빅분기 2유형 문제연습(회귀-2)

대학원 입학가능성 데이터¶

데이터 설명 : 대학원 입학 가능성 예측( 종속변수 : Chance of Admit)¶

x_train: https://raw.githubusercontent.com/Datamanim/datarepo/main/admission/x_train.csv¶

x_test: https://raw.githubusercontent.com/Datamanim/datarepo/main/admission/x_test.csv¶

y_train: https://raw.githubusercontent.com/Datamanim/datarepo/main/admission/y_train.csv¶

y_test(평가용) : https://raw.githubusercontent.com/Datamanim/datarepo/main/admission/y_test.csv¶

In [80]:

# 데이터 불러오기
import pandas as pd
x_train = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/admission/x_train.csv')
x_test = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/admission/x_test.csv')
y_train = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/admission/y_train.csv')
y_test = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/admission/y_test.csv')


# id, Serial No. 컬럼은 제거해주자
X_train = x_train.drop(columns=['ID', 'Serial No.'])
X_test = x_test.drop(columns=['ID', 'Serial No.'])
display(X_train)
display(y_train.head())
y_train = y_train['Chance of Admit']
y_test = y_test['Chance of Admit']

#print(X_train.info())
#print(X_train.nunique())
#print(X_train.isnull().sum()) # 널 값은 없음
#print(X_test.isnull().sum()) # 널 값은 없음

# Research, SOP, LOR, University Rating 이 네 개의 컬럼은
# 수치형 변수이지만 범주형 데이터 이므로 object로 바꿔주자
for i in ['University Rating', 'SOP', 'LOR', 'Research']:
    X_train[i] = X_train[i].astype('object')
    X_test[i] = X_test[i].astype('object')
    


# 수치형 데이터 이상치를 평균으로 대체해주자
def solution(data, col):
    q3 = data[col].quantile(0.75)
    q1 = data[col].quantile(0.25)
    iqr = q3 - q1
    
    upper = q3 + (1.5 * iqr)
    lower = q1 - (1.5 * iqr)
    data[col] = data[col].map(lambda x: data[col].mean() if x > upper or x < lower else x)
    return data

for i in X_train.select_dtypes(exclude='object').columns:
    solution(X_train, i)
    solution(X_test, i)
    
    
    
# 데이터 분할 (학습 데이터 0.67, 검증 데이터 0.33)
from sklearn.model_selection import train_test_split
X_train, X_validation, Y_train, Y_validation = train_test_split(X_train, y_train, test_size=0.33, random_state=43)
X_train.reset_index(drop=True, inplace=True)
X_validation.reset_index(drop=True, inplace=True)



# 수치형 데이터 스케일링 해주자 (StandardScaler)
from sklearn.preprocessing import StandardScaler
sds = StandardScaler()
obj_col = X_train.select_dtypes(include='object').columns
sds.fit(X_train.drop(columns=obj_col))

X_train_sc = sds.transform(X_train.drop(columns=obj_col))
X_train_sc = pd.DataFrame(X_train_sc, columns=X_train.drop(columns=obj_col).columns)

X_validation_sc = sds.transform(X_validation.drop(columns=obj_col))
X_validation_sc = pd.DataFrame(X_validation_sc, columns=X_validation.drop(columns=obj_col).columns)

X_test_sc = sds.transform(X_test.drop(columns=obj_col))
X_test_sc = pd.DataFrame(X_test_sc, columns=X_test.drop(columns=obj_col).columns)

for i in obj_col:
    X_train_sc[i] = X_train[i]
    X_validation_sc[i] = X_validation[i]
    X_test_sc[i] = X_test[i]

    
    
# 수치형 데이터 로그 변환을 수행해주자
import numpy as np
for i in X_train_sc.select_dtypes(exclude='object').columns:
    if X_train_sc[i].min() < 0:
        X_train_sc[i] = X_train_sc[i].map(lambda x: x + abs(X_train_sc[i].min()))
    if X_validation_sc[i].min() < 0:
        X_validation_sc[i] = X_validation_sc[i].map(lambda x: x + abs(X_validation_sc[i].min()))
    if X_test_sc[i].min() < 0:
        X_test_sc[i] = X_test_sc[i].map(lambda x: x + abs(X_test_sc[i].min()))
        
