Dear @CatalinaAlbornoz , @daniela.murcillo ,
I highly appreciate your and your team’s help! Thank you a lot!
1 Like
Hi @a.rozgonyi96 ,
I wanted to let you know that I’ve tested your original problem with the branch that David created to fix this problem (PR 7345) and it worked well, no errors! You can install this branch with
pip install git+https://github.com/PennyLaneAI/pennylane.git@fix-class-proc-trainabilirt
It will be soon merged into Master so once it’s in you can use the master branch instead.
I’ve made a minimal working example for the problem you encountered with ApproxTimeEvolution. I’ll add it to Forum topic 8118 which is more related to this problem.
1 Like
Dear @CatalinaAlbornoz ,
Do I understand correctly, the fix proposed in the PR for the metric tensor doesn’t fix the root of the ancestors related error? So when I apply ApproxTimeEvolution with multiple trotter steps and calculating metric tensor, this doesn’t work for you either while using the branch of PR?
I really appreciate your effort and attention to my problem. Thanks.
That’s correct @a.rozgonyi96 .
The PR fixes the shape mismatch issue that you had reported. Unfortunately, as it stands now, the PR doesn’t fix the problem where ApproxTimeEvolution with multiple trotter steps and calculating metric tensor causes the ancestors error. This is why I want to open a new issue for this.
1 Like