机器学习小练习

机器学习

完整代码

import pandas as pd
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

df=pd.read_csv('G:/liu/python/music.csv') # 导入数据
X = df.drop(columns=['genre'])		#切割数据，成两个变量，输入变量X和输出的因变量Y
Y = df['genre']						# 输出的因变量Y
X_tr, X_te, Y_tr, Y_te = train_test_split( X,Y,test_size = 0.2 )	#模型训练和测试数据分配，训练：测试=8：2

model = DecisionTreeClassifier()	#创建模型
model.fit(X_tr,Y_tr)			# 输入数据X_tr与结果数据Y_tr，训练模型
pre = model.predict(X_te)		#输入测试数据X_te，得到预测数据
score = accuracy_score(Y_te, pre)#对比预测数据与正确结果Y_te的准确度打分
print(score)			# 打印准确度

一、导入python库

import pandas as pd   # 数据分析库
from sklearn.tree import DecisionTreeClassifier # sklearn 算法，决策算法
from sklearn.model_selection import train_test_split #模型训练和测试
from sklearn.metrics import accuracy_score #模型的准确度

二、导入数据

import pandas as pd
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

df=pd.read_csv('G:/liu/python/music.csv')

三、数据处理

X = df.drop(columns=['genre'])
Y = df['genre']
X_tr, X_te, Y_tr, Y_te = train_test_split( X,Y,test_size = 0.2 )

## 四、模型建立

```python
model = DecisionTreeClassifier()
model.fit(X_tr,Y_tr)

五、模型测试

pre = model.predict(X_te)

六、模型准确度

score = accuracy_score(Y_te, pre)

模型保存和加载

import pandas as pd
from sklearn.tree import DecisionTreeClassifier
from sklearn.externals import joblib	#

df=pd.read_csv('G:/liu/python/music.csv')
X = df.drop(columns=['genre'])
Y = df['genre']

model = DecisionTreeClassifier()
model.fit(X, Y)
joblib.dump(model, 'G:/liu/python/music.joblib') #把训练好的模型保存到G:/liu/python/music.joblib

模型的加载和使用

import pandas as pd
from sklearn.tree import DecisionTreeClassifier
from sklearn.externals import joblib
model = joblib.load('G:/liu/python/music.joblib')	#加载模型
pre = model.predict( [[21, 1],[22,0 ]] )	#使用模型预测结果
print(pre)	#打印预测值

导出模型成可视化dot文件

import pandas as pd
from sklearn.tree import DecisionTreeClassifier
from sklearn import tree
df=pd.read_csv('G:/liu/python/music.csv')
X = df.drop(columns=['genre'])
Y = df['genre']

model = DecisionTreeClassifier()
model.fit(X, Y)
#输出文件可视化dot文件，设置可视化属性
tree.export_graphviz(model, out_file = 'G:/liu/python/music.dot', feature_names = ['age', 'gender'], class_names = sorted(Y.unique()), label='all', rounded = True ,filled=True)