Question

我正在尝试在张量流中训练单层感知器（基于this上的代码）在以下数据文件中：

1,1,0.05,-1.05
1,1,0.1,-1.1
....

其中最后一列是标签（3个参数的功能），前三列是函数参数。读取数据并训练模型的代码（为了便于阅读，我将其简化）：

import tensorflow as tf

... # some basics to read the data
example, label = read_file_format(filename_queue)
... # model construction and parameter setting

# Launch the graph
with tf.Session() as sess:
    sess.run(init)
    for epoch in range(training_epochs):
        _, c = sess.run([optimizer, cost], feed_dict={x: example, y: label})
    print("Optimization Finished!")

但是当我运行它时，会出现以下错误：

Traceback (most recent call last):
  File "nn.py", line 85, in <module>
    _, c = sess.run([optimizer, cost], feed_dict={x: example, y: label})
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 710, in run
    run_metadata_ptr)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 887, in _run
    % (np_val.shape, subfeed_t.name, str(subfeed_t.get_shape())))
ValueError: Cannot feed value of shape (3,) for Tensor u'Placeholder:0', which has shape '(?, 3)'

Answer 1

你的图表期望X是一个形状的张量（？，3）。您的示例数据具有形状（3，），即长度为3的1维向量。可以将示例重塑为（1,3），或者在一次拍摄中传递一批示例（例如，10，形状为（10， 3））

Tensorflow - 从文件读取数据时形状错误

1 个答案: