Question

我正在使用使用Tensorflow的线性回归模型。我在sess.run部分中遇到一些错误。请帮助我解决这个问题。

第一部分：

for epoch in range(training_epochs):
    #for (x,y) in zip(train_X,train_Y):
        sess.run(training_step,feed_dict={X:train_X,Y:train_Y})

错误：不兼容的形状：[5,1]与[40,1]

第二部分：

pred_Y = sess.run(pred, feed_dict={X: test_X})

错误：不兼容的形状：[10,1]与[5,1]

我的整个代码是：

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


# Importing the dataset
dataset = pd.read_csv("50_Startups.csv")
X_data = dataset.iloc[:, :-1].values
Y_data = dataset.iloc[:, 4].values

# Encoding categorical data
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
labelencoder = LabelEncoder()
X_data[:, 3] = labelencoder.fit_transform(X_data[:, 3])
onehotencoder = OneHotEncoder(categorical_features = [3])
X_data = onehotencoder.fit_transform(X_data).toarray()

# Avoiding the Dummy Variable Trap
X_data = X_data[:, 1:] #excluding the 1st dummy column


from sklearn.cross_validation import train_test_split
train_X, test_X, train_Y, test_Y = train_test_split(X_data, Y_data, test_size = 0.2, random_state = 0)
train_Y = np.array(train_Y).reshape(40,1)
test_Y = np.array(test_Y).reshape(10,1)

# Parameters
learning_rate = 0.01
training_epochs = 1000
display_step = 50
cost_history = np.empty(shape=[1],dtype=float)
m_dim = train_X.shape[0]
n_dim = train_X.shape[1]

X = tf.placeholder(tf.float32,[None,n_dim])
Y = tf.placeholder(tf.float32,[None,1])
W = tf.Variable(tf.ones((n_dim,1)))
B = tf.Variable(tf.ones((n_dim,1)))

init = tf.global_variables_initializer()


pred = tf.add(tf.matmul(X, W),B)
cost = tf.reduce_mean(tf.square(pred - Y))
training_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)

sess = tf.Session()
sess.run(init)

for epoch in range(training_epochs):
#for (x,y) in zip(train_X,train_Y):
sess.run(training_step,feed_dict={X:train_X,Y:train_Y})
# Display logs per epoch step
if (epoch+1) % display_step == 0:
c = sess.run(cost, feed_dict={X: train_X, Y:train_Y}) 
print("Epoch:", '%04d' % (epoch+1), "cost=", "{:.9f}".format(c), \
"W=", sess.run(W), "b=", sess.run(B))


pred_Y = sess.run(pred, feed_dict={X: test_X}) 
mse = tf.reduce_mean(tf.square(pred_Y - test_Y))
print("MSE: %.4f" % sess.run(mse))

如何在sess.run中正确feed_dict？

0 个答案: