인프런 커뮤니티 질문&답변

train['환불금액'] = train['환불금액'].fillna(train['환불금액'].median())

test['환불금액'] = test['환불금액'].fillna(test['환불금액'].median())

train = train.drop('회원ID', axis=1)

target = train.pop('총구매액')

test_id = test.pop('회원ID')

cols = train.select_dtypes(include ='O').columns

df = pd.concat([train, test])

from sklearn.preprocessing import LabelEncoder

for col in cols:

le = LabelEncoder()

# train[col] = le.fit_transform(train[col])

# test[col] = le.transform(test[col])

df[col] = le.fit_transform(df[col])

# df = pd.get_dummies(df[cols])

train = df[:len(train)]

test = df[len(train):]

from sklearn.model_selection import train_test_split

X_tr, X_val, y_tr, y_val = train_test_split(

train,

target,

test_size = 0.2,

random_state=0)

# from sklearn.ensemble import RandomForestRegressor

# rf = RandomForestRegressor(random_state=0, max_depth=5, n_estimators=500)

# rf.fit(X_tr,y_tr)

# pred = rf.predict(X_val)

from xgboost import XGBRegressor

xgb = XGBRegressor(random_state=0, max_depth=5, n_estimators=400, learning_rate=0.01)

xgb.fit(X_tr, y_tr)

pred = xgb.predict(X_val)

from sklearn.metrics import mean_squared_error, r2_score

mse = mean_squared_error(y_val, pred)

rmse = mse**(1/2)

r2 = r2_score(y_val, pred)

print(mse, rmse, r2)

이렇게 작성했을때 이렇게 나오는데 잘된게 맞는건지 아닌건지 모르겠습니다. 선생님께서는 레이블 인코딩했을때 rmse가 813.22가 나오셨다는데 왜이렇게 차이가 많이나는지 모르겠습니다.

672765.0076766363 820.2225354601252 0.7020188276193622