分类5：机器学习处理帕尔默企鹅数据

1 介绍

帕尔默企鹅数据集是分类问题，每一列的含义如下：

• species 离散值 标签信息，值为 Adelie|Chinstrap|Gentoo 之一
• island 离散值 岛屿，值为 Torgersen|Biscoe|Dream 之一
• culmen_length_mm 连续值 喙的长度（mm）
• culmen_depth_mm 连续值 喙的高度（mm）
• flipper_length_mm 连续值 脚蹼长度（mm）
• body_mass_g 连续值 体重（克）
• sex 离散值 性别，值为 MALE| FEMALE 之一
• 数据集地址：https://download.csdn.net/download/ww596520206/87512899

2 导入包

import numpy as np
import pandas as pd
from sklearn.tree import DecisionTreeClassifier
from sklearn.svm import SVC
from sklearn.neighbors import KNeighborsClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split
np.random.seed(123)

3 导入数据，打印数据信息

data = pd.read_csv("penguins_size.csv")

print(data.head())
print(data.info())
print(data.shape)

# 查看相关系数
print(data.corr())

4 缺失值处理，使用impute

print(data.isnull().sum())

#缺失值填补
from sklearn.impute import SimpleImputer
imputer = SimpleImputer(strategy='most_frequent')
data.iloc[:,:] = imputer.fit_transform(data)

处理前：

处理后：

5 特征编码

from sklearn.preprocessing import LabelEncoder
data["species"] = LabelEncoder().fit_transform(data["species"])
data["island"] = LabelEncoder().fit_transform(data["island"])
data["sex"] = LabelEncoder().fit_transform(data["sex"])

6 划分标签特征，训练集测试集

x = data.drop(["species"],axis=1)
y = data["species"]
X_train, X_test, y_train, y_test = train_test_split(x, y,train_size=0.7,shuffle = True, random_state=123)

7 归一化

from sklearn.preprocessing import  MinMaxScaler
scaler = MinMaxScaler() #归一化
scaler.fit(X_train)
X_train = scaler.transform(X_train)
X_test = scaler.transform(X_test)

8 模型训练和验证

model_list = [DecisionTreeClassifier(), SVC(),KNeighborsClassifier(),RandomForestClassifier()]
for model in model_list:
    model.fit(X_train, y_train)
    y_pred = model.predict(X_test)
    acc = round(accuracy_score(y_test, y_pred), 4)
    print("model:{} acc:{}".format(model, acc))
model:DecisionTreeClassifier() acc:0.9327
model:SVC() acc:0.9904
model:KNeighborsClassifier() acc:0.9904
model:RandomForestClassifier() acc:0.9904