https://github.com/catalystcode/project-estrella

Parametrized ML inference
https://github.com/catalystcode/project-estrella

Last synced: 10 months ago
JSON representation

Parametrized ML inference

• Host: GitHub
• URL: https://github.com/catalystcode/project-estrella
• Owner: CatalystCode
• License: mit
• Created: 2017-06-20T06:58:21.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2017-06-27T13:36:44.000Z (over 8 years ago)
• Last Synced: 2025-01-22T15:48:01.399Z (11 months ago)
• Size: 14.6 KB
• Stars: 1
• Watchers: 7
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# project-estrella

Project-Estrella goal is to operationalize the execution of coefficient based calculations. 

# cloning the repo

To automatically deploy the solution and to benefit from continuous delivery, you need to fork the repo. To work with the repos, you also need to fetch the submodules:

```

git clone

git submodule init 

git submodule update --init --remote

git submodule foreach git checkout master

git submodule foreach git pull origin

```

# deploying the resources

[![Deploy to Azure](http://azuredeploy.net/deploybutton.png)](https://azuredeploy.net/)

or deploy 

```bash

azure group create -n  -l "West Europe"

azure group deployment create -f "azuredeploy.json" -e "azuredeploy.parameters.json" -g  -n 

```

## integrate with Fortis

To integrate the Estrella predictions into Fortis, one simply adds a prediction message to the `locationinferenceinput` Azure queue.

Fortis input message format:

```js

{

    'source': 'ewars-messages',

    'created_at': moment().toISOString(),

    'message' : {

        'message' : "",

        'title' : "",

        'link' : "",

        'user_id' : ""

        'lang' :"",

        'id' : "",

        'originalSources' : ["EWARS"],

        'geo' : {

            lat/lon or lon/lat

        }

    }

}

```