我是张量流的新手，我试图重现网站上发布的实验https://github.com/jiegzhan/image-classification-rnn
我的TensorFlow版本是1.0.1，所以我稍微修改了他的代码。我使用的代码如下。
import os
import sys
import json
import time
import tensorflow as tf
from tensorflow.contrib.rnn.python.ops import rnn, rnn_cell
from tensorflow.examples.tutorials.mnist import input_data
n_input = 28 # MNIST data input (img shape: 28*28)
n_steps = 28 # timesteps
n_hidden = 128 # hidden layer num of features
n_classes = 10 # MNIST total classes (0-9 digits)
#coding:utf-8
def rnn_model(x, weights, biases):
"""RNN (LSTM or GRU) model for image"""
x = tf.transpose(x, [1, 0, 2])
x = tf.reshape(x, [-1, n_input])
x = tf.split(x, n_steps, 0)
# Define a lstm cell with tensorflow
#lstm_cell = rnn_cell.BasicLSTMCell(n_hidden, forget_bias=1.0)
lstm_cell = tf.contrib.rnn.BasicLSTMCell(n_hidden, forget_bias=1.0)
# Get lstm cell output
outputs, states = tf.contrib.rnn.static_rnn(lstm_cell, x, dtype=tf.float32)
return tf.matmul(outputs[-1], weights) + biases
def train():
"""Train an image classifier"""
"""Step 0: load image data and training parameters"""
mnist = input_data.read_data_sets("./data/", one_hot=False)# change one_hot to false by myself ---------
#parameter_file = sys.argv[1]
#params = json.loads(open(parameter_file).read())
params = json.loads(open('parameters.json').read())
"""Step 1: build a rnn model for image"""
x = tf.placeholder("float", [None, n_steps, n_input])
y = tf.placeholder("float", [None, n_classes])
#y = tf.placeholder("float", [n_classes,])
weights = tf.Variable(tf.random_normal([n_hidden, n_classes]), name='weights')
biases = tf.Variable(tf.random_normal([n_classes]), name='biases')
pred = rnn_model(x, weights, biases)
# Define loss and optimizer
#you will get the dreaded 'No gradients provided for any variable' if you switch the args between y and pred
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y, logits=pred))
optimizer = tf.train.AdamOptimizer(learning_rate=params['learning_rate']).minimize(cost)
# Evaluate model
correct_pred = tf.equal(tf.argmax(pred,1), tf.argmax(y,1))
accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))
"""Step 2: train the image classification model"""
with tf.Session() as sess:
#sess.run(tf.initialize_all_variables())# deprecated
sess.run(tf.global_variables_initializer())
step = 1
"""Step 2.0: create a directory for saving model files"""
timestamp = str(int(time.time()))
out_dir = os.path.abspath(os.path.join(os.path.curdir, "trained_model_" + timestamp))
checkpoint_dir = os.path.abspath(os.path.join(out_dir, "checkpoints"))
checkpoint_prefix = os.path.join(checkpoint_dir, "model")
if not os.path.exists(checkpoint_dir):
os.makedirs(checkpoint_dir)
saver = tf.train.Saver(tf.all_variables())
"""Step 2.1: train the image classifier batch by batch"""
while step * params['batch_size'] < params['training_iters']:
batch_x, batch_y = mnist.train.next_batch(params['batch_size'])
# Reshape data to get 28 seq of 28 elements
print(batch_x)
print(batch_y)
batch_x = batch_x.reshape((params['batch_size'], n_steps, n_input))
#batch_y = batch_y.reshape((-1, n_classes))
sess.run(optimizer, feed_dict={x: batch_x, y: batch_y})#**bug is here**
"""Step 2.2: save the model"""
if step % params['display_step'] == 0:
path = saver.save(sess, checkpoint_prefix, global_step=step)
acc = sess.run(accuracy, feed_dict={x: batch_x, y: batch_y})
loss = sess.run(cost, feed_dict={x: batch_x, y: batch_y})
print('Iter: {}, Loss: {:.6f}, Accuracy: {:.6f}'.format(step * params['batch_size'], loss, acc))
step += 1
print("The training is done")
"""Step 3: test the model"""
test_len = 128
test_data = mnist.test.images[:test_len].reshape((-1, n_steps, n_input))
test_label = mnist.test.labels[:test_len]
print("Testing Accuracy:", sess.run(accuracy, feed_dict={x: test_data, y: test_label}))
if __name__ == '__main__':
train()
警告：
WARNING:tensorflow:From D:/TensorFlow/image-classification-rnn-master/train.py:68: all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02.
Instructions for updating:
Please use tf.global_variables instead.
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
[7 3 4 6 1 8 1 0 9 8 0 3 1 2 7 0 2 9 6 0 1 6 7 1 9 7 6 5 5 8 8 3 4 4 8 7 3
6 4 6 6 3 8 8 9 9 4 4 0 7 8 1 0 0 1 8 5 7 1 7 5 5 9 9]
Traceback (most recent call last):
File "D:/TensorFlow/image-classification-rnn-master/train.py", line 98, in <module>
train()
File "D:/TensorFlow/image-classification-rnn-master/train.py", line 80, in train
sess.run(optimizer, feed_dict={x: batch_x, y: batch_y})
File "C:\Users\***\AppData\Local\Programs\Python\Python35\lib\site-packages\tensorflow\python\client\session.py", line 767, in run
run_metadata_ptr)
File "C:\Users\***\AppData\Local\Programs\Python\Python35\lib\site-packages\tensorflow\python\client\session.py", line 944, in _run
% (np_val.shape, subfeed_t.name, str(subfeed_t.get_shape())))
ValueError: Cannot feed value of shape (64,) for Tensor 'Placeholder_1:0', which has shape '(?, 10)'
Process finished with exit code 1
错误在这里：
sess.run(optimizer, feed_dict={x: batch_x, y: batch_y})
print(batch_y ) --> Tensor("ExpandDims:0", shape=(64, 1), dtype=uint8)
我试图在谷歌上找到一个相关的答案并得到一些相似的答案，但我无法让它们中的任何一个起作用。
有人帮忙吗？非常感谢！
答案 0 :(得分：5)
问题是batch_y是一个包含标签（数字从0到9）的张量，指定了正确的类，而你的占位符
y = tf.placeholder(tf.float32, [None, n_classes])
每个示例都需要一个长度为n_classes的向量，用于指定每个类的预期概率。
在这个问题中，一次只有一个类是正确的，因此代替正确的标签n_classes - 长度向量，您可以将y作为标签，例如
y = tf.placeholder(tf.float32, [None])
然后您还必须将cost更改为
cost = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=pred))
sparse表示只有一个类是正确的，因此参数labels是一个大小为[batch_size]的张量，其数字代表每个示例的正确类。