sc_neurocore_engine/bindings/
adex_neuron.rs

1// SPDX-License-Identifier: AGPL-3.0-or-later
2// Commercial license available
3// © Concepts 1996–2026 Miroslav Šotek. All rights reserved.
4// © Code 2020–2026 Miroslav Šotek. All rights reserved.
5// ORCID: 0009-0009-3560-0851
6// Contact: www.anulum.li | protoscience@anulum.li
7// SC-NeuroCore — Adaptive exponential neuron PyO3 binding
8
9use numpy::{IntoPyArray, PyArray1};
10use pyo3::exceptions::PyFloatingPointError;
11use pyo3::prelude::*;
12use pyo3::types::PyDict;
13
14use crate::neuron;
15
16type CompleteTracePacket<'py> = (
17    Bound<'py, PyArray1<f64>>,
18    Bound<'py, PyArray1<f64>>,
19    Bound<'py, PyArray1<u8>>,
20    f64,
21    f64,
22);
23
24#[pyclass(
25    name = "AdExNeuron",
26    module = "sc_neurocore_engine.sc_neurocore_engine"
27)]
28#[derive(Clone)]
29pub struct PyAdExNeuron {
30    inner: neuron::AdExNeuron,
31}
32
33#[pymethods]
34impl PyAdExNeuron {
35    #[new]
36    fn new() -> Self {
37        Self {
38            inner: neuron::AdExNeuron::new(),
39        }
40    }
41    fn step(&mut self, current: f64) -> i32 {
42        self.inner.step(current)
43    }
44    fn reset(&mut self) {
45        self.inner.reset();
46    }
47    fn get_state(&self, py: Python<'_>) -> PyResult<Py<PyAny>> {
48        let d = PyDict::new(py);
49        d.set_item("v", self.inner.v)?;
50        d.set_item("w", self.inner.w)?;
51        Ok(d.into_any().unbind())
52    }
53}
54
55/// Register the adaptive exponential neuron class.
56pub(crate) fn register(module: &Bound<'_, PyModule>) -> PyResult<()> {
57    module.add_class::<PyAdExNeuron>()?;
58    module.add_function(wrap_pyfunction!(adex_simulate_complete, module)?)?;
59    Ok(())
60}
61
62/// Run the checked full-parameter AdEx batch and return every state/event row.
63#[pyfunction]
64#[pyo3(signature = (
65    v, w, v_rest, v_reset, v_threshold, v_rh, delta_t, tau, tau_w,
66    a, b, c_m, dt, n_steps, current
67))]
68#[allow(clippy::too_many_arguments)]
69fn adex_simulate_complete<'py>(
70    py: Python<'py>,
71    v: f64,
72    w: f64,
73    v_rest: f64,
74    v_reset: f64,
75    v_threshold: f64,
76    v_rh: f64,
77    delta_t: f64,
78    tau: f64,
79    tau_w: f64,
80    a: f64,
81    b: f64,
82    c_m: f64,
83    dt: f64,
84    n_steps: usize,
85    current: f64,
86) -> PyResult<CompleteTracePacket<'py>> {
87    let mut model = neuron::AdExNeuron {
88        v,
89        w,
90        v_rest,
91        v_reset,
92        v_threshold,
93        v_rh,
94        delta_t,
95        tau,
96        tau_w,
97        a,
98        b,
99        c_m,
100        dt,
101    };
102    let (v_trace, w_trace, event_trace) = model
103        .simulate_complete(n_steps, current)
104        .map_err(PyFloatingPointError::new_err)?;
105    Ok((
106        v_trace.into_pyarray(py),
107        w_trace.into_pyarray(py),
108        event_trace.into_pyarray(py),
109        model.v,
110        model.w,
111    ))
112}
sc_neurocore_engine/bindings/adex_neuron.rs

sc_neurocore_engine/bindings/
adex_neuron.rs