    X_train_sc[i] = np.log1p(X_train_sc[i])
    X_validation_sc[i] = np.log1p(X_validation_sc[i])
    X_test_sc[i] = np.log1p(X_test_sc[i])



# 모델링 하기 전에 object 컬럼 인코딩을 해주자
# 범주의 개수가 많지 않기 때문에 원핫 인코딩을 수행
from sklearn.preprocessing import LabelEncoder
labelencoder = LabelEncoder()
x_full = pd.concat([X_train_sc, X_validation_sc, X_test_sc])

for i in x_full.select_dtypes(include='object').columns:
    labelencoder.fit(x_full[i])
    X_train_sc[i] = labelencoder.transform(X_train_sc[i])
    X_validation_sc[i] = labelencoder.transform(X_validation_sc[i])
    X_test_sc[i] = labelencoder.transform(X_test_sc[i])



# RandomForestRegressor을 사용하여 모델링을 하자
from sklearn.ensemble import RandomForestRegressor
RFR = RandomForestRegressor(random_state=40)
RFR.fit(X_train_sc, Y_train)

pred_train = RFR.predict(X_train_sc)
pred_validation = RFR.predict(X_validation_sc)



# 실제 데이터와 예측한 데이터를 비교해 성능평가를 해보자
from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score

mse_train = mean_squared_error(Y_train, pred_train)
mse_validation = mean_squared_error(Y_validation, pred_validation)

mae_train = mean_absolute_error(Y_train, pred_train)
mae_validation = mean_absolute_error(Y_validation, pred_validation)

rmse_train = np.sqrt(mse_train)
rmse_validation = np.sqrt(mse_validation)

r2_score_train = r2_score(Y_train, pred_train)
r2_score_validation = r2_score(Y_validation, pred_validation)

print('mse_train', mse_train)
print('mae_train', mae_train)
print('rmse_train', rmse_train)
print('r2_score_train', r2_score_train)
print('\n')
print('mse_validation', mse_validation)
print('mae_validation', mae_validation)
print('rmse_validation', rmse_validation)
print('r2_score_validation', r2_score_validation)


#최종적으로 test데이터를 예측하고 성능평가를 진행하고
#예측 결과를 csv로 만들어 제출하자

pred_test = RFR.predict(X_test_sc)

mse_test = mean_squared_error(y_test, pred_test)
mae_test = mean_absolute_error(y_test, pred_test)
rmse_test = np.sqrt(mse_test)
r2_score_test = r2_score(y_test, pred_test)

print('\n')
print('mse_test',mse_test)
print('mae_test',mae_test)
print('rmse_test',rmse_test)
print('r2_score_test',r2_score_test)


pd.DataFrame({'ID':x_test['ID'], 'Chance of Admit':pred_test}).to_csv('school.csv', index=False)

	GRE Score	TOEFL Score	University Rating	SOP	LOR	CGPA	Research
0	327	114	3	3.0	3.0	9.02	0
1	321	109	4	4.0	4.0	8.68	1
2	301	99	3	2.5	2.0	8.45	1
3	317	106	2	2.0	3.5	8.12	0
4	321	111	3	3.5	4.0	8.83	1
...	...	...	...	...	...	...	...
395	312	109	2	2.5	4.0	9.02	0
396	316	104	3	3.0	3.5	8.00	1
397	295	99	2	2.5	3.0	7.65	0
398	296	95	2	3.0	2.0	7.54	1
399	304	103	5	5.0	3.0	7.92	0

400 rows × 7 columns

	ID	Chance of Admit
0	0	0.61
1	1	0.69
2	2	0.68
3	3	0.73
4	4	0.77

mse_train 0.0005432898507462684
mae_train 0.016814179104477644
rmse_train 0.023308578908768085
r2_score_train 0.9729979375154096


mse_validation 0.0073422296969696986
mae_validation 0.0646742424242424
rmse_validation 0.08568681168633653
r2_score_validation 0.6351237590655042


mse_test 0.0024170792999999944
mae_test 0.036169
rmse_test 0.04916380070743102
r2_score_test 0.8666167456988487

Out[80]:

