Python Tensorflow:MNIST的无效辩论对手[55000]对[10000]

我正在处理此演示文稿中的代码，并收到以下错误：

InvalidArgumentError（回溯请参见上文）：不兼容的形状：[55000]与[10000]

我已经解决了一些关于张量形状/维度的错误，但不知道如何具体理解，更不用说纠正了

我是tf的新手，非常感谢您的建议，以下是代码：

# 1 ~ import tf + data
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data

mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

#2 ~ paras + init
X = tf.placeholder(tf.float32, [None, 784])
W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))

init = tf.initialize_all_variables()

#3 ~ model + correct answers
Y = tf.nn.softmax(tf.matmul(X, W) + b)
Y_ = tf.placeholder(tf.float32, [None, 10]) # one-hot encoding

#4 ~ loss function
cross_entropy = -tf.reduce_sum(Y_ * tf.log(Y))

#5 ~ correct answer + % accuracy
is_correct = tf.equal(tf.argmax(Y, 1), tf.argmax(Y_, 1)) # one-hot decoding
accuracy = tf.reduce_mean(tf.cast(is_correct, tf.float32))

#6 ~ optimizer and training step
optimizer = tf.train.GradientDescentOptimizer(0.003) # learning-rate
train_step = optimizer.minimize(cross_entropy)

#7 ~ session and training loop
sess = tf.Session()
sess.run(init)

for i in range(1000):
    # load a batch of images and correct answers
    batch_X, batch_Y = mnist.train.next_batch(100)
    train_data = {X: batch_X, Y_: batch_Y}

    # train
    sess.run(train_step, feed_dict=train_data)

    # success?
    a,c = sess.run([accuracy, cross_entropy], feed_dict=train_data)

    # success on test data?
    test_data = {X: mnist.train.images, Y_: mnist.test.labels}
    a,c = sess.run([accuracy, cross_entropy], feed_dict=test_data)

总错误输出为：

Traceback (most recent call last):
      File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1021, in _do_call
    return fn(*args)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1003, in _run_fn
    status, run_metadata)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/contextlib.py", line 66, in __exit__
    next(self.gen)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/errors_impl.py", line 469, in raise_exception_on_not_ok_status
    pywrap_tensorflow.TF_GetCode(status))
tensorflow.python.framework.errors_impl.InvalidArgumentError: Incompatible shapes: [55000] vs. [10000]
     [[Node: Equal = Equal[T=DT_INT64, _device="/job:localhost/replica:0/task:0/cpu:0"](ArgMax, ArgMax_1)]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "deep_nn1.py", line 71, in <module>
    a,c = sess.run([accuracy, cross_entropy], feed_dict=test_data)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 766, in run
    run_metadata_ptr)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 964, in _run
    feed_dict_string, options, run_metadata)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1014, in _do_run
    target_list, options, run_metadata)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1034, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Incompatible shapes: [55000] vs. [10000]
     [[Node: Equal = Equal[T=DT_INT64, _device="/job:localhost/replica:0/task:0/cpu:0"](ArgMax, ArgMax_1)]]

Caused by op 'Equal', defined at:
  File "deep_nn1.py", line 47, in <module>
    is_correct = tf.equal(tf.argmax(Y, 1), tf.argmax(Y_, 1)) # one-hot decoding
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/ops/gen_math_ops.py", line 728, in equal
    result = _op_def_lib.apply_op("Equal", x=x, y=y, name=name)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py", line 759, in apply_op
    op_def=op_def)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 2240, in create_op
    original_op=self._default_original_op, op_def=op_def)
  File "/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1128, in __init__
    self._traceback = _extract_stack()

InvalidArgumentError (see above for traceback): Incompatible shapes: [55000] vs. [10000]
     [[Node: Equal = Equal[T=DT_INT64, _device="/job:localhost/replica:0/task:0/cpu:0"](ArgMax, ArgMax_1)]]

回溯（最近一次呼叫最后一次）：
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py”，第1021行，在调用中
返回fn（*args）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/client/session.py”，第1003行，在
状态，运行（元数据）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/contextlib.py”，第66行，在退出时__
下一个（self.gen）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/framework/errors\u impl.py”，第469行，处于引发异常打开状态
pywrap_tensorflow.TF_GetCode（状态））
tensorflow.python.framework.errors\u impl.InvalidArgumentError:不兼容的形状：[55000]与[10000]
[[Node:Equal=Equal[T=DT_INT64，_device=“/job:localhost/replica:0/task:0/cpu:0”]（ArgMax，ArgMax_1）]]
在处理上述异常期间，发生了另一个异常：
回溯（最近一次呼叫最后一次）：
文件“deep_nn1.py”，第71行，在
a、 c=测试运行（[精度、交叉熵]，输入数据=测试数据）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/client/session.py”，第766行，正在运行
运行_元数据_ptr）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/client/session.py”，第964行，正在运行
提要（dict字符串、选项、运行元数据）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/client/session.py”，第1014行，在运行中
目标\u列表、选项、运行\u元数据）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py”，第1034行，在
提升类型（e）（节点定义、操作、消息）
tensorflow.python.framework.errors\u impl.InvalidArgumentError:不兼容的形状：[55000]与[10000]
[[Node:Equal=Equal[T=DT_INT64，_device=“/job:localhost/replica:0/task:0/cpu:0”]（ArgMax，ArgMax_1）]]
由op“Equal”引起，定义为：
文件“deep_nn1.py”，第47行，在
是否正确=tf.equal（tf.argmax（Y，1），tf.argmax（Y，1））#一个热解码
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/ops/gen_math_ops.py”，第728行，等号
结果=_op_def_lib.apply_op（“相等”，x=x，y=y，name=name）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/framework/op_def_library.py”，第759行，在apply_op
op_def=op_def）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/framework/ops.py”，第2240行，在create_op中
初始值=自身值。\默认值\初始值，初始值=初始值）
文件“/Users/joelmcleod/anaconda3/lib/python3.5/site packages/tensorflow/python/framework/ops.py”，第1128行，在__
self.\u traceback=\u extract\u stack（）
InvalidArgumentError（回溯见上文）：不兼容的形状：[55000]与[10000]
[[Node:Equal=Equal[T=DT_INT64，_device=“/job:localhost/replica:0/task:0/cpu:0”]（ArgMax，ArgMax_1）]]

您在此行中输入的是培训数据（行数：55000），而不是测试数据（行数：10000）：

test_data = {X: mnist.train.images, Y_: mnist.test.labels}

只需使用以下工具进行修复：

test_data = {X: mnist.test.images, Y_: mnist.test.labels}

---

非常高兴，奥利弗