NumPy Backend Example

NumPy Backend Example#

This example demonstrates compiling LaTeX-defined models to dense NumPy matrices by reusing the QuTiP backend and converting Qobj objects to arrays. It is useful for small systems and quick prototyping.

What you’ll learn#

  • How to request the NumPy backend and inspect the resulting ndarray.

  • How time-dependent terms are represented for NumPy backends.

Source#

 1# flake8: noqa
 2"""
 3Using the NumPy backend to build a dense matrix instead of QuTiP objects.
 4
 5What this shows:
 6- Selecting `backend="numpy"` to bypass QuTiP.
 7- Providing a `config` explicitly via `make_config`.
 8- Result is a plain NumPy array; useful for custom workflows or JIT-wrapping.
 9"""
10
11from __future__ import annotations
12
13import sys
14from pathlib import Path
15
16ROOT = Path(__file__).resolve().parents[1]
17if str(ROOT) not in sys.path:
18    sys.path.insert(0, str(ROOT))
19
20from latex_parser.latex_api import compile_latex_model, make_config
21
22
23def main() -> None:
24    cfg = make_config(qubits=[1], bosons=[])
25    H_latex = r"\delta \sigma_{x,1}"
26    params = {"delta": 0.4}
27    model = compile_latex_model(
28        H_latex=H_latex,
29        params=params,
30        backend="numpy",
31        config=cfg,
32    )
33    print("NumPy backend matrix:\n", model)
34
35
36if __name__ == "__main__":
37    main()

Run#

python examples/example_numpy_backend.py

Notes#

  • The example converts QuTiP objects to NumPy arrays; it is intended for clarity and education rather than performance.

Contents