microlens-submit 0.16.5__tar.gz → 0.17.0__tar.gz

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/CHANGELOG.md

@@ -5,6 +5,34 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.17.0] - 2026-02-04
+
+### Added
+- **Physical Parameters**: Added 24 physical parameters (Mtot, M1-M4, D_L, D_S, thetaE, piE components, mu_rel components, phi) to parameter spec
+- **Physical Parameter Validation**: Comprehensive validation including mass consistency checks, vector magnitude verification, distance constraints, and unit confusion warnings
+- **Uncertainty Metadata**: Added three new Solution fields:
+  - `uncertainty_method`: Method used to derive uncertainties (mcmc_posterior, fisher_matrix, bootstrap, propagation, inference, literature, other)
+  - `confidence_level`: Confidence level for uncertainties (default: 0.68 for 1-sigma)
+  - `physical_parameter_uncertainties`: Uncertainties for physical parameters (symmetric or asymmetric)
+- **CLI Options**: Added `--physical-param-uncertainty`, `--uncertainty-method`, and `--confidence-level` options to `add-solution` command
+- **Tests**: Added 3 new comprehensive BIC calculation tests verifying parameter counting
+- **Version Management**: Enhanced `bump_version.py` to update and validate `parameter_spec.yaml` (includes drift check)
+
+### Fixed
+- **BIC Calculation Bug**: Fixed critical bug where BIC calculation counted ALL parameters including metadata (t_ref, limb_darkening_coeffs) and physical parameters (Mtot, D_L, etc.) as "free parameters". Now correctly counts only fitted model parameters using new `count_model_parameters()` function. This affects relative probability calculations during export and solution comparison.
+
+### Changed
+- Physical parameters now validated with `validate_physical_parameters()` automatically when present
+- Solution metadata validation now includes uncertainty metadata checks
+- BIC calculation in `submission.py` and `validation.py` now uses `count_model_parameters()` instead of `len(s.parameters)`
+
+### Documentation
+- Added comprehensive examples of physical parameters and uncertainty metadata to tutorial and usage examples
+- Tutorial emphasizes uncertainties are **optional but strongly recommended** for evaluation readiness
+- Created `PHYSICAL_PARAMS_SUMMARY.md` with complete implementation guide
+
+## [Unreleased]
+
 ## [0.16.5] - 2026-02-04
 
 ### Fixed

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/CITATION.cff

@@ -1,9 +1,9 @@
 cff-version: 1.2.0
 message: "If you use microlens-submit, please cite it as below."
 title: "microlens-submit"
-version: "0.16.5"
+version: "0.17.0"
 authors:
   - family-names: Malpas
     given-names: Amber
 url: "https://github.com/AmberLee2427/microlens-submit"
-doi: "10.5281/zenodo.18246117"
+doi: "10.5281/zenodo.18488304"

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: microlens-submit
-Version: 0.16.5
+Version: 0.17.0
 Summary: A tool for managing and submitting microlensing solutions
 Home-page: https://github.com/AmberLee2427/microlens-submit
 Author: Amber Malpas
@@ -210,15 +210,15 @@ import this file directly.
 
 Bibtex:
 ```
-@software{malpas_2025_18246117,
+@software{malpas_2025_18488304,
   author       = {Malpas, Amber},
   title        = {microlens-submit},
   month        = oct,
   year         = 2025,
   publisher    = {Zenodo},
   version      = {v0.16.3},
-  doi          = {10.5281/zenodo.18246117},
-  url          = {https://doi.org/10.5281/zenodo.18246117},
+  doi          = {10.5281/zenodo.18488304},
+  url          = {https://doi.org/10.5281/zenodo.18488304},
 }
 ```
 

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/README.md

@@ -160,15 +160,15 @@ import this file directly.
 