	ID	Chance of Admit
0	13	0.6771
1	16	0.7251
2	23	0.8339
3	25	0.5957
4	29	0.5699
...	...	...
95	467	0.5985
96	484	0.9164
97	485	0.6600
98	489	0.8742
99	499	0.5354

100 rows × 2 columns

레드 와인 퀄리티 예측 데이터¶

데이터 설명 : 레드 와인 퀼리티 예측문제 (종속변수 : quality)¶

x_train: https://raw.githubusercontent.com/Datamanim/datarepo/main/redwine/x_train.csv¶

x_test: https://raw.githubusercontent.com/Datamanim/datarepo/main/redwine/x_test.csv¶

y_train: https://raw.githubusercontent.com/Datamanim/datarepo/main/redwine/y_train.csv¶

y_test(평가용) : https://raw.githubusercontent.com/Datamanim/datarepo/main/redwine/y_test.csv¶

In [134]:

#데이터 불러오기
x_train = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/redwine/x_train.csv')
x_test = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/redwine/x_test.csv')
y_train = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/redwine/y_train.csv')
y_test = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/redwine/y_test.csv')

#id컬럼은 제거해주자
X_train = x_train.drop(columns='ID')
X_test = x_test.drop(columns='ID')

display(X_train)
display(y_train.head())

y_train = y_train['quality']
y_test = y_test['quality']
#print(X_train.info())
#print(X_train.nunique()) #범주형 컬럼이 없다??? 다수치형
#print(X_train.isnull().sum()) #널 값은 없다
#print(X_test.isnull().sum()) #널 값은 없다



#일단은 수치형 컬럼 이상치를 제거해주자
def solution(data, col):
    q3 = data[col].quantile(0.75)
    q1 = data[col].quantile(0.25)
    iqr = q3 - q1
    
    upper = q3 + (1.5*iqr)
    lower = q1 - (1.5*iqr)
    data[col] = data[col].map(lambda x: data[col].mean() if (x>upper) or (x<lower) else x)
    return data

for i in X_train.select_dtypes(exclude='object').columns:
    solution(X_train,i)
    solution(X_test,i)
    

#데이터를 분할해 주자!!!
from sklearn.model_selection import train_test_split
X_train, X_validation, Y_train, Y_validation = train_test_split(X_train, y_train, test_size=0.33, random_state = 43)
#print(X_train.shape, X_validation.shape, Y_train.shape, Y_validation.shape)
X_train.reset_index(drop=True, inplace=True)
X_validation.reset_index(drop=True, inplace=True)



#수치형 데이터 스케일링 해주자(StandardScaler)
obj_col = X_train.select_dtypes(include='object').columns
from sklearn.preprocessing import StandardScaler
sds = StandardScaler()
sds.fit(X_train.drop(columns = obj_col))

X_train_sc = sds.transform(X_train.drop(columns=obj_col))
X_train_sc = pd.DataFrame(X_train_sc, columns = X_train.drop(columns=obj_col).columns)

X_validation_sc = sds.transform(X_validation.drop(columns=obj_col))
X_validation_sc = pd.DataFrame(X_validation_sc, columns = X_validation.drop(columns=obj_col).columns)

X_test_sc = sds.transform(X_test.drop(columns=obj_col))
X_test_sc = pd.DataFrame(X_test_sc, columns = X_test.drop(columns=obj_col).columns)

for i in obj_col:
    X_train_sc[i] = X_train[i]
    X_validation_sc[i] = X_validation[i]
    X_test_sc[i] = X_test[i]

    

#수치형 데이터 로그 변환 해주자
import numpy as np
for i in X_train_sc.select_dtypes(exclude='object').columns:
    if X_train_sc[i].min()<0:
        X_train_sc[i] = X_train_sc[i].map(lambda x : x+abs(X_train_sc[i].min()))
    X_train_sc[i] = np.log1p(X_train_sc[i])
        
    if X_validation_sc[i].min()<0:
        X_validation_sc[i] = X_validation_sc[i].map(lambda x : x+abs(X_validation_sc[i].min()))
    X_validation_sc[i] = np.log1p(X_validation_sc[i])
    
    if X_test_sc[i].min()<0:
        X_test_sc[i] = X_test_sc[i].map(lambda x : x+abs(X_test_sc[i].min()))
    X_test_sc[i] = np.log1p(X_test_sc[i])    

