https://github.com/dbbrandt/mm-api-py

Python FLASK API for handling requests to machine learning ml-mm-api
https://github.com/dbbrandt/mm-api-py

Last synced: 3 months ago
JSON representation

Python FLASK API for handling requests to machine learning ml-mm-api

• Host: GitHub
• URL: https://github.com/dbbrandt/mm-api-py
• Owner: dbbrandt
• Created: 2019-05-11T21:26:22.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2019-05-12T20:21:37.000Z (about 6 years ago)
• Last Synced: 2025-01-03T19:22:17.402Z (5 months ago)
• Language: Python
• Homepage:
• Size: 2.93 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
This is a light wrapper for mlmmapi to support web requests to correct_answer.

Setup:

1. sudo to precidix

2. cd srv

3. git clone https://[gitusername]@github.com/dbbrandt/mm-api-py.git

    * Provided password at prompt

4. Install python 3.6

    * Sudo yum install 3.6

    * yum install python36-devel. (For buing gcc packages like python-Levenshtein

5. Install pip

    * python3 -m pip install --user --upgrade pip

6. Setup venv

    * python3 -m pip install --user virtualenv

7. Create a virtual environment

    * python3 -m venv env

8. Attache to the venv

    * source env/bin/activate

9. Install the needed packages

    * fuzzywuzzy

    * python-Levenshtein  

    * numpy

    * pandas

    * scikit-learn version 0.20.3. (for the model dump).   Pip3 install scikit-learn==0.20.3

    * xgboost

    * Joblib

    * flask

    * mlmmapi

    * uwsgi

10. Update nginx config

    * Edit /etc/nginx/conf.d/mm-api_prodution.conf

    * Add the lines under the location @puma_mm-api_production used for the normal mm-api rails app.

    

           location /mm-api-py {

             include uwsgi_params;

             uwsgi_pass unix:/tmp/mm-api-py.sock;

           }

           

11. Run the uwsgi application

    * Got the /users/precidix/srv/mm-api-py directory

    * Run:

    * uwsgi —ini wsgi.ini &

12. Test and Restart nginx

    * sudo /etc/init.d/nginx configtest

    * sudo /etc/init.d/nginx restart