python - SKlearn：高斯过程回归预测对于任何输入都是相同的

Question

我正在使用 Sklearn 测试高斯过程回归，但我在拟合日期后的 GP 预测在任何地方都是相同的。有什么我想念的吗？

import numpy as np
from sklearn.gaussian_process.kernels import Matern
from sklearn.gaussian_process import GaussianProcessRegressor

y=[  60.3295131,    53.56137623,    6.18362039,   43.27383624,   73.63256713]

x=[[ 0.02690405, -0.76987722],
 [ 0.51031994, -0.46978989],
 [-1.4756594,   2.15417603],
 [ 0.02674444,  0.6513043 ],
 [ 0.31203479,  0.95269764]]

gp = GaussianProcessRegressor(
            kernel=Matern(nu=2.5),
            alpha=1e-6,
            normalize_y=True,
            n_restarts_optimizer=5,
            random_state=125,
        )

gp.fit(x, y)

print(gp.predict([(-1.3,1.54),(.3,.54)]))

print(gp.predict([(3111,.54)])) ```

标签： pythonmachine-learningscikit-learnregressiongaussian