svm中gamma的确定

from __future__ import print_function
from sklearn.learning_curve import  validation_curve
from sklearn.datasets import load_digits
from sklearn.svm import SVC
import matplotlib.pyplot as plt
import numpy as np

digits = load_digits()
X = digits.data
y = digits.target
# np.logspace(-6, -2.3, 5)的值为[1.00000000e-06  8.41395142e-06  7.07945784e-05  5.95662144e-04 5.01187234e-03]
param_range = np.logspace(-6, -2.3, 5)
train_loss, test_loss = validation_curve(
        SVC(), X, y, param_name='gamma', param_range=param_range, cv=10,
        scoring='mean_squared_error')
train_loss_mean = -np.mean(train_loss, axis=1)
test_loss_mean = -np.mean(test_loss, axis=1)

plt.plot(param_range, train_loss_mean, 'o-', color="r",
             label="Training")
plt.plot(param_range, test_loss_mean, 'o-', color="g",
             label="Cross-validation")

plt.xlabel("gamma")
plt.ylabel("Loss")
plt.legend(loc="best")

plt.show()

如图所示gamma取0.006效果最好