Feature/comparison #550
Feature/comparison #550
Conversation
…ation, reducing code duplication while maintaining the same functionality (data preparation, color resolution from components, PlotResult wrapping).
… area(), and duration_curve() methods in both DatasetPlotAccessor and DataArrayPlotAccessor
2. scatter() method - Plots two variables against each other with x and y parameters
3. pie() method - Creates pie charts from aggregated (scalar) dataset values, e.g. ds.sum('time').fxplot.pie()
4. duration_curve() method - Sorts values along the time dimension in descending order, with optional normalize parameter for percentage x-axis
5. CONFIG.Plotting.default_line_shape - New config option (default 'hv') that controls the default line shape for line(), area(), and duration_curve() methods
1. X-axis is now determined first using CONFIG.Plotting.x_dim_priority
2. Facets are resolved from remaining dimensions (x-axis excluded)
x_dim_priority expanded:
x_dim_priority = ('time', 'duration', 'duration_pct', 'period', 'scenario', 'cluster')
- Time-like dims first, then common grouping dims as fallback
- variable stays excluded (it's used for color, not x-axis)
_get_x_dim() refactored:
- Now takes dims: list[str] instead of a DataFrame
- More versatile - works with any list of dimension names
- Add `x` parameter to bar/stacked_bar/line/area for explicit x-axis control - Add CONFIG.Plotting.x_dim_priority for auto x-axis selection order - X-axis determined first, facets from remaining dimensions - Refactor _get_x_column -> _get_x_dim (takes dim list, not DataFrame) - Support scalar data (no dims) by using 'variable' as x-axis
- Add `x` parameter to bar/stacked_bar/line/area for explicit x-axis control
- Add CONFIG.Plotting.x_dim_priority for auto x-axis selection
Default: ('time', 'duration', 'duration_pct', 'period', 'scenario', 'cluster')
- X-axis determined first, facets resolved from remaining dimensions
- Refactor _get_x_column -> _get_x_dim (takes dim list, more versatile)
- Support scalar data (no dims) by using 'variable' as x-axis
- Skip color='variable' when x='variable' to avoid double encoding
- Fix _dataset_to_long_df to use dims (not just coords) as id_vars
- Add `x` parameter to bar/stacked_bar/line/area for explicit x-axis control
- Add CONFIG.Plotting.x_dim_priority for auto x-axis selection
Default: ('time', 'duration', 'duration_pct', 'period', 'scenario', 'cluster')
- X-axis determined first, facets resolved from remaining dimensions
- Refactor _get_x_column -> _get_x_dim (takes dim list, more versatile)
- Support scalar data (no dims) by using 'variable' as x-axis
- Skip color='variable' when x='variable' to avoid double encoding
- Fix _dataset_to_long_df to use dims (not just coords) as id_vars
- Ensure px_kwargs properly overrides all defaults (color, facets, etc.)
…wargs} so user can override 2. scatter unused colors - Removed the unused parameter 3. to_duration_curve sorting - Changed [::-1] to np.flip(..., axis=time_axis) for correct multi-dimensional handling 4. DataArrayPlotAccessor.heatmap - Same kwarg merge fix
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the 📝 WalkthroughWalkthroughIntroduces a new Changes
Sequence Diagram(s)sequenceDiagram
participant User as User/Notebook
participant Comp as Comparison
participant FSList as FlowSystem List
participant Stats as ComparisonStatistics
participant Plotter as ComparisonStatisticsPlot
participant Plot as Plotting Output
User->>Comp: __init__(flow_systems, names)
Comp->>Comp: Validate dimensions & duplicates
Comp->>FSList: Aggregate solutions & combine
Comp->>Comp: Create 'case' dimension
User->>Comp: .solution
Comp-->>User: Combined xarray Dataset
User->>Comp: .statistics
Comp->>Stats: Initialize with Comparison
Stats->>FSList: Extract & cache per-system stats
Stats->>Stats: Concatenate across cases
Stats-->>Comp: ComparisonStatistics instance
User->>Stats: .flow_rates / .sizes / .charge_states
Stats-->>User: Aggregated Dataset with 'case' dim
User->>Stats: .plot.balance(...)
