feat(models): add IREQ model and related documentation (ISO 11079) #260
Conversation
…BuiltEnvironment/development Development
…BuiltEnvironment/development Development
…BuiltEnvironment/development docs(changelog): Improve naming and formatting in changelog
WalkthroughAdds ISO 11079 IREQ computation: new vectorised calc_ireq() and IREQ dataclass, extends input surface with duration, t_re, t_sk and new input classes, updates docs and changelog, and adds comprehensive unit tests for scalar/array/broadcast and invalid cases. Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant calc_ireq as calc_ireq()
participant Validate as _validate_inputs
participant Physics as _compute_physics
participant Solver as _solve_iterative
participant Result as IREQ
User->>calc_ireq: call calc_ireq(m, w_work, tdb, tr, p_air, v_walk, v, rh, clo)
calc_ireq->>Validate: validate shapes, ranges, broadcastability
Validate-->>calc_ireq: OK / raise ValueError
calc_ireq->>Physics: unit conversions, air/skin properties, heat-transfer coeffs
Physics-->>calc_ireq: physical params (i_air, h_c, vp, ...)
calc_ireq->>Solver: iterative solve (neutral condition)
Solver-->>calc_ireq: ireq_neutral, icl_neutral, dle_neutral
calc_ireq->>Solver: iterative solve (minimum condition)
Solver-->>calc_ireq: ireq_min, icl_min, dle_min
calc_ireq->>Result: assemble IREQ dataclass
Result-->>User: return structured results
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes
Possibly related PRs
Suggested reviewers
Pre-merge checks and finishing touches✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
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 |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 13
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
pythermalcomfort/classes_return.py (1)
313-334: PHS API regression vs CHANGELOG 3.6.0 — revert or document breaking change.3.6.0 entry states water_loss and water_loss_watt were removed in favour of sweat_loss_g, sweat_rate_watt, evap_load_wm2_min. This change reintroduces water_loss* without updating 3.7.0 notes, creating an API inconsistency for users migrating from 3.6.x.
Please either:
- Revert to the 3.6.0 fields, or
- Explicitly document the reintroduction in 3.7.0 (with migration notes) and ensure the models that produce these values are aligned.
Happy to draft the changelog/migration snippet if you confirm the intended API.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (6)
CHANGELOG.rst(1 hunks)docs/documentation/models.rst(1 hunks)pythermalcomfort/classes_input.py(2 hunks)pythermalcomfort/classes_return.py(6 hunks)pythermalcomfort/models/ireq.py(1 hunks)tests/test_ireq.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (6)
docs/**/*.rst
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
docs/**/*.rst: Add an '.. autofunction:: pythermalcomfort.models..' entry to the docs for public APIs
Update the appropriate docs reference file that aggregates API docs when adding a public function
Files:
docs/documentation/models.rst
tests/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
tests/**/*.py: Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Use numpy.testing or pytest.approx for numeric assertions
Ensure tests are deterministic, small, and pass locally with pytest before PR
Files:
tests/test_ireq.py
CHANGELOG.*
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Update the CHANGELOG for public API changes
Files:
CHANGELOG.rst
pythermalcomfort/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/**/*.py: Use type hints and NumPy-style docstrings for all library code
Validate inputs in dataclass post_init (call super().post_init() if inheriting BaseInputs)
Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in post_init
Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Use numpy operations for numerical work (e.g., np.log, np.asarray) and avoid math.log when arrays are accepted
Validate domains before logs/roots (arguments must be > 0)
Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Run ruff (lint/format) and docformatter; ensure code is formatted before PR
When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_return.pypythermalcomfort/classes_input.py
pythermalcomfort/models/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/models/*.py: Place new domain model functions in pythermalcomfort/models/.py
Include applicability limits and minimal runnable examples in model function docstrings
Files:
pythermalcomfort/models/ireq.py
pythermalcomfort/classes_input.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Files:
pythermalcomfort/classes_input.py
🧠 Learnings (2)
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate inputs in dataclass __post_init__ (call super().__post_init__() if inheriting BaseInputs)
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
pythermalcomfort/classes_input.py
🧬 Code graph analysis (3)
tests/test_ireq.py (1)
pythermalcomfort/models/ireq.py (1)
calc_ireq(7-186)
pythermalcomfort/models/ireq.py (2)
pythermalcomfort/classes_input.py (1)
IREQInputs(1059-1083)pythermalcomfort/classes_return.py (1)
IREQ(949-974)
pythermalcomfort/classes_input.py (1)
pythermalcomfort/utilities.py (4)
Units(36-40)Postures(517-526)Sex(43-47)validate_type(393-401)
🔇 Additional comments (4)
pythermalcomfort/classes_return.py (2)
15-24: Nice, clearer str formatting for result classes.Width-aligned listing improves readability for multi-line/numpy values. LGTM.
758-759: Minor: np.array vs np.asarray.Functionally equivalent here. If you want to preserve subclass types (rare for dataclasses), np.asarray is slightly more general; otherwise np.array is fine. No action needed.
docs/documentation/models.rst (1)
168-175: Good docs wiring for the new IREQ model.Autofunction and autoclass entries are correctly added under Models and match module paths. LGTM.
pythermalcomfort/classes_input.py (1)
19-69: LGTM! BaseInputs field definitions are well-structured.The explicit field definitions with appropriate default values and type hints provide a solid foundation for input validation. The inclusion of both scalar and array types for numeric fields supports vectorized operations as per the coding guidelines.
CHANGELOG.rst
Outdated
| 3.7.0 (2025-10-16) | ||
| ------------------ | ||
|
|
||
| * Added the `calc_ireq()` function implementing the ISO 11079 Required Clothing Insulation (IREQ) model. | ||
| * The model estimates required clothing insulation for both minimum (`IREQ_min`) and neutral (`IREQ_neutral`) thermal balance conditions, as well as Duration Limited Exposure (DLE) limits. | ||
| * The function supports scalar and vectorized numpy array inputs, automatic broadcasting, and full input validation (type, shape, and physical range checks). | ||
| * Added comprehensive unit tests (`test_ireq.py`) covering scalar, array, broadcasting, invalid, and edge-case inputs. | ||
| * Added validation cases from ISO 11079 Table F.1 for deterministic reference comparison. | ||
| * Added `IREQ` dataclass to standardize model outputs and documentation under `docs/documentation/models.rst`. | ||
|
|
There was a problem hiding this comment.
Changelog missing note about PHS field changes introduced in this PR.
This PR also alters PHS outputs (replaces 3.6.0’s sweat_* fields with water_loss_*), which is a breaking surface change relative to 3.6.x. Please add an explicit bullet under 3.7.0 with migration guidance, or revert the PHS change for a non‑breaking release.
I can propose the wording once the intended direction is confirmed.
🤖 Prompt for AI Agents
In CHANGELOG.rst around lines 4 to 13, the release notes for 3.7.0 omit mention
of the breaking PHS output change (replacing sweat_* fields with water_loss_*);
add an explicit bullet under the 3.7.0 section describing the change, mark it as
breaking, and include brief migration guidance (map old sweat_* names to new
water_loss_* names and note any semantic or unit differences), or alternatively
note that the PHS change was reverted if you choose a non‑breaking release;
update the changelog language to be concise and prescriptive for users upgrading
from 3.6.x.
|
@LianqiWang111 as explained in the documentation the pull request should not be made to Could you also please also address all the open comments from coderabbit? There are many things that need to be fixed. I am also tagging @KristinaM93 who will also review your code. |
…BuiltEnvironment/development development
KristinaM93
left a comment
KristinaM93
left a comment
There was a problem hiding this comment.
Just some suggestions for test_ireg.py
LianqiWang111
force-pushed
the
feature/ireq
branch
from
b24193a to
e617cbd
Compare
There was a problem hiding this comment.
Actionable comments posted: 3
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (5)
CHANGELOG.rst(1 hunks)pythermalcomfort/classes_input.py(4 hunks)pythermalcomfort/classes_return.py(2 hunks)pythermalcomfort/models/ireq.py(1 hunks)tests/test_ireq.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (5)
tests/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
tests/**/*.py: Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Use numpy.testing or pytest.approx for numeric assertions
Ensure tests are deterministic, small, and pass locally with pytest before PR
Files:
tests/test_ireq.py
CHANGELOG.*
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Update the CHANGELOG for public API changes
Files:
CHANGELOG.rst
pythermalcomfort/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/**/*.py: Use type hints and NumPy-style docstrings for all library code
Validate inputs in dataclass post_init (call super().post_init() if inheriting BaseInputs)
Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in post_init
Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Use numpy operations for numerical work (e.g., np.log, np.asarray) and avoid math.log when arrays are accepted
Validate domains before logs/roots (arguments must be > 0)
Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Run ruff (lint/format) and docformatter; ensure code is formatted before PR
When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.pypythermalcomfort/classes_return.py
pythermalcomfort/classes_input.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Files:
pythermalcomfort/classes_input.py
pythermalcomfort/models/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/models/*.py: Place new domain model functions in pythermalcomfort/models/.py
Include applicability limits and minimal runnable examples in model function docstrings
Files:
pythermalcomfort/models/ireq.py
🧠 Learnings (14)
📓 Common learnings
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to tests/**/*.py : Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Applied to files:
tests/test_ireq.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Applied to files:
tests/test_ireq.pypythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate domains before logs/roots (arguments must be > 0)
Applied to files:
tests/test_ireq.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Applied to files:
tests/test_ireq.pypythermalcomfort/models/ireq.pypythermalcomfort/classes_return.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
CHANGELOG.rstpythermalcomfort/classes_input.pypythermalcomfort/models/ireq.pypythermalcomfort/classes_return.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
Applied to files:
CHANGELOG.rstpythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Applied to files:
CHANGELOG.rstpythermalcomfort/classes_input.pypythermalcomfort/classes_return.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate inputs in dataclass __post_init__ (call super().__post_init__() if inheriting BaseInputs)
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in __post_init__
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use type hints and NumPy-style docstrings for all library code
Applied to files:
pythermalcomfort/classes_return.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Applied to files:
pythermalcomfort/classes_return.py
🧬 Code graph analysis (2)
tests/test_ireq.py (1)
pythermalcomfort/models/ireq.py (1)
calc_ireq(6-177)
pythermalcomfort/models/ireq.py (1)
pythermalcomfort/classes_return.py (1)
IREQ(1049-1074)
There was a problem hiding this comment.
Actionable comments posted: 3
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (2)
pythermalcomfort/classes_input.py(5 hunks)pythermalcomfort/models/ireq.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
pythermalcomfort/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/**/*.py: Use type hints and NumPy-style docstrings for all library code
Validate inputs in dataclass post_init (call super().post_init() if inheriting BaseInputs)
Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in post_init
Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Use numpy operations for numerical work (e.g., np.log, np.asarray) and avoid math.log when arrays are accepted
Validate domains before logs/roots (arguments must be > 0)
Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Run ruff (lint/format) and docformatter; ensure code is formatted before PR
When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
pythermalcomfort/classes_input.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Files:
pythermalcomfort/classes_input.py
pythermalcomfort/models/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/models/*.py: Place new domain model functions in pythermalcomfort/models/.py
Include applicability limits and minimal runnable examples in model function docstrings
Files:
pythermalcomfort/models/ireq.py
🧠 Learnings (11)
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate inputs in dataclass __post_init__ (call super().__post_init__() if inheriting BaseInputs)
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in __post_init__
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to tests/**/*.py : Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate domains before logs/roots (arguments must be > 0)
Applied to files:
pythermalcomfort/models/ireq.py
🧬 Code graph analysis (1)
pythermalcomfort/models/ireq.py (1)
pythermalcomfort/classes_return.py (1)
IREQ(1049-1074)
🔇 Additional comments (7)
pythermalcomfort/models/ireq.py (4)
1-4: LGTM! Imports are clean.The imports are minimal and appropriate. Past issues with unused pandas and IREQInputs imports have been properly addressed.
6-177: LGTM! Well-structured function with comprehensive documentation.The function follows best practices:
- Proper type hints for all parameters
- Comprehensive NumPy-style docstring including applicability limits (ISO 11079:2007)
- Runnable examples demonstrating both scalar and vectorized usage
- Input normalisation with
np.asarrayfor vectorised operations- Clean delegation to helper functions
179-250: LGTM! Comprehensive input validation.The validation functions properly enforce:
- Broadcasting compatibility via
np.broadcast_shapes(addresses past review)- Finite values across all inputs (addresses past review)
- Log-domain safety for
p_air > 0(addresses past review)- Physical constraints appropriate for ISO 11079 cold stress evaluation
All validation checks have clear, specific error messages as per coding guidelines.
252-308: LGTM! Clean, single-purpose helper functions.All helpers follow coding guidelines:
- Small, focused functions with clear purposes
- Fully vectorised numpy operations
- Proper handling of numerical edge cases (e.g., division-by-zero guard in
_calculate_heat_transfer_coefficients)pythermalcomfort/classes_input.py (3)
6-6: LGTM! Import issue resolved.
Optionalhas been properly added to the typing imports, addressing the previous critical issue that would have caused aNameErrorat class definition time.
42-42: LGTM! New fields properly integrated into BaseInputs.The new fields (
duration,t_re,t_sk) are correctly declared with appropriate metadata for type validation. This extends the base input surface cleanly for time-referenced temperature inputs.Also applies to: 114-122
845-951: LGTM! Exemplary input validation implementation.The
RidgeRegressionInputsclass demonstrates excellent adherence to coding guidelines:
- Calls
super().__post_init__()before custom validation- Comprehensive finite value checks with clear error messages
- Broadcasting compatibility validation using
np.broadcast_arrays- Mutual requirement validation for
t_reandt_sk(both or neither)- Physical constraint enforcement (positive duration, valid rh range, positive age/height/weight)
This is a model implementation for input validation classes.
There was a problem hiding this comment.
Actionable comments posted: 2
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (2)
pythermalcomfort/classes_input.py(5 hunks)pythermalcomfort/models/ireq.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
pythermalcomfort/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/**/*.py: Use type hints and NumPy-style docstrings for all library code
Validate inputs in dataclass post_init (call super().post_init() if inheriting BaseInputs)
Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in post_init
Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Use numpy operations for numerical work (e.g., np.log, np.asarray) and avoid math.log when arrays are accepted
Validate domains before logs/roots (arguments must be > 0)
Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Run ruff (lint/format) and docformatter; ensure code is formatted before PR
When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
pythermalcomfort/models/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/models/*.py: Place new domain model functions in pythermalcomfort/models/.py
Include applicability limits and minimal runnable examples in model function docstrings
Files:
pythermalcomfort/models/ireq.py
pythermalcomfort/classes_input.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Files:
pythermalcomfort/classes_input.py
🧠 Learnings (12)
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Place new domain model functions in pythermalcomfort/models/<name>.py
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate inputs in dataclass __post_init__ (call super().__post_init__() if inheriting BaseInputs)
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to tests/**/*.py : Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate domains before logs/roots (arguments must be > 0)
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
PR: CenterForTheBuiltEnvironment/pythermalcomfort#0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in __post_init__
Applied to files:
pythermalcomfort/classes_input.py
🧬 Code graph analysis (1)
pythermalcomfort/models/ireq.py (1)
pythermalcomfort/classes_return.py (1)
IREQ(1049-1074)
…BuiltEnvironment/development development
There was a problem hiding this comment.
Actionable comments posted: 2
♻️ Duplicate comments (1)
pythermalcomfort/models/ireq.py (1)
472-480: Negative DLE needs explicit failure handling.If the solver crosses into negative durations we currently emit
"more than 8", which is misleading and hides that the iteration failed. Please surface this as an error (or at least warn) and only apply the sentinel to genuinely large positive values.- if np.isscalar(dle_result): - if dle_result > 8 or dle_result < 0: - dle_result = "more than 8" - else: - mask = (dle_result > 8) | (dle_result < 0) - if np.any(mask): - dle_result = dle_result.astype(object) - dle_result[mask] = "more than 8" + if np.isscalar(dle_result): + if dle_result < 0: + raise RuntimeError("DLE solver returned a negative exposure time.") + if dle_result > 8: + dle_result = "more than 8" + else: + if np.any(dle_result < 0): + raise RuntimeError("DLE solver returned negative exposure times.") + mask = dle_result > 8 + if np.any(mask): + dle_result = dle_result.astype(object) + dle_result[mask] = "more than 8"
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (2)
pythermalcomfort/classes_input.py(5 hunks)pythermalcomfort/models/ireq.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
pythermalcomfort/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/**/*.py: Use type hints and NumPy-style docstrings for all library code
Validate inputs in dataclass post_init (call super().post_init() if inheriting BaseInputs)
Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in post_init
Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Use numpy operations for numerical work (e.g., np.log, np.asarray) and avoid math.log when arrays are accepted
Validate domains before logs/roots (arguments must be > 0)
Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Run ruff (lint/format) and docformatter; ensure code is formatted before PR
When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
pythermalcomfort/classes_input.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Files:
pythermalcomfort/classes_input.py
pythermalcomfort/models/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/models/*.py: Place new domain model functions in pythermalcomfort/models/.py
Include applicability limits and minimal runnable examples in model function docstrings
Files:
pythermalcomfort/models/ireq.py
🧠 Learnings (13)
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate inputs in dataclass __post_init__ (call super().__post_init__() if inheriting BaseInputs)
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in __post_init__
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/{models,utilities}.py : Ensure docstring sections include Args (with units), Returns, Raises, Examples, Applicability
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Place new domain model functions in pythermalcomfort/models/<name>.py
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to tests/**/*.py : Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate domains before logs/roots (arguments must be > 0)
Applied to files:
pythermalcomfort/models/ireq.py
🧬 Code graph analysis (1)
pythermalcomfort/models/ireq.py (1)
pythermalcomfort/classes_return.py (1)
IREQ(1049-1074)
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (1)
pythermalcomfort/models/ireq.py (1)
477-485: Reconsider negative DLE handling.The current code checks
dle_resultafter rounding (good), but silently converts negative DLE values to"more than 8". A negative DLE indicates the iterative solver in_solve_dle_iterationfailed (likely whenq_storechanged sign during bisection), which is a computational error rather than a valid "long duration" result.Consider raising a warning or exception when
dle_result < 0is detected, or at minimum document this behaviour so users understand that"more than 8"can also mean "solver failed".dle_result = np.round(dle * 10.0) / 10.0 if np.isscalar(dle_result): - if dle_result > 8 or dle_result < 0: + if dle_result > 8: + dle_result = "more than 8" + elif dle_result < 0: + import warnings + warnings.warn("DLE calculation produced negative result (solver instability); returning 'more than 8'") dle_result = "more than 8" else: - mask = (dle_result > 8) | (dle_result < 0) + mask_long = dle_result > 8 + mask_failed = dle_result < 0 + if np.any(mask_failed): + import warnings + warnings.warn(f"DLE calculation produced {np.sum(mask_failed)} negative result(s) (solver instability)") + mask = mask_long | mask_failed if np.any(mask): dle_result = dle_result.astype(object) dle_result[mask] = "more than 8"
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (2)
pythermalcomfort/classes_input.py(5 hunks)pythermalcomfort/models/ireq.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
pythermalcomfort/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/**/*.py: Use type hints and NumPy-style docstrings for all library code
Validate inputs in dataclass post_init (call super().post_init() if inheriting BaseInputs)
Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in post_init
Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Use numpy operations for numerical work (e.g., np.log, np.asarray) and avoid math.log when arrays are accepted
Validate domains before logs/roots (arguments must be > 0)
Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Run ruff (lint/format) and docformatter; ensure code is formatted before PR
When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
pythermalcomfort/classes_input.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Files:
pythermalcomfort/classes_input.py
pythermalcomfort/models/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/models/*.py: Place new domain model functions in pythermalcomfort/models/.py
Include applicability limits and minimal runnable examples in model function docstrings
Files:
pythermalcomfort/models/ireq.py
🧠 Learnings (15)
📓 Common learnings
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate inputs in dataclass __post_init__ (call super().__post_init__() if inheriting BaseInputs)
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in __post_init__
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/{models,utilities}.py : Ensure docstring sections include Args (with units), Returns, Raises, Examples, Applicability
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
Applied to files:
pythermalcomfort/classes_input.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use type hints and NumPy-style docstrings for all library code
Applied to files:
pythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Place new domain model functions in pythermalcomfort/models/<name>.py
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to tests/**/*.py : Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate domains before logs/roots (arguments must be > 0)
Applied to files:
pythermalcomfort/models/ireq.py
🧬 Code graph analysis (2)
pythermalcomfort/classes_input.py (1)
pythermalcomfort/models/jos3.py (10)
tdb(1403-1414)tdb(1417-1418)rh(1460-1462)rh(1465-1466)tr(1421-1423)tr(1426-1427)v(1469-1471)v(1474-1475)clo(1505-1507)clo(1510-1511)
pythermalcomfort/models/ireq.py (1)
pythermalcomfort/classes_return.py (1)
IREQ(1049-1074)
🔇 Additional comments (8)
pythermalcomfort/models/ireq.py (4)
179-187: LGTM!The broadcasting validation correctly uses
np.broadcast_shapesto ensure all inputs can be broadcast together, following repository patterns for vectorised behaviour.
189-250: LGTM!The range validation is comprehensive and addresses previous review concerns:
- Finite value checks prevent NaN/Inf issues
- Physical constraints enforced with clear error messages
- p_air > 0 ensures log() domain validity
As per coding guidelines.
354-365: LGTM!The final IREQ calculation correctly recomputes all dependent quantities (q_evap, q_resp, t_clothing, heat fluxes) using the converged
ireqvalue, ensuring consistency. This addresses the previous review concern about stale intermediate values.
422-429: LGTM!The zero guard correctly handles the edge case where
q_storeapproaches zero by replacing it with a small non-zero value that preserves the sign. The implementation ensures1e-10 * signis never zero, preventing division-by-zero when computing-40.0 / q_store.pythermalcomfort/classes_input.py (4)
6-6: LGTM!The new fields (
duration,t_re,t_sk) and theOptionalimport follow the repository patterns for BaseInputs. Type metadata is correctly specified for validation.Also applies to: 42-42, 114-122
845-951: LGTM!The
RidgeRegressionInputsclass demonstrates excellent validation practices:
- Finite value checks prevent NaN/Inf issues
- Broadcasting validation ensures input compatibility
- Mutual requirement for
t_reandt_skis enforced- Physical constraints (non-negative age/height/weight, rh range) have clear error messages
As per coding guidelines and learnings.
1242-1290: LGTM!The
IREQInputsclass documentation and field definitions address previous review concerns:
- Units correctly documented (m in W/m², w_work in W/m², p_air in l/(m²·s))
Optionalproperly imported and used in type hints- Clear notes about unit conversion for met values
1292-1334: LGTM!The
__post_init__validation addresses all previous review concerns:
- Calls
super().__post_init__()for base validation (no redundant pandas conversion)- Validates and normalises all IREQ-specific parameters with
np.asarray- Enforces physical constraints (m > 0, w_work ≥ 0, p_air > 0, rh in [0,100], etc.)
- Stores normalised arrays back to instance variables
As per coding guidelines and learnings.
