Reinforcement Learning: From Zero to State of the Art with Pytorch 4

Post by __bee » Fri, Jun 08, 2018, 9:39 AM UTC

zxcvvcxz wrote: post #2

Hate to be that person, but - If you can go from zero to SOTA from a web tutorial in anything less than, I don't know, a year, then the field is severely underdeveloped. But let that be an opportunity: what it really means is that the field is quickly growing and there aren't well-established experts or leaders.

Post by poppingtonic » Fri, Jun 08, 2018, 10:49 AM UTC

poiuytqwer wrote: post #13

Anyone got anything similar to this for deep learning?

Post by RobertoG » Fri, Jun 08, 2018, 1:51 PM UTC

Sean1708 wrote: post #15

RobertoG wrote: post #10

Pytorch 4? Pytorch 1 is not available yet. I suppose it means Pytorch 0.4. but it doesn't sounds so good. We should leave marketing outside technical explanations. What will happen in the future when there is a Pytorch 4 and somebody find this repo?

As far as I can tell it's only the Hacker News title that mentions PyTorch 4 (in fact the repo description says "PyTorch0.4 tutorial ..."), and I think it's far more likely that that was just a typo/misunderstanding that some sort of insidious marketing technique.