python–tensorlow . math . ndtri()
哎哎哎:# t0]https://www . geeksforgeeks . org/python-tensorlow-math-ndtri/
TensorFlow 是谷歌为开发机器学习模型和深度学习神经网络而设计的开源 python 库。 ndtri() 用于寻找标准正态的分位数。
语法: tf.math.ndtri(x，name)
参数:
- x: 是输入张量。此张量允许的数据类型是浮点型或双精度型..
- 名称(可选):定义操作的名称。
返回: 返回 x 的逆误差函数
例 1:
Python 3
# Importing the library
import tensorflow as tf
# Initializing the input tensor
a = tf.constant([.2, .5, .7, 1], dtype = tf.float64)
# Printing the input tensor
print('Input: ', a)
# Calculating result
res = tf.math.ndtri(x = a)
# Printing the result
print('Result: ', res)
输出:
Input: tf.Tensor([0.2 0.5 0.7 1\. ], shape=(4, ), dtype=float64)
Result: tf.Tensor([-0.84162123 0\. 0.52440051 inf], shape=(4, ), dtype=float64)
例 2:
Python 3
# importing the library
import tensorflow as tf
# Initializing the input tensor
a = tf.constant([-.1, -.2, -.5, .5, .2, .1], dtype = tf.float64)
# Printing the input tensor
print('Input: ', a)
# Calculating result
res = tf.math.ndtri(x = a)
# Printing the result
print('Result: ', res)
输出:
Input: tf.Tensor([-0.1 -0.2 -0.5 0.5 0.2 0.1], shape=(6, ), dtype=float64)
Result: tf.Tensor([ -inf -inf -inf 0\. -0.84162123 -1.28155157], shape=(6, ), dtype=float64)
版权属于：月萌API www.moonapi.com，转载请注明出处
