Korbinian Kottmann

@josh146 Is there a way to import `optax` for this demo? I am following a comment of yours when optimizing with jax from an older forum post: https://discuss.pennylane.ai/t/jax-and-pennylane-optimizers/1752/4

I'm a bit at a loss with the issue upon importing `jax` in the CI actions. Locally everything works fine. I also tried updating to `jax==3.1.14` in the requirements file,...

This hacky solution would allow for sensible gradients: ```python wires=range(2) dev = qml.device("default.qubit", wires=wires, shots=10000) H = qml.PauliZ(0)@qml.PauliZ(1) @qml.qnode(dev) def qnode(x): qml.RY(x, wires=0) return classical_shadow_expval(H) @qml.qnode(dev) def qnode0(x): qml.RY(x, wires=0)...

Using the vectorized `default.qubit` version gradient evaluation is, as expected, much faster: ```con >>> %timeit qml.grad(qnode)(x) 123 ms ± 2.04 ms per loop (mean ± std. dev. of 7 runs,...

I think the name could do some tuning, how about `shadow_expval` ?

Currently there is a big overlap between the class and module doc-string. I am leaning towards shrinking the class one to have everything concentrated in the module description. Thoughts? Other...

Reconstructing a Bell state seems to work: ```python wires = range(2) dev = qml.device("default.qubit", wires=wires, shots=1000) @qml.qnode(dev) def qnode(): qml.Hadamard(0) qml.CNOT(wires=[0, 1]) return qml.classical_shadow(wires=wires, n_snapshots=1000) bitstrings, recipes = qnode() shadow...

In the current version of this numpy based prototype you can now: * reconstruct local snapshots (reduced density matrices) * reconstruct global snapshots (density matrices) * compute expectation values of...

I think hitting a problem with dtypes such as ```python import tensorflow as tf qml.operation.enable_new_opmath() a = tf.constant(0.5) b = tf.constant(0.5j) x = qml.s_prod(a, X(0)) y = qml.s_prod(b, Y(0)) res...

Can I merge to master right now or should I wait for decoupling the release candidate branch?