DT Regression

Decision Tree Regression

import pandas as pd

df = pd.read_csv("mins-played.csv")

df

	Day	Outlook	Temp	Humidity	Wind	Minutes Played
0	D1	Sunny	Hot	High	Weak	20
1	D2	Sunny	Hot	High	Strong	24
2	D3	Overcast	Hot	High	Weak	40
3	D4	Rain	Mild	High	Weak	50
4	D5	Rain	Cool	Normal	Weak	60
5	D6	Rain	Cool	Normal	Strong	10
6	D7	Overcast	Cool	Normal	Strong	4
7	D8	Sunny	Mild	High	Weak	10
8	D9	Sunny	Cool	Normal	Weak	60
9	D10	Rain	Mild	Normal	Weak	40
10	D11	Sunny	Mild	High	Strong	45
11	D12	Overcast	Mild	High	Strong	40
12	D13	Overcast	Hot	Normal	Weak	35
13	D14	Rain	Mild	High	Strong	20

df["Minutes Played"].std()

18.3111087402348

import numpy as np
# np.std(df["Minutes Played"].values)

df.query("Wind=='Weak'")["Minutes Played"].std()*len(df.query("Wind=='Weak'"))/len(df)

10.180585192846463

df.query("Wind=='Strong'")["Minutes Played"].std()*len(df.query("Wind=='Strong'"))/len(df)

6.933944897151599

out = {}
for temp in df["Temp"].unique():
    print(temp)
    out[temp] = df.query("Temp==@temp")["Minutes Played"].std()*len(df.query("Temp==@temp"))/len(df)
    print(out[temp])
    print()

Hot
2.6636888135137133

Mild
6.696785704762413

Cool
8.770699519880226

df["Minutes Played"].std() - pd.Series(out).sum()

0.17993470207844808