Test on Simulated Annealing

[AzAINN]

• The notebook contains the graphs generated from loading in already bootstrapped results and running interpolation on them.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[AzAINN]

The reset_index() is implemented in run_bootstrap() when the parameter reduce_mem is False, and hence interpolate() doesn't have reset_index() it wouldn't cause an indexing issue in the dataframe when we bootstrap using the package.

But when using pre-bootstrapped data and then using interpolation without reduce_mem = False, this causes an issue. So adding that line in the interpolate() prevents that issue.

I don't think this is a required change, as the goal is to bootstrap using the package, but it helps when we already have bootstrapped data to interpolate.

[AzAINN]

SA_test_v2.ipynb takes in raw .p files in the raw_runs folder and formats the files expected by the native bootstrapping (aka run_Bootstrap() ).

Creates _run_runs_pk_l folder. Then these instances are bootstrapped to populated the checkpoints folder

[bernalde]

Good first approach, but there is a lot of cleaning up to be done here.

• Let's compress all the individual runs in a single file and remove from the history each independent file
• In the notebooks, let's spend space describing the structure of the data and how we recover parameters from there
• There are two versions of the notebooks; we need to have a single one. One of them has errors in the middle
• Uploading the QUBONotebooks as it is should not be done. Let's clean it up and have it as a hand version implementation of the method, even better if we can have them both in the same notebook to show how the package behaves
• Number of shots is not a parameter if you already have the number of reads

[bernalde]

Closed in favor of #30. Put all the comments from here, there