Stats->>Plotter: Initialize plotter
Plotter->>FSList: Delegate to per-system plotters
Plotter->>Plotter: Concatenate results by 'case'
Plotter->>Plot: Combine into faceted figure
Plot-->>User: Multi-case visualization
User->>Comp: .diff(reference='case_name')
Comp->>Comp: Compute differences vs reference
Comp-->>User: Difference Dataset
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
Suggested labels
Suggested reviewers
Poem
Pre-merge checks and finishing touches❌ Failed checks (1 warning, 1 inconclusive)
✅ Passed checks (1 passed)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
1. __getattr__ - dynamically delegates any method to the underlying accessor 2. _wrap_plot_method - single method that handles all the data collection and concatenation 3. _recreate_figure - infers plot type from the original figure and recreates with combined data Tradeoffs: - Less explicit type hints on method signatures (but still works the same) - Infers plot type from original figure rather than hardcoding per method - Automatically supports any new methods added to StatisticsPlotAccessor in the future
1. __getattr__ - dynamically delegates any method to the underlying accessor 2. _wrap_plot_method - single method that handles all the data collection and concatenation 3. _recreate_figure - infers plot type from the original figure and recreates with combined data Tradeoffs: - Less explicit type hints on method signatures (but still works the same) - Infers plot type from original figure rather than hardcoding per method - Automatically supports any new methods added to StatisticsPlotAccessor in the future
…igure creation. The _DATA_KWARGS mapping defines which kwargs affect data processing - everything else passes through to plotly.
…igure creation. The _DATA_KWARGS mapping defines which kwargs affect data processing - everything else passes through to plotly.
…igure creation. The _DATA_KWARGS mapping defines which kwargs affect data processing - everything else passes through to plotly.
# Conflicts: # .github/workflows/docs.yaml # mkdocs.yml
…mmary of the changes made:
Changes Made
1. flixopt/config.py
- Replaced three separate config attributes (extra_dim_priority, dim_slot_priority, x_dim_priority) with a single unified dim_priority tuple
- Updated CONFIG.Plotting class docstring and attribute definitions
- Updated to_dict() method to use the new attribute
- The new priority order: ('time', 'duration', 'duration_pct', 'variable', 'cluster', 'period', 'scenario')
2. flixopt/dataset_plot_accessor.py
- Created new assign_slots() function that centralizes all dimension-to-slot assignment logic
- Fixed slot fill order: x → color → facet_col → facet_row → animation_frame
- Updated all plot methods (bar, stacked_bar, line, area, heatmap, scatter, pie) to use assign_slots()
- Removed old _get_x_dim() and _resolve_auto_facets() functions
- Updated docstrings to reference dim_priority instead of x_dim_priority
3. flixopt/statistics_accessor.py
- Updated _resolve_auto_facets() to use the new assign_slots() function internally
- Added import for assign_slots from dataset_plot_accessor
Key Design Decisions
- Single priority list controls all auto-assignment
- Slots are filled in fixed order based on availability
- None means a slot is not available for that plot type
- 'auto' triggers auto-assignment from priority list
- Explicit string values override auto-assignment
…e/comparison # Conflicts: # docs/notebooks/08c-clustering.ipynb # flixopt/config.py
…e/comparison # Conflicts: # docs/notebooks/08c-clustering.ipynb # flixopt/config.py
1. Removed _resolve_facets method - fxplot handles 'auto' resolution internally
2. Updated all methods to pass facet params directly to fxplot
3. sizes now uses ds.fxplot.bar() instead of px.bar
4. effects now uses ds.fxplot.bar() with proper column naming
statistics_accessor.py:
1. Simplified effects method significantly:
- Works directly with Dataset (no DataArray concat/conversion)
- Uses dict.get for aspect selection
- Cleaner aggregation logic
- Returns Dataset with effects as data variables
- Uses fxplot.bar instead of px.bar
The code is now consistent - all plotting methods in both StatisticsPlotAccessor and ComparisonStatisticsPlot use fxplot for centralized dimension/slot handling.
