I am trying to follow this example/lambada-train_model.ipynb for my dataset. I am able to run smoothly until the last cell. (my dataset has only two classes 0 and 1) When I check the source code where this error occurs and found this strange thing: As I understand, self.gen_model is the original...

aug.augment("Your goal is to separate those (group) into Separate strings.")[0] 'Our goal is to separate those ( group ) into Separate strings.' Is there any way to get rid of these extra whitespaces?

Code import nlpaug.augmenter.word as naw original = """ “Fury said to a mouse, That he met in the house, ‘Let us both go to law: I will prosecute you.—Come, I’ll take no denial; We must have a...

Hi, I'm trying to create a word augmenter using glove. This is my code: import nlpaug.augmenter.word.word_embs as nawwe aug = nawwe.WordEmbsAug(model_type='glove',model_path='/content/drive/MyDrive/data/Glove/glove.6B.300d.txt', action = 'insert') (I'm reading the glove .txt file from my google driv...

import nlpaug.augmenter.word as naw aug = naw.ContextualWordEmbsAug() text="Transformers are the most popular toys" print(f"original text:{text}") print(f"Augmented text: {aug.augment(text)}") original text:Transformers are the most popular toys RuntimeError Traceback (m...

Thank you for creating this library, quite helpful! Just curious, it doesn't look like this paper is covered in this library: https://arxiv.org/pdf/2003.02245.pdf its Implementation is here Is that right? Any plans to incorporate it?

Hi! I am trying to use nlpaug.augmenter.word on a big corpus (500k sentences). Is muti-GPU supported? And if so, is it transparent, or it needs to be enabled somehow? Thanks and great work!

Hi, I was using nlpaug.augmenter.word.context_word_embs.ContextualWordEmbsAug to augment my text with bert embeddings. Here is my code: import torch import nlpaug.augmenter.word.context_word_embs as naw device = torch.device("cuda" if torch.cuda.is_available() else "cpu") aug = naw.ContextualWordEm...

aug = nas.ContextualWordEmbsForSentenceAug(model_path='xlnet-base-cased') text='am not interested in linguistics that does not address race racism is about power raciolinguistics brings' augmented_texts = aug.augment(text) print("Original:") print(text) print("Augmented Texts:") print(augmented_text...

It seems that candidates in RandomCharAug is not working as expected from the docs, but is instead ignored. As expected, this also shows up as spec_char being ignored. For example, letters that are not included in candidates are still swapped or deleted: In: nac.RandomCharAug(action="swap", candidat...