Skip to content

A Low latency search engine for products with elasticsearch backend and USE embeddings.Entire Application is running on docker containers deployed on AWS instances and search results were given by Elasticsearch, Images are retrieved from S3.

Notifications You must be signed in to change notification settings

sunilbelde/apparel-search-engine-Elasticsearch

Repository files navigation

apparel-search-engine-Elasticsearch

A Simple search engine for products.

Why Elasticsearch?

Elasticsearch is widely used in many Companies because of it's simplicity and speed in searching. It stores the data in the form of documents(files) in a Index (similar to table in SQL). Searching is done with the method of Inverted Indices which makes elasticsearch fast.

Dataset:

Dataset can be obtained from here.

Docker

Note : keep the docker installed and running before the next steps.

Steps to set Elasticsearch container

-docker pull docker.elastic.co/elasticsearch/elasticsearch:7.11.0

-docker run --name CONTAINER_NAME -d -p 9200:9200 IMAGE_NAME

Steps to run the Flask application.

-Execute the commands where the Dockerfile is present.

-docker build -t IMAGE_NAME .

-docker run --name CONTAINER_NAME -d -p 5000:5000 IMAGE_NAEME

Run the command to check the containers are up and running

docker ps

Go to the browser and hit http://localhost:5000 you will see the home page of the application and search for the products Results :

plot

Architecture of the environment:

plot

Tools Used:

plot

About

A Low latency search engine for products with elasticsearch backend and USE embeddings.Entire Application is running on docker containers deployed on AWS instances and search results were given by Elasticsearch, Images are retrieved from S3.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published