Question

在尝试udacity课程深度学习任务时，我遇到了将模型的预测与训练集标签进行比较的问题

from __future__ import print_function
import numpy as np
import tensorflow as tf
from six.moves import cPickle as pickle
from six.moves import range

pickle_file = 'notMNIST.pickle'

with open(pickle_file, 'rb') as f:
  save = pickle.load(f)
  train_dataset = save['train_dataset']
  train_labels = save['train_labels']
  valid_dataset = save['valid_dataset']
  valid_labels = save['valid_labels']
  test_dataset = save['test_dataset']
  test_labels = save['test_labels']
  del save  # hint to help gc free up memory
  print('Training set', train_dataset.shape, train_labels.shape)
  print('Validation set', valid_dataset.shape, valid_labels.shape)
  print('Test set', test_dataset.shape, test_labels.shape)

输出为：
训练集（200000,28,28）（200000，）
验证集（10000,28,28）（10000，）
测试集（10000,28,28）（10000，）

# With gradient descent training, even this much data is prohibitive.
# Subset the training data for faster turnaround.
train_subset = 10000

graph = tf.Graph()
with graph.as_default():

  # Input data.
  # Load the training, validation and test data into constants that are
  # attached to the graph.
  tf_train_dataset = tf.constant(train_dataset[:train_subset, :])
  tf_train_labels = tf.constant(train_labels[:train_subset])
  tf_valid_dataset = tf.constant(valid_dataset)
  tf_test_dataset = tf.constant(test_dataset)

  # Variables.
  # These are the parameters that we are going to be training. The weight
  # matrix will be initialized using random values following a (truncated)
  # normal distribution. The biases get initialized to zero.
  weights = tf.Variable(
    tf.truncated_normal([image_size * image_size, num_labels]))
  biases = tf.Variable(tf.zeros([num_labels]))

  # Training computation.
  # We multiply the inputs with the weight matrix, and add biases. We compute
  # the softmax and cross-entropy (it's one operation in TensorFlow, because
  # it's very common, and it can be optimized). We take the average of this
  # cross-entropy across all training examples: that's our loss.
  logits = tf.matmul(tf_train_dataset, weights) + biases
  loss = tf.reduce_mean(
    tf.nn.softmax_cross_entropy_with_logits(labels=tf_train_labels, logits=logits))

  # Optimizer.
  # We are going to find the minimum of this loss using gradient descent.
  optimizer = tf.train.GradientDescentOptimizer(0.5).minimize(loss)

  # Predictions for the training, validation, and test data.
  # These are not part of training, but merely here so that we can report
  # accuracy figures as we train.
  train_prediction = tf.nn.softmax(logits)
  valid_prediction = tf.nn.softmax(
    tf.matmul(tf_valid_dataset, weights) + biases)
  test_prediction = tf.nn.softmax(tf.matmul(tf_test_dataset, weights) + biases)


num_steps = 801

def accuracy(predictions, labels):
    return (100.0 * np.sum(np.argmax(predictions, 1) == np.argmax(labels, 1))
          / predictions.shape[0])


with tf.Session(graph=graph) as session:
  # This is a one-time operation which ensures the parameters get initialized as
  # we described in the graph: random weights for the matrix, zeros for the
  # biases. 
  tf.global_variables_initializer().run()
  print('Initialized')
  for step in range(num_steps):
    # Run the computations. We tell .run() that we want to run the optimizer,
    # and get the loss value and the training predictions returned as numpy
    # arrays.
    _, l, predictions = session.run([optimizer, loss, train_prediction])
    if (step % 100 == 0):
      print('Loss at step %d: %f' % (step, l))
      print('Training accuracy: %.1f%%' % accuracy(
        predictions, train_labels[:train_subset, :]))
      # Calling .eval() on valid_prediction is basically like calling run(), but
      # just to get that one numpy array. Note that it recomputes all its graph
      # dependencies.
      print('Validation accuracy: %.1f%%' % accuracy(
        valid_prediction.eval(), valid_labels))
  print('Test accuracy: %.1f%%' % accuracy(test_prediction.eval(), test_labels))

输出：

C：\ Users \ Arslan \ Anaconda3 \ lib \ site-packages \ ipykernel_launcher.py：5：DeprecationWarning：elementwise ==比较失败;这将在未来引发错误。＆＃34;＆＃34;＆＃34;

它为所有数据集提供了0％的准确度。我想我们无法使用＆＃39; ==＆＃39;来比较数组任何帮助将不胜感激

Answer 1

我认为此表达式中出现错误：

np.sum(np.argmax(predictions, 1) == np.argmax(labels, 1))

你能告诉我们一些关于2个数组predictions，labels的信息吗？通常的东西 - dtype，shape，一些样本值。也许去额外的步骤并为每个显示np.argmax(...)。

在numpy中，您可以比较相同大小的数组，但是在比较大小不匹配的数组时会更加挑剔：

In [522]: np.arange(10)==np.arange(5,15)
Out[522]: array([False, False, False, False, False, False, False, False, False, False], dtype=bool)
In [523]: np.arange(10)==np.arange(5,14)
/usr/local/bin/ipython3:1: DeprecationWarning: elementwise == comparison failed; this will raise an error in the future.
  #!/usr/bin/python3
Out[523]: False

Answer 2

我通过将python升级到3.6.4（最新）来解决这个问题

conda update python

Answer 3

此错误告诉您正在执行的比较实际上没有意义，因为两个数组的形状不同，因此无法执行逐元素比较。这是一个示例：

x = np.random.randint(0,5,(3,2))
y = np.random.randint(0,5,(5,7))

尝试x==y的地方会产生：

DeprecationWarning：逐元素比较失败；将来会出现错误。 x == y

正确的方法是使用np.array_equal，它可以检查形状和元素是否相等：

np.array_equal(x,y)
# False

对于浮点数，np.allclose更适合，因为它可以控制比较结果的相对和绝对公差。这是一个示例：

x = np.random.random((400,34))
y = x.round(6)

np.array_equal(x,y)
# False
np.allclose(x,y)
# False
np.allclose(x,y, atol=1e-05)
# True

Python：DeprecationWarning：elementwise ==比较失败;这将在未来引发错误

3 个答案: