numpy_neural_network icon indicating copy to clipboard operation
numpy_neural_network copied to clipboard

多分类损失函数 梯度那里没看懂

Open 2hip3ng opened this issue 4 years ago • 2 comments

https://github.com/yizt/numpy_neural_network/blob/master/nn/losses.py dy = y_probability - y_true 这个地方的推导不是很理解。大佬这个多分类的损失函数求梯度咋来的啊。

2hip3ng avatar Apr 25 '21 13:04 2hip3ng

@2hip3ng https://blog.csdn.net/csuyzt/article/details/81839388 这里有推导

yizt avatar Apr 26 '21 02:04 yizt

嗯嗯,昨天晚上找到了,十分感谢🙏


该邮件从移动设备发送

--------------原始邮件-------------- 发件人:"mick.yi @.>; 发送时间:2021年4月26日(星期一) 上午10:35 收件人:"yizt/numpy_neural_network" @.>; 抄送:"Zhipeng Wang @.>;"Mention @.>; 主题:Re: [yizt/numpy_neural_network] 多分类损失函数 梯度那里没看懂 (#7)

@2hip3ng https://blog.csdn.net/csuyzt/article/details/81839388 这里有推导

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

2hip3ng avatar Apr 26 '21 02:04 2hip3ng