[데이터처리와분석] 데이터 준비하기

개요¶

1. ML을 위한 첫 데이터를 다운받고 데이터에 대해서 확인해보는 과정¶

2. 사용 메소드(전처리)¶

1. read_csv(파일,인코딩)
2. train_set.copy()
3. df_copy.corr()
4. train_x = train_set.drop(["Salary"], axis=1), train_y = df_copy["Salary"]
5. train_set, test_set = train_test_split(df, test_size=0.2, random_state=42)

3. 사용 메소드(학습)¶

1. LinearRegression()
2. lin_reg.fit(train_x, train_y)
3. lin_reg.predict(train_x)

4. 사용 메소드(Plot)¶

1. df_copy.plot.scatter(x='YearsExperience', y='Salary')
2. plt.plot(train_x,train_y_hat, color='red', linewidth=2)

In [ ]:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import statsmodels.api as sm
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import r2_score

In [ ]:

#sns.set()
%matplotlib inline

df = pd.read_csv("SalaryData.csv")

In [ ]:

train_set, test_set = train_test_split(df, test_size=0.2, random_state=42)

In [ ]:

df_copy = train_set.copy()

In [ ]:

df_copy.plot.scatter(x='YearsExperience', y='Salary')
# plot을 이용해서 .plot.scatter(x,y)를 설정해준 것

Out[ ]:

<Axes: xlabel='YearsExperience', ylabel='Salary'>

In [ ]:

df_copy.corr()
# .corr을 통해서 각기 변수에 대해서 correlation을 구한 것이다.

Out[ ]:

	YearsExperience	Salary
YearsExperience	1.00000	0.98211
Salary	0.98211	1.00000

In [ ]:

train_x = train_set.drop(["Salary"], axis=1)
train_y = df_copy["Salary"]
# 학습하기 위해서 데이터를 가져오기 (x는 행으로 가져와야함)

In [ ]:

lin_reg = LinearRegression()
# LinearRegression을 통해서 데이터만 넣으면 바로 값이 나온다.
lin_reg.fit(train_x, train_y)
# .fit으로 학습하기

Out[ ]:

LinearRegression()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

Linearregression의 값을 확인해 볼 수 있는 변수

1. coef_
2. intercept_

In [ ]:

print("Coefficients: ", lin_reg.coef_)
# 계수
print("Intercept: ", lin_reg.intercept_)
# 그 값

Coefficients:  [9423.81532303]
Intercept:  25321.583011776813

In [ ]:

plt.scatter(train_x, train_y,  color='blue')
# plot.scatter를 통해서 값을 넣기
train_y_hat = lin_reg.predict(train_x)
plt.plot(train_x,train_y_hat, color='red', linewidth=2)

Out[ ]:

[<matplotlib.lines.Line2D at 0x152870570d0>]