ValueError：无法为具有形状“（？，1）”的张量“Placeholder_1：0”提供形状（6165、5）的值

Question

> WARNING: The TensorFlow contrib module will not be included in TensorFlow 2.0.
For more information, please see:
  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md
  * https://github.com/tensorflow/addons
If you depend on functionality not listed there, please file an issue.

WARNING:tensorflow:From C:/Users/SONSANGWOO/Desktop/Euroaquae/The_third_semester_at_BCN/ANN/Exercise/TimeSeriespy_RNN.py:74: BasicLSTMCell.__init__ (from tensorflow.python.ops.rnn_cell_impl) is deprecated and will be removed in a future version.
Instructions for updating:
This class is equivalent as tf.keras.layers.LSTMCell, and will be replaced by that in Tensorflow 2.0.
WARNING:tensorflow:From C:/Users/SONSANGWOO/Desktop/Euroaquae/The_third_semester_at_BCN/ANN/Exercise/TimeSeriespy_RNN.py:75: dynamic_rnn (from tensorflow.python.ops.rnn) is deprecated and will be removed in a future version.
Instructions for updating:
Please use `keras.layers.RNN(cell)`, which is equivalent to this API
WARNING:tensorflow:From C:\Users\SONSANGWOO\Anaconda3\lib\site-packages\tensorflow\python\ops\tensor_array_ops.py:162: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
Traceback (most recent call last):

  File "<ipython-input-1-7716630f4e29>", line 1, in <module>
    runfile('C:/Users/SONSANGWOO/Desktop/Euroaquae/The_third_semester_at_BCN/ANN/Exercise/TimeSeriespy_RNN.py', wdir='C:/Users/SONSANGWOO/Desktop/Euroaquae/The_third_semester_at_BCN/ANN/Exercise')

  File "C:\Users\SONSANGWOO\Anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 704, in runfile
    execfile(filename, namespace)

  File "C:\Users\SONSANGWOO\Anaconda3\lib\site-packages\spyder_kernels\customize\spydercustomize.py", line 108, in execfile
    exec(compile(f.read(), filename, 'exec'), namespace)

  File "C:/Users/SONSANGWOO/Desktop/Euroaquae/The_third_semester_at_BCN/ANN/Exercise/TimeSeriespy_RNN.py", line 97, in <module>
    X: trainX, Y: trainY})

  File "C:\Users\SONSANGWOO\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 929, in run
    run_metadata_ptr)

  File "C:\Users\SONSANGWOO\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1128, in _run
    str(subfeed_t.get_shape())))

ValueError: Cannot feed value of shape (6165, 5) for Tensor 'Placeholder_1:0', which has shape '(?, 1)'

我收到一个错误，我只是检查了每个变量的维度，它看起来都一样，没有任何问题...你能告诉我哪里出了问题以及如何解决吗？

我想做的是天气预报。 输入的形状是( xxxx , 5)，这里 xxxx 是输入数据的行数，5 是输入的类型，包括平均温度等。

输出形状必须是 (yyyy, 1)，因为它的列将预测降水量。

奇怪的是，当程序读取文件时，Data_Y 有一个形状 (hhhh, 5)，本来应该是 (yyyy, 1)。

我认为这导致了这里的所有错误。

输入文件的链接如下

Input file

我该如何解决这个问题？请给我你的帮助。

import tensorflow as tf
import pandas as pd
import numpy as np
import matplotlib
from matplotlib import pyplot as plt

tf.reset_default_graph()
tf.set_random_seed(777)  # reproducibility





def MinMaxScaler(data):

    numerator = data - np.min(data, 0)
    denominator = np.max(data, 0) - np.min(data, 0)
    # noise term prevents the zero division
    return numerator / (denominator + 1e-7)


# train Parameters
seq_length = 6
data_dim = 5
hidden_dim = 10
output_dim = 1
learning_rate = 0.01
iterations = 500




# Open, High, Low, Volume, Close
#df = pd.read_csv("precipitation_post.csv", quotechar='"', decimal=".")
#df = df.interpolate(method ='linear', limit_direction ='forward')
#xy = df.reindex(index=df.index[::-1])
xy = np.loadtxt('df.txt', dtype='double', delimiter=' ', skiprows=1)
#xy = xy[::-1]  

