Kemal Kurniawan

I am having the same error, but mine is `KeyError: 29770`. Tried @FinchPowers's solution but didn't work. UPDATE: I switched to [coveralls-python](https://github.com/coveralls-clients/coveralls-python) and it just works.

@GuillaumeSmaha On pip it's listed as `coveralls`. You should check the link I posted above.

Hi, thanks for the PR. The tests are failing because `torch.jit` doesn't seem to have attribute `export`. I suspect this is because the PyTorch version this library uses is too...

Awesome, thank you! I think we can start simple: 1. Figure out what the lowest version of PyTorch that has `torch.jit.export` is 2. Update the PyTorch version in `.github/workflows/run_tests.yml` so...

Thanks for the detailed write up! Really appreciate the time you spent investigating this. It's surprising how half-baked the JIT support feels with all these unsupported operations. With your suggestion,...

Sorry for my very late reply. If you're still interested in updating this PR, perhaps you could put the link to the request form on the dataset name?

I'm guessing you have in-place operations inside your model `forward()`. That error is usually caused by such operations.

Hi, Your usage seems alright. The examples are meant to show how to use the CRF layer given that one has produced the emission scores, i.e. (unnormalized) log P(y_t |...

> What was confusing to me originally was the fact that your CRF layer is actually a loss that one can minimize, whereas other PyTorch implementations had a separately-defined Viterbi...

@Huijun-Cui Thanks for letting me know. Next time please open a separate issue.