ValueError：模型的特征数必须与输入匹配。型号n_特征为947，输入n_特征为15

Question

所以，我使用随机林来创建这个数据集：https://archive.ics.uci.edu/ml/datasets/auto+mpg
但当我试图预测某件事时，它抛出了一个错误：

ValueError: Number of features of the model must match the input. Model n_features is 947 and input n_features is 15

这是我的文件：

import joblib  # para salvar o modelo
import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler  # Para Normalizar
from sklearn.preprocessing import OneHotEncoder
from sklearn.preprocessing import LabelEncoder
from sklearn.metrics import r2_score

data = pd.read_csv('auto-mpg.csv',sep = ',')

data['horsepower'] = data['horsepower'].replace('?','100')

x = data.iloc[:,1:].values
y = data.iloc[:,0].values

lb = LabelEncoder()
x[:,7] = lb.fit_transform(x[:,7])

onehot = OneHotEncoder()
x = onehot.fit_transform(x).toarray()

xtrain,xtest,ytrain,ytest = train_test_split(x,y,test_size = 0.2,random_state = 0)

sc = StandardScaler()
x = sc.fit_transform(x)

rfr = RandomForestRegressor(n_estimators = 200,random_state = 0)
rfr.fit(xtrain,ytrain)

ypred_rfr = rfr.predict(xtest)
print('Acuracia:',round(r2_score(ytest,ypred_rfr)*100,2),'%')

joblib.dump(rfr,'randon-forest.model')

这里是错误：

import joblib  # para salvar o modelo
import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import OneHotEncoder

data = pd.read_csv('teste.csv',sep = ',')
print(data.columns);
logit = joblib.load('randon-forest.model')

onehot = OneHotEncoder()
data = onehot.fit_transform(data).toarray()

sc = StandardScaler()
data = sc.fit_transform(data)

# montar um vetor de dados
dados_vet = pd.DataFrame(data)
print(data)
# classificar esse vetor com o logit_bank
result_predict = logit.predict(dados_vet)

print('Logit Bank')
print(result_predict)

Answer 1

TL；博士：
用于预测的数据必须与用于训练模型的数据具有相同数量的特征

更详细的回答：
您在auto-mpg.csv上训练模型，我猜其中有947个特性（列）（1列作为标记值）。但是，teste.csv处的数据（我不知道它是什么）可能只有15个特征，因此您无法使用在不同数量特征上训练的模型预测15个特征数据

一个简单的例子：假设你有一个模型，该模型使用以下公式预测房屋建造年份的房价和房间数量：

price = number_of_rooms * 5 + (year-2000) * 20

您试图用代码做的事情，相当于只提供这个示例模型的房间数量。模特也必须“知道”年份