我有以下代码:
df = load_data()
pd.set_option('display.max_columns', None)
df.dtypes
intBillID object
chBillChargeCode object
chBillNo object
chOriginalBillNo object
sdBillDate datetime64[ns]
sdDueDate datetime64[ns]
sdDatePaidCancelled datetime64[ns]
sdBillCancelledDate object
totalDaysToPay int64
paidInDays int64
paidOnTime int64
chBillStatus object
chBillType object
chDebtorCode object
chBillGroupCode int64
dcTotFeeBilledAmt float64
dcFinalBillExpAmt float64
dcTotProgBillAmt float64
dcTotProgBillExpAmt float64
dcReceiveBillAmt float64
dcTotWipHours float64
dcTotWipTargetAmt float64
vcReason object
OperatingUnit object
BusinessUnit object
LosCode object
dcTotNetBillAmt float64
dtype: object
然后我有这个:
# Separate features and labels
X, y = df[['totalDaysToPay', 'paidOnTime','dcTotFeeBilledAmt','dcFinalBillExpAmt','dcTotProgBillAmt', 'dcTotProgBillExpAmt','dcTotProgBillExpAmt','dcReceiveBillAmt','dcTotWipHours','dcTotWipTargetAmt']].values, df['paidInDays'].values
print('Features:',X[:10], '\nLabels:', y[:10], sep='\n')
然后我拆分 X,Y
从 sklearn.model_selection 导入 train_test_split
# Split data 70%-30% into training set and test set
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.30, random_state=0)
print ('Training Set: %d rows\nTest Set: %d rows' % (X_train.shape[0], X_test.shape[0]))
然后我想转换数字和类别特征:
# Train the model
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
from sklearn.impute import SimpleImputer
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.linear_model import LinearRegression
import numpy as np
from sklearn.ensemble import GradientBoostingRegressor
# Define preprocessing for numeric columns (scale them)
numeric_features = [8,9,10,11,12,13,15,16,17,18,19,20,21,26]
numeric_transformer = Pipeline(steps=[
('scaler', StandardScaler())])
# Define preprocessing for categorical features (encode them)
categorical_features = [1,23,24,25]
categorical_transformer = Pipeline(steps=[
('onehot', OneHotEncoder(handle_unknown='ignore'))])
# Combine preprocessing steps
preprocessor = ColumnTransformer(
transformers=[
('num', numeric_transformer, numeric_features),
('cat', categorical_transformer, categorical_features)])
# Create preprocessing and training pipeline
pipeline = Pipeline(steps=[('preprocessor', preprocessor),
('regressor', GradientBoostingRegressor())])
# fit the pipeline to train a linear regression model on the training set
model = pipeline.fit(X_train, (y_train))
print (model)
但是我收到此错误:
ValueError: all features must be in [0, 9] or [-10, 0]
答案 0 :(得分:2)
在这一行中,您为 X 选择了 10 个特征,因此现在更改了 X 的形状。
# Separate features and labels
X, y = df[['totalDaysToPay', 'paidOnTime','dcTotFeeBilledAmt','dcFinalBillExpAmt','dcTotProgBillAmt', 'dcTotProgBillExpAmt','dcTotProgBillExpAmt','dcReceiveBillAmt','dcTotWipHours','dcTotWipTargetAmt']].values, df['paidInDays'].values
现在,您需要根据范围 [0-9] 给出 'numeric_features
' 的索引。
更具体地说,您在“numeric features
”中传递的索引应该反映这个数组。
['totalDaysToPay', 'paidOnTime','dcTotFeeBilledAmt','dcFinalBillExpAmt','dcTotProgBillAmt', 'dcTotProgBillExpAmt','dcTotProgBillExpAmt','dcReceiveBillAmt','dcTotWipHours','dcTotWipTargetAmt']
此数组对于原始“df
”是正确的:[8,9,10,11,12,13,15,16,17,18,19,20,21,26]
不适用于 X
。