svr_sklearn 0.0.4 svr_sklearn: ^0.0.4 copied to clipboard
Implementation of sklearn svr model to Flutter using an pre-trained model .
Support machine Vector For Regression (SVR)-Sklearn-dart #
- Implementation of sklearn svr model to Flutter
Introduction #
This library uses pre-trained models using sklearn-learn and svr-tojson for python.You have to build it and export it in the asset directory of your flutter application.
Steps to follows #
- You create your SVR model in python using sklearn package,once done you use svr-tojson package to create your json file who will contains all your attributs of your model.
- You install this package by doing this:
dependencies:
svr_sklearn:0.0.3
-You Export your json file(model) into the asset like this:
flutter:
uses-material-design: true
assets:
- assets/file.json
- Import the library in your application and load the model:
import 'package:flutter/material.dart';
import 'package:svr_sklearn/svr_sklearn.dart';
import 'dart:convert';
void main() => runApp(new MaterialApp(
home: new HomePage(),
debugShowCheckedModeBanner: false)
);
class HomePage extends StatefulWidget {
@override
_HomePageState createState() {
return new _HomePageState();
}
}
class _HomePageState extends State<HomePage> {
SVR svr;
_HomePageState() {
loadModel("assets/svrmodel.json").then((x) {
this.svr = SVR.fromMap(json.decode(x));
});
}
@override
Widget build(BuildContext context) {
return Scaffold(
// add any widget with svr.predict() callback
);
}