Ranter
Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
![](/static/devrant/img/pipeless-devrant-banner-white.png)
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Comments
-
@treeroot good thinking. I have to learn how to be useful to society and not a lazy piece of shit... ;) lol
-
@SteffTek It is called "Magic". There is no way to actually "understand" a neural network. Try to think of it like this: using multiple weighted scoring algorithems that use other weighted sciring algorithems as input. as long as you "understand" a scoring algorithem, you can "undrstand" neural networks.
-
@SteffTek Scoring algorithem is really simple. imagine you see something that has a color. that color has "scores" for green, red and blue components. the "sum" of those "scores" gives you a representation of the color you see.
-
@magicMirror jeah I know that. The only thing I can imagine doing with a neural network is analysing a picture and replicate it
-
neurals can be used to solve a more general problems then that. it is very good at classification problems. like looking at a photo and telling if there is a cat in it, and where. The most usefull thing about it, that you don't need to know how it actually work in order to use it. This is true for all ML algorithems.
-
@magicMirror I know the different usages, but not the HOWWW :C and Wikipedia isn't helping :0
-
@magicMirror and I also know the theory of weight bias hiddenlayer output and input and everything... Just not how to actually use it
-
@SteffTek heh. you first need a problem, and then you need data. then you can apply ML. I suggest you go take a look at http://www.pyimagesearch.com for examples on ML use. read about projects that were done with tensorflow. Then look at this, for a real world application of ML: https://youtube.com/watch/...
-
@SteffTek See I think you want to do the heavy weight lifting stuff without using any framework like tensorflow. Am I right?
-
@SteffTek Let us consider tenserflow. It has models repository in it. This models repository contains various learning models like inception etc. Just google it how they are implemented. My suggestion start with models like SVM, perceptron etc.
-
@DLMousey Thats what I tried to clarify first. @SteffTek wants to learn it by diving deep.
Related Rants
Me: knows how to program a neural network
Also me: doesn't know how to use it
Again me: bad at math so doesn't know how to use the outputs of a neural network
undefined
am
i
a
fucking
idiot