LogisticRegression 多分类实例

import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt
import plotly.graph_objs as go
from plotly.offline import init_notebook_mode, iplot
init_notebook_mode(connected = True)
from sklearn import metrics

df = pd.read_csv(filepath_or_buffer='./data/iris/iris.csv',header=None)
df.columns = ['sepal_len', 'sepal_width', 'petal_len', 'petal_width', 'class']
df.tail()

	sepal_len	sepal_width	petal_len	petal_width	class
145	6.7	3.0	5.2	2.3	Iris-virginica
146	6.3	2.5	5.0	1.9	Iris-virginica
147	6.5	3.0	5.2	2.0	Iris-virginica
148	6.2	3.4	5.4	2.3	Iris-virginica
149	5.9	3.0	5.1	1.8	Iris-virginica

df.loc[df['class'] == 'Iris-setosa','class'] = 0
df.loc[df['class'] == 'Iris-versicolor','class'] = 1
df.loc[df['class'] == 'Iris-virginica','class'] = 2

df.tail()

	sepal_len	sepal_width	petal_len	petal_width	class
145	6.7	3.0	5.2	2.3	2
146	6.3	2.5	5.0	1.9	2
147	6.5	3.0	5.2	2.0	2
148	6.2	3.4	5.4	2.3	2
149	5.9	3.0	5.1	1.8	2

df.head()

	sepal_len	sepal_width	petal_len	petal_width	class
0	5.1	3.5	1.4	0.2	0
1	4.9	3.0	1.4	0.2	0
2	4.7	3.2	1.3	0.2	0
3	4.6	3.1	1.5	0.2	0
4	5.0	3.6	1.4	0.2	0

labels = df.groupby('class').size().index
values = df.groupby('class').size()
trace = go.Pie(labels=labels, values=values)
layout = go.Layout(width=350, height=350)
fig = go.Figure(data=[trace], layout=layout)
iplot(fig)

var gd = document.getElementById(‘fa8f013f-33df-43da-b3ca-0d40cc41d3a7’);
var x = new MutationObserver(function (mutations, observer) {{
var display = window.getComputedStyle(gd).display;
if (!display || display === ‘none’) {{
console.log([gd, ‘removed!’]);
Plotly.purge(gd);
observer.disconnect();
}}
}});

// Listen for the removal of the full notebook cells
var notebookContainer = gd.closest(’#notebook-container’);
if (notebookContainer) {{
x.observe(notebookContainer, {childList: true});
}}

// Listen for the clearing of the current output cell
var outputEl = gd.closest(’.output’);
if (outputEl) {{
x.observe(outputEl, {childList: true});
}}

                    })                };                });            </script>        </div>

X = df.iloc[:, :4].values
Y = df['class'].values.astype('int')

x_train, x_test, y_train, y_test = train_test_split(X,Y, test_size = 0.3, random_state = 0)

x_train.shape,y_train.shape,x_test.shape, y_test.shape

((105, 4), (105,), (45, 4), (45,))

模型搭建与分类器训练

1、导入模型，调用逻辑回归LogisticRegression()函数。

• penalty: 正则化选择参数（惩罚项的种类），默认方式为L2正则化
• C: 正则项系数的倒数
• solver: 对于多分类任务， 使用‘newton-cg’, ‘sag’, ‘saga’ and ‘lbfgs’ 来解决多项式loss
• multi_class: 默认值‘ovr’适用于二分类问题，对于多分类问题，用‘multinomial’在全局的概率分布上最小化损失

2、训练LogisticRegression分类器

• 调用fit(x,y)的方法来训练模型，其中x为数据的属性，y为所属类型。

3、利用训练得到的模型对数据集进行预测 predict()，返回预测结果。

lr = LogisticRegression(penalty='l2',solver='newton-cg',multi_class='multinomial')

lr.fit(x_train,y_train)

LogisticRegression(multi_class='multinomial', solver='newton-cg')

print("Logistic Regression模型训练集的准确率：%.3f" %lr.score(x_train, y_train))

Logistic Regression模型训练集的准确率：0.981

y_hat = lr.predict(x_test)
accuracy  = metrics.accuracy_score(y_test,y_hat)
print("Logistic Regression模型正确率：%.3f" %accuracy)

Logistic Regression模型正确率：0.978