pytorch-crf
pytorch-crf copied to clipboard
(Linear-chain) Conditional random field in PyTorch.
### Description Supports Torchscripting the CRF model. Fixes issue #93 without changing the current interface. Added additional tests to verify that Torchscripted CRF model outputs for forward and decode are...
I would like to know the the parameter "num_tags" of the CRF()? should the tagset={'O':0, 'B':1, 'I':2, '': 3, '': 4} include these two labels:'', ''? because I need to...
### Description Upgrade library to support Pytorch 1.2 which will enable us to Torchscript the forward and decode method.
Hi, @kmkurn I had a similar problem that @vijay120 had. I also have this issue and the start, end, and transition scores do not update. ``` for epoch_i in range(epochs):...
When I Calculate the loss using this moddle,the error is `RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation` . But When I...
This fixes #92
Besides the toy examples listed in the docs and tests, are there actual examples of this library available anywhere? I'm interested in using this library for a sequence labeling project,...
Hello, I am trying to use a BERTCRF model. Unfortunately, the following error message appears: **IndexError: index -100 is out of bounds for dimension 0 with size 9** I have...
Hi, the author, is it right to think the end_transitions can be deleted in def _compute_score??? Thanks.
I want to build a CRF that represents a specific state graph where I know that transitions between certain states are impossible (e.g. I have states A, B, and C,...