Question

我无法理解为什么这不起作用，基本上我正在尝试检索m和q的值只是为了打印它们但我总是得到[nan，nan]

import tensorflow as tf
import pandas as pd
import matplotlib.pyplot as plt

m = tf.Variable(tf.random_uniform(shape=()), dtype=tf.float32)
q = tf.Variable(tf.random_uniform(shape=()), dtype=tf.float32)

X = tf.placeholder(tf.float32)
y = tf.placeholder(tf.float32)

linear_model = m * X + q

cost = tf.reduce_sum(tf.square(linear_model - y))
optimizer = tf.train.GradientDescentOptimizer(0.01)
gdescent = optimizer.minimize(cost)

train, test = [pd.read_csv(file) for file in ["train.csv", "test.csv"]]

if False:  # Scatter training points
    plt.scatter(train['x'], train['y'], s=1)
    plt.show()

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    _, m, q = sess.run([gdescent, m, q], feed_dict={X: train['x'].values, y: train['y'].values})
    print(m, q)

train.csv和test.csv都是带有标题行和两列值x和y的文件

前10行train.csv

x,y
24,21.54945196
50,47.46446305
15,17.21865634
38,36.58639803
87,87.28898389
36,32.46387493
12,10.78089683
81,80.7633986
25,24.61215147

Answer 1

解决了，错误是train.csv错过了第215行的y值

为什么在训练后运行变量时张量流会返回NaN？

1 个答案: