# Tract
Tiny, no-nonsense, self contained, portable SharedTensor and ONNX inference.
## Example
```
# extern crate tract_core;
# extern crate ndarray;
# fn main() {
use tract_core::*;
use tract_core::model::*;
use tract_core::model::dsl::*;
// build a simple model that just add 3 to each input component
let mut model = Model::default();
let input = model.add_source("input").unwrap();
let three = model.add_const("three".to_string(), 3f32.into()).unwrap();
let add = model.add_node("add".to_string(),
Box::new(tract_core::ops::math::Add::default())).unwrap();
model.add_edge(OutletId::new(input, 0), InletId::new(add, 0)).unwrap();
model.add_edge(OutletId::new(three, 0), InletId::new(add, 1)).unwrap();
// we build an execution plan. default input and output are inferred from
// the model graph
let plan = SimplePlan::new(&model).unwrap();
// run the computation.
let input = ndarray::arr1(&[1.0f32, 2.5, 5.0]);
let mut outputs = plan.run(tvec![input.into()]).unwrap();
// take the first and only output tensor
let mut tensor = outputs.pop().unwrap();
// unwrap it as array of f32
let tensor = tensor.to_array_view::().unwrap();
assert_eq!(tensor, ndarray::arr1(&[4.0, 5.5, 8.0]).into_dyn());
# }
```
While creating a model from Rust code is usefull for testing the library,
real-life use-cases will usually load a SharedTensor or ONNX model using
tract-tf or tract-onnx crates.