Label Propagation Algorithm by Rust.
Label propagation (LP) is graph-based semi-supervised learning (SSL).
A simple LGC and a more advanced CAMLP have been implemented.
You can find the examples in the examples directory.
The label is a continuous value of [0, class_n]
, and the result of predict_proba
is the value of the label.
```rust use std::result::Result; use std::error::Error;
extern crate labelpropagation; extern crate ndarray; extern crate ndarraystats;
use ndarray::prelude::*; use label_propagation::{CAMLP, LGC}; use ndarray::Array;
pub fn main() -> Result<(), Box
// node index for label
let x = array![0, 1];
// label
let y = array![0, 1];
// make model
let mut model = CAMLP::new(graph).iter(30).beta(0.1);
// let mut model = LGC::new(graph).iter(30).alpha(0.99);
model.fit(&x, &y)?;
let target = array![0, 1];
let result = model.predict_proba(&target);
println!("{:?}", result);
Ok(())
} ```
sh
docker build -t graph .
docker run -it -v $PWD:/app graph bash