Computes exp(x) - 1 element-wise.
tf.math.expm1(
    x, name=None
)
  i.e. exp(x) - 1 or e^(x) - 1, where x is the input tensor. e denotes Euler's number and is approximately equal to 2.718281.
x = tf.constant(2.0) tf.math.expm1(x) ==> 6.389056 x = tf.constant([2.0, 8.0]) tf.math.expm1(x) ==> array([6.389056, 2979.958], dtype=float32) x = tf.constant(1 + 1j) tf.math.expm1(x) ==> (0.46869393991588515+2.2873552871788423j)
| Args | |
|---|---|
| x | A Tensor. Must be one of the following types:bfloat16,half,float32,float64,complex64,complex128. | 
| name | A name for the operation (optional). | 
| Returns | |
|---|---|
| A Tensor. Has the same type asx. | 
    © 2022 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 4.0.
Code samples licensed under the Apache 2.0 License.
    https://www.tensorflow.org/versions/r2.9/api_docs/python/tf/math/expm1