wyliec

It's always good to own your music, I've wanted to for along time (instead of relying on Spotify/Apple Music). You can download a list of your songs that are currently on music services using www.tunemymusic.com. Once you've got a list, copy and paste the names with arists (or just type a few manually) into the box below. Make sure each song/artist is on a new line.

When going travelling without a phone, or wanting to use a 100 hour battery life mp3 player, this the project I decided to do!

It is legal, just like other youtube converter sites.

No need to take the name so literally though - it's rare names are taken literally at all!

I plan to make many different types/colours of pins for different topics (social, informative, advertisement, etc). Cheers.

It's a mimic - there isn't a method in apple maps that render it. I made the 3D versions myself. I use scene libraries that overlay the camera view (that is basically what AR is).

when you say output models are you referring to the optimized ANNs, or the organisation of output neurons?

...Hence why I've called my "feedforward artificial neural network model that maps sets of input data onto a set of appropriate outputs" builder Perceptron...

The pin gets rendered in a 3D scene, and positioned via finding the difference between your current coordinates and the pins. The building stuff is so that people cannot see a pin when there should be a building in the way. I do this by using computer vision that analyses the map imagery.

Why exactly?

These are most likely more matured than Perceptron at the moment. More work to do!

Not in particular. Just a satisfiying name, and relevant enough.

Really happy to hear it's living up to its purpose.

I started this project when I was 16 and it is now 5 months into development. I'm interested to know what people think of the idea, and how I've decided to go about the project (see README for screenshots and details). Depending on the interest I receive, I hope to get it onto the app store soon. Thanks for taking a look.

Cheers for feed back. The cost/error function is simply the 1/2(a-y)^2 ...(I think its called the square cost function). I wrote all the code myself (no libs) so people like yourself can see all steps required to train an ANN. I was thinking of getting things working for GPU, which should even make cython look slow. All depends on how much attention I get. Right now it's still got quite a few bugs, and in general is under done.

great to hear! The reason I don't use libraries is because I want the code, in itself, to be educational. So people can read the code and see what algorithms are being used for every step. This project is less about trying to compete with amazing performance, and more of an educational play ground.