07wk-035: 아이스크림(이상치) / 의사결정나무

최규빈
2023-10-16

1. 강의영상

https://youtu.be/playlist?list=PLQqh36zP38-weIAPJ89acBOc2V3_0HMcX&si=Lqw2fusURrQqchzH

2. Imports

import numpy as np
import pandas as pd
import sklearn.model_selection
import sklearn.linear_model
import sklearn.tree
import matplotlib.pyplot as plt 
import seaborn as sns

3. Data

np.random.seed(43052)
temp = pd.read_csv('https://raw.githubusercontent.com/guebin/DV2022/master/posts/temp.csv').iloc[:100,3].to_numpy()
temp.sort()
ice_sales = 10 + temp * 0.5 + np.random.randn(100)
ice_sales[0] = 200
df_train = pd.DataFrame({'temp':temp,'ice_sales':ice_sales})
df_train

	temp	ice_sales
0	-4.1	200.000000
1	-3.7	9.234175
2	-3.0	9.642778
3	-1.3	9.657894
4	-0.5	9.987787
...	...	...
95	12.4	17.508688
96	13.4	17.105376
97	14.7	17.164930
98	15.0	18.555388
99	15.2	18.787014

100 rows × 2 columns

plt.plot(df_train.temp,df_train.ice_sales,'o')

4. 분석

- 분석: 의사결정나무

# step1
X = df_train[['temp']]
y = df_train['ice_sales']
# step2 
predictr = sklearn.tree.DecisionTreeRegressor()
# step3 
predictr.fit(X,y)
# step4 
df_train['ice_sales_hat'] = predictr.predict(X)

# plt.plot(df_train.temp,df_train.ice_sales,'o')
# plt.plot(df_train.temp,df_train.ice_sales_hat,'--')
plt.plot(df_train.temp[1:],df_train.ice_sales[1:],'o')
plt.plot(df_train.temp[1:],df_train.ice_sales_hat[1:],'--')

- 12.5~18 구간사이의 unseen data를 가상으로 만들고 예측값을 살펴보자.

XX = df_test = pd.DataFrame({'temp':np.linspace(12.5,18,100)})

df_test['ice_sales_hat'] = predictr.predict(XX)

plt.plot(df_train.temp[1:],df_train.ice_sales[1:],'o',color='C0',alpha=0.5)
plt.plot(df_train.temp[1:],df_train.ice_sales_hat[1:],'--',color='C1',alpha=0.5)
plt.plot(df_test.temp,df_test.ice_sales_hat,'--',color='C2',linewidth=2)

- -15~0 구간사이의 unseen data를 가상으로 만들고 예측값을 살펴보자.

XX = df_test = pd.DataFrame({'temp':np.linspace(-15,0,100)})

df_test['ice_sales_hat'] = predictr.predict(XX)

plt.plot(df_train.temp[1:],df_train.ice_sales[1:],'o',color='C0',alpha=0.5)
plt.plot(df_train.temp[1:],df_train.ice_sales_hat[1:],'--',color='C1',alpha=0.5)
plt.plot(df_test.temp,df_test.ice_sales_hat,'--',color='C2',linewidth=2)

뭐 이 데이터에서는 최선이지 않을까?

6. HW

- 없어요. 다른과목 중간고사 준비 잘하세요!