# Conflicts: # docs/notebooks/02-heat-system.ipynb # docs/notebooks/03-investment-optimization.ipynb # docs/notebooks/04-operational-constraints.ipynb # docs/notebooks/05-multi-carrier-system.ipynb # docs/notebooks/06a-time-varying-parameters.ipynb # docs/notebooks/07-scenarios-and-periods.ipynb # docs/notebooks/08c-clustering.ipynb # flixopt/config.py # flixopt/statistics_accessor.py
|
@coderabbitai review |
✅ Actions performedReview triggered.
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
Fix all issues with AI Agents 🤖
In @docs/notebooks/08a-aggregation.ipynb:
- Line 276: Remove the stray XML tag artifact "<cell_type>markdown</cell_type>"
from the markdown cell so the header renders correctly; replace the entire line
that currently reads "<cell_type>markdown</cell_type>## Visual Comparison: Heat
Balance" with a plain markdown header "## Visual Comparison: Heat Balance"
(ensure no surrounding XML fragments remain in the notebook cell content).
In @flixopt/comparison.py:
- Around line 70-76: In __init__, ensure FlowSystem.name may be None: when names
is None, build self._names by mapping each FlowSystem in flow_systems to a
non-None fallback (e.g., use fs.name if truthy else f"flow_system_{index}" or
another deterministic fallback like fs.__class__.__name__ + f"_{index}"), then
validate uniqueness of the resulting self._names (raise ValueError if
duplicates). Update the code that currently sets self._names = names or [fs.name
for fs in flow_systems] to perform this defaulting and uniqueness check (refer
to the __init__ method and self._names symbol to locate the change).
🧹 Nitpick comments (1)
flixopt/comparison.py (1)
354-365: Silent exception handling may hide legitimate errors.The
except (KeyError, ValueError): continueon lines 359-360 silently skips systems when data extraction fails. While this allows partial data when systems have different components, it could also hide genuine configuration errors or bugs.Consider logging a warning when a system is skipped so users are aware that not all cases are included in the plot.
🔎 Suggested improvement
+import logging +logger = logging.getLogger('flixopt') + for fs, case_name in zip(self._comp._systems, self._comp._names, strict=True): try: result = getattr(fs.statistics.plot, method_name)(*args, **kwargs) datasets.append(result.data.expand_dims(case=[case_name])) title = result.figure.layout.title.text or title except (KeyError, ValueError): + logger.debug(f"Skipping case '{case_name}' for {method_name}: data not available") continue
📜 Review details
Configuration used: defaults
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (16)
CHANGELOG.mddocs/notebooks/02-heat-system.ipynbdocs/notebooks/03-investment-optimization.ipynbdocs/notebooks/04-operational-constraints.ipynbdocs/notebooks/05-multi-carrier-system.ipynbdocs/notebooks/06a-time-varying-parameters.ipynbdocs/notebooks/07-scenarios-and-periods.ipynbdocs/notebooks/08a-aggregation.ipynbdocs/notebooks/08b-rolling-horizon.ipynbdocs/notebooks/08c-clustering.ipynbdocs/notebooks/08c2-clustering-storage-modes.ipynbdocs/notebooks/08d-clustering-multiperiod.ipynbdocs/user-guide/results/index.mdflixopt/__init__.pyflixopt/comparison.pyflixopt/statistics_accessor.py
🧰 Additional context used
🧬 Code graph analysis (1)
flixopt/__init__.py (1)
flixopt/comparison.py (1)
Comparison(23-165)
🪛 markdownlint-cli2 (0.18.1)
CHANGELOG.md
65-65: Code block style
Expected: indented; Actual: fenced
(MD046, code-block-style)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Build documentation
🔇 Additional comments (30)
docs/notebooks/06a-time-varying-parameters.ipynb (1)
311-311: Verify notebook metadata stripping intent and implications.Line 311 shows the notebook metadata reduced to
{}(empty). Per the AI summary, this removeskernelspecandlanguage_infosections. While consistent with broader PR-wide notebook refactoring, confirm:
- This metadata removal is intentional as part of a documentation consistency effort (not accidental).
