李宏毅机器学习 Machine_Learning_2019_Task3

作业1：预测PM2.5的值

在这个作业中，我们将用梯度下降方法预测PM2.5的值

• hw1要求：
  1、要求python3.5+
  2、只能用（1）numpy（2）scipy（3）pandas
  3、请用梯度下降手写线性回归
  4、最好的公共简单基线
  5、对于想加载模型而并不想运行整个训练过程的人：
• hw_best要求：
  1、要求python3.5+
  2、任何库都可以用
  3、在kaggle上获得你选择的更高的分

读取数据：
import numpy as np
import scipy 
import pandas as pd

lr = 0.001
# lr = 0.001

def get_train():
    # 获得PM2.5的训练数据
    with open("train.csv") as f:
        data = f.readlines()
    data = [i.strip().split(',') for i in data]
    train_data = [];train_label = [];
    for i in data[10::18]:
        for j in range(len(i[3:]) - 10):
            train_data.append(i[3:][j : j + 9])
            train_label.append(i[3:][j + 10])
    return np.array(train_data,dtype = np.float64),np.array(train_label,dtype = np.float64)

def get_test():
    # 获得PM2.5的测试数据
    with open("test(1).csv") as f:
        data = f.readlines()
    data = [i.strip().split(',') for i in data]
    test_data = [];test_label = [];
    for i in data[9::18]:
        test_data.append(i[2:])
    with open("answer.csv") as f:
        data = f.readlines()
    data = [i.strip().split(',') for i in data]
    for i in data[1:]:
        test_label.append(i[1])
    return np.array(test_data,dtype = np.float64),np.array(test_label,dtype = np.float64)

import numpy as np
import math
import pandas as pd
from sklearn.preprocessing import StandardScaler

# 损失函数
def lossfunc(x_train,y_train,theta):
    x_train = np.array(x_train,float)
    y_train = np.array(y_train, float)
    return np.sum((y_train - x_train.dot(theta)) ** 2) / len(y_train)
# '损失函数求导
def dlossfunc(x_train,y_train,theta):
    x_train = np.array(x_train,float)
    y_train = np.array(y_train, float)
    return x_train.T.dot(x_train.dot(theta) - y_train) * 2. / len(y_train)
# 归一化
def standardfunc(data):
    standard = StandardScaler()
    standard.fit(data)
    data =standard.transform(data)
    return data

• 待补充：