机器学习之随机森林算法

第1关：Bagging

import numpy as np
from collections import Counter
from sklearn.tree import DecisionTreeClassifier
class BaggingClassifier():
    def __init__(self, n_model=10):
        '''
        初始化函数
        '''
        #分类器的数量，默认为10
        self.n_model = n_model
        #用于保存模型的列表，训练好分类器后将对象append进去即可
        self.models = []
    def fit(self, feature, label):
        '''
        训练模型
        :param feature: 训练数据集所有特征组成的ndarray
        :param label:训练数据集中所有标签组成的ndarray
        :return: None
        '''
        #************* Begin ************#
        for i in range(self.n_model):
            m = len(feature)
            index = np.random.choice(m, m)
            sample_data = feature[index]
            sample_lable = label[index]
            model = DecisionTreeClassifier()
            model = model.fit(sample_data, sample_lable)
            self.models.append(model)
        #************* End **************#
    def predict(self, feature):
        '''
        :param feature:训练数据集所有特征组成的ndarray
        :return:预测结果，如np.array([0, 1, 2, 2, 1, 0])
        '''
        #************* Begin ************#
        result = []
        vote = []
        for model in self.models:
            r = model.predict(feature)
            vote.append(r)
        vote = np.array(vote)
        for i in range(len(feature)):
            v = sorted(Counter(vote[:, i]).items(), key=lambda x: x[1], reverse=True)
            result.append(v[0][0])
        return np.array(result)
        #************* End **************#

第2关：随机森林算法流程

import numpy as np
from collections import  Counter
from sklearn.tree import DecisionTreeClassifier
class RandomForestClassifier():
    def __init__(self, n_model=10):
        '''
        初始化函数
        '''
        #分类器的数量，默认为10
        self.n_model = n_model
        #用于保存模型的列表，训练好分类器后将对象append进去即可
        self.models = []
        #用于保存决策树训练时随机选取的列的索引
        self.col_indexs = []
    def fit(self, feature, label):
        '''
        训练模型
        :param feature: 训练数据集所有特征组成的ndarray
        :param label:训练数据集中所有标签组成的ndarray
        :return: None
        '''
        #************* Begin ************#
        for i in range(self.n_model):
            m = len(feature)
            index = np.random.choice(m, m)
            col_index = np.random.permutation(len(feature[0]))[:int(np.log2(len(feature[0])))]
            sample_data = feature[index]
            sample_data = sample_data[:, col_index]
            sample_lable = label[index]
            model = DecisionTreeClassifier()
            model = model.fit(sample_data, sample_lable)
            self.models.append(model)
            self.col_indexs.append(col_index)
        #************* End **************#
    def predict(self, feature):
        '''
        :param feature:训练数据集所有特征组成的ndarray
        :return:预测结果，如np.array([0, 1, 2, 2, 1, 0])
        '''
        #************* Begin ************#
        result = []
        vote = []
        for i, model in enumerate(self.models):
            f = feature[:, self.col_indexs[i]]
            r = model.predict(f)
            vote.append(r)
        vote = np.array(vote)
        for i in range(len(feature)):
            v = sorted(Counter(vote[:, i]).items(), key=lambda x: x[1], reverse=True)
            result.append(v[0][0])
        return np.array(result)
        #************* End **************#

第3关：手写数字识别

from sklearn.ensemble import RandomForestClassifier
import numpy as np
def digit_predict(train_image, train_label, test_image):
    '''
    实现功能：训练模型并输出预测结果
    :param train_image: 包含多条训练样本的样本集，类型为ndarray,shape为[-1, 8, 8]
    :param train_label: 包含多条训练样本标签的标签集，类型为ndarray
    :param test_image: 包含多条测试样本的测试集，类型为ndarry
    :return: test_image对应的预测标签，类型为ndarray
    '''

    #************* Begin ************#
    X = np.reshape(train_image, newshape=(-1, 64))
    clf=RandomForestClassifier(n_estimators=500, max_depth=10)
    clf.fit(X, y=train_label)
    return clf.predict(test_image)
    #************* End **************#