从零开始机器学习之旅2【神经网络】

模块1 准备阶段

1.1第三方包准备

#encoding=utf8
import pandas as pd
import numpy as np
from sklearn import model_selection 
from sklearn.metrics import accuracy_score
from sklearn.metrics import classification_report
from sklearn.preprocessing import LabelEncoder
from io import StringIO

1.2自定义模型&函数准备

#构建感知机算法
class Perceptron(object):
    def __init__(self, learning_rate = 0.01, max_iter = 200):
        self.lr = learning_rate
        self.max_iter = max_iter
    def fit(self, data, label):
        '''
        input:data(ndarray):训练数据特征
              label(ndarray):训练数据标签
        output:w(ndarray):训练好的权重
               b(ndarry):训练好的偏置
        '''
        #编写感知机训练方法，w为权重，b为偏置
        self.w = np.array([1.]*data.shape[1])
        self.b = np.array([1.])
        #********* Begin *********#
        for _ in range(self.max_iter):
            has_wrong = False
            for idx in range(len(data)):
                x = data[idx]
                y = label[idx]                
                if y*(self.w.dot(x)+self.b) <= 0:
                    has_wrong = True
                    self.w += self.lr*y*x
                    self.b += self.lr*y
            if not has_wrong:
                return None
        return None
        #********* End *********#

    def predict(self, data):
        '''
        input:data(ndarray):测试数据特征
        output:predict(ndarray):预测标签
        '''
        #********* Begin *********#
        predict = []
        for i in range(data.shape[0]):
            reuslt = self.w.dot(data[i])+self.b
            if reuslt >= 0:
                result = 1
            else:
                result = -1
            predict.append(result)
        predict = np.array(predict)
        #********* End *********#
        return predict
# 读入数据
def createDataSet():
    """ 数据读入 """
    rawData = StringIO(
        """编号,色泽,根蒂,敲声,纹理,脐部,触感,好瓜
    1,青绿,蜷缩,浊响,清晰,凹陷,硬滑,是
    2,乌黑,蜷缩,沉闷,清晰,凹陷,硬滑,是
    3,乌黑,蜷缩,浊响,清晰,凹陷,硬滑,是
    4,青绿,蜷缩,沉闷,清晰,凹陷,硬滑,是
    5,浅白,蜷缩,浊响,清晰,凹陷,硬滑,是
    6,青绿,稍蜷,浊响,清晰,稍凹,软粘,是
    7,乌黑,稍蜷,浊响,稍糊,稍凹,软粘,是
    8,乌黑,稍蜷,浊响,清晰,稍凹,硬滑,是
    9,乌黑,稍蜷,沉闷,稍糊,稍凹,硬滑,否
    10,青绿,硬挺,清脆,清晰,平坦,软粘,否
    11,浅白,硬挺,清脆,模糊,平坦,硬滑,否
    12,浅白,蜷缩,浊响,模糊,平坦,软粘,否
    13,青绿,稍蜷,浊响,稍糊,凹陷,硬滑,否
    14,浅白,稍蜷,沉闷,稍糊,凹陷,硬滑,否
    15,乌黑,稍蜷,浊响,清晰,稍凹,软粘,否
    16,浅白,蜷缩,浊响,模糊,平坦,硬滑,否
    17,青绿,蜷缩,沉闷,稍糊,稍凹,硬滑,否
""")
    df = pd.read_csv(rawData, sep=",")
    return df


# 基于LabelEncoder的类别变量重编码
class MultiColumnLabelEncoder:
    def __init__(self, columns=None):
        self.columns = columns  # array of column names to encode

    def fit(self, X, y=None):
        return self  # not relevant here

    def transform(self, X):
        """
        Transforms columns of X specified in self.columns using
        LabelEncoder(). If no columns specified, transforms all
        columns in X.
        """
        output = X.copy()
        if self.columns is not None:
            for col in self.columns:
                output[col] = LabelEncoder().fit_transform(output[col])
        else:
            for colname, col in output.iteritems():
                output[colname] = LabelEncoder().fit_transform(col)
        return output

    def fit_transform(self, X, y=None):
        return self.fit(X, y).transform(X)

模块2 实验数据处理

2.1读入数据

df =  createDataSet()
# 输出数据
print(type(df)) # 
df

2.2 类别变量重编码

为什么要进行类别变量重编码呢?

答:机器学习中要求所有输入和输出变量均为数字。这意味着，如果你的数据包含分类数据，则必须先将其编码为数字，然后才能拟合和评估模型。

df = MultiColumnLabelEncoder(columns=['色泽', '根蒂', '敲声', '纹理', '脐部', '触感', '好瓜']).fit_transform(df)
# 输出
print(type(df)) # 
df

2.3数据划分训练集和测试集

feature_names = ['色泽', '根蒂', '敲声', '纹理', '脐部', '触感']
target_names = ['是', '否']
feature = df[feature_names]
label = df['好瓜']
# df[['色泽', '根蒂', '敲声', '纹理', '脐部', '触感']]
x = feature
print('df[[\'色泽\', \'根蒂\', \'敲声\', \'纹理\', \'脐部\', \'触感\']]')
print(x)
# df['好瓜']
y = label
print('df[\'好瓜\']')
print(y)
# 测试集占 0.2
x_train, x_test, y_train, y_test = model_selection.train_test_split(x, y, random_state=1, test_size=0.2)

模块3 模型训练

3.1 选择模型

myPerceptron = Perceptron()

3.2 代入数据模型训练

myPerceptron.fit(np.array(x_train), np.array(y_train))

3.3 输入x 预测y

y_te_pred = myPerceptron.predict(np.array(x_test))
print('感知模型的参数为w={0}, b={1}'.format(myPerceptron.w, myPerceptron.b)

print('y的真实值')
print(y_test)

3.4 模型评估 P,R,F1

# 基于P、R和F1的模型评估
# target_names 以 y的label分类为准
target_names = ['否', '是']
print("myPerceptron 模型评估")
print(classification_report(np.array(y_test), y_te_pred, target_names=target_names))
# print(y_te_pred)
# print(y_test)

myPerceptron 模型评估
查准率 查全率 f1-score 当前行的类别在测试数据中的样本总量

  否       0.00      0.00      0.00               1
   是       0.75      1.00      0.86               3

准确率 0.75 4
平均值 0.38 0.50 0.43 4
weighted avg 0.56 0.75 0.64 4

查准分母是预测 查全分母是真实
否的查准率 0/0 否的查全率 0/1
是的查准率 3/4 是的查全率 3/3 //所以会报错

3.5 基于精确率的模型评估

print("myPerceptron 模型评估")
accuracy = accuracy_score(y_te_pred, np.array(y_test))
print('accuracy:{}'.format(round(accuracy,2)))
print(y_te_pred)
print(y_test)