Tom Carchrae

given the pain of getting the correct version of nvidia drivers on a system all lined up, i wonder if a docker image of this repo would help. (or alternatively,...

@diffractometer - sounds right. if you do have a cuda supported nvidia card (but not the cuda libs installed) you could probably still run the docker + gpu extensions locally....

i suspect you'd hit the same error in jupyter as it seems the code requires cuda/nccl https://github.com/openai/jukebox/blob/master/jukebox/utils/dist_utils.py#L42 i guess this is really a bug in documentation (a common one) that...

hmm - or maybe not? and there is a cpu only flag. (i bet it'll be damn slow tho!) https://github.com/openai/jukebox/blob/b4dc344d0e293b9a713d9cfe15be7c66ba472902/jukebox/utils/dist_utils.py#L77

did you get any more useful error from this output? ``` print(f"Caught error during NCCL init (attempt {attempt_idx} of {n_attempts}): {e}") ``` also, love the comment on the next line...

so, i went through the install steps, and the sample seems to work for me (it is still running/downloading stuff) my system: ubuntu 18.04, cuda lib installed 10.2.89, gtx 1060...

adding to my wishlist. - source ref to the original template (a line number would be ok - but sourcemaps would be oh so sexy! http://www.html5rocks.com/en/tutorials/developertools/sourcemaps/ )

also, noted that this behaviour goes back to the original dust.js : http://akdubya.github.io/dustjs/ - you can see it in the 'try it out boxes' if you copy the example above.

Ok, here is the issue. Despite the docs saying that a chunk created by map must _always_ call Chunk.end(), it just isn't true if you call map again on the...

I think the issue here is that the preferred context order often changes depending on use, so either default is always going to confuse/annoy someone. Perhaps forcing section name scoping...