him1411 commited on
Commit
0e909a2
1 Parent(s): 352e2f7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +80 -0
README.md CHANGED
@@ -1,3 +1,83 @@
1
  ---
2
  license: apache-2.0
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
+ datasets:
4
+ - him1411/EDGAR10-Q
5
+ language:
6
+ - en
7
+ metrics:
8
+ - rouge
9
  ---
10
+ license: mit
11
+ language:
12
+ - en
13
+ tags:
14
+ - finance
15
+ - ContextNER
16
+ - language models
17
+ datasets:
18
+ - him1411/EDGAR10-Q
19
+ metrics:
20
+ - rouge
21
+ ---
22
+
23
+ EDGAR-Tk-Instruct-Large
24
+ =============
25
+
26
+ T5 Large model finetuned on [EDGAR10-Q dataset](https://huggingface.co/datasets/him1411/EDGAR10-Q)
27
+
28
+ You may want to check out
29
+ * Our paper: [CONTEXT-NER: Contextual Phrase Generation at Scale](https://arxiv.org/abs/2109.08079/)
30
+ * GitHub: [Click Here](https://github.com/him1411/edgar10q-dataset)
31
+
32
+
33
+
34
+ Direct Use
35
+ =============
36
+
37
+ It is possible to use this model to generate text, which is useful for experimentation and understanding its capabilities. **It should not be directly used for production or work that may directly impact people.**
38
+
39
+ How to Use
40
+ =============
41
+
42
+ You can very easily load the models with Transformers, instead of downloading them manually. The [Tk-Instruct-Large model](https://huggingface.co/allenai/tk-instruct-large-def-pos) is the backbone of our model. Here is how to use the model in PyTorch:
43
+
44
+ ```python
45
+ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
46
+ tokenizer = AutoTokenizer.from_pretrained("him1411/EDGAR-Tk-Instruct-Large")
47
+ model = AutoModelForSeq2SeqLM.from_pretrained("him1411/EDGAR-Tk-Instruct-Large")
48
+ ```
49
+ Or just clone the model repo
50
+ ```
51
+ git lfs install
52
+ git clone https://huggingface.co/him1411/EDGAR-Tk-Instruct-Large
53
+ ```
54
+
55
+ Inference Example
56
+ =============
57
+
58
+ Here, we provide an example for the "ContextNER" task. Below is an example of one instance.
59
+
60
+ ```python
61
+ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
62
+ tokenizer = AutoTokenizer.from_pretrained("him1411/EDGAR-Tk-Instruct-Large")
63
+ model = AutoModelForSeq2SeqLM.from_pretrained("him1411/EDGAR-Tk-Instruct-Large")
64
+ # Input shows how we have appended instruction from our file for HoC dataset with instance.
65
+ input = "14.5 years . The definite lived intangible assets related to the contracts and trade names had estimated weighted average useful lives of 5.9 years and 14.5 years, respectively, at acquisition."
66
+ tokenized_input= tokenizer(input)
67
+ # Ideal output for this input is 'Definite lived intangible assets weighted average remaining useful life'
68
+ output = model(tokenized_input)
69
+ ```
70
+
71
+
72
+ BibTeX Entry and Citation Info
73
+ ===============
74
+ If you are using our model, please cite our paper:
75
+
76
+ ```bibtex
77
+ @article{gupta2021context,
78
+ title={Context-NER: Contextual Phrase Generation at Scale},
79
+ author={Gupta, Himanshu and Verma, Shreyas and Kumar, Tarun and Mishra, Swaroop and Agrawal, Tamanna and Badugu, Amogh and Bhatt, Himanshu Sharad},
80
+ journal={arXiv preprint arXiv:2109.08079},
81
+ year={2021}
82
+ }
83
+ ```