Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Nicely done but wait, 15MB?? Clearly the model isn't big enough :) Are you in under or overfitting regime?

As I was listening through the samples it seemed to me that it would start out quite energetic and then converge on repetitive, slow chords. Any ideas on why this could be happening? Or perhaps it's not true.

Also, you should label the samples with numbers so that it's possible to refer to them easily. I liked 4th from bottom quite a bit in the beginning.



What's wrong with 15MB? A Googlenet model is only ~50MB while being able to recognize objects in real world images with good accuracy.


Yes, and if you look at the images deep dream generates from that data, it's not even close. This isn't about recognizing music.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: