I made a neural network to classify fake animals || Devlog

[@teddieo7647]

Keep making videos!

[@nextProgram]

Teddie O Thanks I will

[@DevDrawer]

Very clever. I am working on my Python skills. I have not used it that much but I am learning it so this video looks cool to me and shows what I can do with it in the future. Thanks.

[@nextProgram]

Thanks! Yeah, Python is pretty fun

[@erikaordog8288]

What characteristics would the program use to determine if there is a cat in a picture? Or would the program determine those itself, by feeding it a bunch of pictures of cats, or something like that?

[@nextProgram]

Erika Ordog it would determine those itself. For example if you wanted to make a neural network that could recognize digits that someone drew, the easiest way to do it is to convert the image into an array of 0’s and 1’s that represent whether the pixel is closer to white or black. Each of those are fed into the network, so you might have 16*16 data points fed into it instead of 2 in my example. Then you’d also need something called hidden layers, which are basically just extra nodes that all of the data from the previous layer flows into and that each have their own weights and biases. I think I got this far when I tried it, but what’s really hard is trying to backpropagate all of the information because the network has to be able to update itself to learn. But yeah, if you made one for identifying cats then it should be able to determine the characteristics itself based on the pixel data

[@jiraibozo]

i think i understand so math math msth math math math math and BOOM you have a neural network that wins dark souls

[@nextProgram]

Pretty much lol

[@teddieo7647]

What coding language is this?

[@nextProgram]

Teddie O It’s Python

[@pilotsupreme7574]

Teddie O z