Scribd
Upload
117 views5 pages

TensorFlow Linear Regression Guide

This document summarizes using linear regression in TensorFlow to model the relationship between age, weight, and blood fat content using sample data. It defines the input data, model, loss function, training operation, and evaluates the trained model on some test inputs. The model is trained over 1000 steps to minimize the loss and fit the weights and bias between the input features and blood fat content target variable.

Uploaded by

LalithaJyothi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as RTF, PDF, TXT or read online on Scribd
117 views5 pages

TensorFlow Linear Regression Guide

This document summarizes using linear regression in TensorFlow to model the relationship between age, weight, and blood fat content using sample data. It defines the input data, model, loss function, training operation, and evaluates the trained model on some test inputs. The model is trained over 1000 steps to minimize the loss and fit the weights and bias between the input features and blood fat content target variable.

Uploaded by

LalithaJyothi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as RTF, PDF, TXT or read online on Scribd
You are on page 1/ 5

https://codability.

in/a-guide-to-tensorflow-linear-regression-part-5/

Reference:

# D G Kleinbaum and L L Kupper,

# Applied Regression Analysis and Other Multivariable Methods,

# Duxbury Press, 1978, page 149.

# Helmut Spaeth,

# Mathematical Algorithms for Linear Regression,

# Academic Press, 1991,

# ISBN 0-12-656460-4.

# Discussion:

# Age and weight are to be related to blood fat content.

# There are 25 rows of data. The data columns include:

# I, the index;

# A0, 1;

# A1, the weight;

# A2, the age;

# B, the blood fat content.

#
# We seek a model of the form:

# B = A0 * X0 + A1 * X1 + A2 * X2.

5 columns

25 rows

Index

One

Weight (kilograms)

Age (Years)

Blood fat content

1 1 84 46 354

2 1 73 20 190

3 1 65 52 405

4 1 70 30 263

5 1 76 57 451

6 1 69 25 302

7 1 63 28 288

8 1 72 36 385

9 1 79 57 402

10 1 75 44 365

11 1 27 24 209

12 1 89 31 290

13 1 65 52 346

14 1 57 23 254
15 1 59 60 395

16 1 69 48 434

17 1 60 34 220

18 1 79 51 374

19 1 75 50 308

20 1 82 34 220

21 1 59 46 311

22 1 67 23 181

23 1 85 37 274

24 1 55 40 303

25 1 63 30 244

Solution in Tensorflow -

import tensorflow as t

W = t.Variable(t.zeros([2, 1]), name="weights")

b = t.Variable(0., name="bias")

def inputs():

weight_age = [[84, 46], [73, 20], [65, 52], [70, 30], [76, 57], [69, 25], [63, 28], [72, 36], [79, 57],
[75, 44], [27, 24], [89, 31], [65, 52], [57, 23], [59, 60], [69, 48], [60, 34], [79, 51], [75, 50], [82,
34], [59, 46], [67, 23], [85, 37], [55, 40], [63, 30]]

blood_fat_content = [354, 190, 405, 263, 451, 302, 288, 385, 402, 365, 209, 290, 346, 254,
395, 434, 220, 374, 308, 220, 311, 181, 274, 303, 244]

return t.to_float(weight_age), t.to_float(blood_fat_content)

def inference(X):

return t.matmul(X, W) + b
def loss(X,Y):

Y_predicted = inference(X)

return t.reduce_sum(t.squared_difference(Y, Y_predicted))

def train(total_loss):

learning_rate = 0.0000001

return t.train.GradientDescentOptimizer(learning_rate).minimize(total_loss)

def evaluates(sess, X, Y):

print(sess.run(inference([[80. ,25.]])))

print(sess.run(inference([[65. ,25.]])))

with t.Session() as sess:

t.global_variables_initializer().run()

X, Y = inputs()

total_loss = loss(X, Y)

train_op = train(total_loss)

coord = t.train.Coordinator()

training_steps = 1000

for step in range(training_steps):

sess.run([train_op])

evaluates(sess, X, Y)
coord.request_stop()

sess.close()

You might also like