Lensity

Finding connection with a human lens

Team

• Product Owner: Josphine Eng
• Scrum Master: Julie Truong
• Development Team Members: Brian Kilrain

Table of Contents

1. Micro-Services
2. Requirements
3. Usage
4. Docker Development
   1. Build Image
   2. Launch Docker Container
   3. Train Corpus
5. Development
   1. Installing Dependencies
   2. Compile React
   3. Seed Database Data
   4. Train Data
6. Team
7. Contributing

Micro-Services

Photo-Processor

Curator

Similarity Server

Requirements

• Node 0.10.x
• Postgresql 6.1.x
• Webpack 1.13.x

Usage

• POST /upload Request to upload a photo

• GET /photos Request to get all of users photos

• GET /stacks Request to get stacks from curator micro-service

• POST /createPair Create a pair in the postgreSQL database

• GET /getPairs Request to get the 5 most recently created pairs

• GET /getRandStacks Request to curator to grab the 6 most recently created stacks for community view

Docker Development

Build Image

In root folder run:

docker build -t app .

Launch Docker Container

docker-compose up

Train Corpus

• Use Postman to send POST request to http://0.0.0.0:5000/train

Development

Installing Dependencies

From within the root directory:

npm install

Compile React

From within the root directory run webpack:

webpack -watch

Seed Database Data

From db/seeds directory:

Make sure postgres is running Drop database and create empty database

drop database app;
// DROP DATABASE
create database app;
// CREATE DATABASE

Restart main server

Run seed.js using node

run node seeds.js

email: [email protected] pw: 123

• If pushing to redis server, uncomment out the redis line. Note* this has not been tested.

Train Data

• Upload training photos to S3

• naming convention: http://url.com/path/to/photo/im1342.jpg... im1343.jpg... im1344.jpg

• In curator/routes.js -> set training counter to first image file number (line 16)
• Uncomment volume to trainingCorpus.json in yml file
• Send POST request to main-web-server/kickoffTraining with postman
• Copy and paste data in trainingCorpus.json in curator to trainingCorpus.json in simserver

Contributing

See CONTRIBUTING.md for contribution guidelines.