- Removing
kernelspecandlanguage_infodoes not break notebook execution or CI/documentation pipeline assumptions (e.g., if Jupyter Book, JupyterHub, or doc-build tools rely on these fields).docs/notebooks/02-heat-system.ipynb (1)
35-35: LGTM! Clean migration to structured data presentation.The pandas import and DataFrame-based output improve readability and align with the comparison-oriented workflows introduced in this PR.
Also applies to: 287-294
docs/notebooks/08c-clustering.ipynb (3)
104-104: LGTM! Consistent FlowSystem labeling for comparison workflows.Setting explicit names on FlowSystem instances enables clear identification in the new Comparison API.
Also applies to: 146-146, 306-306, 355-355
254-260: LGTM! Structured metrics comparison.The DataFrame-based comparison of clustering algorithms is more readable than the previous print-based approach.
311-312: LGTM! Appropriate use of the Comparison class.Demonstrates comparing two FlowSystem configurations (clustered vs. modified clustered) with the new Comparison API.
flixopt/__init__.py (1)
20-20: LGTM! Properly exposes the new Comparison class.The import and all export correctly add the Comparison class to the public API.
Also applies to: 43-43
docs/notebooks/08b-rolling-horizon.ipynb (2)
97-97: LGTM! Consistent labeling for workflow variants.Setting names on the FlowSystem copies clearly distinguishes the "Full Optimization" and "Rolling Horizon" approaches.
Also applies to: 137-137
184-197: LGTM! Effective use of Comparison for side-by-side visualization.Consolidates two separate balance plots into a single, unified comparison view using the new Comparison class.
docs/notebooks/08d-clustering-multiperiod.ipynb (3)
143-143: LGTM! Consistent FlowSystem labeling.The names clearly identify the optimization methods being compared.
Also applies to: 349-349
439-448: LGTM! Demonstrates side-by-side comparison in multi-period context.Shows how the Comparison class can be used to compare full optimization vs. two-stage workflow results with automatic faceting by period and scenario.
467-471: LGTM! Demonstrates solution expansion workflow.Shows how to map clustered solutions back to full resolution using
expand_solution(), which is valuable for understanding the expanded timestep mapping.docs/notebooks/08a-aggregation.ipynb (2)
175-175: Good practice: naming FlowSystem for comparison clarity.Setting
fs_dispatch.name = 'Two-Stage'ensures the system is clearly labeled in subsequent comparison plots and outputs.
288-290: Effective use of the new Comparison API.The side-by-side comparison workflow is concise and aligns with the new Comparison class capabilities described in the PR objectives.
docs/notebooks/08c2-clustering-storage-modes.ipynb (2)
145-145: Consistent FlowSystem naming pattern.The naming workflow (
fs_full.name = 'Full Optimization'andfs_expanded.name = f'Mode: {mode}') aligns with the comparison-based approach introduced across the notebook ecosystem.Also applies to: 281-283
335-337: Clear demonstration of recommended storage mode.The comparison between full optimization and the
intercluster_cyclicmode effectively demonstrates the recommended approach for seasonal storage systems.docs/user-guide/results/index.md (1)
280-428: Comprehensive documentation for the Comparison API.The new section is well-structured, covering:
- Clear basic usage examples
- Explicit dimension matching requirements
- Complete property and method listings
- Practical comparison scenarios
The examples are runnable and align with the notebook usage patterns demonstrated elsewhere in the PR.
docs/notebooks/03-investment-optimization.ipynb (1)
233-240: Improved output presentation with DataFrames.Converting print-based outputs to pandas DataFrames provides:
- Structured, tabular presentation
- Better readability for single-row results (via
.T)- Professional appearance consistent with data analysis best practices
These changes enhance the notebook's pedagogical value.
Also applies to: 283-290, 348-362
docs/notebooks/07-scenarios-and-periods.ipynb (2)
103-115: Effective xarray-based visualization of scenario data.The approach of constructing an xarray Dataset from scenario-specific demand profiles and rendering with
fxplot.linealigns with the project's shift toward xarray-based visualization, providing a clean and maintainable pattern for multi-scenario data.