 Bibtex:
 ```
-@software{malpas_2025_18246117,
+@software{malpas_2025_18488304,
   author       = {Malpas, Amber},
   title        = {microlens-submit},
   month        = oct,
   year         = 2025,
   publisher    = {Zenodo},
   version      = {v0.16.3},
-  doi          = {10.5281/zenodo.18246117},
-  url          = {https://doi.org/10.5281/zenodo.18246117},
+  doi          = {10.5281/zenodo.18488304},
+  url          = {https://doi.org/10.5281/zenodo.18488304},
 }
 ```
 

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/microlens_submit/__init__.py

@@ -5,7 +5,7 @@ validate, and export a challenge submission using either the Python API or
 the command line interface.
 """
 
-__version__ = "0.16.5"
+__version__ = "0.17.0"
 
 from .models import Event, Solution, Submission
 from .utils import load

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/microlens_submit/cli/commands/solutions.py

@@ -187,7 +187,31 @@ def add_solution(
     physical_param: Optional[List[str]] = typer.Option(
         None,
         "--physical-param",
-        help=("Physical parameters (M_L, D_L, M_planet, a, etc.) " "derived from model parameters [ADVANCED]"),
+        help=("Physical parameters (Mtot, D_L, thetaE, etc.) " "derived from model parameters [ADVANCED]"),
+    ),
+    physical_param_uncertainty: Optional[List[str]] = typer.Option(
+        None,
+        "--physical-param-uncertainty",
+        help=(
+            "Physical parameter uncertainties as key=value. "
+            "Can be single value (symmetric) or [lower,upper] (asymmetric) [ADVANCED]"
+        ),
+    ),
+    uncertainty_method: Optional[str] = typer.Option(
+        None,
+        "--uncertainty-method",
+        help=(
+            "Method used to derive uncertainties [ADVANCED]. "
+            "Options: mcmc_posterior, fisher_matrix, bootstrap, propagation, inference, literature, other"
+        ),
+    ),
+    confidence_level: Optional[float] = typer.Option(
+        None,
+        "--confidence-level",
+        help=(
+            "Confidence level for uncertainties (default: 0.68 = 1σ) [ADVANCED]. "
+            "Standard values: 0.68 (1σ), 0.95 (2σ), 0.997 (3σ)"
+        ),
     ),
     relative_probability: Optional[float] = typer.Option(
         None,
@@ -281,6 +305,10 @@ def add_solution(
     sol.limb_darkening_coeffs = _parse_pairs(limb_darkening_coeff)
     sol.parameter_uncertainties = _parse_pairs(parameter_uncertainty) or uncertainties
     sol.physical_parameters = _parse_pairs(physical_param)
+    sol.physical_parameter_uncertainties = _parse_pairs(physical_param_uncertainty)
+    sol.uncertainty_method = uncertainty_method
+    if confidence_level is not None:
+        sol.confidence_level = confidence_level
     sol.log_likelihood = log_likelihood
     sol.relative_probability = relative_probability
     sol.n_data_points = n_data_points

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/microlens_submit/cli/commands/validation.py

@@ -12,6 +12,7 @@ from rich.table import Table
 from microlens_submit.error_messages import enhance_validation_messages
 from microlens_submit.text_symbols import symbol
 from microlens_submit.utils import load
+from microlens_submit.validate_parameters import count_model_parameters
 
 console = Console()
 
@@ -196,13 +197,17 @@ def compare_solutions(
         need_calc = [s for s in solutions if s.relative_probability is None]
         if need_calc:
             can_calc = all(
-                s.log_likelihood is not None and s.n_data_points and s.n_data_points > 0 and len(s.parameters) > 0
+                s.log_likelihood is not None
+                and s.n_data_points
+                and s.n_data_points > 0
+                and count_model_parameters(s.parameters) > 0
                 for s in need_calc
             )
             remaining = max(1.0 - provided_sum, 0.0)
             if can_calc:
                 bic_vals = {
-                    s.solution_id: len(s.parameters) * math.log(s.n_data_points) - 2 * s.log_likelihood
+                    s.solution_id: count_model_parameters(s.parameters) * math.log(s.n_data_points)
+                    - 2 * s.log_likelihood
                     for s in need_calc
                 }
                 bic_min = min(bic_vals.values())
@@ -219,7 +224,7 @@ def compare_solutions(
 
     rows = []
     for sol in solutions:
-        k = len(sol.parameters)
+        k = count_model_parameters(sol.parameters)
         bic = k * math.log(sol.n_data_points) - 2 * sol.log_likelihood
         rp = sol.relative_probability if sol.relative_probability is not None else rel_prob_map.get(sol.solution_id)
         rows.append(

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/microlens_submit/models/solution.py

