# Paraphrasing for NLU Data Augmentation\[Experimental\] **URL:** \ **Category:** Release Announcements\ **Created:** [April 20, 2020, 11:24am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744 "2020-04-20T11:24:06Z")\ **Posts on this page:** 11\ **Page:** 2
**Author:** ![tn.son](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/tn.son/32/8747_2.png) [@tn.son](https://forum.rasa.com/u/tn.son)\ **Post date:** [May 13, 2020, 7:31am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/21 "2020-05-13T07:31:32Z")
- With WhiteSpace tokenizer and CountVectorsFeaturizer, We generated more data and chatbot’s classification performance increases significantly. - When we have pre-trained BERT embedding, we are arguing about 2 approaches: 1. Generate more data with fine-tuning BERT language model, then feed into data-training to improve classification performance? 2. No need generating more training data because NLU with pre-trained BERT embedding can understand sentence and similar sentences already. How do you think about it? Which approach is better? ---
**Author:** ![dakshvar22](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/dakshvar22/32/3993_2.png) [@dakshvar22](https://forum.rasa.com/u/dakshvar22)\ **Post date:** [May 13, 2020, 7:46am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/22 "2020-05-13T07:46:31Z")
@tn.son Regarding your question on BERT, here is one quantitative evaluation evidence on how data augmentation can still be useful with BERT in the pipeline - [Paraphrasing for NLU Data Augmentation[Experimental] - #4 by JulianGerhard](http://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/4) > No need generating more training data because NLU with pre-trained BERT embedding can understand sentence and similar sentences already. Pre-trained BERT is good at understanding similar word contexts but actually not that great at understanding similar sentences. Hence doing a sentence level data augmentation still makes sense. ---
**Author:** ![tn.son](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/tn.son/32/8747_2.png) [@tn.son](https://forum.rasa.com/u/tn.son)\ **Post date:** [May 14, 2020, 4:06am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/24 "2020-05-14T04:06:07Z")
Thank you very much! ---
**Author:** ![tn.son](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/tn.son/32/8747_2.png) [@tn.son](https://forum.rasa.com/u/tn.son)\ **Post date:** [May 28, 2020, 1:51am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/25 "2020-05-28T01:51:39Z")
Could I ask some questions? - When do your team plan to release this feature for Rasa? - Currently, this feature only supports for English, could you share how to fine-tune GPT-2 for this task? - For other language like Vietnamese, we don’t have kind of [ParaNMT-50M dataset]. So I think about create the same ParaNMT for Vietnamese using Google Translate from [ParaNMT-50M dataset] or back translation. How do you think about this idea? - Could you suggest for me some ways to do this job in Vietnamese or non-English? ---
**Author:** ![nuszki](https://avatars.discourse-cdn.com/v4/letter/n/df705f/32.png) [@nuszki](https://forum.rasa.com/u/nuszki)\ **Post date:** [June 15, 2020, 10:16pm UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/26 "2020-06-15T22:16:13Z")
Thanks @dakshvar22, this is very helpful. Just a tiny bug: since June 3rd when transformers 2.11.0 came out the model initialising throws an error, so maybe just change install code to `transformers==2.10.0`? ---
**Author:** ![dakshvar22](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/dakshvar22/32/3993_2.png) [@dakshvar22](https://forum.rasa.com/u/dakshvar22)\ **Post date:** [June 16, 2020, 5:46am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/27 "2020-06-16T05:46:38Z")
@nuszki Thanks for pointing it out. I have pinned the version and should now work. ---
**Author:** ![vishu1994](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/vishu1994/32/4970_2.png) [@vishu1994](https://forum.rasa.com/u/vishu1994)\ **Post date:** [January 29, 2021, 9:50am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/31 "2021-01-29T09:50:36Z")
Is there any link to download this model.Downloading this from colab is very time consuming. ---
**Author:** ![iamsid](https://avatars.discourse-cdn.com/v4/letter/i/ecae2f/32.png) [@iamsid](https://forum.rasa.com/u/iamsid)\ **Post date:** [September 8, 2021, 4:42am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/32 "2021-09-08T04:42:28Z")
@dakshvar22 hello, is it now available for non-English models. ---
**Author:** ![natalieEn](https://avatars.discourse-cdn.com/v4/letter/n/7bcc69/32.png) [@natalieEn](https://forum.rasa.com/u/natalieEn)\ **Post date:** [May 13, 2022, 8:51am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/33 "2022-05-13T08:51:51Z")
Hi @dakshvar22, the download link in colab doesn’t work for me. I have the same problem as @indranil180 but even the new link you posted as a reply doesn’t work. is there any other way or link to get the model? ---
**Author:** ![AravindNani](https://avatars.discourse-cdn.com/v4/letter/a/c0e974/32.png) [@AravindNani](https://forum.rasa.com/u/AravindNani)\ **Post date:** [November 8, 2022, 7:49am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/34 "2022-11-08T07:49:51Z")
I am also facing the same issue as you @dakshvar22 could you please help us. We are getting this error when running the model. Archive: model.zip End-of-central-directory signature not found. Either this file is not a zipfile, or it constitutes one disk of a multi-part archive. In the latter case the central directory and zipfile comment will be found on the last disk(s) of this archive. unzip: cannot find zipfile directory in one of model.zip or model.zip.zip, and cannot find model.zip.ZIP, period. Please help us in resolving this issue. ---
**Author:** ![vaidehi](https://avatars.discourse-cdn.com/v4/letter/v/9fc348/32.png) [@vaidehi](https://forum.rasa.com/u/vaidehi)\ **Post date:** [January 3, 2023, 4:52am UTC](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744/35 "2023-01-03T04:52:20Z")
Hi Team ,@alexweidauer Please help us we are having issue in using this colab. And it been running since long as well as far as I see in this post thread [Previous page](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744.md?page=1)