prithivida commited on
Commit
d3c1083
1 Parent(s): 15a97d2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -3
README.md CHANGED
@@ -6,15 +6,17 @@ Parrot is a paraphrase based utterance augmentation framework purpose built to a
6
 
7
  ### Installation
8
  ```python
9
- pip install parrot
10
  ```
11
 
12
  ### Quickstart
13
  ```python
14
 
 
15
  import warnings
16
  warnings.filterwarnings("ignore")
17
- parrot = Parrot(model_tag="prithivida/parrot_paraphraser_on_T5", use_gpu=False)
 
18
  phrases = ["Can you recommed some upscale restaurants in Rome?",
19
  "What are the famous places we should not miss in Russia?"
20
  ]
@@ -60,7 +62,6 @@ Input_phrase: What are the famous places we should not miss in Russia
60
  ```python
61
 
62
  para_phrases = parrot.augment(input_phrase=phrase,
63
- diversity_ranker="levenshtein",
64
  do_diverse=False,
65
  max_return_phrases = 10,
66
  max_length=32,
@@ -72,6 +73,7 @@ Input_phrase: What are the famous places we should not miss in Russia
72
 
73
 
74
 
 
75
  ## 2. Why Parrot?
76
  **Huggingface** lists [12 paraphrase models,](https://huggingface.co/models?pipeline_tag=text2text-generation&search=paraphrase) **RapidAPI** lists 7 fremium and commercial paraphrasers like [QuillBot](https://rapidapi.com/search/paraphrase?section=apis&page=1), Rasa has discussed an experimental paraphraser for augmenting text data [here](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744), Sentence-transfomers offers a [paraphrase mining utility](https://www.sbert.net/examples/applications/paraphrase-mining/README.html) and [NLPAug](https://github.com/makcedward/nlpaug) offers word level augmentation with a [PPDB](http://paraphrase.org/#/download) (a multi-million paraphrase database). While these attempts at paraphrasing are great, there are still some gaps and paraphrasing is NOT yet a mainstream option for text augmentation in building NLU models....Parrot is a humble attempt to fill some of these gaps.
77
 
 
6
 
7
  ### Installation
8
  ```python
9
+ pip install git+https://github.com/PrithivirajDamodaran/Parrot.git
10
  ```
11
 
12
  ### Quickstart
13
  ```python
14
 
15
+ from parrot import Parrot
16
  import warnings
17
  warnings.filterwarnings("ignore")
18
+
19
+ parrot = Parrot(diversity_ranker="euclidean", model_tag="prithivida/parrot_paraphraser_on_T5", use_gpu=False)
20
  phrases = ["Can you recommed some upscale restaurants in Rome?",
21
  "What are the famous places we should not miss in Russia?"
22
  ]
 
62
  ```python
63
 
64
  para_phrases = parrot.augment(input_phrase=phrase,
 
65
  do_diverse=False,
66
  max_return_phrases = 10,
67
  max_length=32,
 
73
 
74
 
75
 
76
+
77
  ## 2. Why Parrot?
78
  **Huggingface** lists [12 paraphrase models,](https://huggingface.co/models?pipeline_tag=text2text-generation&search=paraphrase) **RapidAPI** lists 7 fremium and commercial paraphrasers like [QuillBot](https://rapidapi.com/search/paraphrase?section=apis&page=1), Rasa has discussed an experimental paraphraser for augmenting text data [here](https://forum.rasa.com/t/paraphrasing-for-nlu-data-augmentation-experimental/27744), Sentence-transfomers offers a [paraphrase mining utility](https://www.sbert.net/examples/applications/paraphrase-mining/README.html) and [NLPAug](https://github.com/makcedward/nlpaug) offers word level augmentation with a [PPDB](http://paraphrase.org/#/download) (a multi-million paraphrase database). While these attempts at paraphrasing are great, there are still some gaps and paraphrasing is NOT yet a mainstream option for text augmentation in building NLU models....Parrot is a humble attempt to fill some of these gaps.
79