Auto-parallelism tutorial #288
Conversation
|
|
|
||
| # %% | ||
| circuit4_parallel = utils.parallelize(circuit=circuit4) | ||
| circuit4_parallel = utils.no_similarity(circuit4_parallel) |
There was a problem hiding this comment.
This has been changed to utils.remove_tags https://github.com/QuEraComputing/bloqade-circuit/blame/main/src/bloqade/cirq_utils/parallelize.py#L139
There was a problem hiding this comment.
Also, it's now run at the end of parallelize, so you can just remove all calls to no_similarity and it will work the same.
|
|
||
|
|
||
| # %% | ||
| def visualize_grid_interactions(circuit, L=4): |
There was a problem hiding this comment.
I wonder if this should be some sort of stdlib-- it's a lot of code to drop into a tutorial. Another option would be to "hide" it in the webpage to give a better flow of the tutorial. @david-pl .
There was a problem hiding this comment.
I agree, the code that produces these plots has little benefit for the user. Also, we are currently not running notebooks when building the documentation, so you'd have to add the image manually anyway. My suggestion would be to show the images, but not the source code.
As a side note: those graphs could do with some more detailed explanations.
|
One general comment here: about a third of the example is just reproducing the noisy GHZ result for which we already have an example and a blog post. Granted, you show parallelization benefits the fidelity, but this can be readily seen from other examples that use automated parallelism too. Can we just remove the manual parallelization GHZ part? I feel like it's mostly redundant at this point. Another small thing: maybe a few sentences explaining the Steane code and a reference or two wouldn't hurt :) |
3 participants
No description provided.