mindspore.Tensor.log2

Tensor.log2() → Tensor

Returns the logarithm to the base 2 of a tensor element-wise

\[y_i = \log_2(x_i)\]

Warning

  • This is an experimental API that is subject to change or deletion.

  • If the self value of operator Log2 is within the range (0, 0.01] or [0.95, 1.05], the output accuracy may be affacted.

Note

The value of self must be greater than 0.

Returns

Tensor, has the same shape as the self, and the dtype changes according to the self.dtype.

  • if self.dtype is in [float16, float32, float64, complex64, complex128], the output dtype is the same as the self.dtype.

  • if self.dtype is double type, the output dtype is float64.

  • if self.dtype is integer or boolean type on Ascend, the output dtype is float32.

Raises

  • TypeError – If dtype of self is not one of bool, int8, int32, int64, uint8, uint32, uint64, float16, float32, float64, double, complex64, complex128.

  • TypeError – If dtype of self is integer or boolean type on CPU and GPU.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> import numpy as np
>>> from mindspore import Tensor
>>> x = Tensor(np.array([3.0, 5.0, 7.0]), mindspore.float32)
>>> output = Tensor.log2(x)
>>> print(output)
[1.5849625 2.321928  2.807355 ]
>>> x = Tensor(np.array([2, 4, 8]).astype(np.float16))
>>> output = Tensor.log2(x)
>>> print(output)
[1. 2. 3.]