Tensorflow:无需运行任何会话即可将 Tensor 转换为 numpy 数组 [英] Tensorflow: Tensor to numpy array conversion without running any session

问题描述

我在 tensorflow 中创建了一个 OP，其中对于某些处理，我需要将我的数据从张量对象转换为 numpy 数组.我知道我们可以使用 tf.eval() 或 sess.run 来评估任何张量对象.我真正想知道的是，有没有办法在不运行任何会话的情况下将张量转换为数组，因此我们又避免使用 .eval() 或 .run().

非常感谢任何帮助！

def tensor_to_array(tensor1):'''将张量对象转换为numpy数组'''array1 = SESS.run(tensor1) **#====== 需要绕过这一行**返回 array1.astype("uint8")def array_to_tensor(array):'''将numpy数组转换为张量对象'''tensor_data = tf.convert_to_tensor(数组，dtype=tf.float32)返回张量数据

解决方案

更新

#必须在eagar模式下def tensor_to_array(tensor1):返回 tensor1.numpy()

示例

<预><代码>>>>将张量流导入为 tf>>>tf.enable_eager_execution()>>>def tensor_to_array(tensor1):...返回 tensor1.numpy()...>>>x = tf.constant([1,2,3,4])>>>tensor_to_array(x)数组([1, 2, 3, 4], dtype=int32)

---

我相信你可以在没有 tf.eval() 或 sess.run 的情况下使用 tf.enable_eager_execution()

>

示例

 将 tensorflow 导入为 tf将 numpy 导入为 nptf.enable_eager_execution()x = np.array([1,2,3,4])c = tf.constant([4,3,2,1])c+x<tf.Tensor: id=5, shape=(4,), dtype=int32, numpy=array([5, 5, 5, 5], dtype=int32)>

有关张量流急切模式的更多详细信息，请在此处查看:张量流急切

如果没有tf.enable_eager_execution():

 将 tensorflow 导入为 tf将 numpy 导入为 npc = tf.constant([4,3,2,1])x = np.array([1,2,3,4])c+x<tf.Tensor 'add:0' shape=(4,) dtype=int32>

I have created an OP in tensorflow where for some processing I need my data to be converted from tensor object to numpy array. I know we can use tf.eval() or sess.run to evaluate any tensor object. What I really want to know is, Is there any way to convert tensor to array without any session running, so in turn we avoid use of .eval() or .run().

Any help is highly appreciated!

def tensor_to_array(tensor1):
    '''Convert tensor object to numpy array'''
    array1 = SESS.run(tensor1) **#====== need to bypass this line**
    return array1.astype("uint8")

def array_to_tensor(array):
    '''Convert numpy array to tensor object'''
    tensor_data = tf.convert_to_tensor(array, dtype=tf.float32)
    return tensor_data

解决方案

Updated

# must under eagar mode
def tensor_to_array(tensor1):
    return tensor1.numpy()

example

>>> import tensorflow as tf
>>> tf.enable_eager_execution()
>>> def tensor_to_array(tensor1):
...     return tensor1.numpy()
...
>>> x = tf.constant([1,2,3,4])
>>> tensor_to_array(x)
array([1, 2, 3, 4], dtype=int32)

---

I believe you can do it without tf.eval() or sess.run by using tf.enable_eager_execution()

example

import tensorflow as tf
import numpy as np
tf.enable_eager_execution()
x = np.array([1,2,3,4])
c = tf.constant([4,3,2,1])
c+x
<tf.Tensor: id=5, shape=(4,), dtype=int32, numpy=array([5, 5, 5, 5], dtype=int32)>

For more details about tensorflow eager mode, checkout here:Tensorflow eager

If without tf.enable_eager_execution():

import tensorflow as tf
import numpy as np
c = tf.constant([4,3,2,1])
x = np.array([1,2,3,4])
c+x
<tf.Tensor 'add:0' shape=(4,) dtype=int32>

这篇关于Tensorflow:无需运行任何会话即可将 Tensor 转换为 numpy 数组的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！