Commit
·
bffb91a
1
Parent(s):
ca3ceeb
README.md
CHANGED
@@ -1,59 +1,4 @@
|
|
1 |
-
---
|
2 |
-
language: en
|
3 |
-
license: mit
|
4 |
-
pipeline_tag: document-question-answering
|
5 |
-
tags:
|
6 |
-
- layoutlm
|
7 |
-
- document-question-answering
|
8 |
-
- pdf
|
9 |
-
widget:
|
10 |
-
- text: "What is the invoice number?"
|
11 |
-
src: "https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png"
|
12 |
-
- text: "What is the purchase amount?"
|
13 |
-
src: "https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/contract.jpeg"
|
14 |
-
---
|
15 |
-
|
16 |
-
# LayoutLM for Visual Question Answering
|
17 |
-
|
18 |
-
This is a fine-tuned version of the multi-modal [LayoutLM](https://aka.ms/layoutlm) model for the task of question answering on documents. It has been fine-tuned using both the [SQuAD2.0](https://huggingface.co/datasets/squad_v2) and [DocVQA](https://www.docvqa.org/) datasets.
|
19 |
-
|
20 |
-
## Getting started with the model
|
21 |
-
|
22 |
-
To run these examples, you must have [PIL](https://pillow.readthedocs.io/en/stable/installation.html), [pytesseract](https://pypi.org/project/pytesseract/), and [PyTorch](https://pytorch.org/get-started/locally/) installed in addition to [transformers](https://huggingface.co/docs/transformers/index).
|
23 |
-
|
24 |
-
```python
|
25 |
-
from transformers import pipeline
|
26 |
-
|
27 |
-
nlp = pipeline(
|
28 |
-
"document-question-answering",
|
29 |
-
model="impira/layoutlm-document-qa",
|
30 |
-
)
|
31 |
-
|
32 |
-
nlp(
|
33 |
-
"https://templates.invoicehome.com/invoice-template-us-neat-750px.png",
|
34 |
-
"What is the invoice number?"
|
35 |
-
)
|
36 |
-
# {'score': 0.9943977, 'answer': 'us-001', 'start': 15, 'end': 15}
|
37 |
-
|
38 |
-
nlp(
|
39 |
-
"https://miro.medium.com/max/787/1*iECQRIiOGTmEFLdWkVIH2g.jpeg",
|
40 |
-
"What is the purchase amount?"
|
41 |
-
)
|
42 |
-
# {'score': 0.9912159, 'answer': '$1,000,000,000', 'start': 97, 'end': 97}
|
43 |
-
|
44 |
-
nlp(
|
45 |
-
"https://www.accountingcoach.com/wp-content/uploads/2013/10/[email protected]",
|
46 |
-
"What are the 2020 net sales?"
|
47 |
-
)
|
48 |
-
# {'score': 0.59147286, 'answer': '$ 3,750', 'start': 19, 'end': 20}
|
49 |
-
```
|
50 |
-
|
51 |
-
**NOTE**: This model and pipeline was recently landed in transformers via [PR #18407](https://github.com/huggingface/transformers/pull/18407) and [PR #18414](https://github.com/huggingface/transformers/pull/18414), so you'll need to use a recent version of transformers, for example:
|
52 |
-
|
53 |
-
```bash
|
54 |
-
pip install git+https://github.com/huggingface/transformers.git@2ef774211733f0acf8d3415f9284c49ef219e991
|
55 |
-
```
|
56 |
|
57 |
## About us
|
58 |
|
59 |
-
This model was created by the team at
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
|
2 |
## About us
|
3 |
|
4 |
+
This model was created by the team at (DarwinBox|Vinay Bhushan|Rushikesh|Koushik).
|