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (1)
CHANGELOG.rst (1)
4-8: Past review comment regarding PHS breaking change documentation remains unaddressed.A previous review noted that the 3.7.0 changelog should document breaking PHS output changes (if applicable to this release). The current 3.7.0 entry contains no such documentation or migration guidance. If PHS changes are new to or corrected in 3.7.0, add a warning section similar to the 3.6.0 entry (lines 20–40); otherwise, if PHS changes remain from 3.6.0, no additional action is needed as that version already documents them.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (1)
CHANGELOG.rst(1 hunks)
🧰 Additional context used
📓 Path-based instructions (1)
CHANGELOG.*
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Update the CHANGELOG for public API changes
Files:
CHANGELOG.rst
🧠 Learnings (1)
📓 Common learnings
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
8468981 to
220b34e
Compare
… and clarify conversion from met
LianqiWang111
force-pushed
the
feature/ireq
branch
from
220b34e to
ab084b2
Compare
There was a problem hiding this comment.
Actionable comments posted: 3
♻️ Duplicate comments (1)
pythermalcomfort/models/ireq.py (1)
476-485: Don’t mask negative DLE values as “more than 8”.Negative DLE means the storage equation never crossed zero and the iteration failed; rewriting that as “more than 8” tells users the exposure is safe when it’s actually undefined. Raise instead (or propagate NaNs) so downstream code can handle the failure explicitly, and only map genuinely > 8 h values to the sentinel.
- if np.isscalar(dle_result): - if dle_result > 8 or dle_result < 0: - dle_result = "more than 8" - else: - mask = (dle_result > 8) | (dle_result < 0) - if np.any(mask): - dle_result = dle_result.astype(object) - dle_result[mask] = "more than 8" + if np.isscalar(dle_result): + if dle_result < 0: + raise ValueError("Computed DLE is negative; iteration failed to converge.") + if dle_result > 8: + dle_result = "more than 8" + else: + if np.any(dle_result < 0): + raise ValueError("Computed DLE contains negative values; iteration failed to converge.") + mask = dle_result > 8 + if np.any(mask): + dle_result = dle_result.astype(object) + dle_result[mask] = "more than 8"
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (6)
CHANGELOG.rst(1 hunks)docs/documentation/models.rst(1 hunks)pythermalcomfort/classes_input.py(2 hunks)pythermalcomfort/classes_return.py(1 hunks)pythermalcomfort/models/ireq.py(1 hunks)tests/test_ireq.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (6)
CHANGELOG.*
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Update the CHANGELOG for public API changes
Files:
CHANGELOG.rst
pythermalcomfort/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/**/*.py: Use type hints and NumPy-style docstrings for all library code
Validate inputs in dataclass post_init (call super().post_init() if inheriting BaseInputs)
Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in post_init
Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Use numpy operations for numerical work (e.g., np.log, np.asarray) and avoid math.log when arrays are accepted
Validate domains before logs/roots (arguments must be > 0)
Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Run ruff (lint/format) and docformatter; ensure code is formatted before PR
When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_return.pypythermalcomfort/classes_input.py
pythermalcomfort/models/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/models/*.py: Place new domain model functions in pythermalcomfort/models/.py
Include applicability limits and minimal runnable examples in model function docstrings
Files:
pythermalcomfort/models/ireq.py
pythermalcomfort/classes_input.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Files:
pythermalcomfort/classes_input.py
docs/**/*.rst
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
docs/**/*.rst: Add an '.. autofunction:: pythermalcomfort.models..' entry to the docs for public APIs
Update the appropriate docs reference file that aggregates API docs when adding a public function
Files:
docs/documentation/models.rst
tests/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
tests/**/*.py: Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Use numpy.testing or pytest.approx for numeric assertions
Ensure tests are deterministic, small, and pass locally with pytest before PR
Files:
tests/test_ireq.py
🧠 Learnings (16)
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
Applied to files:
CHANGELOG.rstpythermalcomfort/models/ireq.pypythermalcomfort/classes_input.pydocs/documentation/models.rst
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/{models,utilities}.py : Ensure docstring sections include Args (with units), Returns, Raises, Examples, Applicability
Applied to files:
CHANGELOG.rstpythermalcomfort/classes_input.pydocs/documentation/models.rst
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
CHANGELOG.rstpythermalcomfort/models/ireq.pypythermalcomfort/classes_return.pypythermalcomfort/classes_input.pydocs/documentation/models.rsttests/test_ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Place new domain model functions in pythermalcomfort/models/<name>.py
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in __post_init__
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.pytests/test_ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate inputs in dataclass __post_init__ (call super().__post_init__() if inheriting BaseInputs)
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_return.pytests/test_ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to tests/**/*.py : Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Applied to files:
pythermalcomfort/models/ireq.pytests/test_ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate domains before logs/roots (arguments must be > 0)
Applied to files:
pythermalcomfort/models/ireq.pytests/test_ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Applied to files:
pythermalcomfort/models/ireq.pypythermalcomfort/classes_return.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use type hints and NumPy-style docstrings for all library code
Applied to files:
pythermalcomfort/classes_return.pypythermalcomfort/classes_input.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Applied to files:
pythermalcomfort/classes_return.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to docs/**/*.rst : Add an '.. autofunction:: pythermalcomfort.models.<module>.<func>' entry to the docs for public APIs
Applied to files:
docs/documentation/models.rst
🧬 Code graph analysis (3)
pythermalcomfort/models/ireq.py (1)
pythermalcomfort/classes_return.py (1)
IREQ(1049-1074)
pythermalcomfort/classes_input.py (1)
pythermalcomfort/models/jos3.py (10)
tdb(1403-1414)tdb(1417-1418)tr(1421-1423)tr(1426-1427)v(1469-1471)v(1474-1475)rh(1460-1462)rh(1465-1466)clo(1505-1507)clo(1510-1511)
tests/test_ireq.py (1)
pythermalcomfort/models/ireq.py (1)
calc_ireq(6-177)
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (1)
CHANGELOG.rst(1 hunks)
🧰 Additional context used
📓 Path-based instructions (1)
CHANGELOG.*
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
Update the CHANGELOG for public API changes
Files:
CHANGELOG.rst
🧠 Learnings (4)
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
Applied to files:
CHANGELOG.rst
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/{models,utilities}.py : Ensure docstring sections include Args (with units), Returns, Raises, Examples, Applicability
Applied to files:
CHANGELOG.rst
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
CHANGELOG.rst
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Applied to files:
CHANGELOG.rst
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (1)
pythermalcomfort/classes_return.py (1)
1048-1074: Docstring prose still inconsistent with ArrayLike type hints.Several attribute descriptions (lines 1057, 1059, 1061, 1065) still say "float or list of floats" while the type annotations correctly use
npt.ArrayLike. This creates confusion since the runtime actually returns NumPy arrays for vectorised inputs. Lines 1055 and 1063 already use the correct prose ("float, list[float], or numpy.ndarray"), so the remaining fields should match that pattern for consistency.Apply this diff to align the docstring with the type hints:
- ireq_neutral : float or list of floats + ireq_neutral : float, list[float], or numpy.ndarray Upper bound of required insulation [clo]. - icl_min : float or list of floats + icl_min : float, list[float], or numpy.ndarray Lower bound of required basic clothing insulation (ISO 9920) [clo]. - icl_neutral : float or list of floats + icl_neutral : float, list[float], or numpy.ndarray Upper bound of required basic clothing insulation (ISO 9920) [clo]. - dle_neutral : float or list of floats or str + dle_neutral : float, numpy.ndarray, or str Upper bound of duration limited exposure [hours] (values > 8 represented as "more than 8").
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (2)
pythermalcomfort/classes_return.py(1 hunks)pythermalcomfort/models/ireq.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
pythermalcomfort/**/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/**/*.py: Use type hints and NumPy-style docstrings for all library code
Validate inputs in dataclass post_init (call super().post_init() if inheriting BaseInputs)
Convert pandas Series to list/np.array and normalize numeric inputs with np.asarray in post_init
Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Use numpy operations for numerical work (e.g., np.log, np.asarray) and avoid math.log when arrays are accepted
Validate domains before logs/roots (arguments must be > 0)
Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Run ruff (lint/format) and docformatter; ensure code is formatted before PR
When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Files:
pythermalcomfort/classes_return.pypythermalcomfort/models/ireq.py
pythermalcomfort/models/*.py
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
pythermalcomfort/models/*.py: Place new domain model functions in pythermalcomfort/models/.py
Include applicability limits and minimal runnable examples in model function docstrings
Files:
pythermalcomfort/models/ireq.py
🧠 Learnings (15)
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Follow repository patterns: dataclasses, BaseInputs metadata, validate_type, Units/Enums (Postures, WorkIntensity, Sex)
Applied to files:
pythermalcomfort/classes_return.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/classes_input.py : Add or update input dataclasses in pythermalcomfort/classes_input.py (or adjacent to the function when appropriate)
Applied to files:
pythermalcomfort/classes_return.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use type hints and NumPy-style docstrings for all library code
Applied to files:
pythermalcomfort/classes_return.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Support vectorized behavior; use np.atleast_1d, np.broadcast_to, and check broadcast compatibility
Applied to files:
pythermalcomfort/classes_return.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : When accepting arrays, ensure computations are vectorized and avoid per-element Python loops
Applied to files:
pythermalcomfort/classes_return.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Functions should be small, single-purpose, and include concise, runnable examples in NumPy-style docstrings
Applied to files:
pythermalcomfort/classes_return.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/{models,utilities}.py : Ensure docstring sections include Args (with units), Returns, Raises, Examples, Applicability
