graykode / xlnet-Pytorch

Simple XLNet implementation with Pytorch Wrapper

https://arxiv.org/pdf/1906.08237.pdf

graykode/xlnet-Pytorch Issues

Confusion about the relative position embedding with attn_type='bi' but bsz=1
Updated 3 years ago1
Parameter initialized with torch.randn may be not a good choice
Updated 3 years ago
position-wise feedforward only one linear layer?
Updated 3 years ago
Runtime Error in colab
Updated 4 years ago
Batch training
Updated 4 years ago
Error and general question
Updated 4 years ago1
RuntimeError: Expected object of scalar type Byte but got scalar type Bool for argument #2 'other'
Updated 4 years ago3
how to save the pre-trained language model?
Updated 5 years ago1
how to do inference?
Updated 5 years ago1
Model name 'bert-large-uncased' was not found in model name list (bert-base-uncased, bert-large-uncased, bert-base-cased, bert-large-cased, bert-base-multilingual-uncased, bert-base-multilingual-cased, bert-base-chinese).
Updated 5 years ago
TypeError:can't convert np.ndarray of type numpy.bool_
Updated 5 years ago2
about first men‘s bug
Updated 5 years ago
Reimplement training time and the performance on each task?
Updated 5 years ago5
Low accuracy on sample task
Updated 5 years ago1
Re-implementation Performance
Updated 5 years ago1
do we need to create_data for every epoch?
Closed 5 years ago
The permutation function " _local_perm" is confused.
Closed 5 years ago