12
BadFox
5y

I'm reinventing the wheel by making yet another neural network library. It's not any good yet but I learn as I go along.

The only documentation that exists now is the admittedly quite comprehensive code comments. I'm it because Keras (using TensorFlow) requires a 3.5 compute ability rating for CUDA acceleration (which I don't have) and it doesn't support OpenCL. Eventually, I will make my implementation support both with varying levels of acceleration for different compute capabilities with the oldest supported being my hardware. If I ever get around to it.

I'd say wish me luck but determination would be infinitely more useful.

Comments
  • 3
    www.paperspace.com/ml gets you the cuda power you need. And via ssh it's like on your machine.
    good luck 👍
  • 5
    @heyheni I could use that. At the cost of losing learning about everything from nothing. What I want to eventually use it for will not tolerate that kind of latency anyway. Thanks tho.
Add Comment