Hello PennyLane Community, I’m trying to compute the Quantum Fisher Information (QFI) matrix for specific parameters of the quantum circuit using the qml.gradients.quantum_fisher function. in PennyLane. My goal is to: 1: Prepare a Bell state. 2: Evolve it under the Hamiltonian: H = \theta_1 Z_1 …

Hi @a.rozgonyi96 , It look like part of the issue is that you’re mixing Jax and Numpy. I’ve modified your code to use only Numpy and it works. For this I’ve changed the coeffs from using jnp to pnp and I’ve removed the interface in the declaration of the qnode (this way the qnode will detect it auto…

Hi @a.rozgonyi96 , it looks like this might be related to an issue that was surfaced for Torch a while ago. Initially this didn’t seem to be an issue when using JAX but as you demonstrated the problem occurs with JAX too. My colleague David will take a look at this next week. We’ll let you know if …

Dear @CatalinaAlbornoz , I’m trying to adapt the single argument method for my master code, but I’m encountering a trainability issue when computing qml.gradient.quantum_fisher for a circuit with a few layers of parameters using JAX. I believe this is the same issue described and fixed here . Despit…

Hi @a.rozgonyi96 , I can replicate the error. It does look related to the issue that you referenced. Let me check with the team and get back to you on this.

Hi @a.rozgonyi96 , Actually I realized that you’re using Numpy instead of JAX for your parameters here, so even if you set interface=“jax” it’s not actually using JAX because your parameters are Numpy arrays. So the first thing I did was moving everything to actually using JAX, and and I removed t…

Thank you @CatalinaAlbornoz , your alternative works! My question would be, then, is the StronglyEntanglingLayers function not compatible with JAX?

Hi @a.rozgonyi96 , quantum_fisher + JAX + StronglyEntanglingLayers is indeed throwing an error. I’m not sure whether it’s an easy fix or a more fundamental problem. However there’s a workaround. You can see in the documentation for quantum_fisher that it coincides with the metric_tensor with a pre…

Hey @a.rozgonyi96 , I didn’t mention this yesterday but the reason why quantum_fisher fails but metric_tensor() works in this case is because internally, adjoint_metric_tensor() is used when executing on "default.qubit" with exact expectations (shots=None ). In all other cases, metric_tensor() is u…

Hi @CatalinaAlbornoz , Thank you very much! It does work indeed, however only on the test code, because when I run it on my main code, it throws the same error as the one I mentioned in a separate topic here . The problem arises when I set more than one Trotter step. I run into this error message: …

That’s good progress @a.rozgonyi96 ! Let me know how it goes with the suggested PR.

Dear @CatalinaAlbornoz , I apologise for the delayed response. I’ve been going through an intensive period since my PhD defence. I’ve now resumed my project that I asked for help with it. However, I still have the issue and don’t know how to solve it. The problem arises when my circuit contains St…

Shape mismatch issue when calculating QFI matrix

PennyLane Help

a.rozgonyi96 June 16, 2025, 7:20am 21

Dear @CatalinaAlbornoz , @daniela.murcillo ,

I highly appreciate your and your team’s help! Thank you a lot!

1 Like

Topic		Replies	Views
Issue with QFIM for time evolved mixed state PennyLane Help	25	301	July 16, 2025
Issue with calculating `qml.qinfo.transforms.quantum_fisher` PennyLane Help	6	337	February 14, 2024
Qfim metric tensor verify diagonal elements PennyLane Help	19	157	December 13, 2024
Error in Quantum Fisher information with noise PennyLane Help	6	430	February 20, 2024
The quantum fisher information matrix calculated by my own code is inconsistent with that by qml.metric_tensor PennyLane Help	2	1229	October 28, 2021

Shape mismatch issue when calculating QFI matrix

Related topics