Python sklearn 朴素贝叶斯之高斯贝叶斯

一、

import numpy as np
import pandas as pd
from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import GaussianNB
from sklearn.metrics import confusion_matrix
from sklearn.model_selection import learning_curve
from sklearn.model_selection import ShuffleSplit

digits = load_digits()  # 手写数字数据集
X, y = digits.data, digits.target
print(pd.DataFrame(X).head())
print(np.unique(y))

Xtrain, Xtest, ytrain, ytest = train_test_split(X, y, test_size=0.3, random_state=42)
print("Train:", Xtrain.shape, "Test:", Xtest.shape)

gnb = GaussianNB()

gnb.fit(Xtrain, ytrain)
acc = gnb.score(Xtest, ytest)
print("Accuracy:", acc)
prob = gnb.predict_proba(Xtest)
print(prob.shape, prob[0, :].sum())  # [540, 10]  1
y_pred = gnb.predict(Xtest)
print((prob.argmax(axis=1) == y_pred).sum())  # 540

cm = confusion_matrix(ytest, y_pred)
print("Confusion Matrix", cm)  # 混淆矩阵

gnb = GaussianNB()
cv = ShuffleSplit(n_splits=50, test_size=0.2, random_state=42)
train_sizes, train_scores, test_scores = learning_curve(gnb, X, y, cv=cv, n_jobs=4)
print(train_sizes)  # 每次分训练集和测试集建模之后，训练集上的样本数量
print(np.mean(train_scores, axis=1), np.mean(test_scores, axis=1))

二、

• (score and time)

import datetime
from time import time

import numpy as np
import matplotlib.pyplot as plt
from sklearn.naive_bayes import GaussianNB
from sklearn.svm import SVC
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.datasets import load_digits
from sklearn.model_selection import learning_curve  # 导入画学习曲线的类
from sklearn.model_selection import ShuffleSplit  # 导入设定交叉验证模式的类


def plot_learning_curve(model, title, X, y,
                        ax,  # 选择子图
                        ylim=None,  # 设置纵坐标的取值范围
                        cv=None,  # 交叉验证
                        n_jobs=None  # 设定所要使用的线程
                        ):
    train_sizes, train_scores, test_scores = learning_curve(model, X, y, cv=cv, n_jobs=n_jobs)

    ax.set_title(title)

    if ylim is not None:
        ax.set_ylim(*ylim)

    ax.set_xlabel("Training examples")
    ax.set_ylabel("Score")
    ax.grid()  # 显示网格作为背景
    ax.plot(train_sizes, np.mean(train_scores, axis=1), 'o-'
            , color="r", label="Training score")
    ax.plot(train_sizes, np.mean(test_scores, axis=1), 'o-'
            , color="g", label="Test score")
    ax.legend(loc="best")
    return ax


digits = load_digits()
X, y = digits.data, digits.target

titles = ["Naive Bayes", "DecisionTree", "SVM", "RandomForest", "Logistic"]
models = [GaussianNB(), DecisionTreeClassifier(), SVC(gamma=0.001), RandomForestClassifier(n_estimators=50), LogisticRegression(C=.1, multi_class="auto", solver='lbfgs', max_iter=5000)]
cv = ShuffleSplit(n_splits=50, test_size=0.2, random_state=42)

fig, axes = plt.subplots(1, 5, figsize=(30, 6))
for index, title, model in zip(range(len(titles)), titles, models):
    time_start = time()

    plot_learning_curve(model, title, X, y, ax=axes[index], ylim=[0.7, 1.05], cv=cv, n_jobs=4)

    time_end = time()
    print("{}:{}".format(title, datetime.datetime.fromtimestamp(time_end-time_start).strftime("%M:%S:%f")))
plt.show()

time

score

三、

• (log loss)

from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import GaussianNB
from sklearn.svm import SVC
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import log_loss
from sklearn.metrics import brier_score_loss  # 布里尔分数


digits = load_digits()  # 手写数字数据集
X, y = digits.data, digits.target
Xtrain, Xtest, ytrain, ytest = train_test_split(X, y, test_size=0.3, random_state=42)


gnb = GaussianNB()
gnb.fit(Xtrain, ytrain)
gnb_prob = gnb.predict_proba(Xtest)
print(gnb_prob.shape)  # (540,10)

lr = LogisticRegression(C=1., solver='lbfgs', max_iter=5000, multi_class="auto")
lr.fit(Xtrain, ytrain)
lr_prob = lr.predict_proba(Xtest)
print(lr_prob.shape)  # (540,10)

svc = SVC(kernel="linear", gamma=1)
svc.fit(Xtrain, ytrain)
svc_df = svc.decision_function(Xtest)
svc_prob = (svc_df - svc_df.min())/(svc_df.max() - svc_df.min())
print(svc_prob.shape)  # (540,10)

print(log_loss(ytest, gnb_prob), log_loss(ytest, lr_prob), log_loss(ytest, svc_prob))  # 对比多个模型
# print(brier_score_loss(y_true=ytest, y_prob=gnb_prob))  # error