shayneobrien / coreference-resolution

Efficient and clean PyTorch reimplementation of "End-to-end Neural Coreference Resolution" (Lee et al., EMNLP 2017).

https://arxiv.org/pdf/1707.07045.pdf

shayneobrien/coreference-resolution Issues

RuntimeError: received an empty list of sequences
Updated a year ago1
What should I do with the data?
Updated 3 years ago4
list index out of range in pad_sequence of torch implementation.
Updated 3 years ago8
error when evaluating
Updated 3 years ago2
hello,author ,i am try to train a model with your code,the data used conll2012,i found it precise loss decrease, precise approximately equal to 0.
Updated 4 years ago7
A bug, can you fix that?
Updated 4 years ago1
Error during training
Updated 4 years ago4
Error in sentence truncation
Updated 4 years ago
Is there a problem in function:"remove_overlapping(sorted_spans):" ?
Updated 4 years ago5
Gamma in the LR scheduler is too small
Updated 4 years ago
probs = [F.softmax(tensr) for tensr in with_epsilon] may be wrong?
Updated 4 years ago
Do you figure out your TODO list?
Updated 4 years ago
train issue
Updated 5 years ago2
Pretrained model?
Closed 5 years ago
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation
Updated 5 years ago2
hello,author, I am
Closed 5 years ago
the performance of model ?
Updated 5 years ago1
How to preprocess the Data ?
Closed 5 years ago2
model does not predict clusters
Closed 5 years ago3
need data set
Closed 6 years ago1