    
    
#오프젝트 컬럼이 없기 떄문에 인코딩은 해줄 필요가 없음 바로 모델링하자
from sklearn.ensemble import RandomForestRegressor
RFR = RandomForestRegressor(random_state=40)
RFR.fit(X_train_sc, Y_train)

pred_train = RFR.predict(X_train_sc)
pred_validation = RFR.predict(X_validation_sc)


from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score

mse_train = mean_squared_error(Y_train, pred_train)
mse_validation = mean_squared_error(Y_validation, pred_validation)

mae_train = mean_absolute_error(Y_train, pred_train)
mae_validation = mean_absolute_error(Y_validation, pred_validation)

rmse_train = np.sqrt(mse_train)
rmse_validation = np.sqrt(mse_validation)

r2_score_train = r2_score(Y_train, pred_train)
r2_score_validation = r2_score(Y_validation, pred_validation)


print('mse_train',mse_train)
print('mae_train',mae_train)
print('rmse_train',rmse_train)
print('r2_score_train',r2_score_train)
print('\n')
print('mse_validation',mse_validation)
print('mae_validation',mae_validation)
print('rmse_validation',rmse_validation)
print('r2_score_validation',r2_score_validation)


#최종적으로 test데이터 예측해 성능평가를 해보고 
pred_test = RFR.predict(X_test_sc)

mse_test = mean_squared_error(y_test, pred_test)
mae_test = mean_absolute_error(y_test, pred_test)
rmse_test = np.sqrt(mse_test)
r2_score_test = r2_score(y_test, pred_test)


print('\n')
print('mse_test',mse_test)
print('mae_test',mae_test)
print('rmse_test',rmse_test)
print('r2_score_test',r2_score_test)


pd.DataFrame({'ID':x_test['ID'], 'quality':pred_test}).to_csv('wine.csv', index=False)

	fixed acidity	volatile acidity	citric acid	residual sugar	chlorides	free sulfur dioxide	total sulfur dioxide	density	pH	sulphates	alcohol
0	10.6	0.44	0.68	4.1	0.114	6.0	24.0	0.99700	3.06	0.66	13.4
1	7.0	0.60	0.30	4.5	0.068	20.0	110.0	0.99914	3.30	1.17	10.2
2	8.0	0.43	0.36	2.3	0.075	10.0	48.0	0.99760	3.34	0.46	9.4
3	7.9	0.53	0.24	2.0	0.072	15.0	105.0	0.99600	3.27	0.54	9.4
4	8.0	0.45	0.23	2.2	0.094	16.0	29.0	0.99620	3.21	0.49	10.2
...	...	...	...	...	...	...	...	...	...	...	...
1274	9.4	0.40	0.47	2.5	0.087	6.0	20.0	0.99772	3.15	0.50	10.5
1275	11.0	0.30	0.58	2.1	0.054	7.0	19.0	0.99800	3.31	0.88	10.5
1276	9.3	0.39	0.44	2.1	0.107	34.0	125.0	0.99780	3.14	1.22	9.5
1277	7.0	0.50	0.14	1.8	0.078	10.0	23.0	0.99636	3.53	0.61	10.4
1278	6.5	0.67	0.00	4.3	0.057	11.0	20.0	0.99488	3.45	0.56	11.8

1279 rows × 11 columns

	ID	quality
0	1	6
1	2	5
2	3	5
3	4	6
4	5	6

mse_train 0.05081869158878505
mae_train 0.1622663551401869
rmse_train 0.22543001483561378
r2_score_train 0.9207777306883845


mse_validation 0.4251274231678487
mae_validation 0.5014420803782507
rmse_validation 0.6520179623046045
r2_score_validation 0.37446445264958395


mse_test 0.3656053125
mae_test 0.44378124999999996
rmse_test 0.6046530513443226
r2_score_test 0.3963173374613004

Out[134]:

	ID	quality
0	0	5.98
1	7	6.28
2	12	5.06
3	13	4.66
4	25	5.08
...	...	...
315	1587	5.24
316	1589	5.59
317	1591	5.02
318	1592	5.13
319	1595	6.28

