AttributeError of compound kernel of Gaussian processes

Describe the bug

I am trying to use GaussianProcess classifier or regressor using the compound kernel (comprised of RBF and white kernels). Fit method of Gaussian process generates an error regarding its kernel, declaring that ‘CompoundKernel’ object has no attribute ‘k1’. I regenerated the error using the following simpler code:

Steps/Code to Reproduce

import numpy as np
import pandas as pd
from sklearn.gaussian_process import GaussianProcessRegressor
from sklearn.gaussian_process.kernels import CompoundKernel, WhiteKernel, RBF
X1 = np.random.exponential(range(0,10), size = 10)
X2 = np.random.poisson(range(0,10), size = 10)
y = np.random.normal(size = 10)
df = pd.DataFrame()
df["X1"] = X1
df["X2"] = X2
df["y"] = y
df1 = df.iloc[:,0:2]
df2 = df.iloc[:,2]

kernel = CompoundKernel([WhiteKernel(noise_level=2), RBF(length_scale=3)])
gaus = GaussianProcessRegressor(kernel = kernel)
gaus.fit(df1, df2) # executing this line generates the error

Expected Results

AttributeError                            Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_2056/1607188154.py in <module>
     15 kernel = CompoundKernel([WhiteKernel(noise_level=2), RBF(length_scale=3)])
     16 gaus = GaussianProcessRegressor(kernel = kernel)
---> 17 gaus.fit(df1, df2) # executing this line generates the error
......
......
    540             The non-fixed, log-transformed hyperparameters of the kernel
    541         """
--> 542         k_dims = self.k1.n_dims
    543         for i, kernel in enumerate(self.kernels):
    544             kernel.theta = theta[i * k_dims:(i + 1) * k_dims]

AttributeError: 'CompoundKernel' object has no attribute 'k1'

Actual Results

The code works well without using the compound kernel (e.g., RBF only), and the way I built the compound kernel is exactly according to the provided format of sklearn. Not sure why I receive this attribute error.

Versions

System:
    python: 3.7.12 (default, Sep 10 2021, 00:21:48)  [GCC 7.5.0]
executable: /usr/bin/python3
   machine: Linux-5.4.104+-x86_64-with-Ubuntu-18.04-bionic

Python dependencies:
          pip: 21.1.3
   setuptools: 57.4.0
      sklearn: 1.0.1
        numpy: 1.19.5
        scipy: 1.4.1
       Cython: 0.29.24
       pandas: 1.1.5
   matplotlib: 3.2.2
       joblib: 1.1.0
threadpoolctl: 3.0.0

Built with OpenMP: True

[glemaitre]

this CompoundKernel module seems to generate error for a classification problem in the same manner

I was probably not clear, you should not use it even in classification. Basically, it is only used in the internal of scikit-learn:

In [1]: from sklearn.datasets import load_iris
   ...: from sklearn.gaussian_process import GaussianProcessClassifier
   ...: from sklearn.gaussian_process.kernels import RBF
   ...: X, y = load_iris(return_X_y=True)
   ...: kernel = 1.0 * RBF(1.0)
   ...: gpc = GaussianProcessClassifier(kernel=kernel,
   ...:         random_state=0).fit(X, y)
In [2]: gpc.kernel_
Out[2]: CompoundKernel(7.71, 1.36, 5.26, 0.669, 6.13, 1.15)
In [3]: gpc.kernel_.get_params()
Out[3]: 
{'kernels': [47.2**2 * RBF(length_scale=3.92),
  13.8**2 * RBF(length_scale=1.95),
  21.4**2 * RBF(length_scale=3.15)]}

Here we indeed have 3 kernels, because the model is trained in a 1-vs-rest manner, thus 1 estimator + 1 kernel for each. But it is transparent to the user.

[marcozzxx810]

take