《机器学习实战》学习记录-ch2

PS: 个人笔记，建议不看
原书资料：https://github.com/ageron/handson-ml2

2.1数据获取

import pandas as pd
data = pd.read_csv(r"C:\Users\cyan\Desktop\AI\ML\handson-ml2\datasets\housing\housing.csv")

data.head()
data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 20640 entries, 0 to 20639
Data columns (total 10 columns):
 #   Column              Non-Null Count  Dtype  
---  ------              --------------  -----  
 0   longitude           20640 non-null  float64
 1   latitude            20640 non-null  float64
 2   housing_median_age  20640 non-null  float64
 3   total_rooms         20640 non-null  float64
 4   total_bedrooms      20433 non-null  float64
 5   population          20640 non-null  float64
 6   households          20640 non-null  float64
 7   median_income       20640 non-null  float64
 8   median_house_value  20640 non-null  float64
 9   ocean_proximity     20640 non-null  object 
dtypes: float64(9), object(1)
memory usage: 1.6+ MB

data.columns

Index(['longitude', 'latitude', 'housing_median_age', 'total_rooms',
       'total_bedrooms', 'population', 'households', 'median_income',
       'median_house_value', 'ocean_proximity'],
      dtype='object')

data['ocean_proximity'].value_counts().plot()

data.describe()

import matplotlib.pyplot as plt

%matplotlib inline # 这是IPython的内置绘图命令，PyCharm用不了，可以省略plt.show()
#data.hist(bins=100,figsize=(20,15),column = 'longitude') # 选一列
# 绘制直方图
data.hist(bins=50,figsize=(20,15)) # bins 代表柱子的数目，高度为覆盖宽度内取值数目之和

# plt.show()

# 划分数据集与测试集
import numpy as np
# 自定义划分函数
def split_train_test(data, test_ratio):
    shuffled_indices = np.random.permutation(len(data)) # 将 0 ~ len(data) 随机打乱
    test_set_size = int(len(data) * test_ratio)
    test_indices = shuffled_indices[:test_set_size]
    train_indices = shuffled_indices[test_set_size:]
    return data.iloc[train_indices], data.iloc[test_indices]

train_data,test_data = my_split_train_test(data,.2)
len(train_data),len(test_data)

(16512, 4128)

from sklearn.model_selection import train_test_split
# 利用 sklean的包 切分数据集，random_state 类似 np.random.seed(42), 保证了每次运行切分出的测试集相同
train_set, test_set = train_test_split(data, test_size=0.2, random_state=42)
len(train_set),len(test_set)

(16512, 4128)

# 但是仅仅随机抽取作为测试集是不合理的，要保证测试集的数据分布跟样本一致
# 创建收入类别属性，为了服从房价中位数的分布对数据进行划分
data["income_cat"] = pd.cut(data["median_income"],
                               bins=[0., 1.5, 3.0, 4.5, 6., np.inf],
                               labels=[1, 2, 3, 4, 5])

# 分层抽样
from sklearn.model_selection import StratifiedShuffleSplit
split = StratifiedShuffleSplit(n_splits=1, test_size=0.2, random_state=42) # 
for train_index, test_index in split.split(data, data["income_cat"]):
    strat_train_set = data.loc[train_index]
    strat_test_set = data.loc[test_index]

# 查看测试集数据分布比例
strat_test_set["income_cat"].value_counts() / len(strat_test_set),data["income_cat"].value_counts() / len(data)

(3    0.350533
 2    0.318798
 4    0.176357
 5    0.114341
 1    0.039971
 Name: income_cat, dtype: float64,
 3    0.350581
 2    0.318847
 4    0.176308
 5    0.114438
 1    0.039826
 Name: income_cat, dtype: float64)

# 删除添加的 income_cat 属性
strat_test_set.drop("income_cat",axis=1,inplace=True)
strat_train_set.drop("income_cat",axis=1,inplace=True)
# 或者如此删除，可能效率更高，或者更美观吧
for set_ in (strat_train_set, strat_test_set):
    set_.drop("income_cat", axis=1, inplace=True)