320 rows × 2 columns

현대 차량 가격 문제 데이터¶

데이터 설명 : 현대 차량 가격문제 (종속변수 : price)¶

x_train: https://raw.githubusercontent.com/Datamanim/datarepo/main/hyundai/x_train.csv¶

x_test: https://raw.githubusercontent.com/Datamanim/datarepo/main/hyundai/x_test.csv¶

y_train: https://raw.githubusercontent.com/Datamanim/datarepo/main/hyundai/y_train.csv¶

y_test(평가용) : https://raw.githubusercontent.com/Datamanim/datarepo/main/hyundai/y_test.csv¶

In [189]:

#데이터 불러오기
import pandas as pd
x_train = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/hyundai/x_train.csv')
x_test = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/hyundai/x_test.csv')
y_train = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/hyundai/y_train.csv')
y_test = pd.read_csv('https://raw.githubusercontent.com/Datamanim/datarepo/main/hyundai/y_test.csv')

#ID컬럼은 삭제하자
X_train = x_train.drop(columns='ID')
X_test = x_test.drop(columns='ID')

display(X_train)
display(y_train.head())

y_train = y_train['price']
y_test = y_test['price']

#print(X_train.info())
#print(X_train.nunique())
#print(X_test.isnull().sum()) #널 값은 없다
#print(X_train.isnull().sum()) #널 값은 없다

## 수치형 컬럼이지만 범주형인 데이터 타입을 object로 변경
X_train['year'] = X_train['year'].astype('object')
X_test['year'] = X_test['year'].astype('object')



#수치형 데이터 이상치를 제거해주자
def solution(data, col):
    q3 = data[col].quantile(0.75)
    q1 = data[col].quantile(0.25)
    iqr = q3 - q1
    
    upper = q3 + (1.5*iqr)
    lower = q1 - (1.5*iqr)
    data[col] = data[col].map(lambda x : data[col].mean() if (x>upper) or (x<lower) else x)
    return data

for i in X_train.select_dtypes(exclude='object').columns:
    solution(X_train, i)
    solution(X_test, i)
    
    
    
#데이터 분할을 해주자(학습데이터 0.67, 검증데이터 0.33)
from sklearn.model_selection import train_test_split
X_train, X_validation, Y_train, Y_validation = train_test_split(X_train, y_train, test_size=0.33, random_state=43)
#print(X_train.shape, X_validation.shape, Y_train.shape, Y_validation.shape)
X_train.reset_index(drop=True, inplace=True)
X_validation.reset_index(drop=True, inplace=True)


#수치형 데이터 스케일링 해주자
obj_col = X_train.select_dtypes(include='object').columns
#print(obj_col)
from sklearn.preprocessing import StandardScaler
sds = StandardScaler()
sds.fit(X_train.drop(columns=obj_col))
X_train_sc = sds.transform(X_train.drop(columns=obj_col))
X_train_sc = pd.DataFrame(X_train_sc, columns = X_train.drop(columns=obj_col).columns)

X_validation_sc = sds.transform(X_validation.drop(columns=obj_col))
X_validation_sc = pd.DataFrame(X_validation_sc, columns = X_validation.drop(columns=obj_col).columns)

X_test_sc = sds.transform(X_test.drop(columns=obj_col))
X_test_sc = pd.DataFrame(X_test_sc, columns = X_test.drop(columns=obj_col).columns)


for i in obj_col:
    X_train_sc[i] = X_train[i]
    X_validation_sc[i] = X_validation[i]
    X_test_sc[i] = X_test[i]
    

#수치형 데이터 로그변환 해주자
import numpy as np
for i in X_train_sc.select_dtypes(exclude='object').columns:
    if X_train_sc[i].min() < 0:
        X_train_sc[i] = X_train_sc[i].map(lambda x : x+abs(X_train_sc[i].min()))
    X_train_sc[i] = np.log1p(X_train_sc[i])
    
    if X_validation_sc[i].min() < 0:
        X_validation_sc[i] = X_validation_sc[i].map(lambda x : x+abs(X_validation_sc[i].min()))
    X_validation_sc[i] = np.log1p(X_validation_sc[i])
    
    if X_test_sc[i].min() < 0:
        X_test_sc[i] = X_test_sc[i].map(lambda x : x+abs(X_test_sc[i].min()))
    X_test_sc[i] = np.log1p(X_test_sc[i])

