Qblox Backend: Acquisition memory on the device is not cleared if running the same experiment twice

• Quantify-Scheduler version: cb7fbdae or later
• Python version: 3.10
• Python install method (anaconda/virtualenv/other?): conda
• Operating System: windows

Description

Describe what you were trying to get done.

What I Did

Run the same schedule twice/Run the same compiled schedule twice

What Was Expected

The results should be independent, as I want to process the results externally (with quantify or something else)

What Actually Happened

We get the average between both experiments

Proposed Solution

Add a flag to QRMComponent.prepare which clears acquisition memory. Default should be true.

Background

This behavior is old. But as the json files generated by QbloxBaseModule.compile() have random names, the cache never hit and the program was always reuploaded by prepare(), clearing acquisition memory. Since cb7fbdae, we no longer use the json files but the program dict directly. If the same program is run twice in a row, we hit the cache and there is no upload, therefore the acquisition memory is not cleared.