Add getting learner's data as pandas.DataFrame; add learner.to_dataframe method

[basnijholt]

Description

This code in this PR allows exporting a learner to a pandas.DataFrame. This will be useful when running many learners (e.g, in a BalancingLearner) and then being able to just concatenate the DataFrames to get the combined result.

Currently saving the learner creates a pickle file and while this is fine for immediate use, for long term storage this is a bad practice.

I chose to use pandas over xarray because we are dealing with tabular data instead of ND dense or sparse arrays.

Ping @CameronKing

Example:

Checklist

• Fixed style issues using pre-commit run --all (first install using pip install pre-commit)
• pytest passed

Type of change

Check relevant option(s).

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• (Code) style fix or documentation update
• This change requires a documentation update

[akhmerov]

What problem does this PR solve? I believe a pandas DataFrame isn't a go-to choice for numerical data; is there a use case which is sufficiently broadly relevant?

[akhmerov]

OK, I see: the entries indeed have different meanings and possibly dimensions (x, y, value). For that reason pandas seems appropriate.

[codecov-commenter]

Codecov Report

Merging #358 (e091772) into master (2c8d2ad) will decrease coverage by 0.16%.
The diff coverage is 76.00%.

@@            Coverage Diff             @@
##           master     #358      +/-   ##
==========================================
- Coverage   80.67%   80.50%   -0.17%     
==========================================
  Files          38       38              
  Lines        5071     5336     +265     
  Branches      948      997      +49     
==========================================
+ Hits         4091     4296     +205     
- Misses        835      873      +38     
- Partials      145      167      +22     

Impacted Files	Coverage Δ
adaptive/learner/integrator_learner.py	88.51% <33.33%> (-2.65%)	⬇️
adaptive/learner/average_learner1D.py	77.51% <62.50%> (-1.30%)	⬇️
adaptive/learner/learnerND.py	60.26% <65.21%> (+0.14%)	⬆️
adaptive/learner/average_learner.py	83.76% <71.42%> (-3.11%)	⬇️
adaptive/learner/learner1D.py	89.47% <71.42%> (-0.50%)	⬇️
adaptive/learner/learner2D.py	78.75% <72.72%> (-0.55%)	⬇️
adaptive/learner/sequence_learner.py	85.29% <75.00%> (-3.17%)	⬇️
adaptive/learner/balancing_learner.py	75.00% <80.00%> (+0.52%)	⬆️
adaptive/learner/data_saver.py	86.66% <80.00%> (-3.34%)	⬇️
adaptive/utils.py	78.49% <84.00%> (+2.02%)	⬆️
... and 4 more

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more