Multi yoy integration - combines PRs #456, #460 and #464 at once.

[cdeline]

See individual PR's #456, #460 and #464 for discussion.

Issues & PRs

• Closes 394 multi yoy - allows multiple slope calculations across multi-year periods #464
• Closes 459 yoy label='center' #460
• Closes Degradation timeseries plot - add left or right label option #456
• Closes degradation_year_on_year should allow center-labeled YoY_values #459
• Closes degradation_timeseries_plot should have different label= options #455
• Closes Multi-YoY analysis based on Hugo Quest et al 2023 #394

• Code changes are covered by tests
• Code changes have been evaluated for compatibility/integration with TrendAnalysis
• New functions added to __init__.py
• API.rst is up to date, along with other sphinx docs pages
• Example notebooks are rerun and differences in results scrutinized
• Updated changelog

[Copilot]

Pull request overview

This PR merges prior work to support multi-Year-on-Year (multi-YoY) degradation analysis and improved time-labeling options across calculations and plots, plus accompanying tests and documentation/CI updates.

Changes:

• Add label and multi_yoy options to degradation_year_on_year, including new calc metadata and updated point-usage handling.
• Add label option to degradation_timeseries_plot and extend plotting tests (including a multi-YoY plotting scenario).
• Update docs/changelog and notebook/nbval workflow configuration for notebook testing.

Reviewed changes

Copilot reviewed 11 out of 11 changed files in this pull request and generated 9 comments.

Show a summary per file

File	Description
setup.py	Unpins nbval in test requirements.
rdtools/degradation.py	Adds label + multi_yoy support and related calc-info changes.
rdtools/plotting.py	Adds label support to the degradation time-series plot and updates summary plot coloring behavior.
rdtools/test/degradation_test.py	Adds tests for label='center', _avg_timestamp_old_Pandas, and multi_yoy=True.
rdtools/test/plotting_test.py	Extends degradation time-series plotting tests to cover labeling and multi-YoY behavior.
docs/system_availability_example.ipynb	Notebook output/display tweak.
docs/sphinx/source/changelog/pending.rst	Adds pending changelog entries for new parameters/features.
docs/sphinx/source/changelog.rst	Includes the pending changelog into the main changelog.
docs/TrendAnalysis_example_NSRDB.ipynb	Minor notebook code formatting tweak.
docs/TrendAnalysis_example.ipynb	Minor notebook code formatting tweaks.
.github/workflows/nbval.yaml	Updates nbval sanitize flag usage and command invocation.

Comments suppressed due to low confidence (1)

rdtools/degradation.py:237

• The usage_of_points docstring describes only values 0/1/2, but multi_yoy=True can produce larger counts. Please update the docstring to reflect that counts may exceed 2 in multi-YoY mode (and what higher values mean).

          probability of `exceedance_prob`
        * `usage_of_points` - number of times each point in energy_normalized
          is used to calculate a degradation slope. 0: point is never used. 1:
          point is either used as a start or endpoint. 2: point is used as both
          start and endpoint for an Rd calculation.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

For label='center', the code sets center=True on the rolling window and shifts the plotted x-values by offset_days=182. Because center=True already shifts the result index to the window midpoint, subtracting an additional 182 days effectively double-shifts the timestamps and is inconsistent with the docstring’s “center of rolling median interval”. Consider either (a) not shifting when center=True, or (b) using center=False and shifting by half the window, but not both.

Suggested change

	center = False
	offset_days = 0
	elif label == "center":
	center = True
	offset_days = 182
	elif label == "left":
	center = True
	offset_days = 365
	# Right-labeled: rolling index at right edge, no offset
	center = False
	offset_days = 0
	elif label == "center":
	# Center-labeled: rolling index at midpoint, no additional offset
	center = True
	offset_days = 0
	elif label == "left":
	# Left-labeled: rolling index at right edge, shift by full window to left edge
	center = False
	offset_days = rolling_days

[martin-springer]

@cdeline - If you get a chance, could you take a look at this suggestion?

[cdeline]

@martin-springer - yeah I'm going to need to change this logic because now the YoY-Info dict being passed in is no longer explicitly right-labeled. I'm going to revert it back to just be center=True or center=False. We'll need to update the TA object as well to allow it to have correct label options.

[martin-springer]

users would need to pass label=None explicitly to trigger those checks but end up with label='right'. I think it's best to rely on the default handling.

I'll remove the label is None handling in degradation.py and plotting.py. I'll also update the tests that specifically trigger this. Commit 0fe8ebd

[Copilot]

Pull request overview

Copilot reviewed 10 out of 11 changed files in this pull request and generated 5 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[martin-springer]

With pandas three now released, should we think about dropping support for pandas 1.x.x?

[cdeline]

yeah, maybe we start with a deprecation warning for pandas < 2.0 in our next release. Pandas 2.0 was only put out in 2023 so we can probably hang on to this functionality for a few more years - its not really causing any harm. PVLib is still on pandas > 1.3.3 so we should probably still support it too...