Question

program for applying linear regression on dataset using tensorflow
# About housing_data: 
# dataset characterisitics:multivariate
#Associate task:          regression
#Number of instance:      506
#number of attribut:      14

#Attribute Information:

#1. CRIM: per capita crime rate by town
#2. ZN: proportion of residential land zoned for lots over 25,000 sq.ft.
#3. INDUS: proportion of non-retail business acres per town
#4. CHAS: Charles River dummy variable
#5. NOX: nitric oxides concentration (parts per 10 million)
#6. RM: average number of rooms per dwelling
#7. AGE: proportion of owner-occupied units built prior to 1940
#8. DIS: weighted distances to five Boston employment centres
#9. RAD: index of accessibility to radial highways
#10. TAX: full-value property-tax rate per $10,000
#11. PTRATIO: pupil-teacher ratio by town
#12. B: 1000(Bk - 0.63)^2 where Bk is the proportion of blacks by town
#13. LSTAT: % lower status of the population
#14. MEDV: Median value of owner-occupied homes in $1000's

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

batch_size=50

#make queue of file
filename_queue = tf.train.string_input_producer(
tf.train.match_filenames_once("housing_data.csv"),
shuffle=True)

line_reader = tf.TextLineReader(skip_header_lines=1)

_, csv_row = line_reader.read(filename_queue)

record_defaults = [[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]]

CRIM,ZN,INDUS,CHAS,NOX,RM,AGE,DIS,RAD,TAX,PTRATIO,B,LSTAT,MEDV = tf.decode_csv(
                        csv_row, record_defaults=record_defaults)

# Turn the features back into a tensor.
features =    tf.pack([CRIM,ZN,INDUS,CHAS,NOX,RM,AGE,DIS,RAD,TAX,PTRATIO,B,LSTAT,MEDV])

#variable which we need to fill when we are ready to comput the graph
x=tf.placeholder(tf.float32)
y=tf.placeholder(tf.float32)

W = tf.Variable(tf.random_normal([1]), name='weight')
b = tf.Variable(tf.random_normal([1]), name='bias')
y_pred = tf.add(tf.mul(x, W), b)

error=tf.reduce_sum((y-y_pred)**2/506)

learning_rate = 0.01
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(error)

#create the session to use the graph
with tf.Session() as sess:
  # Here we tell tensorflow that we want to initialize all
  # the variables in the graph so we can use them 
  #tf.initialize_all_variables().run()
  sess.run(tf.initialize_all_variables())
  sess.run([features,MEDV])

  coord = tf.train.Coordinator()
  threads = tf.train.start_queue_runners(coord=coord)

  sess.run([features,MEDV])

  #gradient descent loop for 500 iteration
  for _ in range(500):
      #select random mini batch
      indices=np.random.choice(506,batch_size)
      x_batch,y_batch=features[indices],MEDV[indices]

      #do gradient descent
    _,loss_val=sess.run([optimizer,error],feed_dict={x:x_batch,y:y_batch})

    print _,loss_val




coord.request_stop()
coord.join(threads)

---------------------------错误------------------- ---------------------

When I am running this code I am getting following error.
error:
 Traceback (most recent call last):
 File "linear_reg_tf.py", line 92, in <module>
 x_batch,y_batch=features[indices],MEDV[indices]
 File "/usr/local/lib/python2.7/dist-packages/tensorflow/pytho/ops/array_ops.py", line 161, in _SliceHelper
raise TypeError("Bad slice index %s of type %s" % (s, type(s)))
TypeError: Bad slice index [468 105 218 124 492 428 464 194 110  76 165   127 480 414  88 332  54  68

195 162 223 65 340 279 390 390 1 127 290 82 430 154 202 389 478 42 410 25 161 42 143 256 481 417 263 92 335 101 430 334]类型

Answer 1

也许代替x_batch,y_batch=features[indices],MEDV[indices] 你可以尝试

x_batch,y_batch = [], []
for index in indices:
    x_batch.append(features[index])
    y_batch.append(MEDV[index])

Answer 2

此代码正在运行，但我想知道我们是否要测试一些新数据，然后我们将如何测试或如何将数据拆分为训练集和测试集

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

batch_size=50

filename_queue = tf.train.string_input_producer(
    tf.train.match_filenames_once("housing_data.csv"),
    shuffle=True)


# for each column below.
line_reader = tf.TextLineReader(skip_header_lines=1)

_, csv_row = line_reader.read(filename_queue)

# Type information and column names based on the decoded CSV.
 record_defaults = [[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]]

CRIM,ZN,INDUS,CHAS,NOX,RM,AGE,DIS,RAD,TAX,PTRATIO,B,LSTAT,MEDV =     tf.decode_csv(
                        csv_row, record_defaults=record_defaults)

# Turn the features back into a tensor.
features =    tf.pack([CRIM,ZN,INDUS,CHAS,NOX,RM,AGE,DIS,RAD,TAX,PTRATIO,B,LSTAT,MEDV])

#variable which we need to fill when we are ready to comput the graph
x=tf.placeholder(dtype=features.dtype)
y=tf.placeholder(dtype=MEDV.dtype)

W = tf.Variable(tf.random_normal([1]), name='weight')
b = tf.Variable(tf.random_normal([1]), name='bias')
y_pred = tf.add(tf.mul(x, W), b)


#loss function will measure the distance between our observations and    predictions
#and average over them.Here housing data have 506 instances so divide by 506.
error=tf.reduce_sum((y-y_pred)**2/506)


learning_rate = 0.01
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(error)

#create会话以使用图表使用tf.Session（）作为sess：

  sess.run(tf.initialize_all_variables())

  coord = tf.train.Coordinator()
  threads = tf.train.start_queue_runners(coord=coord)

  x_data =np.array(sess.run(features))
  y_data =np.array(sess.run(MEDV))

  #gradient descent loop for 10 iteration
  for _ in range(10):
      sess.run([features,MEDV])

      #do gradient descent
      _,loss_val=sess.run([optimizer,error],feed_dict={x:x_data,y:y_data})

      print _,loss_val


coord.request_stop()
coord.join(threads)

谁能帮我解决运行这个tensorflow程序时遇到的错误

2 个答案: