search

radioML / dataset

Open RadioML Synthetic Benchmark Dataset
244 stars 140 forks source link

Sampling improvements #3

Closed n-west closed 8 years ago

n-west commented 8 years ago

This builds up dict datasets with 1000 vectors per (mod, snr) key that are:

A cldnn trained on this looks like figure_1-2

n-west commented 8 years ago

accomanpying sountrack for PR: https://youtu.be/JpiCOClZDlk?list=PLofULVzdp2rGyMOE8e1PBPVOPJZDAW67n&t=15