Predict with PHP
1. Install.
composer require modelfox/modelfox
2. Predict.
First, import the modelfox library and load the model file. Then, make an associative array with info for a new patient that matches the CSV, excluding the diagnosis column. Finally, call predict and print out the result.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
<?php
namespace modelfox\modelfox;
require_once(dirname(dirname(__FILE__)) . '/vendor/autoload.php');
$model_path = dirname(dirname(__FILE__)) . '/heart_disease.modelfox';
$model = Model::from_path($model_path);
// Create an example input matching the schema of the CSV file the model was trained on.
// Here the data is just hard-coded, but in your application you will probably get this from a database or user input.
$input = [
'age' => 63.0,
'gender' => 'male',
'chest_pain' => 'typical angina',
'resting_blood_pressure' => 145.0,
'cholesterol' => 233.0,
'fasting_blood_sugar_greater_than_120' => 'true',
'resting_ecg_result' => 'probable or definite left ventricular hypertrophy',
'exercise_max_heart_rate' => 150.0,
'exercise_induced_angina' => 'no',
'exercise_st_depression' => 2.3,
'exercise_st_slope' => 'downsloping',
'fluoroscopy_vessels_colored' => '0',
'thallium_stress_test' => 'fixed defect'
];
$output = $model->predict($input);
var_dump($output);