1. 程式人生 > >tf.placeholder使用說明

tf.placeholder使用說明

lai random alt 多維 atm NPU multiply amp view

tf.placeholder(dtype, shape=None, name=None)

placeholder,占位符,在tensorflow中類似於函數參數,運行時必須傳入值。

  • dtype:數據類型。常用的是tf.float32,tf.float64等數值類型。
  • shape:數據形狀。默認是None,就是一維值,也可以是多維,比如[2,3], [None, 3]表示列是3,行不定。
  • name:名稱。

代碼片段-1(計算3*4=12)

[python] view plain copy
  1. #!/usr/bin/env python
  2. # _*_ coding: utf-8 _*_
  3. import tensorflow as tf
  4. import numpy as np
  5. input1 = tf.placeholder(tf.float32)
  6. input2 = tf.placeholder(tf.float32)
  7. output = tf.multiply(input1, input2)
  8. with tf.Session() as sess:
  9. print sess.run(output, feed_dict = {input1:[3.], input2: [4.]})


代碼片段-2(計算矩陣相乘,x*x)

[python]
view plain copy
    1. #!/usr/bin/env python
    2. # _*_ coding: utf-8 _*_
    3. import tensorflow as tf
    4. import numpy as np
    5. x = tf.placeholder(tf.float32, shape=(1024, 1024))
    6. y = tf.matmul(x, x)
    7. with tf.Session() as sess:
    8. # print(sess.run(y)) # ERROR: x is none now
    9. rand_array = np.random.rand(1024, 1024)
    10. print(sess.run(y, feed_dict={x: rand_array})) # Will succeed.

tf.placeholder使用說明