#display(X_validation_sc)
#모델링 하기전 인코딩을 해주자
#fuelType, transmission 컬럼은 원핫인코딩
#model, year 컬름은 범주의 수가 많으므로 라벨 인코딩 해주자

X_full = pd.concat([X_train_sc, X_validation_sc, X_test_sc])
from sklearn.preprocessing import LabelEncoder
labelencoder = LabelEncoder()
for i in ['model','year']:
    X_full[i] = labelencoder.fit_transform(X_full[i])
    
X_full = pd.get_dummies(X_full)
X_train_sc = X_full[:len(X_train_sc)]
X_validation_sc = X_full[len(X_train_sc):len(X_train_sc)+len(X_validation_sc)]
X_test_sc = X_full[len(X_train_sc)+len(X_validation_sc):]


#RandomForestRegressor을 사용한 모델링
from sklearn.ensemble import RandomForestRegressor
RFR = RandomForestRegressor()
RFR.fit(X_train_sc, Y_train)
pred_train = RFR.predict(X_train_sc)
pred_validation = RFR.predict(X_validation_sc)



#예측데이터와 실제데이터를 비교해 성능평가지표를 사용해 평가 ㄱㄱ
from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score

train_mse = mean_squared_error(Y_train, pred_train)
validation_mse = mean_squared_error(Y_validation, pred_validation)

train_mae = mean_absolute_error(Y_train, pred_train)
validation_mae = mean_absolute_error(Y_validation, pred_validation)

train_rmse = np.sqrt(train_mse)
validation_rmse = np.sqrt(validation_mse)

train_r2_score = r2_score(Y_train, pred_train)
validation_r2_score = r2_score(Y_validation, pred_validation)


print('train_mse',train_mse)
print('train_mae',train_mae)
print('train_rmse',train_rmse)
print('train_r2_score',train_r2_score)
print('\n')
print('validation_mse',validation_mse)
print('validation_mae',validation_mae)
print('validation_rmse',validation_rmse)
print('validation_r2_score',validation_r2_score)


#최종적으로 test데이터 예측후 성능평가해보고 
#결과를 csv파일로 저장하기
pred_test = RFR.predict(X_test_sc)

test_mse = mean_squared_error(y_test, pred_test)
test_mae = mean_absolute_error(y_test, pred_test)
test_rmse = np.sqrt(test_mse)
test_r2_score = r2_score(y_test, pred_test)

print('\n')
print('test_mse',test_mse)
print('test_mae',test_mae)
print('test_rmse',test_rmse)
print('test_r2_score',test_r2_score)

pd.DataFrame({'ID':x_test['ID'], 'price':pred_test}).to_csv('handa_car.csv', index=False)

	model	year	transmission	mileage	fuelType	tax(£)	mpg	engineSize
0	I30	2019	Manual	21	Petrol	150	34.0	2.0
1	Santa Fe	2018	Semi-Auto	10500	Diesel	145	39.8	2.2
2	Tucson	2017	Manual	29968	Diesel	30	61.7	1.7
3	Kona	2018	Manual	27317	Petrol	145	52.3	1.0
4	Tucson	2018	Semi-Auto	31459	Diesel	145	57.7	1.7
...	...	...	...	...	...	...	...	...
3883	Tucson	2019	Manual	10200	Petrol	145	34.9	1.6
3884	I10	2019	Manual	9487	Petrol	145	60.1	1.0
3885	Tucson	2016	Manual	25224	Diesel	30	61.7	1.7
3886	Kona	2019	Manual	8550	Petrol	145	44.1	1.0
3887	Kona	2018	Manual	26615	Petrol	145	52.3	1.0

3888 rows × 8 columns

	ID	price
0	0	23995
1	1	28490
2	2	13251
3	3	14990
4	4	17591

train_mse 245345.66791365875
train_mae 326.63128535810705
train_rmse 495.32380107729404
train_r2_score 0.9929084080931919


validation_mse 1631374.6119890925
validation_mae 857.5842187138968
validation_rmse 1277.2527596326001
validation_r2_score 0.9526308751437623


test_mse 8179892.77387731
test_mae 765.227514489026
test_rmse 2860.0511837862814
test_r2_score 0.8009698904317185

In [ ]: