引言

想通过随机森林来获取数据的主要特征

1、理论

随机森林是一个高度灵活的机器学习方法，拥有广泛的应用前景，从市场营销到医疗保健保险。 既可以用来做市场营销模拟的建模，统计客户来源，保留和流失。也可用来预测疾病的风险和病患者的易感性。

根据个体学习器的生成方式，目前的集成学习方法大致可分为两大类，即个体学习器之间存在强依赖关系，必须串行生成的序列化方法，以及个体学习器间不存在强依赖关系，可同时生成的并行化方法；

前者的代表是Boosting，后者的代表是Bagging和“随机森林”（Random
Forest）

随机森林在以决策树为基学习器构建Bagging集成的基础上，进一步在决策树的训练过程中引入了随机属性选择（即引入随机特征选择）。

简单来说，随机森林就是对决策树的集成，但有两点不同：

（2）特征选取的差异性：每个决策树的n个分类特征是在所有特征中随机选择的（n是一个需要我们自己调整的参数）
随机森林，简单理解， 比如预测salary，就是构建多个决策树job，age，house，然后根据要预测的量的各个特征（teacher，39，suburb）分别在对应决策树的目标值概率（salary<5000,salary>=5000），从而，确定预测量的发生概率（如，预测出P(salary<5000)=0.3）.

随机森林是一个可做能够回归和分类。 它具备处理大数据的特性，而且它有助于估计或变量是非常重要的基础数据建模。

参数说明：

最主要的两个参数是n_estimators和max_features。

n_estimators：表示森林里树的个数。理论上是越大越好。但是伴随着就是计算时间的增长。但是并不是取得越大就会越好，预测效果最好的将会出现在合理的树个数。

max_features：随机选择特征集合的子集合，并用来分割节点。子集合的个数越少，方差就会减少的越快，但同时偏差就会增加的越快。根据较好的实践经验。如果是回归问题则：

max_features＝n_features，如果是分类问题则max_features＝sqrt(n_features)。

如果想获取较好的结果，必须将max_depth＝None,同时min_sample_split=1。
同时还要记得进行cross_validated（交叉验证），除此之外记得在random forest中，bootstrap=True。但在extra-trees中，bootstrap=False。

2、随机森林python实现

2.1Demo1

实现随机森林基本功能

#随机森林from sklearn.tree import DecisionTreeRegressor from sklearn.ensemble import RandomForestRegressor import numpy as np from sklearn.datasets import load_iris iris=load_iris() #print iris#iris的４个属性是：萼片宽度　萼片长度　花瓣宽度　花瓣长度　标签是花的种类：setosa versicolour virginica print(iris['target'].shape)rf=RandomForestRegressor()#这里使用了默认的参数设置 rf.fit(iris.data[:150],iris.target[:150])#进行模型的训练 #随机挑选两个预测不相同的样本 instance=iris.data[[100,109]] print(instance)rf.predict(instance[[0]])print('instance 0 prediction；',rf.predict(instance[[0]]))print( 'instance 1 prediction；',rf.predict(instance[[1]]))print(iris.target[100],iris.target[109])

运行结果

(150,)
[[ 6.3 3.3 6. 2.5]
[ 7.2 3.6 6.1 2.5]]
instance 0 prediction； [ 2.]
instance 1 prediction； [ 2.]
2 2

2.2 Demo2

3种方法的比较

#random forest testfrom sklearn.model_selection import cross_val_scorefrom sklearn.datasets import make_blobsfrom sklearn.ensemble import RandomForestClassifierfrom sklearn.ensemble import ExtraTreesClassifierfrom sklearn.tree import DecisionTreeClassifierX, y = make_blobs(n_samples=10000, n_features=10, centers=100,random_state=0)clf = DecisionTreeClassifier(max_depth=None, min_samples_split=2,random_state=0)scores = cross_val_score(clf, X, y)print(scores.mean()) clf = RandomForestClassifier(n_estimators=10, max_depth=None,min_samples_split=2, random_state=0)scores = cross_val_score(clf, X, y)print(scores.mean()) clf = ExtraTreesClassifier(n_estimators=10, max_depth=None,min_samples_split=2, random_state=0)scores = cross_val_score(clf, X, y)print(scores.mean())

运行结果：

0.979408793821
0.999607843137
0.999898989899

2.3 Demo3-实现特征选择

#随机森林2from sklearn.tree import DecisionTreeRegressor from sklearn.ensemble import RandomForestRegressor import numpy as np from sklearn.datasets import load_iris iris=load_iris() from sklearn.model_selection import cross_val_score, ShuffleSplit X = iris["data"] Y = iris["target"] names = iris["feature_names"] rf = RandomForestRegressor() scores = [] for i in range(X.shape[1]): score = cross_val_score(rf, X[:, i:i+1], Y, scoring="r2", cv=ShuffleSplit(len(X), 3, .3)) scores.append((round(np.mean(score), 3), names[i])) print(sorted(scores, reverse=True))

运行结果：

[(0.89300000000000002, 'petal width (cm)'), (0.82099999999999995, 'petal length
(cm)'), (0.13, 'sepal length (cm)'), (-0.79100000000000004, 'sepal width (cm)')]

2.4 demo4-随机森林

本来想利用以下代码来构建随机随机森林决策树，但是，遇到的问题是，程序一直在运行，无法响应，还需要调试。

#随机森林4#coding:utf-8 import csv from random import seed from random import randrange from math import sqrt def loadCSV(filename):#加载数据，一行行的存入列表 dataSet = [] with open(filename, 'r') as file: csvReader = csv.reader(file) for line in csvReader: dataSet.append(line) return dataSet # 除了标签列，其他列都转换为float类型 def column_to_float(dataSet): featLen = len(dataSet[0]) - 1 for data in dataSet: for column in range(featLen): data[column] = float(data[column].strip()) # 将数据集随机分成N块，方便交叉验证，其中一块是测试集，其他四块是训练集 def spiltDataSet(dataSet, n_folds): fold_size = int(len(dataSet) / n_folds) dataSet_copy = list(dataSet) dataSet_spilt = [] for i in range(n_folds): fold = [] while len(fold) < fold_size: # 这里不能用if，if只是在第一次判断时起作用，while执行循环，直到条件不成立 index = randrange(len(dataSet_copy)) fold.append(dataSet_copy.pop(index)) # pop() 函数用于移除列表中的一个元素（默认最后一个元素），并且返回该元素的值。 dataSet_spilt.append(fold) return dataSet_spilt # 构造数据子集 def get_subsample(dataSet, ratio): subdataSet = [] lenSubdata = round(len(dataSet) * ratio)#返回浮点数 while len(subdataSet) < lenSubdata: index = randrange(len(dataSet) - 1) subdataSet.append(dataSet[index]) # print len(subdataSet) return subdataSet # 分割数据集 def data_spilt(dataSet, index, value): left = [] right = [] for row in dataSet: if row[index] < value: left.append(row) else: right.append(row) return left, right # 计算分割代价 def spilt_loss(left, right, class_values): loss = 0.0 for class_value in class_values: left_size = len(left) if left_size != 0: # 防止除数为零 prop = [row[-1] for row in left].count(class_value) / float(left_size) loss += (prop * (1.0 - prop)) right_size = len(right) if right_size != 0: prop = [row[-1] for row in right].count(class_value) / float(right_size) loss += (prop * (1.0 - prop)) return loss # 选取任意的n个特征，在这n个特征中，选取分割时的最优特征 def get_best_spilt(dataSet, n_features): features = [] class_values = list(set(row[-1] for row in dataSet)) b_index, b_value, b_loss, b_left, b_right = 999, 999, 999, None, None while len(features) < n_features: index = randrange(len(dataSet[0]) - 1) if index not in features: features.append(index) # print 'features:',features for index in features:#找到列的最适合做节点的索引，（损失最小） for row in dataSet: left, right = data_spilt(dataSet, index, row[index])#以它为节点的，左右分支 loss = spilt_loss(left, right, class_values) if loss < b_loss:#寻找最小分割代价 b_index, b_value, b_loss, b_left, b_right = index, row[index], loss, left, right # print b_loss # print type(b_index) return {'index': b_index, 'value': b_value, 'left': b_left, 'right': b_right} # 决定输出标签 def decide_label(data): output = [row[-1] for row in data] return max(set(output), key=output.count) # 子分割，不断地构建叶节点的过程对对对 def sub_spilt(root, n_features, max_depth, min_size, depth): left = root['left'] # print left right = root['right'] del (root['left']) del (root['right']) # print depth if not left or not right: root['left'] = root['right'] = decide_label(left + right) # print 'testing' return if depth > max_depth: root['left'] = decide_label(left) root['right'] = decide_label(right) return if len(left) < min_size: root['left'] = decide_label(left) else: root['left'] = get_best_spilt(left, n_features) # print 'testing_left' sub_spilt(root['left'], n_features, max_depth, min_size, depth + 1) if len(right) < min_size: root['right'] = decide_label(right) else: root['right'] = get_best_spilt(right, n_features) # print 'testing_right' sub_spilt(root['right'], n_features, max_depth, min_size, depth + 1) # 构造决策树 def build_tree(dataSet, n_features, max_depth, min_size): root = get_best_spilt(dataSet, n_features) sub_spilt(root, n_features, max_depth, min_size, 1) return root # 预测测试集结果 def predict(tree, row): predictions = [] if row[tree['index']] < tree['value']: if isinstance(tree['left'], dict): return predict(tree['left'], row) else: return tree['left'] else: if isinstance(tree['right'], dict): return predict(tree['right'], row) else: return tree['right'] # predictions=set(predictions) def bagging_predict(trees, row): predictions = [predict(tree, row) for tree in trees] return max(set(predictions), key=predictions.count) # 创建随机森林 def random_forest(train, test, ratio, n_feature, max_depth, min_size, n_trees): trees = [] for i in range(n_trees): train = get_subsample(train, ratio)#从切割的数据集中选取子集 tree = build_tree(train, n_features, max_depth, min_size) # print 'tree %d: '%i,tree trees.append(tree) # predict_values = [predict(trees,row) for row in test] predict_values = [bagging_predict(trees, row) for row in test] return predict_values # 计算准确率 def accuracy(predict_values, actual): correct = 0 for i in range(len(actual)): if actual[i] == predict_values[i]: correct += 1 return correct / float(len(actual)) if __name__ == '__main__': seed(1) dataSet = loadCSV(r'G:\0研究生\tianchiCompetition\训练小样本2.csv') column_to_float(dataSet) n_folds = 5 max_depth = 15 min_size = 1 ratio = 1.0 # n_features=sqrt(len(dataSet)-1) n_features = 15 n_trees = 10 folds = spiltDataSet(dataSet, n_folds)#先是切割数据集 scores = [] for fold in folds: train_set = folds[ :] # 此处不能简单地用train_set=folds，这样用属于引用,那么当train_set的值改变的时候，folds的值也会改变，所以要用复制的形式。（L[:]）能够复制序列，D.copy() 能够复制字典，list能够生成拷贝 list(L) train_set.remove(fold)#选好训练集 # print len(folds) train_set = sum(train_set, []) # 将多个fold列表组合成一个train_set列表 # print len(train_set) test_set = [] for row in fold: row_copy = list(row) row_copy[-1] = None test_set.append(row_copy) # for row in test_set: # print row[-1] actual = [row[-1] for row in fold] predict_values = random_forest(train_set, test_set, ratio, n_features, max_depth, min_size, n_trees) accur = accuracy(predict_values, actual) scores.append(accur) print ('Trees is %d' % n_trees) print ('scores:%s' % scores) print ('mean score:%s' % (sum(scores) / float(len(scores))))

2.5 随机森林分类sonic data

# CART on the Bank Note datasetfrom random import seedfrom random import randrangefrom csv import reader# Load a CSV filedef load_csv(filename): file = open(filename, "r") lines = reader(file) dataset = list(lines) return dataset# Convert string column to floatdef str_column_to_float(dataset, column): for row in dataset: row[column] = float(row[column].strip())# Split a dataset into k foldsdef cross_validation_split(dataset, n_folds): dataset_split = list() dataset_copy = list(dataset) fold_size = int(len(dataset) / n_folds) for i in range(n_folds): fold = list() while len(fold) < fold_size: index = randrange(len(dataset_copy)) fold.append(dataset_copy.pop(index)) dataset_split.append(fold) return dataset_split# Calculate accuracy percentagedef accuracy_metric(actual, predicted): correct = 0 for i in range(len(actual)): if actual[i] == predicted[i]: correct += 1 return correct / float(len(actual)) * 100.0# Evaluate an algorithm using a cross validation splitdef evaluate_algorithm(dataset, algorithm, n_folds, *args): folds = cross_validation_split(dataset, n_folds) scores = list() for fold in folds: train_set = list(folds) train_set.remove(fold) train_set = sum(train_set, []) test_set = list() for row in fold: row_copy = list(row) test_set.append(row_copy) row_copy[-1] = None predicted = algorithm(train_set, test_set, *args) actual = [row[-1] for row in fold] accuracy = accuracy_metric(actual, predicted) scores.append(accuracy) return scores# Split a data set based on an attribute and an attribute valuedef test_split(index, value, dataset): left, right = list(), list() for row in dataset: if row[index] < value: left.append(row) else: right.append(row) return left, right# Calculate the Gini index for a split datasetdef gini_index(groups, class_values): gini = 0.0 for class_value in class_values: for group in groups: size = len(group) if size == 0: continue proportion = [row[-1] for row in group].count(class_value) / float(size) gini += (proportion * (1.0 - proportion)) return gini# Select the best split point for a datasetdef get_split(dataset): class_values = list(set(row[-1] for row in dataset)) b_index, b_value, b_score, b_groups = 999, 999, 999, None for index in range(len(dataset[0])-1): for row in dataset: groups = test_split(index, row[index], dataset) gini = gini_index(groups, class_values) if gini < b_score: b_index, b_value, b_score, b_groups = index, row[index], gini, groups print ({'index':b_index, 'value':b_value}) return {'index':b_index, 'value':b_value, 'groups':b_groups}# Create a terminal node valuedef to_terminal(group): outcomes = [row[-1] for row in group] return max(set(outcomes), key=outcomes.count)# Create child splits for a node or make terminaldef split(node, max_depth, min_size, depth): left, right = node['groups'] del(node['groups']) # check for a no split if not left or not right: node['left'] = node['right'] = to_terminal(left + right) return # check for max depth if depth >= max_depth: node['left'], node['right'] = to_terminal(left), to_terminal(right) return # process left child if len(left) <= min_size: node['left'] = to_terminal(left) else: node['left'] = get_split(left) split(node['left'], max_depth, min_size, depth+1) # process right child if len(right) <= min_size: node['right'] = to_terminal(right) else: node['right'] = get_split(right) split(node['right'], max_depth, min_size, depth+1)# Build a decision treedef build_tree(train, max_depth, min_size): root = get_split(train) split(root, max_depth, min_size, 1) return root# Make a prediction with a decision treedef predict(node, row): if row[node['index']] < node['value']: if isinstance(node['left'], dict): return predict(node['left'], row) else: return node['left'] else: if isinstance(node['right'], dict): return predict(node['right'], row) else: return node['right']# Classification and Regression Tree Algorithmdef decision_tree(train, test, max_depth, min_size): tree = build_tree(train, max_depth, min_size) predictions = list() for row in test: prediction = predict(tree, row) predictions.append(prediction) return(predictions)# Test CART on Bank Note datasetseed(1)# load and prepare datafilename = r'G:\0pythonstudy\决策树\sonar.all-data.csv'dataset = load_csv(filename)# convert string attributes to integersfor i in range(len(dataset[0])-1): str_column_to_float(dataset, i)# evaluate algorithmn_folds = 5max_depth = 5min_size = 10scores = evaluate_algorithm(dataset, decision_tree, n_folds, max_depth, min_size)print('Scores: %s' % scores)print('Mean Accuracy: %.3f%%' % (sum(scores)/float(len(scores))))

运行结果：

{'index': 38, 'value': 0.0894}
{'index': 36, 'value': 0.8459}
{'index': 50, 'value': 0.0024}
{'index': 15, 'value': 0.0906}
{'index': 16, 'value': 0.9819}
{'index': 10, 'value': 0.0785}
{'index': 16, 'value': 0.0886}
{'index': 38, 'value': 0.0621}
{'index': 5, 'value': 0.0226}
{'index': 8, 'value': 0.0368}
{'index': 11, 'value': 0.0754}
{'index': 0, 'value': 0.0239}
{'index': 8, 'value': 0.0368}
{'index': 29, 'value': 0.1671}
{'index': 46, 'value': 0.0237}
{'index': 38, 'value': 0.0621}
{'index': 14, 'value': 0.0668}
{'index': 4, 'value': 0.0167}
{'index': 37, 'value': 0.0836}
{'index': 12, 'value': 0.0616}
{'index': 7, 'value': 0.0333}
{'index': 33, 'value': 0.8741}
{'index': 16, 'value': 0.0886}
{'index': 8, 'value': 0.0368}
{'index': 33, 'value': 0.0798}
{'index': 44, 'value': 0.0298}
Scores: [48.78048780487805, 70.73170731707317, 58.536585365853654, 51.2195121951
2195, 39.02439024390244]
Mean Accuracy: 53.659%
请按任意键继续. . .

知识点：

1.load CSV file

from csv import reader# Load a CSV filedef load_csv(filename): file = open(filename, "r") lines = reader(file) dataset = list(lines) return datasetfilename = r'G:\0pythonstudy\决策树\sonar.all-data.csv'dataset=load_csv(filename)print(dataset)

2.把数据转化成float格式

# Convert string column to floatdef str_column_to_float(dataset, column): for row in dataset: row[column] = float(row[column].strip()) # print(row[column])# convert string attributes to integersfor i in range(len(dataset[0])-1): str_column_to_float(dataset, i)

3.把最后一列的分类字符串转化成0、1整数

def str_column_to_int(dataset, column): class_values = [row[column] for row in dataset]#生成一个class label的list # print(class_values) unique = set(class_values)#set 获得list的不同元素 print(unique) lookup = dict()#定义一个字典 # print(enumerate(unique)) for i, value in enumerate(unique): lookup[value] = i # print(lookup) for row in dataset: row[column] = lookup[row[column]] print(lookup['M'])

4、把数据集分割成K份

# Split a dataset into k foldsdef cross_validation_split(dataset, n_folds): dataset_split = list()#生成空列表 dataset_copy = list(dataset) print(len(dataset_copy)) print(len(dataset)) #print(dataset_copy) fold_size = int(len(dataset) / n_folds) for i in range(n_folds): fold = list() while len(fold) < fold_size: index = randrange(len(dataset_copy)) # print(index) fold.append(dataset_copy.pop(index))#使用.pop()把里边的元素都删除（相当于转移），这k份元素各不相同。 dataset_split.append(fold) return dataset_splitn_folds=5 folds = cross_validation_split(dataset, n_folds)#k份元素各不相同的训练集

5.计算正确率

# Calculate accuracy percentagedef accuracy_metric(actual, predicted): correct = 0 for i in range(len(actual)): if actual[i] == predicted[i]: correct += 1 return correct / float(len(actual)) * 100.0#这个是二值分类正确性的表达式

6.二分类每列

# Split a data set based on an attribute and an attribute valuedef test_split(index, value, dataset): left, right = list(), list()#初始化两个空列表 for row in dataset: if row[index] < value: left.append(row) else: right.append(row) return left, right #返回两个列表，每个列表以value为界限对指定行（index）进行二分类。

7.使用gini系数来获得最佳分割点

# Calculate the Gini index for a split datasetdef gini_index(groups, class_values): gini = 0.0 for class_value in class_values: for group in groups: size = len(group) if size == 0: continue proportion = [row[-1] for row in group].count(class_value) / float(size) gini += (proportion * (1.0 - proportion)) return gini# Select the best split point for a datasetdef get_split(dataset): class_values = list(set(row[-1] for row in dataset)) b_index, b_value, b_score, b_groups = 999, 999, 999, None for index in range(len(dataset[0])-1): for row in dataset: groups = test_split(index, row[index], dataset) gini = gini_index(groups, class_values) if gini < b_score: b_index, b_value, b_score, b_groups = index, row[index], gini, groups # print(groups) print ({'index':b_index, 'value':b_value,'score':gini}) return {'index':b_index, 'value':b_value, 'groups':b_groups}

这段代码，在求gini指数，直接应用定义式，不难理解。获得最佳分割点可能比较难看懂，这里用了两层迭代，一层是对不同列的迭代，一层是对不同行的迭代。并且，每次迭代，都对gini系数进行更新。

8、决策树生成

# Create child splits for a node or make terminaldef split(node, max_depth, min_size, depth): left, right = node['groups'] del(node['groups']) # check for a no split if not left or not right: node['left'] = node['right'] = to_terminal(left + right) return # check for max depth if depth >= max_depth: node['left'], node['right'] = to_terminal(left), to_terminal(right) return # process left child if len(left) <= min_size: node['left'] = to_terminal(left) else: node['left'] = get_split(left) split(node['left'], max_depth, min_size, depth+1) # process right child if len(right) <= min_size: node['right'] = to_terminal(right) else: node['right'] = get_split(right) split(node['right'], max_depth, min_size, depth+1)

这里使用了递归编程，不断生成左叉树和右叉树。

9.构建决策树

# Build a decision treedef build_tree(train, max_depth, min_size): root = get_split(train) split(root, max_depth, min_size, 1) return root tree=build_tree(train_set, max_depth, min_size)print(tree)

10、预测test集

# Build a decision treedef build_tree(train, max_depth, min_size): root = get_split(train)#获得最好的分割点,下标值，groups split(root, max_depth, min_size, 1) return root # tree=build_tree(train_set, max_depth, min_size)# print(tree) # Make a prediction with a decision treedef predict(node, row): print(row[node['index']]) print(node['value']) if row[node['index']] < node['value']:#用测试集来代入训练的最好分割点，分割点有偏差时，通过搜索左右叉树来进一步比较。 if isinstance(node['left'], dict):#如果是字典类型，执行操作 return predict(node['left'], row) else: return node['left'] else: if isinstance(node['right'], dict): return predict(node['right'], row) else: return node['right']tree = build_tree(train_set, max_depth, min_size)predictions = list()for row in test_set: prediction = predict(tree, row) predictions.append(prediction)

11.评价决策树

# Evaluate an algorithm using a cross validation splitdef evaluate_algorithm(dataset, algorithm, n_folds, *args): folds = cross_validation_split(dataset, n_folds) scores = list() for fold in folds: train_set = list(folds) train_set.remove(fold) train_set = sum(train_set, []) test_set = list() for row in fold: row_copy = list(row) test_set.append(row_copy) row_copy[-1] = None predicted = algorithm(train_set, test_set, *args) actual = [row[-1] for row in fold] accuracy = accuracy_metric(actual, predicted) scores.append(accuracy) return scores

以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持。