Hi, Is it possible to use backpropagation with default.qubit on Pytorch? Or only tensorflow? Also, does lightning.qubit support backpropagation? Thanks James

Hi @James_Ellis ! Is it possible to use backpropagation with default.qubit on Pytorch? Or only tensorflow? Currently, standard backpropagation is supported using both TensorFlow and the default Autograd: import pennylane as qml from pennylane import numpy as np dev = qml.device("default.qubit",…

Wow… using the transfer learning demo ( Quantum transfer learning — PennyLane ) I am seeing more than 3x improvement in training speed. In your opinion, do you think more speed up could be witnessed using backprop on Tensorflow vs adjoint on Pytorch? Thanks for your help! :slight_smile:

[image] James_Ellis: Wow… using the transfer learning demo ( Quantum transfer learning | PennyLane Demos ) I am seeing more than 3x improvement in training speed. That’s awesome to hear! What interface, device, and differentiation method are you using? In your opinion, do you think more speed…

Using the torch interface I tried: 1.) ‘lightning.qubit’ with finite-difference differentiation method (I assume this it the default?). Time: 20s 2.) ‘lightning.qubit’ with adjoint backpropagation. Time: 6s The code I used is exactly the same as the demo. I will try and convert the demo into Ten…

Oh and just for reference, ‘default.qubit’ with finite-difference had a time of 40s for that demo on my computer.

This is fantastic James! Excellent work! :fire: :rocket: The PennyLane dev team has put in a lot of work recently in improving speed and performance. It’s great to see actual numbers for user runtimes!

I have run a similar experiment on Tensorflow. I basically followed a tutorial ( انتقال یادگیری و تنظیم دقیق | TensorFlow Core ) because I am not familiar with TF. I modified it to use ResNet50. (For each I give the time for ‘lightning.qubit’ and ‘default.qubit’ respectively) Using ‘adjoint’ as t…

Hi @James_Ellis , Thanks for the results! It seems odd indeed, that you saw poorer performance using backprop than with the finite-diff with default qubit :thinking: Locally backprop seems to be faster, though this might mean that the difference arises when run on a GPU. As lightning.qubit doesn’t…

Hi @James_Ellis , Have two things to add after a chat with @josh : For a couple of parameters, using finite differences or the parameter shift rule can actually be more efficient than backpropagation. This is due to lower overheads with those techniques. Backpropagation performs particularly well f…

Backpropagation with Pytorch

PennyLane Help

Topic		Replies	Views
Backpropagation with lightning.qubit and PyTorch PennyLane Help	3	1890	March 15, 2022
Pytorch benchmarks, different devices, and computing resources PennyLane Help	1	461	March 15, 2022
Execution time very long. Options to speed up? PennyLane Plugins	9	2451	April 26, 2021
Adjoint and backpropagation differentiation methods give different results PennyLane Help	13	857	April 13, 2022
Differentiating quantum circuits using backprop PennyLane Help	3	166	June 14, 2024

Backpropagation with Pytorch

Related topics