• Semantic Search
  • Setup
    • Setup using Docker
      • 1. Clone Repository
      • 2. Configure
      • 3. Run
        • Troubleshooting
    • Setup on Local Machine
      • 1. Install Dependencies
      • 2. Install Semantic Search
      • 3. Configure
      • 4. Run
  • Use
  • Upgrade
    • On Docker
    • On Local Machine
  • Miscellaneous
  • Acknowledgments

Semantic Search

Allow natural language search on user content like notes, images, transactions using transformer based models

All search is done locally. User can interface with semantic-search app via Emacs, API or Commandline

Setup

Setup using Docker

1. Clone Repository

  git clone https://github.com/debanjum/semantic-search && cd semantic-search

2. Configure

• Add Content Directories for Semantic Search to Docker-Compose
• Update docker-compose.yml to mount your images, org-mode notes, ledger/beancount directories
• If required, edit config settings in docker_sample_config.yml.

3. Run

docker-compose up -d

Troubleshooting

• The first run will take time. Let it run, it's mostly not hung

• Symptom: Errors out with "Killed" in error message

  • Fix: Increase RAM available to Docker Containers in Docker Settings
  • Refer: StackOverflow Solution, Configure Resources on Docker for Mac

• Symptom: Errors out complaining about Tensors mismatch, null etc

  • Mitigation: Delete content-type > image section from docker_sample_config.yml

Setup on Local Machine

1. Install Dependencies

1. Install Python3 [Required[
2. Install Conda [Required]

3. Install Exiftool [Optional]

   sudo apt-get -y install libimage-exiftool-perl

2. Install Semantic Search

git clone https://github.com/debanjum/semantic-search && cd semantic-search
conda env create -f environment.yml
conda activate semantic-search

3. Configure

• Configure files/directories to search in content-type section of sample_config.yml

• To run application on test data, update file paths containing /data/ to tests/data/ in sample_config.yml

  • Example replace /data/notes/*.org with tests/data/notes/*.org

4. Run

Load ML model, generate embeddings and expose API to query notes, images, transactions etc specified in config YAML

python3 -m src.main -c=sample_config.yml -vv

Use

• Semantic Search via Emacs

  • Install semantic-search.el
  • Run M-x semantic-search <user-query>

• Semantic Search via API

  • Query: GET http://localhost:8000/search?q="What is the meaning of life"&t=notes
  • Regenerate Embeddings: GET http://localhost:8000/regenerate
  • Semantic Search API Docs

• UI to Edit Config

  • Config UI

Upgrade

On Docker

  docker-compose build

On Local Machine

  cd semantic-search
  git pull origin master
  conda deactivate semantic-search
  conda env update -f environment.yml
  conda activate semantic-search

Miscellaneous

• The experimental /chat API endpoint uses the OpenAI API

  • It is disabled by default
  • To use it add your openai-api-key to config.yml

Acknowledgments

• MiniLM Model for Asymmetric Text Search. See SBert Documentation
• OpenAI CLIP Model for Image Search. See SBert Documentation
• Charles Cave for OrgNode Parser
• Sven Marnach for PyExifTool