catalyst
catalyst copied to clipboard
PennyLaneAI
A JIT compiler for hybrid quantum programs in PennyLane
Running: ```python @pytest.mark.parametrize("arg", [jnp.array([[0.1, 0.2], [0.3, 0.4]])]) @pytest.mark.parametrize("order", ["good", "bad"]) def test_vjp_as_residual(arg, order): """See https://github.com/PennyLaneAI/catalyst/issues/852""" def jax_callback(fn, result_type): @pure_callback def callback_fn(*args) -> result_type: return fn(*args) @callback_fn.fwd def callback_fn_fwd(*args): ans, vjp_func...
**Context:** Lightning-Kokkos tests may oversubscribe a runner. **Description of the Change:** Limit the number of OpenMP threads for Lightning-Kokkos tests. **Benefits:** **Possible Drawbacks:** **Related GitHub Issues:**
There are several places where we modify JAX during compilation. Just to list some: ```python # Required for JAX tracer objects as PennyLane wires. # pylint: disable=unnecessary-lambda setattr(jax.interpreters.partial_eval.DynamicJaxprTracer, "__hash__", lambda...
The following program raises an error: ```py import jax.numpy as jnp from catalyst import * @qjit def f(n: int, m: int): x = jnp.ones((n, m), dtype=float) y = jnp.ones((n, m),...
It turns out that there are several bugs involved in attempting to reshape dynamic shaped arrays. In particular: - `jnp.reshape` of a dynamically-shaped array with new compile-time shape results in...
## Context Thread-Level Speculation is a technique that has been used in various research to speed up general purpose programs by speculatively executing code downstream of a function call. The...
When using float32 as a result type in callbacks the program crashes. I think this is because at some point we assume float64 will be used but float32 was used....
#### Issue description Using ``@qjit(autograph=True)`` decorator to loop over Pytree types that are not ``jax.numpy`` (like python builtin list, tuple, and numpy array) fails. No conversion will be applied. Converting...
```python dev = qml.device("lightning.qubit", wires=5, shots=20) @qml.qjit @qml.qnode(dev, mcm_method="one-shot", postselect_mode="fill-shots") def one_shot(x): qml.RX(x, wires=0) m_0 = catalyst.measure(0, postselect=1) return qml.sample(wires=0) @qml.qjit @qml.qnode(dev, mcm_method="single-branch-statistics", postselect_mode="fill-shots") def sbs(x): qml.RX(x, wires=0) m_0 =...
```python dev = qml.device("lightning.qubit", wires=1, shots=20) @qml.qjit @qml.qnode(dev, mcm_method="one-shot") def func(x): qml.RX(x, wires=0) m_0 = catalyst.measure(0, postselect=1) return {"hi": qml.expval(qml.Z(0))} ``` ```pycon >>> func(0.9) array(1.) ```
Metadata
Owner
Metadata
A JIT compiler for hybrid quantum programs in PennyLane