Mini Mori

Simple and Powerful image engine search, that know NLP and use Vector search for find similar results.

Demo

Features

• vector search
• keyword search
• dockerized
• user-friendly interface

BackEnd

In backend use Fastapi framework, meilisearch, qdrant, pytest

first set DataBase and virtualenv

Set Qdrant database:

docker pull qdrant/qdrant docker run -p 6333:6333 qdrant/qdrant

Set Meilisearch:

docker run -d --rm
--name meili-search
-p 7700:7700
-e MEILI_ENV='development'
-v $(pwd)/meili_data:/meili_data
getmeili/meilisearch

Now Set virtualenv:

python -m venv venv_name

active it in Linux:

source venv_name/bin/activate

Run Locally

Clone the project

  git clone https://github.com/sina-mobarez/mini-Mori.git

Go to the project directory

  cd mini-Mori

Install dependencies

  cd backend
  pip install requirements.txt

Fill Qdrant Database (exist a json file that contain 10000 products information by this command process images of products and store in vectordb):

  python scripts/process_images.py products.json batch-size

Fill Meilisearch Database [exist a json file that contain 10000 products information by this command indexing all products in database]:

  python scripts/load_products_to_meilisearch.py products.json

Start the server

  fastapi run

• Now you can use swagger or postman to call apis

FrontEnd

User interface use Vue.js for run it:

  cd ../frontend

Install dependencies

  npm install

Start the server

  npm run dev

• [just for production] Compile and Minify for Production

  npm run build

Running Tests

To run tests, run the following command

  pytest /tests

make sure you are in /backend.

Contributing

Contributions are always welcome!

Feedback

If you have any feedback, please reach out to us at [email protected]

License

MIT