https://github.com/unstructured-io/irs-manual-demo

https://github.com/unstructured-io/irs-manual-demo

Last synced: 4 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/unstructured-io/irs-manual-demo
• Owner: Unstructured-IO
• Created: 2023-04-09T21:32:13.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2023-06-09T19:17:25.000Z (about 3 years ago)
• Last Synced: 2025-06-14T14:04:41.837Z (about 1 year ago)
• Language: Python
• Size: 63.8 MB
• Stars: 15
• Watchers: 5
• Forks: 7
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
## Chat with IRS Manuals

This directory contains an application for chatting with IRS manuals. Once data is available, the chat application only uses self-hosted models and can be run in a disconnected environment. Here's how to get started with the chatbot:

### Installation

```bash

pip install -r requirements.txt

```

### Environment Variables

**Note there are other options for these connections, but these are the ones referenced in this implementation**

[OpenAI](https://platform.openai.com/docs/api-reference)

[Pinecone](https://docs.pinecone.io/)

```python

PINECONE_API_KEY

PINECONE_API_ENV

OPENAI_API_KEY

PINECONE_INDEX_NAME

```

### Download PDFs from IRS website

```bash

python download_data.py    

```

![Download](./gifs/down.gif)

### Run PDFs against unstructured-ingest

```bash

PYTHONPATH=. ./unstructured/ingest/main.py \

  --local-input-path  \

  --structured-output-dir  \

  # optional parameter -> this will hit the *NEW* API vs. processing locally

  --partition-by-api

  ```

![Download gif](./gifs/process.gif)

Here's an example of the structured json output

![JSON](./gifs/sbys.gif)

### Seed and utilize vector db

```bash

python ingest_data.py 

```

### Run the chat CLI

```bash

python cli_app.py

```

![Chat](./gifs/chat.gif)

### Chat with our hosted instance [here](https://huggingface.co/spaces/unstructuredio/irs-manuals)