import numpy as np
import matplotlib.pyplot as plt

from sklearn.svm import LinearSVC
from sklearn.datasets import load_digits
from sklearn.model_selection import learning_curve

#模型選擇學習曲線learning_curve模型
def test_learning_curve():
    ### 加載數據
    digits = load_digits()
    X,y=digits.data,digits.target
    
 
#### 獲取學習曲線 ######
    train_sizes=np.linspace(0.1,1.0,endpoint=True,dtype=‘float‘)
    abs_trains_sizes,train_scores, test_scores = learning_curve(LinearSVC(),X, y,cv=10, scoring="accuracy",train_sizes=train_sizes)
    ###### 對每個 C ，獲取 10 折交叉上的預測得分上的均值和方差 #####
    train_scores_mean = np.mean(train_scores, axis=1)
    train_scores_std 
 
= np.std(train_scores, axis=1)
    test_scores_mean = np.mean(test_scores, axis=1)
    test_scores_std = np.std(test_scores, axis=1)
    ####### 繪圖 ######
    fig=plt.figure()
    ax=fig.add_subplot(1,1,1)

    ax.plot(abs_trains_sizes, train_scores_mean, label="Training Accuracy", color="r")
    ax.fill_between(abs_trains_sizes, train_scores_mean 
 
- train_scores_std,train_scores_mean + train_scores_std, alpha=0.2, color="r")
    ax.plot(abs_trains_sizes, test_scores_mean, label="Testing Accuracy", color="g")
    ax.fill_between(abs_trains_sizes, test_scores_mean - test_scores_std,test_scores_mean + test_scores_std, alpha=0.2, color="g")

    ax.set_title("Learning Curve with LinearSVC")
    ax.set_xlabel("Sample Nums")
    ax.set_ylabel("Score")
    ax.set_ylim(0,1.1)
    ax.legend(loc=‘best‘)
    plt.show()
    
#調用test_learning_curve()
test_learning_curve()

吳裕雄 python 機器學習——模型選擇學習曲線learning_curve模型