时间戳

首页 > 解决方案 > 执行高斯回归时出错。如何解决？

python - 执行高斯回归时出错。如何解决？

问题描述

我在 Python 中运行高斯回归。我的数据集的形状为 (10000,5)。但是当我尝试拟合模型时出现错误：

AttributeError: 'list' object has no attribute 'n_dims'

我该如何解决这个问题？

我最初认为这个错误是由于我的因变量的维度可能与自变量不同而引起的。但即使将它们更改为相同的维度，我也无法找到代码的问题。任何帮助都感激不尽。

import numpy as np
import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn.gaussian_process import GaussianProcessRegressor
from sklearn.gaussian_process.kernels import (RBF, Matern, RationalQuadratic,
                                              ExpSineSquared, DotProduct,
                                              ConstantKernel)

data_set = pd.read_excel(r'XXXXX', sheet = 'Worksheet', header = 0)
data_set.head()
test_set = data_set

y = test_set.iloc[:,4]
test_set.drop(test_set.columns[4], axis = 1, inplace = True)    
X = test_set 
x=StandardScaler().fit_transform(X)   

X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.3, random_state=0)
y_train = np.array(y_train)
y_test = np.array(y_test)
y_train = np.reshape(y_train, (7000,1))
y_test = np.reshape(y_test, (3000,1))

kernels = [1.0 * RBF(length_scale=1.0, length_scale_bounds=(1e-1, 10.0))]

gp = GaussianProcessRegressor(kernel=kernels)   
gp.fit(X_train, y_train)


File "<ipython-input-23-5a576449fdb6>", line 1, in <module>
    gp.fit(X_train, y_train)

  File "C:\Program Files\Anaconda\lib\site-packages\sklearn\gaussian_process\gpr.py", line 203, in fit
    if self.optimizer is not None and self.kernel_.n_dims > 0:

AttributeError: 'list' object has no attribute 'n_dims'

标签： pythonmachine-learningscikit-learn

解决方案

初始化时GaussianProcessRegressor(kernel=kernels)传递的参数kernel必须是内核对象。你正在传递一个列表。

此处的文档中的更多信息。

推荐阅读