【实战】线性回归问题-初遇

线性回归问题—解析法实现一元线性回归

步骤1. 导入库

import tensorflow as tf
import pandas as pd

步骤2. 导入数据集，查看数据集

dataset = pd.read_csv('./lianxi1.csv')
dataset.head()

	x	y
0	32.502345	31.707006
1	53.426804	68.777596
2	61.530358	62.562382
3	47.475640	71.546632
4	59.813208	87.230925

dataset.shape

(100, 2)

dataset.info

<bound method DataFrame.info of             x          y
0   32.502345  31.707006
1   53.426804  68.777596
2   61.530358  62.562382
3   47.475640  71.546632
4   59.813208  87.230925
..        ...        ...
95  50.030174  81.536991
96  49.239765  72.111832
97  50.039576  85.232007
98  48.149859  66.224958
99  25.128485  53.454394

[100 rows x 2 columns]>

dataset.describe()

	x	y
count	100.000000	100.000000
mean	48.958341	72.735051
std	9.746379	16.658187
min	25.128485	31.707006
25%	41.526164	60.788586
50%	49.634970	72.179542
75%	56.762631	83.215080
max	70.346076	118.591217

步骤3. 提取特征和标签

feature_column = ['x']
label_column = ['y']

feature = dataset[feature_column]
label = dataset[label_column]

feature.head()

	x
0	32.502345
1	53.426804
2	61.530358
3	47.475640
4	59.813208

X = feature.values

Y = label.values

X.shape

(100, 1)

步骤4. 计算w,b

meanX = tf.reduce_mean(X)
meanY = tf.reduce_mean(Y)

sumXY = tf.reduce_sum((X-meanX)*(Y-meanY))
sumX = tf.reduce_sum((X-meanX)**2)

w = sumXY/sumX
b = meanY - w*meanX

print('w=',w.numpy())
print('w=',b.numpy())

w= 1.3224310226878684
w= 7.991020985734451