QData/TextAttack
TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model training in NLP https://textattack.readthedocs.io/en/master/
PythonMIT
Issues
- 0
Why does relaxing the constraint (RepeatModification) lead to less successful augmentation?
#789 opened by YanghaoZYH - 6
- 0
a bugs that occurred during the evaluation
#788 opened by jiangdaolime - 0
- 1
Chinese HowNet replacement words are too slow
#785 opened by liuyuyan2717 - 1
- 1
Can not add --constraint!
#774 opened by HuichiZhou - 0
USE score questions
#775 opened by liuyuyan2717 - 1
Perturbations for NER dataset
#773 opened by avishwa23 - 0
Issue of attacking seq2seq model
#772 opened by StarWatch27 - 2
How to use textattack on generative models?
#725 opened by pratik2358 - 0
Issues attacking se2seq model (seq2sick)
#771 opened by SachJbp - 2
Error encountering while following tutorial
#716 opened by sh0416 - 1
Installation error
#760 opened by Y-0919 - 1
Textattack augment command line usage reports error: unexpected keyword argument 'high_yield'
#730 opened by Falanke21 - 0
中文对抗样本为什么报错
#762 opened by liuyuyan2717 - 0
argument error? in PSO attack->perform_search->_turn
#753 opened by nohuma - 1
Potential bug of textattack.datasets.dataset.Dataset after calling filter_by_labels_
#745 opened by wenh06 - 2
ValueError with attack_dataset.
#738 opened by weihan1 - 1
- 2
Can't run TextAttack on PSO recipe using CLI
#743 opened by spneshaei - 0
Improve training speed by allow padding to batch-wise max length instead of model supported max length
#737 opened by zodiacg - 0
ValueError: high is out of bounds for int64 (word_swap_change_number._alter_number)
#741 opened by fabriceyhc - 1
Why is datasets constrained to 2.4.0 ?
#736 opened by Alex-S-H-P - 4
[WRONG augmentation] pos_list.append(token.annotation_layers["pos"][0]._value) ->KeyError: 'pos'
#727 opened by code-chendl - 1
Custom Model API & Custom Dataset
#729 opened by saadat-r-ahmed - 1
Compatibility issue w/ Transformers model
#722 opened by svenhendrikx - 1
Implement custom tokenizer trouble
#724 opened by zhy605420954 - 1
How to use textattack on generative models?
#726 opened by pratik2358 - 1
Could TextAttack only attack substring of input.
#733 opened by ludybupt - 1
- 0
- 4
- 1
Issue with using AttackArgs class outside of the command line (in my Python script)
#728 opened by sabrinagoellner - 2
- 0
Bug of utils/string.py/words_from_text()
#723 opened by Immortalise - 0
ImportError: cannot import name'HuggingFaceDataset' from'textattack.datasets
#721 opened by zhy605420954 - 1
- 0
- 0
TypeError:predict() got an unexpected keyword argument 'force_token_predictions'
#715 opened by nlpnewone - 3
IndexError when use WordSwapQWERTY to augment data
#712 opened by zzk0 - 4
Setup issue: fatal error C1083: Cannot open compiler generated file: '': Invalid argument
#711 opened by raulkviana - 2
- 3
Support for token classification
#704 opened by vanboefer - 2
Attacking T5 model doesn't work
#697 opened by poojithat512 - 1
Typo in Help String
#702 opened by JosephTLucas - 5
PWWSRen2019 attack run much time
#696 opened by Jiang010 - 4
The attacked_text is recursively called with no stop when using TextBugger attack.
#690 opened by plasmashen - 3
`pip install textattack` with error message `ERROR: Cannot uninstall 'certifi'.`
#692 opened by otterlover24 - 3