Installation Guide

1. Make sure python 3.7 installed. Run sudo apt install python3.7 (ubuntu os)
2. Clone this repo git clone https://github.com/DoraMatching/DoraMatching-API-ML.git
3. cd DoraMatching-API-ML
4. Install virtualenv: pip3 install --user virtualenv
5. Set env python37: virtualenv --python=/usr/bin/python3.7 python37
6. Active python37: source python37/bin/activate
7. Install dependencies:

pip3 install flask flask_cors requests nltk bs4 pandas matplotlib regex scipy sklearn underthesea

8. Copy Vietnamese stopwords to corpora stopwords folder

cp ./vietnamese_dash $HOME/nltk_data/corpora/stopwords/vietnamese_dash

9. Run the ML API python3 app.py
10. Test

For Raspbian OS

# Install dependencies
sudo apt-get install libatlas-base-dev

For Vagrant box

Recommend: cpus>=1, memory>=1GB

# Vagranfile
# -*- mode: ruby -*-
# vi: set ft=ruby :

Vagrant.configure("2") do |config|
    config.vm.box = "tranphuquy19/doramatching-api-ml"
    config.vm.box_version = "1.0.0"
    config.vm.network "private_network", ip: "172.16.10.107"
    config.vm.hostname = "ml.dora"
  
    config.vm.provider "virtualbox" do |vb|
        vb.name = "ml.dora"
        vb.cpus = 2
        vb.memory = "2048"
    end
end

Init VM: vagrant up then vagrant ssh

Start API Server:

cd DoraMatching-API-ML/
source python38/bin/activate # active env
python3 app.py # start server