Table of Contents

Installation Guide - Community Edition

This document applies only to SWIRL AI Connect, Community Edition.

SWIRL's start-up process no longer starts redis. You must now have redis installed and running before starting SWIRL. Refer to Install Redis for more information.

System Requirements

Platform - Ubuntu, RHEL, or OS/X; Microsfot Windows is not supported for local installation due to Celery support
8+ VCPU, 16+ GB of memory
At least 500 GB of free disk space
Python 3.11 or newer

MacOS

Python 3.12.x (or latest stable) with pip
- If necessary, modify the system PATH so that Python runs when you type python at the Terminal (not python3)
- venv (optional)
- pyenv (optional)
Homebrew installed and updated
Redis installed:
```
 brew install redis
```
jq installed:
```
brew install jq
```

Linux

Python 3.12.x (or latest stable) with pip
Redis and jq installed:
```
sudo apt install jq redis-server -y
```

PostgreSQL (optional)

If you wish to use PostgreSQL as a data source or as the SWIRL back-end database:

Install PostgreSQL
Modify the system PATH so that pg_config from the PostgreSQL distribution is accessible from the command line
Install psycopg2 using pip:
```
pip install psycopg2
```

Installing SWIRL AI Connect

Clone the repo:

git clone https://github.com/swirlai/swirl-search
cd swirl-search

To install SWIRL on MacOS, execute this command from the Console:

./install.sh

To install SWIRL on Linux, execute this command from the Console:

apt-get update --allow-insecure-repositories -y && apt-get install apt-file -y && apt-file update && apt-get install -y python3-dev build-essential
./install.sh

If there are problems running install.sh, proceed manually:

pip install -r requirements.txt
python -m spacy download en_core_web_lg
python -m nltk.downloader stopwords
python -m nltk.downloader punkt

Issues with certifications on OS/X? See: urllib and "SSL: CERTIFICATE_VERIFY_FAILED" Error

Setup SWIRL

Execute the following command from the Console to setup SWIRL:

python swirl.py setup

Setup RAG

To enable SWIRL's Real-Time Retrieval Augmented Generation (RAG) on your localhost, run the following commands from the Console before installing the Galaxy UI:
```
export MSAL_CB_PORT=8000
export MSAL_HOST=localhost
```

For more information, refer to the RAG Configuration guide.

Install the Galaxy UI

To install the Galaxy UI, you must have the latest Docker app for MacOS or Linux installed and running locally.

To install Galaxy, execute the following command the Console (with the Docker app running):

./install-ui.sh

The Galaxy UI components should be installed only after running the ./install.sh and python swirl.py setup commands.

Start SWIRL

Execute the following command from the Console to start SWIRL:

python swirl.py start

Open the SWIRL Homepage

Enter this URL into a browser: http://localhost:8000/swirl/

The following page should appear:

SWIRL Homepage

Open the Galaxy UI

Open this URL with a browser: http://localhost:8000 (or http://localhost:8000/galaxy/)

If the search page appears, click Log Out at the top, right. The SWIRL login page will appear:

SWIRL Login

Enter the username admin and password password, then click Login.
Enter a search in the search box and press the Search button. Ranked results appear in just a few seconds:

SWIRL Results

Click the Generate AI Insight button to RAG using the most relevant results, if you have specified an OpenAI key as noted earlier.

SWIRL Results with RAG

Click the profile avatar in the upper right corner of the Galaxy UI. Then click Manage SWIRL to explore the rest of SWIRL's features.
To view the raw result JSON, click Search under the API section of the Manage SWIRL page linked above, or open http://localhost:8000/swirl/search/

The most recent Search object will be displayed at the top. Click on the result_url link to view the full JSON Response. For example:

SWIRL JSON response

Read the SWIRL User Guide for additional information.