sklearn_NNmodel
2023-02-10 本文已影响0人
Lonelyroots
sklearn_NNmodel
from sklearn.neural_network import MLPClassifier
from sklearn.neural_network import MLPRegressor
import numpy as np
X = np.array([[0.,0.],
[1.,1.]])
y = np.array([0,1])
print(X.shape)
print(y.shape)
clf = MLPClassifier(solver="sgd",alpha=1e-5,activation="relu",hidden_layer_sizes=(5,2),max_iter=2000,tol=1e-4,verbose=True) # solver选择优化算法sgd(随机梯度下降),alpha正则项系数,hidden layer里用激活函数relu,hidden_layer_sizes=(5,2),2个隐藏层,分别有5个隐藏节点,2个隐藏节点,二元组:里面存2个元素的元组,max_iter最大迭代次数,tol忍受度(sklearn默认连续10次小于,停止迭代),verbose多打印一些信息
clf.fit(X,y)
# 参数项
print([coef.shape for coef in clf.coefs_])
print([coef for coef in clf.coefs_])
# 截距项
print([intercept.shape for intercept in clf.intercepts_])
print([intercept for intercept in clf.intercepts_])
predict_value = clf.predict([[2,2],
[-1,-2]])
print(predict_value)
predict_proba = clf.predict_proba([[2.,2.],[-1.,-2.]])
print(predict_proba)
文章到这里就结束了!希望大家能多多支持Python(系列)!六个月带大家学会Python,私聊我,可以问关于本文章的问题!以后每天都会发布新的文章,喜欢的点点关注!一个陪伴你学习Python的新青年!不管多忙都会更新下去,一起加油!
Editor:Lonelyroots