273-280: Consistent DataFrame-based output improvements.Converting results to pandas DataFrames (optimal sizing, CHP operation summary, sensitivity analysis) provides structured, professional output formatting that enhances the notebook's clarity and aligns with data analysis best practices.
Also applies to: 354-362, 389-395
docs/notebooks/05-multi-carrier-system.ipynb (1)
35-40: LGTM! Well-structured demonstration of the new Comparison API.The notebook updates effectively showcase:
- FlowSystem naming with
name='With CHP'andname='No CHP'- Clean DataFrame-based summary outputs for comparison
- Side-by-side visualization using
fx.ComparisonThe progression from individual system analysis to comparison is pedagogically sound.
Also applies to: 130-130, 335-348, 369-369, 413-426, 446-458
docs/notebooks/04-operational-constraints.ipynb (1)
35-41: LGTM! Consistent updates demonstrating the Comparison workflow.The notebook appropriately demonstrates:
- Named FlowSystem instances for clearer comparison
- DataFrame-based cost summaries
comp.statistics.plot.effects()for side-by-side effect visualizationAlso applies to: 112-112, 276-282, 333-333, 363-372, 392-394
flixopt/statistics_accessor.py (3)
1869-1875: LGTM on the dictionary-based aspect selection.Using a dictionary with
.get()is cleaner than the previous approach. TheNonecheck with explicit error message provides clear feedback for invalid aspects.
1891-1916: Dimension handling inbyparameter logic looks correct.The logic correctly:
- Sums over all groupable dimensions when
by=None- Preserves the appropriate dimension for
by='component',by='contributor', orby='time'- Validates that
timedimension exists before allowingby='time'
1886-1889: Dimension state after groupby is incorrectly described; selection filtering provides safety.After
ds.groupby('component').sum()at line 1887,'component'remains as a dimension (it's the grouping key), while'contributor'is removed (summed). The concern aboutcomponentbecoming unavailable is reversed. However, ifselectcontains a'contributor'key, it will be silently ignored after groupby removes that dimension—though_apply_selectionfilters to only valid dimensions, so this is handled gracefully. If intentional (preventing selection on aggregated dimensions), this is fine; otherwise, document the expected behavior for theselectparameter with respect toby.CHANGELOG.md (1)
63-89: LGTM! Comprehensive changelog entry for the new Comparison feature.The documentation clearly covers:
- Constructor options (default names vs custom names)
- Side-by-side plotting with automatic faceting
- Combined data access via
comp.solutionandcomp.statistics.*- Diff computation with reference case selection
- Complete list of mirrored properties and methods
The static analysis hint about code block style (fenced vs indented) can be safely ignored—fenced code blocks are the preferred style for this changelog format.
flixopt/comparison.py (5)
97-123: LGTM! Well-designed dimension validation.The validation:
- Only checks core dimensions (
time,period,scenario), correctly ignoring auxiliary dimensions likecluster_boundary- Provides clear error messages indicating which dimensions are missing or extra
- Suggests using
.transform.sel()to align dimensionsThis is a thoughtful design that allows flexibility while ensuring data compatibility.
148-165: LGTM! Clean diff() implementation.The method:
- Accepts either string name or integer index for flexibility
- Validates that string references exist in names
- Uses xarray broadcasting correctly to subtract reference from all cases
196-209: LGTM on helper methods for data aggregation.
_concat_propertyand_merge_dict_propertyprovide clean abstractions for combining data across cases. The documented behavior of later cases overriding earlier ones in_merge_dict_propertyis reasonable for color/unit merging.
317-346: LGTM on kwargs splitting approach.The
_DATA_KWARGSdictionary and_split_kwargsmethod provide a clean separation between data-processing arguments and plotly styling arguments. This allows the wrapper to pass data kwargs to the underlying methods while forwarding plotly kwargs to the figure creation.
1-21: LGTM! Clean module setup with appropriate exports.The module correctly:
- Uses
__all__to export onlyComparison- Defines type aliases matching
statistics_accessor.py- Uses TYPE_CHECKING for the FlowSystem import to avoid circular imports
docs/notebooks/08a-aggregation.ipynb
Outdated
| "metadata": {}, | ||
| "source": [ | ||
| "## Visual Comparison: Heat Balance" | ||
| "<cell_type>markdown</cell_type>## Visual Comparison: Heat Balance\n", |
There was a problem hiding this comment.
