Pete Walsh

> I assume I should just use the tokens from the first one because the list is sorted based on probability...? That's correct

Hey @MJ2468, this version of CopyNet was built to run with an older version of AllenNLP. Since then I've ported this implementation into the official AllenNLP Models repository. I'd recommend...

Hey, still seeing this issue with the latest version.

I believe we've encountered this before. It's an issue with race conditions in NLTK when it tries to download the same corpora from two processes at the same time. You...

> I did another run like this. This time only one of them failed. The results table points me to this dataset: https://beaker.org/ds/01GC7PCX5M9B357GX6YJFY7C5R/details. > > This is clearly incomplete. Its...

> Currently, the second run will get stuck when it's time to start the step. What do you mean get stuck?

That might be a bug considering you did what it told you to do, and then you still got an error. @dirkgr is more familiar with the local workspace. That...

+1 for this. I ran into a similar situation where the step in question was defined in a dependency of my project (catwalk) so I couldn’t tick the version.

> A run of steps where each one depends on the previous step, and no other dependencies are unfulfilled, those could always be grouped. It would cut the overhead of...

Thanks @chrisjsewell, that worked.