: -1 .

Returns

Tensor, with the same type and shape as the logits.

Raises

• TypeError – If axis is not an int.

• TypeError – If dtype of logits is neither float16 nor float32.

• ValueError – If axis is not in range [-len(logits.shape), len(logits.shape)).

• ValueError – If dimension of logits is less than 1.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> import numpy as np
>>> from mindspore import Tensor, ops
>>> logits = Tensor(np.array([1, 2, 3, 4, 5]), mindspore.float32)
>>> output = ops.log_softmax(logits)
>>> print(output)
[-4.4519143 -3.4519143 -2.4519143 -1.4519144 -0.4519144]