prithivida commited on
Commit
851b42e
1 Parent(s): fd89988

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +0 -98
README.md CHANGED
@@ -3,101 +3,3 @@
3
  # Gramformer
4
  Human and machine-generated text often suffer from grammatical and/or typographical errors. It can be spelling, punctuation, grammatical or word choice errors. Gramformer is a library that exposes 3 separate interfaces to a family of algorithms to **detect, highlight and correct** grammar errors. To make sure the corrections and highlights recommended are of high quality, it comes with a quality estimator. You can use Gramformer in one or more areas mentioned under the "use-cases" section below or any other use case as you see fit. Gramformer stands on the shoulders of giants, it combines some of the top-notch researches in grammar correction. *Note: It works at **sentence levels** and has been trained on 128 length sentences, so not (yet) suitable for long prose or paragraphs (stay tuned for upcoming releases)*
5
 
6
- ## Usecases for Gramformer
7
-
8
- **Area 1: Post-processing machine-generated text**
9
-
10
- Machine-Language generation is becoming mainstream, so will post-processing machine-generated text.
11
-
12
- - Conditioned Text generation output(Text2Text generation).
13
- - NMT: Machine Translated output.
14
- - ASR or STT: Speech to text output.
15
- - HTR: Handwritten text recognition output.
16
- - Paraphrase generation output.
17
- - Controlled Text generation output(Text generation with PPLM) **[TBD]**.
18
- - Free-form text generation output(Text generation)**[TBD]**.
19
-
20
-
21
- **Area 2:Human-In-The-Loop (HITL) text**
22
- <ul>
23
- <li>Most Supervised NLU (Chatbots and Conversational) systems need humans/experts to enter or edit text that needs to be grammatically correct otherwise the quality of HITL data can degrade the model over a period of time </li>
24
- </ul>
25
-
26
- **Area 3:Assisted writing for humans**
27
- <ul>
28
- <li>Integrating into custom Text editors of your Apps. (A Poor man's grammarly, if you will) </li>
29
- </ul>
30
-
31
- **Area 4:Custom Platform integration**
32
-
33
- As of today grammatical safety nets for authoring social contents (Post or Comments) or text in messaging platforms is very little (word level correction) or non-existent.The onus is on the author to install tools like grammarly to proof read.
34
-
35
- - Messaging platforms and Social platforms can highlight / correct grammtical errors automatically without altering the meaning or intent.
36
-
37
- ## Installation
38
- ```python
39
- pip install git+https://github.com/PrithivirajDamodaran/Gramformer.git@v0.1
40
- ```
41
- ## Quick Start
42
-
43
- ### Correcter - [Available now]
44
- ```python
45
- from gramformer import Gramformer
46
- import torch
47
-
48
- def set_seed(seed):
49
- torch.manual_seed(seed)
50
- if torch.cuda.is_available():
51
- torch.cuda.manual_seed_all(seed)
52
-
53
- set_seed(1212)
54
-
55
-
56
- gf = Gramformer(models = 2, use_gpu=False) # 0=detector, 1=highlighter, 2=corrector, 3=all
57
-
58
- influent_sentences = [
59
- "Matt like fish",
60
- "the collection of letters was original used by the ancient Romans",
61
- "We enjoys horror movies",
62
- "Anna and Mike is going skiing",
63
- "I walk to the store and I bought milk",
64
- "We all eat the fish and then made dessert",
65
- "I will eat fish for dinner and drank milk",
66
- "what be the reason for everyone leave the company",
67
- ]
68
-
69
- for influent_sentence in influent_sentences:
70
- corrected_sentence = gf.correct(influent_sentence)
71
- print("[Input] ", influent_sentence)
72
- print("[Correction] ",corrected_sentence[0])
73
- print("-" *100)
74
- ```
75
-
76
- ```text
77
- [Input] Matt like fish
78
- [Correction] Matt likes fish
79
- ----------------------------------------------------------------------------------------------------
80
- [Input] the collection of letters was original used by the ancient Romans
81
- [Correction] The collection of letters was originally used by the ancient Romans.
82
- ----------------------------------------------------------------------------------------------------
83
- [Input] We enjoys horror movies
84
- [Correction] We enjoy horror movies
85
- ----------------------------------------------------------------------------------------------------
86
- [Input] Anna and Mike is going skiing
87
- [Correction] Anna and Mike are going skiing
88
- ----------------------------------------------------------------------------------------------------
89
- [Input] I walk to the store and I bought milk
90
- [Correction] I walked to the store and bought milk.
91
- ----------------------------------------------------------------------------------------------------
92
- [Input] We all eat the fish and then made dessert
93
- [Correction] We all ate the fish and then made dessert
94
- ----------------------------------------------------------------------------------------------------
95
- [Input] I will eat fish for dinner and drank milk
96
- [Correction] I'll eat fish for dinner and drink milk.
97
- ----------------------------------------------------------------------------------------------------
98
- [Input] what be the reason for everyone leave the company
99
- [Correction] what can be the reason for everyone to leave the company.
100
- ----------------------------------------------------------------------------------------------------
101
- ```
102
-
103
-
 
3
  # Gramformer
4
  Human and machine-generated text often suffer from grammatical and/or typographical errors. It can be spelling, punctuation, grammatical or word choice errors. Gramformer is a library that exposes 3 separate interfaces to a family of algorithms to **detect, highlight and correct** grammar errors. To make sure the corrections and highlights recommended are of high quality, it comes with a quality estimator. You can use Gramformer in one or more areas mentioned under the "use-cases" section below or any other use case as you see fit. Gramformer stands on the shoulders of giants, it combines some of the top-notch researches in grammar correction. *Note: It works at **sentence levels** and has been trained on 128 length sentences, so not (yet) suitable for long prose or paragraphs (stay tuned for upcoming releases)*
5