# train/test split
train_size = int(len(xy) * 0.7)
train_set = xy[0:train_size]
test_set = xy[train_size - seq_length:] # Index from [train_size - seq_length] to utilize past sequence

# Scale each
train_set = MinMaxScaler(train_set)
test_set = MinMaxScaler(test_set)
x = xy
y = xy[:, [-1]] # close as label

# build datasets
def build_dataset(time_series, seq_length):
    dataX = []
    dataY = []
    for i in range(0, len(time_series) - seq_length):
        _x = time_series[i:i + seq_length]
        _y = time_series[i + seq_length]
        print(_x, "->", _y)
        dataX.append(_x)
        dataY.append(_y)
    return np.array(dataX), np.array(dataY)

trainX, trainY = build_dataset(train_set, seq_length)
testX, testY = build_dataset(test_set, seq_length)

# input place holders
X = tf.placeholder(tf.float32, shape=[None, seq_length, data_dim])
Y = tf.placeholder(tf.float32, shape=[None, 1])

# build a LSTM network
cell = tf.contrib.rnn.BasicLSTMCell(
    num_units=hidden_dim, state_is_tuple=True, activation=tf.tanh)
outputs, _states = tf.nn.dynamic_rnn(cell, X, dtype=tf.float32)
Y_pred = tf.contrib.layers.fully_connected(
    outputs[:, -1], output_dim, activation_fn=None)  # We use the last cell's output

# cost/loss
loss = tf.reduce_sum(tf.square(Y_pred - Y))  # sum of the squares
# optimizer
optimizer = tf.train.AdamOptimizer(learning_rate)
train = optimizer.minimize(loss)

# RMSE
targets = tf.placeholder(tf.float32, [None, 1])
predictions = tf.placeholder(tf.float32, [None, 1])
rmse = tf.sqrt(tf.reduce_mean(tf.square(targets - predictions)))

with tf.Session() as sess:
    init = tf.global_variables_initializer()
    sess.run(init)

    # Training step
    for i in range(iterations):
        _, step_loss = sess.run([train, loss], feed_dict={
                                X: trainX, Y: trainY})
        print("[step: {}] loss: {}".format(i, step_loss))

    # Test step
    test_predict = sess.run(Y_pred, feed_dict={X: testX})
    rmse_val = sess.run(rmse, feed_dict={
                    targets: testY, predictions: test_predict})
    print("RMSE: {}".format(rmse_val))

    # Plot predictions
plt.plot(testY)
plt.plot(test_predict)
plt.xlabel("Time Period")
plt.ylabel("Precipitation")
plt.show()

Answer 1

鉴于您提供的信息，这里有一个解决方案。很明显，正如您可能已经意识到的那样，问题出在build_dataset 函数中。您需要将您的功能更改为以下内容。

def build_dataset(data, seq_length):
  dataX = []
  dataY = []
  for i in range(seq_length):
    dataX.append(data[i:data.shape[0]-(seq_length-i)].reshape(-1, 1, 5))
  dataX = np.concatenate(dataX, axis=1)
  dataY = data[i+1:train_set.shape[0],4].reshape(-1, 1)
  return dataX, dataY

此函数以下列方式返回数据。假设您有以下几行，

22.90 20.20 31.00 93.00 0.00
22.90 21.20 26.00 91.00 0.00
22.40 20.20 27.40 89.00 0.00
22.40 15.40 29.00 90.00 0.00
21.30 14.40 26.00 82.00 0.00
21.50 20.20 23.00 96.00 0.00
22.10 17.20 23.60 97.00 20.70

它给出X，

22.90 20.20 31.00 93.00 0.00
22.90 21.20 26.00 91.00 0.00
22.40 20.20 27.40 89.00 0.00
22.40 15.40 29.00 90.00 0.00
21.30 14.40 26.00 82.00 0.00
21.50 20.20 23.00 96.00 0.00

输出Y 为20.70

对于完整的数据集，它会产生以下形状。

Input: (6165, 6, 5)
Output: (6165, 1)