Fix corrupted markdown text.
The line contains an XML tag artifact <cell_type>markdown</cell_type> that will render incorrectly in the notebook. This appears to be a merge/edit error.
🔎 Proposed fix
- "<cell_type>markdown</cell_type>## Visual Comparison: Heat Balance\n",
+ "## Visual Comparison: Heat Balance\n",📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| "<cell_type>markdown</cell_type>## Visual Comparison: Heat Balance\n", | |
| "## Visual Comparison: Heat Balance\n", |
🤖 Prompt for AI Agents
In @docs/notebooks/08a-aggregation.ipynb around line 276, Remove the stray XML
tag artifact "<cell_type>markdown</cell_type>" from the markdown cell so the
header renders correctly; replace the entire line that currently reads
"<cell_type>markdown</cell_type>## Visual Comparison: Heat Balance" with a plain
markdown header "## Visual Comparison: Heat Balance" (ensure no surrounding XML
fragments remain in the notebook cell content).
| def __init__(self, flow_systems: list[FlowSystem], names: list[str] | None = None) -> None: | ||
| if len(flow_systems) < 2: | ||
| raise ValueError('Comparison requires at least 2 FlowSystems') | ||
|
|
||
| self._systems = flow_systems | ||
| self._names = names or [fs.name for fs in flow_systems] | ||
|
|
There was a problem hiding this comment.
Potential issue: FlowSystem.name might be None, leading to None in names list.
If names is not provided and a FlowSystem has name=None, line 75 will include None in the names list. This could cause issues:
- The uniqueness check (line 82) would pass with multiple
Nonevalues if only one FS has no name - Using
Noneas a coordinate value in xarray can cause unexpected behavior
Consider adding validation or defaulting to a fallback name pattern.
🔎 Suggested fix
- self._names = names or [fs.name for fs in flow_systems]
+ self._names = names or [fs.name or f'Case {i}' for i, fs in enumerate(flow_systems)]🤖 Prompt for AI Agents
In @flixopt/comparison.py around lines 70-76, In __init__, ensure
FlowSystem.name may be None: when names is None, build self._names by mapping
each FlowSystem in flow_systems to a non-None fallback (e.g., use fs.name if
truthy else f"flow_system_{index}" or another deterministic fallback like
fs.__class__.__name__ + f"_{index}"), then validate uniqueness of the resulting
self._names (raise ValueError if duplicates). Update the code that currently
sets self._names = names or [fs.name for fs in flow_systems] to perform this
defaulting and uniqueness check (refer to the __init__ method and self._names
symbol to locate the change).
…own</cell_type> tag from markdown source
2. flixopt/comparison.py line 75: Added fallback for None names:
# Before
self._names = names or [fs.name for fs in flow_systems]
# After
self._names = names or [fs.name or f'System {i}' for i, fs in enumerate(flow_systems)]
Add comparison functionality for analyzing multiple FlowSystem solutions: - Compare results across different scenarios - Visualization support for comparison data 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
Add comparison functionality for analyzing multiple FlowSystem solutions: - Compare results across different scenarios - Visualization support for comparison data 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <[email protected]>
Summary
Adds a new Comparison class for comparing multiple FlowSystems side-by-side with automatic faceting.
New Feature: fx.Comparison
Compare optimized FlowSystems with unified data access and plotting:
Compare two systems
comp = fx.Comparison([fs_base, fs_modified])
Side-by-side plots (auto-facets by 'case')
comp.statistics.plot.balance('Heat')
comp.statistics.plot.effects()
Access combined data with 'case' dimension
comp.solution # xr.Dataset
comp.statistics.flow_rates # xr.Dataset
Compute differences
comp.diff() # Relative to first case
comp.diff('baseline') # Or specify reference
Features
Additional Improvements
Summary by CodeRabbit
New Features
Documentation
✏️ Tip: You can customize this high-level summary in your review settings.