By what principle are the weights of a conventional neural network distributed when classifying an image?

S

sexyman4202022-03-01 14:34:47

Python

sexyman420, 2022-03-01 14:34:47

I don’t understand a little how the weights are placed in the input layer of the neural network when classifying an image, because how can you determine the importance of a single pixel so that it all works also on test data?

Reply

Answer the question

In order to leave comments, you need to log in

0 answer(s)

Similar questions

N

Nikita2016-06-29 20:59:58

Why does it drive in Safari, but everything is ok in Chrome? 3Reply

K

k0r0g2021-10-02 15:15:40

Why doesn't OR work in re python? 4Reply

A

Alexander APS2021-10-02 16:20:13

Is it true that a neural network in the case of classification is, roughly speaking, an ensemble of logistic regressions? 1Reply

P

pit_ne_spit2021-10-02 17:28:20

Why is laravel_session returning strange value when using get.cookie? 1Reply

A

artgin2021-10-02 19:26:35

Python (aiogram): When clicking on an inline button, how to send the contents of the button to the bot? 0Reply

Didn't find what you were looking for?

Ask your question