Predict with Rust
1. Install.
[dependencies]
modelfox = { git = "https://github.com/modelfoxdotdev/modelfox" }
2. Predict.
First, import the modelfox library and load the model file. Then, make an object 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
fn main() {
// Load the model from the path.
let model: modelfox::Model =
modelfox::Model::from_path("heart_disease.modelfox", None).unwrap();
// 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.
let input = modelfox::predict_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": 0.0,
"thallium_stress_test": "fixed defect",
};
// Make the prediction!
let output = model.predict_one(input, None);
// Print the output.
println!("Output: {:?}", output);
}