PetrochukM/PyTorch-NLP
Basic Utilities for PyTorch Natural Language Processing (NLP)
PythonBSD-3-Clause
Issues
- 0
SpacyWordSplitter: module not found - AllenNLP v1.5.0
#125 opened by vrunm - 0
- 2
wmt_dataset download failed
#119 opened by chloejiwon - 0
PackagesNotFoundError in anaconda
#121 opened by SihanLiuEcho - 2
SpacyEncoder: TypeError: __init__() got an unexpected keyword argument 'language'
#105 opened by TheophileBlard - 0
- 0
Gating for inputs
#118 opened by NehaMotlani - 1
Make `check_files` more generic
#110 opened by PetrochukM - 5
- 5
- 1
MaxTokensBatchSampler
#114 opened by salvacarrion - 0
handling large-scale datasets with distributed dataloaders for iterative datasets
#109 opened by rabeehk - 0
Fix `fork_rng_wrap`
#108 opened by PetrochukM - 0
Could this be improved?
#107 opened by PetrochukM - 1
- 1
OSError on second usage of FastText()
#96 opened by rodinasophie - 3
Inefficient embedding loading code in README.md
#89 opened by JamieCT - 1
ONNX support for Encoders
#95 opened by cjermain - 0
- 1
Why there is a huge result difference between Awd-Lstm Weight_Drop and this one
#93 opened by realiti4 - 5
- 2
Input batch size doesn't match hidden[0] batch size
#62 opened by homelifes - 0
Fix simple typo: donwload -> download
#91 opened by timgates42 - 1
- 2
- 3
- 2
- 2
- 3
- 1
- 2
A little bug in torchnlp/nn/weight_drop.py
#75 opened by rishishian - 1
bleu metric doesn't seem to work properly
#78 opened by HadiZayer - 2
Http 403 when calling FastText()
#70 opened by crleblanc - 2
LockedDropout - Doc Misleading
#71 opened by JiaqiLiu - 1
FastText memory requirements
#65 opened by zbyte64 - 1
Consider adding a batch_first option to follow more closely nn.RNN argument structure
#63 opened by jpilaul - 1
- 1
pip freeze | grep -w cupy
#67 opened by cy69855522 - 1
Consider renaming pad_batch to stack_and_pad
#59 opened by PetrochukM - 4
RuntimeError: Vector for token darang has 230 dimensions, but previously read vectors have 300 dimensions. All vectors must have the same number of dimensions.
#57 opened by aurooj - 2
Ujson reliance doesn't permit successful install
#49 opened by dspoka - 1
Dataset does not return tuple
#50 opened by zplizzi - 1
missing fine-grained option in smt dataset
#54 opened by mourga - 1
- 1
- 1
Example code is outdated, AWD now uses SplitCrossEntropyLoss rather than CrossEntropyLoss
#47 opened by liulu112601 - 4
- 4
- 1
Test on IMDB dataset is failing
#43 opened by pravarmahajan - 2
The SpacyEncoder tokenizes incorrectly
#44 opened by huhk-sysu