[머신러닝]ML lab2. 텐서플로우 시작하기. Linear regression

1. 텐서플로우란 무엇인가?

 Tensor(Node)가 어떤 설계된 그래프를 통해서 흐르는 (Flow) 것.

그 과정을 통해서 연산이된다.

즉, 어떤 데이터 또는 변수 그리고 계산식은 Tensor(Node)라고 이해할 수 있다.

2. 실습코드

# Lab 2 Linear Regression

import tensorflow as tf

tf.set_random_seed(777) # for reproducibility

# X and Y data

x_train = [1, 2, 3]

y_train = [1, 2, 3]

# Try to find values for W and b to compute y_data = x_data * W + b

# We know that W should be 1 and b should be 0

# But let TensorFlow figure it out

# Variable:텐서플로우가 변경시키는 값. 학습하면서 변경하는 값

# 값이 하나인 Random 한 값을 준다.

W = tf.Variable(tf.random_normal([1]), name='weight')

b = tf.Variable(tf.random_normal([1]), name='bias')

# Our hypothesis XW+b

hypothesis = x_train * W + b

# cost/loss function

# reduce_mean : 평균

# cost : 에러의 정도, 오차의 정도

cost = tf.reduce_mean(tf.square(hypothesis - y_train))

# Minimize

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

train = optimizer.minimize(cost)

# Launch the graph in a session.

sess = tf.Session()

# Initializes global variables in the graph.

# variable 사용시 활성화 필요

sess.run(tf.global_variables_initializer())

# Fit the line

for step in range(2001):

sess.run(train)

if step % 20 == 0: # 20번에 한번 씩 출력

print(step, sess.run(cost), sess.run(W), sess.run(b))

# Learns best fit W:[ 1.], b:[ 0.]

'''

0 2.82329 [ 2.12867713] [-0.85235667]

20 0.190351 [ 1.53392804] [-1.05059612]

40 0.151357 [ 1.45725465] [-1.02391243]

...

1920 1.77484e-05 [ 1.00489295] [-0.01112291]

1940 1.61197e-05 [ 1.00466311] [-0.01060018]

1960 1.46397e-05 [ 1.004444] [-0.01010205]

1980 1.32962e-05 [ 1.00423515] [-0.00962736]

2000 1.20761e-05 [ 1.00403607] [-0.00917497]

'''

3. Placeholders

비어 있는 변수로 설정하여 원하는 입력값을 넣어줄 수 있다.

# Lab 2 Linear Regression

import tensorflow as tf

tf.set_random_seed(777) # for reproducibility

# Try to find values for W and b to compute y_data = W * x_data + b

# We know that W should be 1 and b should be 0

# But let's use TensorFlow to figure it out

W = tf.Variable(tf.random_normal([1]), name='weight')

b = tf.Variable(tf.random_normal([1]), name='bias')

# Now we can use X and Y in place of x_data and y_data

# # placeholders for a tensor that will be always fed using feed_dict

# See http://stackoverflow.com/questions/36693740/

# 비어있는 변수로 설정하고 입력값 설정이 가능하다

# [None] 아무 값이나 들어올 수 있다

X = tf.placeholder(tf.float32, shape=[None])

Y = tf.placeholder(tf.float32, shape=[None])

# Our hypothesis XW+b

hypothesis = X * W + b

# cost/loss function

cost = tf.reduce_mean(tf.square(hypothesis - Y))

# Minimize

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

train = optimizer.minimize(cost)

# Launch the graph in a session.

sess = tf.Session()

# Initializes global variables in the graph.

sess.run(tf.global_variables_initializer())

# Fit the line

for step in range(2001):

cost_val, W_val, b_val, _ = \

sess.run([cost, W, b, train],

feed_dict={X: [1, 2, 3], Y: [1, 2, 3]})

if step % 20 == 0:

print(step, cost_val, W_val, b_val)

# Learns best fit W:[ 1.], b:[ 0]

'''

...

1980 1.32962e-05 [ 1.00423515] [-0.00962736]

2000 1.20761e-05 [ 1.00403607] [-0.00917497]

'''

# Testing our model

print(sess.run(hypothesis, feed_dict={X: [5]}))

print(sess.run(hypothesis, feed_dict={X: [2.5]}))

print(sess.run(hypothesis, feed_dict={X: [1.5, 3.5]}))

'''

[ 5.0110054]

[ 2.50091505]

[ 1.49687922 3.50495124]

'''

# Fit the line with new training data

for step in range(2001):

cost_val, W_val, b_val, _ = \

sess.run([cost, W, b, train],

feed_dict={X: [1, 2, 3, 4, 5],

Y: [2.1, 3.1, 4.1, 5.1, 6.1]})

if step % 20 == 0:

print(step, cost_val, W_val, b_val)

# Testing our model

print(sess.run(hypothesis, feed_dict={X: [5]}))

print(sess.run(hypothesis, feed_dict={X: [2.5]}))

print(sess.run(hypothesis, feed_dict={X: [1.5, 3.5]}))

'''

1960 3.32396e-07 [ 1.00037301] [ 1.09865296]

1980 2.90429e-07 [ 1.00034881] [ 1.09874094]

2000 2.5373e-07 [ 1.00032604] [ 1.09882331]

[ 6.10045338]

[ 3.59963846]

[ 2.59931231 4.59996414]

'''