更多机器学习知识请查收于： https://blog.csdn.net/weixin_45316122/article/details/109854595
Trick：纯demo，心在哪里，结果就在那里
# -*- coding: utf-8 -*- import numpy as np from sklearn import datasets from sklearn.linear_model import LogisticRegression iris = datasets.load_iris() # print(list(iris.keys())) # print(iris['DESCR']) # print(iris['feature_names']) X = iris['data'][:, 3:] # print(X) # 二分类，变成0--1 y = (iris['target'] == 2).astype(np.int) print(y) log_reg = LogisticRegression(solver="sag", max_iter=1000) log_reg.fit(X, y) X_new = np.linspace(1, 3, 1000).reshape(-1, 1) # 注意predict_proba.与predict的区别 y_proba = log_reg.predict_proba(X_new) y_hat = log_reg.predict(X_new) print(y_proba) print(y_hat)
