amazon-braket / amazon-braket-containers Goto Github PK

Describe the feature you'd like
An issue with v0.23 pennylane lightning.gpu on QAOA use cases has been detected (PennyLaneAI/pennylane#2745). The bug will be addressed in the incoming lightning.gpu v0.24 release on June 21. As QAOA is an important use case for Braket, do Braket consider it necessary to update the containers correspondingly?

How would this feature be used? Please describe.
With the proposed upgrade, users will be able to run QAOA simulations with lightning.gpu embedded simulator with out-of-the-box containers.

Describe alternatives you've considered
Need to confirm if os.system("pip install git+https://github.com/PennyLaneAI/pennylane-lightning-gpu.git@main") in the job script can serve as a workaround.

Additional context
Add any other context or screenshots about the feature request here.

Describe the bug
We are not able to create successful job when using SV1 device for PennyLane when using Base Containers and were running into import issues like: ImportError: cannot import name 'ObservableReturnTypes' from 'pennylane.operation'

To Reproduce
Steps to reproduce the behavior:

• Go to Using_PennyLane_with_Braket_Jobs.ipynb in braket-examples, locally or by spinning up the Braket Notebook Instance.
• Instead of using the image_uri = Framework.PL_TENSORFLOW or Framework.PL_PYTORCH, use Framework.BASE` to as image_uri.

Expected behavior
Check the CloudWatch logs in the console account for the submitted job, as the job would fail with described bug.

Screenshots
N/A

System information (please complete the following information as applicable):

• Package Versions: amazon-braket-containers==1.6.5 and containers for base image using PennyLane==0.23
• Working Environment: [e.g. Local, Braket Notebook Instances] : All
• Browser [e.g. chrome, safari]: N/A
• Region [e.g. us-west-2]: All regions.

Known workaround (if applicable):

1. Submit the pennylane jobs using tensorflow or pytorch image uris.
2. Create custom base image_uri:
   a. Checkout this package and add pennylane dependency as pennylane==0.22.0 here: https://github.com/aws/amazon-braket-containers/blob/main/base/jobs/docker/1.0/py3/Dockerfile.cpu#L123
   b. Create a custom base image_uri by submitting the container build in your aws_account by following these steps: https://github.com/aws/amazon-braket-containers/blob/main/README.md
   c. Use the created custom image_uri from your ECR repository to run the pennylane job with base containers.

Additional context
Add any other context about the problem here.