Schuhgröße: MAE “zu Fuß”

import pandas as pd

datei = "synthetische_daten_3bis40_mit_BMI_Kategorien.csv"

#with open(datei) as f:
#    mein_text = f.read()
#mein_text.split("\n")

df = pd.read_csv(datei)

df.head()

	ID	Alter_Jahre	Geschlecht	Körpergröße_cm	Gewicht_kg	BMI	BMI_Kategorie	Schuhgröße_EU
0	1	20	w	165.6	58.1	21.2	Normalgewicht	40
1	2	4	m	101.1	11.8	11.5	Untergewichtig	25
2	3	12	m	147.5	35.7	16.4	Normalgewicht	35
3	4	19	m	177.4	68.0	21.6	Normalgewicht	42
4	5	6	m	122.0	22.8	15.3	Normalgewicht	30

from sklearn.model_selection import train_test_split

y = df["Schuhgröße_EU"]

type(y)

pandas.core.series.Series

df.columns

Index(['ID', 'Alter_Jahre', 'Geschlecht', 'Körpergröße_cm', 'Gewicht_kg',
       'BMI', 'BMI_Kategorie', 'Schuhgröße_EU'],
      dtype='object')

#meine_cols = ['ID', 'Alter_Jahre', 'Geschlecht', 'Körpergröße_cm', 'Gewicht_kg',
#       'BMI', 'BMI_Kategorie']

X = df.drop(["Geschlecht" , "Schuhgröße_EU", "BMI_Kategorie" ] , axis=1)

X

	ID	Alter_Jahre	Körpergröße_cm	Gewicht_kg	BMI
0	1	20	165.6	58.1	21.2
1	2	4	101.1	11.8	11.5
2	3	12	147.5	35.7	16.4
3	4	19	177.4	68.0	21.6
4	5	6	122.0	22.8	15.3
...	...	...	...	...	...
1495	1496	22	189.3	88.9	24.8
1496	1497	35	166.1	62.6	22.7
1497	1498	28	165.2	65.5	24.0
1498	1499	39	172.3	68.0	22.9
1499	1500	32	180.6	82.8	25.4

1500 rows × 5 columns

X.describe()

	ID	Alter_Jahre	Körpergröße_cm	Gewicht_kg	BMI
count	1500.000000	1500.000000	1500.000000	1500.000000	1500.000000
mean	750.500000	18.564667	154.977533	53.709733	20.777133
std	433.157015	10.557531	25.751946	23.149581	4.299159
min	1.000000	3.000000	81.300000	7.000000	9.500000
25%	375.750000	10.000000	138.500000	31.975000	17.400000
50%	750.500000	17.000000	164.700000	58.900000	21.100000
75%	1125.250000	27.000000	173.800000	71.150000	23.800000
max	1500.000000	40.000000	194.400000	110.300000	33.500000

from sklearn.model_selection import train_test_split

train_X, val_X, train_y, val_y = train_test_split(X, y, random_state = 0)

train_X.shape

(1125, 5)

val_X.shape

(375, 5)

from sklearn.linear_model import LinearRegression

reg = LinearRegression()

reg.fit(train_X, train_y)

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.

val_X

	ID	Alter_Jahre	Körpergröße_cm	Gewicht_kg	BMI
471	472	15	170.9	65.1	22.3
9	10	8	125.7	21.0	13.3
1499	1500	32	180.6	82.8	25.4
54	55	3	102.8	16.1	15.2
1411	1412	37	165.5	63.5	23.2
...	...	...	...	...	...
75	76	4	105.5	20.3	18.2
481	482	6	107.9	18.9	16.2
1350	1351	22	179.9	85.1	26.3
748	749	7	122.4	20.7	13.8
1362	1363	22	187.7	88.4	25.1

375 rows × 5 columns

predict_y = reg.predict( val_X ) 

len(predict_y), len(val_y)

(375, 375)

#predict_y[:10]

#val_y[:10]

differenz = []

for i in range(0,10):
    v = list(val_y)[i]
    p = list(predict_y)[i]
    
    # print( f"val: {v}, predict: {p}")
    differenz.append( abs(v-p) )
differenz

[np.float64(0.3022913002060861),
 np.float64(0.4896670825697669),
 np.float64(0.1178661698626513),
 np.float64(0.36005438749023),
 np.float64(0.46754436728781457),
 np.float64(0.1530543092586072),
 np.float64(0.30683828995378093),
 np.float64(0.4532737267499307),
 np.float64(0.2105452874775935),
 np.float64(0.04441925830533222)]

MAE = sum( differenz ) / len( differenz )
MAE

np.float64(0.29055541791617934)