Applied to files:
pythermalcomfort/classes_return.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Include applicability limits and minimal runnable examples in model function docstrings
Applied to files:
pythermalcomfort/classes_return.pypythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Run ruff (lint/format) and docformatter; ensure code is formatted before PR
Applied to files:
pythermalcomfort/classes_return.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/models/*.py : Place new domain model functions in pythermalcomfort/models/<name>.py
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Enforce physical constraints in validation (e.g., non-negativity, z2>z0, z1>z0, z2>z1) with clear error messages
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate inputs in dataclass __post_init__ (call super().__post_init__() if inheriting BaseInputs)
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to tests/**/*.py : Write tests covering scalars, arrays, broadcasting, and invalid inputs (TypeError/ValueError)
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Validate domains before logs/roots (arguments must be > 0)
Applied to files:
pythermalcomfort/models/ireq.py
📚 Learning: 2025-09-15T01:23:02.859Z
Learnt from: CR
Repo: CenterForTheBuiltEnvironment/pythermalcomfort PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-09-15T01:23:02.859Z
Learning: Applies to pythermalcomfort/**/*.py : Use validate_type(...) for type checks; raise TypeError for wrong types and ValueError for invalid values
Applied to files:
pythermalcomfort/models/ireq.py
🧬 Code graph analysis (1)
pythermalcomfort/models/ireq.py (1)
pythermalcomfort/classes_return.py (1)
IREQ(1049-1074)
🔇 Additional comments (9)
pythermalcomfort/models/ireq.py (9)
1-16: Imports and function signature look good.The imports are clean with no unused dependencies (pandas/IREQInputs were correctly removed in earlier commits). Type hints are present and appropriate for vectorised inputs, and the function signature clearly indicates required vs. optional parameters.
17-142: Comprehensive NumPy-style docstring with Applicability section.The docstring properly documents the ISO 11079 model with all required sections: Parameters (with units), Returns, Raises, Examples (both scalar and vectorised), Applicability (with validated ranges), Notes, and References. Runnable examples demonstrate the expected usage patterns. Well done following the repository guidelines.
Based on learnings.
144-193: Main function body is well-structured and validates required parameters.The function explicitly checks that
tdbandtrare provided (lines 145-148), converts all inputs to NumPy arrays consistently, validates shapes and ranges through dedicated helpers, and packages results into the structured IREQ dataclass. The logic is clean with good separation of concerns.
195-203: Broadcasting validation correctly implemented.The shape validation properly uses
np.broadcast_shapesto check compatibility, allows scalars, and provides clear error messages when broadcasting fails. This follows the coding guidelines for supporting vectorised behaviour.Based on learnings.
205-266: Comprehensive input validation with finite checks and physical constraints.All inputs are validated for finite values (preventing NaN/Inf issues) and physical constraints matching the ISO 11079 limits. The
p_air > 0check properly guards the log() domain used later in the calculation. Error messages clearly indicate which constraint was violated.Based on learnings.
268-324: Calculation helpers are properly vectorised and single-purpose.All helper functions use NumPy operations for vectorised calculations (avoiding per-element loops) and serve single purposes. The
_calculate_heat_transfer_coefficientsfunction appropriately guards against division by zero (line 313). While the docstrings are minimal one-liners, this is acceptable for private utility functions.Based on learnings.
327-383: Bisection solver for IREQ correctly implements classical algorithm with proper bounds.The iterative solver properly initialises explicit lower (0) and upper (2.0 m²·K/W) bounds, performs classical bisection by tightening the bracket based on the residual sign, and recomputes the final energy balance with the converged
ireqvalue. This addresses previous concerns about hard-capping the solution and using stale values.
388-447: DLE iteration solver with proper zero guard.The iterative solver for Duration Limited Exposure (DLE) correctly guards against division by zero (lines 441-446) by replacing near-zero
q_storevalues with a small non-zero value that preserves sign information. The guard ensures numerical stability when the heat storage rate approaches zero.
505-509: Return structure properly packages results in user-friendly units.The function returns a well-structured dictionary with IREQ and ICL values in clo units and DLE in hours, ready for consumption by the main
calc_ireqfunction to construct the IREQ dataclass.
3 participants
Description
This pull request introduces the implementation of the IREQ (Required Clothing Insulation) model according to ISO 11079.
The update includes:
Type of change
New feature (non-breaking change which adds functionality)
This change requires a documentation update
How Has This Been Tested?
All tests for the calc_ireq() function were conducted in tests/test_ireq.py using pytest.
Test A: Verified scalar input calculations to ensure numerical stability and physically valid outputs.
Test B: Verified vectorized and broadcasting input handling with NumPy arrays to ensure consistent output shapes and array compatibility.
Test Configuration:
Firmware version: N/A
Hardware: macOS (M1)
Toolchain: Python 3.11, pytest 8.4.1
SDK: N/A
Checklist
Summary by CodeRabbit
New Features
Bug Fixes / Quality
Documentation
Tests