@@ -134,7 +134,10 @@ class Solution(BaseModel):
     limb_darkening_model: Optional[str] = None
     limb_darkening_coeffs: Optional[dict] = None
     parameter_uncertainties: Optional[dict] = None
+    uncertainty_method: Optional[str] = None
+    confidence_level: Optional[float] = 0.68
     physical_parameters: Optional[dict] = None
+    physical_parameter_uncertainties: Optional[dict] = None
     log_likelihood: Optional[float] = None
     relative_probability: Optional[float] = None
     n_data_points: Optional[int] = None
@@ -155,6 +158,7 @@ class Solution(BaseModel):
             higher_order_effects = values.get("higher_order_effects", [])
             bands = values.get("bands", [])
             t_ref = values.get("t_ref")
+            limb_darkening_coeffs = values.get("limb_darkening_coeffs")
 
             # Only check for totally broken objects (e.g., wrong types)
             basic_errors = []
@@ -176,6 +180,7 @@ class Solution(BaseModel):
                 higher_order_effects=higher_order_effects,
                 bands=bands,
                 t_ref=t_ref,
+                limb_darkening_coeffs=limb_darkening_coeffs,
             )
             if validation_warnings:
                 warnings.warn(f"Solution created with potential issues: {'; '.join(validation_warnings)}", UserWarning)
@@ -388,6 +393,18 @@ class Solution(BaseModel):
         )
         messages.extend(consistency_messages)
 
+        # Check solution metadata (uncertainties, etc.)
+        from ..validate_parameters import validate_solution_metadata
+
+        metadata_messages = validate_solution_metadata(
+            parameter_uncertainties=self.parameter_uncertainties,
+            physical_parameters=self.physical_parameters,
+            physical_parameter_uncertainties=self.physical_parameter_uncertainties,
+            uncertainty_method=self.uncertainty_method,
+            confidence_level=self.confidence_level,
+        )
+        messages.extend(metadata_messages)
+
         return messages
 
     def _save(self, event_path: Path) -> None:

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/microlens_submit/models/submission.py

@@ -18,6 +18,7 @@ import psutil
 from pydantic import BaseModel, Field
 
 from ..text_symbols import symbol
+from ..validate_parameters import count_model_parameters
 from .event import Event
 from .solution import Solution
 
@@ -500,14 +501,15 @@ class Submission(BaseModel):
                                 s.log_likelihood is None
                                 or s.n_data_points is None
                                 or s.n_data_points <= 0
-                                or len(s.parameters) == 0
+                                or count_model_parameters(s.parameters) == 0
                             ):
                                 can_calc = False
                                 break
                         remaining = max(1.0 - provided_sum, 0.0)
                         if can_calc:
                             bic_vals = {
-                                s.solution_id: len(s.parameters) * math.log(s.n_data_points) - 2 * s.log_likelihood
+                                s.solution_id: count_model_parameters(s.parameters) * math.log(s.n_data_points)
+                                - 2 * s.log_likelihood
                                 for s in need_calc
                             }
                             bic_min = min(bic_vals.values())

{microlens_submit-0.16.5 → microlens_submit-0.17.0}/microlens_submit/tier_validation.py

@@ -38,21 +38,21 @@ Note:
 from typing import Dict, List, Optional, Set
 
 # Tier definitions with their associated event lists
+# --- BEGIN AUTO-GENERATED: TIER_DEFINITIONS ---
 TIER_DEFINITIONS = {
     "beginner": {
         "description": "Beginner challenge tier with limited event set",
         "event_prefix": "rmdc26_",
-        "event_range": [0, 200],
+        "event_range": [0, 253],
     },
     "experienced": {
         "description": "Experienced challenge tier with full event set",
         "event_prefix": "rmdc26_",
-        "event_range": [0, 2000],
+        "event_range": [0, 3000],
     },
     "test": {
         "description": "Testing tier for development",
         "event_list": [
-            # Add test events here
             "evt",
             "test-event",
             "EVENT001",
@@ -68,17 +68,14 @@ TIER_DEFINITIONS = {
         "description": "2018 test events tier",
         "event_prefix": "ulwdc1_",
         "event_range": [0, 293],
-        "event_list": [
-            # Add 2018 test events here
-            "2018-EVENT-001",
-            "2018-EVENT-002",
-        ],
+        "event_list": ["2018-EVENT-001", "2018-EVENT-002"],
     },
     "None": {
         "description": "No validation tier (skips event validation)",
-        "event_list": [],  # Empty list means no validation
+        "event_list": [],
     },
 }
+# --- END AUTO-GENERATED: TIER_DEFINITIONS ---
 
 # Cache for event lists to avoid repeated list creation
 _EVENT_LIST_CACHE: Dict[str, Set[str]] = {}