3 ms·
Hmm a thought. With neural networks it's impossible to actually describe what the network does. Including proving that it has or has not used GPLed code for a
by nottorp 4y ago
Hmm a thought.
With neural networks it's impossible to actually describe what the network does. Including proving that it has or has not used GPLed code for a certain input/output set.
One could argue that all code output is GPL with the associated restrictions unless said network has provably never been trained on GPLed code.