Sprint plan - Githubissues

hotosm / ml-enabler

A service that integrates ML models to mapping applications.

https://www.hotosm.org/updates/the-machine-learning-enabler/

BSD 2-Clause "Simplified" License

40 stars 11 forks source link

Sprint plan #9

Open geohacker opened 5 years ago

geohacker commented 5 years ago

Overall project timeline is May 16 - July 8

Sprint 1910

May 16 - 17

Kickoff
Setup Flask boilerplate

Sprint 1911

May 20 - 31

First iteration of ml-enabler with API end points
Package up looking-glass Docker, and document the process

Sprint 1912

June 3 - 14

With the help of HOT deploy ml-enabler
Complete prediction APIs
Provide a command line interface for looking glass calculation pipeline.
Enable markdown for TM annotations API

Sprint 1913

June 17 - 28

Demonstrate looking-glass integration API
Test TM integration API
Demonstrate task splitting API usage
Implementation of a second extension to support a pre-processed data service API

Sprint 1914

July 1 - 8

Training
Documentation

pantierra commented 5 years ago

@geohacker Can we please include these two specific points:

Provide a command line interface for looking glass calculation pipeline.
Implementation of a second extension to support a pre-processed data service API (I will create an issue for that).