https://github.com/unstructured-io/pipeline-receipts
Preprocessing pipeline notebooks and API supporting text extraction from receipts images
https://github.com/unstructured-io/pipeline-receipts
Last synced: about 1 year ago
JSON representation
Preprocessing pipeline notebooks and API supporting text extraction from receipts images
- Host: GitHub
- URL: https://github.com/unstructured-io/pipeline-receipts
- Owner: Unstructured-IO
- License: apache-2.0
- Created: 2023-03-10T10:15:15.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2023-06-20T22:07:10.000Z (about 3 years ago)
- Last Synced: 2025-02-15T20:56:36.371Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 1.39 MB
- Stars: 2
- Watchers: 3
- Forks: 2
- Open Issues: 6
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE
Awesome Lists containing this project
README
Pre-Processing Pipeline for Receipts
This repo implements a document pre-processing pipeline for receipts. Currently, the pipeline is under development. The pipeline assumes the receipts are in PDF or image formats (JPG, PNG).
The API is hosted at `https://api.unstructured.io`.
## :coffee: Getting Started
* Using `pyenv` to manage virtualenv's is recommended
* Mac install instructions:
* `brew install pyenv-virtualenv`
* `pyenv install 3.8.15`
Create a virtualenv to work in and activate it, e.g. for one named `receipts`:
`pyenv virtualenv 3.8.15 receipts`
`pyenv activate receipts`
* Run `make install`
* Start a local jupyter notebook server with `make run-jupyter`
**OR**
just start the fast-API locally with `make run-web-app`
#### Extracting Structured Text from an Receipt Image
After API starts, you can extract the elements of Receipt files with the command:
```
curl -X 'POST' \
'http://localhost:8000/receipts/v0.1.0/receipts' \
-F 'files=@' \
| jq -C . | less -R
```
### Generating Python files from the pipeline notebooks
You can generate the FastAPI APIs from your pipeline notebooks by running `make generate-api`.
## :guardsman: Security Policy
See our [security policy](https://github.com/Unstructured-IO/pipeline-receipts/security/policy) for
information on how to report security vulnerabilities.
## 🤗 Hugging Face
[Hugging Face Spaces](https://huggingface.co/spaces) offer a simple way to host ML demo apps, models and datasets directly on our organization’s profile. This allows us to showcase our projects and work collaboratively with other people in the ML ecosystem. Visit our space [here](https://huggingface.co/unstructuredio)!
## Learn more
| Section | Description |
|-|-|
| [Company Website](https://unstructured.io) | Unstructured.io product and company info |
| [Fine-tuned Models and Data](https://huggingface.co/naver-clova-ix) | CORD Consolidated Receipt dataset and Donut model |