Bruno Cabral

Debugging I can see that things are not implemented yet. For example, sample_tag.py calls Tagger::viterbi(). At line 527 in crfsuite.hpp it calls "labels->release(labels);", but labels are set to NULL. EDIT:...

With the latest set of commit, things are working now ! Thank you !!

Awesome !! I did got my results at time. However, not with your version of higher order =( It was very slow and not feasible to use. I ended using...

@lessw2020 I know that I´m just a beggar, but the first thing I do every morning is open this issue to check if you got to MABN. Good vibes from...

@StellaAthena I know. I was jk about the mx 250. However, I do see the use case for training a larger model where a cpu check-pointing/ gradient accumulation would not...

Sounds fair @StellaAthena , hopefully in my upcoming vacation I can work on it. But I'm curious, can you share the Nvidia feedback?

Thank you @taoleicn . I did a **very possibly wrong** implementation here https://github.com/bratao/sru/commit/1c614c34713a699451c60986afa2d9b0d3d86cba However, I'm running some tests and apparently it is converging faster

@taoleicn I only tested on toy examples and them converged faster. Unfortunately, I'm too overwhelmed with work to test on something like enwik8 Did you had the opportunity to check...

Just to comment about it, I had the same error, but it was masked because of the try catch. In my case, I was missing the nvcc . Changing the...

Hi @taolei87 , 1. I do super-long, almost infinite sequence labeling on legal documents. Almost all RNN networks are too sensitive to overfitting as my training set is small. On...