Python 如何为 TensorFlow 变量赋值？

How to assign a value to a TensorFlow variable?

提问by abora

I am trying to assign a new value to a tensorflow variable in python.

我正在尝试为 python 中的 tensorflow 变量分配一个新值。

import tensorflow as tf
import numpy as np

x = tf.Variable(0)
init = tf.initialize_all_variables()
sess = tf.InteractiveSession()
sess.run(init)

print(x.eval())

x.assign(1)
print(x.eval())

But the output I get is

但我得到的输出是

0
0

So the value has not changed. What am I missing?

所以价值没有改变。我错过了什么？

采纳答案by mrry

In TF1, the statement x.assign(1)does not actually assign the value 1to x, but rather creates a tf.Operationthat you have to explicitly runto update the variable.* A call to Operation.run()or Session.run()can be used to run the operation:

在 TF1 中，该语句x.assign(1)实际上并未将值分配1给x，而是创建了tf.Operation，您必须显式运行它以更新变量。* 调用Operation.run()或Session.run()可用于运行操作：

assign_op = x.assign(1)
sess.run(assign_op)  # or `assign_op.op.run()`
print(x.eval())
# ==> 1

(* In fact, it returns a tf.Tensor, corresponding to the updated value of the variable, to make it easier to chain assignments.)

（* 实际上，它返回一个tf.Tensor，对应于变量的更新值，以便更容易地进行链式赋值。）

However, in TF2 x.assign(1)will now assign the value eagerly:

但是，在 TF2 中x.assign(1)现在会急切地分配值：

x.assign(1)
print(x.numpy())
# ==> 1

回答by Gordon Erlebacher

There is an easier approach:

有一个更简单的方法：

x = tf.Variable(0)
x = x + 1
print x.eval()

回答by Salvador Dali

First of all you can assign values to variables/constants just by feeding values into them the same way you do it with placeholders. So this is perfectly legal to do:

首先，您可以将值分配给变量/常量，只需像使用占位符一样将值输入它们即可。所以这样做是完全合法的：

import tensorflow as tf
x = tf.Variable(0)
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    print sess.run(x, feed_dict={x: 3})

Regarding your confusion with the tf.assign()operator. In TF nothing is executed before you run it inside of the session. So you always have to do something like this: op_name = tf.some_function_that_create_op(params)and then inside of the session you run sess.run(op_name). Using assign as an example you will do something like this:

关于您对tf.assign()运算符的混淆。在 TF 中，在会话内运行之前不会执行任何操作。所以你总是必须做这样的事情：op_name = tf.some_function_that_create_op(params)然后在你运行的会话内部sess.run(op_name)。以assign为例，您将执行以下操作：

import tensorflow as tf
x = tf.Variable(0)
y = tf.assign(x, 1)
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    print sess.run(x)
    print sess.run(y)
    print sess.run(x)

回答by kmario23

Also, it has to be noted that if you're using your_tensor.assign(), then the tf.global_variables_initializerneed not be called explicitly since the assign operation does it for you in the background.

此外，必须注意的是，如果您正在使用your_tensor.assign()，则tf.global_variables_initializer无需显式调用 ，因为分配操作会在后台为您执行此操作。

Example:

例子：

In [212]: w = tf.Variable(12)
In [213]: w_new = w.assign(34)

In [214]: with tf.Session() as sess:
     ...:     sess.run(w_new)
     ...:     print(w_new.eval())

# output
34 

However, this will not initialize all variables, but it will only initialize the variable on which assignwas executed on.

但是，这不会初始化所有变量，而只会初始化assign执行过的变量。

回答by Robin Dinse

You can also assign a new value to a tf.Variablewithout adding an operation to the graph: tf.Variable.load(value, session). This function can also save you adding placeholders when assigning a value from outside the graph and it is useful in case the graph is finalized.

您还可以在tf.Variable不向图形添加操作的情况下为 a 分配一个新值：tf.Variable.load(value, session)。此功能还可以节省您在从图形外部分配值时添加占位符的情况，并且在图形最终确定时很有用。

import tensorflow as tf
x = tf.Variable(0)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
print(sess.run(x))  # Prints 0.
x.load(1, sess)
print(sess.run(x))  # Prints 1.

Update: This is depricated in TF2 as eager execution is default and graphs are no longer exposed in the user-facing API.

更新：这在 TF2 中已被弃用，因为 Eager Execution 是默认设置，并且图形不再在面向用户的 API 中公开。

回答by prosti

Here is the complete working example:

这是完整的工作示例：

import numpy as np
import tensorflow as tf

w= tf.Variable(0, dtype=tf.float32) #good practice to set the type of the variable
cost = 10 + 5*w + w*w
train = tf.train.GradientDescentOptimizer(0.01).minimize(cost)

init = tf.global_variables_initializer()
session = tf.Session()
session.run(init)

print(session.run(w))

session.run(train)
print(session.run(w)) # runs one step of gradient descent

for i in range(10000):
  session.run(train)

print(session.run(w))

Note the output will be:

请注意，输出将是：

0.0
-0.049999997
-2.499994

This means at the very start the Variable was 0, as defined, then after just one step of gradient decent the variable was -0.049999997, and after 10.000 more steps we are reaching -2.499994 (based on our cost function).

这意味着一开始变量是 0，正如定义的那样，然后在梯度下降一步之后，变量是 -0.049999997，再经过 10.000 步后，我们达到 -2.499994（基于我们的成本函数）。

Note: You originally used the Interactive session. Interactive session is useful when multiple different sessions needed to be run in the same script. However, I used the non interactive session for simplicity.

注意：您最初使用的是交互式会话。当需要在同一个脚本中运行多个不同的会话时，交互式会话非常有用。但是，为了简单起见，我使用了非交互式会话。

回答by Aashish Dahiya

Use Tensorflow eager execution mode which is latest.

使用最新的 Tensorflow 急切执行模式。

import tensorflow as tf
tf.enable_eager_execution()
my_int_variable = tf.get_variable("my_int_variable", [1, 2, 3])
print(my_int_variable)

回答by Prabhant Singh

So i had a adifferent case where i needed to assign values before running a session, So this was the easiest way to do that:

所以我有一个不同的情况，我需要在运行会话之前分配值，所以这是最简单的方法：

other_variable = tf.get_variable("other_variable", dtype=tf.int32,
  initializer=tf.constant([23, 42]))

here i'm creating a variable as well as assigning it values at the same time

在这里，我正在创建一个变量并同时为其赋值

回答by user5931

I answered a similar question here. I looked in a lot of places that always created the same problem. Basically, I did not want to assign a value to the weights, but simply change the weights. The short version of the above answer is:

我在这里回答了一个类似的问题。我查看了很多总是会产生相同问题的地方。基本上，我不想为权重赋值，而是简单地改变权重。上述答案的简短版本是：

tf.keras.backend.set_value(tf_var, numpy_weights)

tf.keras.backend.set_value(tf_var, numpy_weights)