Refactor models list

[alirezazolanvari]

• Centralize models registries in params.py
• Replace abstractmethods in DrugReleaseModel using configs from models registries
• Remove parameter dataclasses of the models
• Set default value 1 for the Entire releasable amount (M_0/M) in the respective models
• Reformat by black

[codecov-commenter]

Codecov Report

❌ Patch coverage is 90.99099% with 10 lines in your changes missing coverage. Please review.
✅ Project coverage is 94.12%. Comparing base (0e582b3) to head (6dc9537).
⚠️ Report is 1 commits behind head on dev.

Files with missing lines	Patch %	Lines
drux/base_model.py	86.67%	3 Missing and 3 partials ⚠️
drux/registry.py	91.84%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##              dev      #39      +/-   ##
==========================================
- Coverage   98.73%   94.12%   -4.60%     
==========================================
  Files           9       10       +1     
  Lines         235      187      -48     
  Branches       26       21       -5     
==========================================
- Hits          232      176      -56     
- Misses          2        7       +5     
- Partials        1        4       +3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[codacy-production]

Not up to standards ⛔

🔴 Issues 10 minor

Alerts:
⚠ 10 issues (≤ 0 issues of at least minor severity)

Results:
10 new issues

Category	Results
Documentation	8 minor
CodeStyle	2 minor

View in Codacy

🟢 Metrics -3 complexity · 0 duplication

Metric	Results
Complexity	-3
Duplication	0

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}

[alirezazolanvari]

@sepandhaghighi, I revised the changes that were proposed in this PR. The summary of changes and the main idea/motivation behind them are presented below. It might look like more than a single PR, so feel free to let me know to split it into different PRs.

New Architecture

Motivation

Keeping registry info of all the models centralized to be able to build on it for future developments.

File Overview

File	Role	Key Exports
registry.py	Single source of truth: equations, params, validation	@equation, register_validation(), register_params(), get_model_config()
messages.py	Error message constants	ERROR_* string constants
base_model.py	Base class + class factory	DrugReleaseModel, create_model_class()
<model>.py	Model entry points	e.g. ZeroOrderModel, WeibullModel
__init__.py	Public API surface	Re-exports all model classes + version

Dependency Graph

__init__.py
 ├── registry.py ── messages.py
 ├── <model>.py ─── base_model.py
 │                   ├── registry.py
 │                   └── messages.py

Runtime Flow

User calls: ZeroOrderModel(k0=0.5)
         │
         ▼
┌─────────────────────────┐
│  <model>.py              │  create_model_class("zero_order", ...)
│  (factory call)          │  ── runs at import time, produces class
└────────┬────────────────┘
         │  __init__(**kwargs)
         ▼
┌─────────────────────────┐
│  base_model.py           │  1. Applies defaults from registry
│  (generated class)       │  2. Builds SimpleNamespace for parameters
│                          │  3. Stores _model_name, _parameters, _label
└────────┬────────────────┘
         │  .simulate(duration, time_step)
         ▼
┌─────────────────────────┐
│  base_model.py           │  1. _validate_parameters() ← validation from registry
│  DrugReleaseModel        │  2. _model_function(t)     ← equation from registry
│                          │  3. Returns release profile (np.ndarray)
└─────────────────────────┘

Before → After

Aspect	Before	After
Equations	Inline lambda in params.py	Named functions in registry.py with @equation
Validation	Inline lambda + message imports in params.py	register_validation() calls co-located with equations
Params metadata	Separate params.py assembling MODELS_REGISTRY	register_params() calls co-located with equations
Parameter storage	Dynamic make_dataclass via utils.py	SimpleNamespace inline in base_model.py
Model classes	~25-line hand-written class per model	1-line create_model_class() call per model
Defaults	Duplicated in registry and __init__ signature	Registry only (single source)
Base class	ABC with no abstract methods	Plain class
Files touched to add a model	5+	2 (see below)
Total core files	8	4

Adding a New Model

All three aspects (params, equation, validation) are added in registry.py under one section:

registry.py                              <model>.py
─────────────                            ─────────────
register_params("name", {...})           from .base_model import create_model_class

@equation("name")                        NameModel = create_model_class(
def name(p, t):                              model_name="name",
    return ...                               class_name="NameModel",
                                             label="Name Model",
register_validation("name", {...})       )

Then add to __init__.py:

from .name import NameModel

Checklist

#	File	What to add
1	messages.py	Error string constants (if new ones needed)
2	registry.py	register_params() + @equation function + register_validation()
3	<model>.py	Single create_model_class() call (can be done in __init__.py)
4	__init__.py	Import the new model class

[alirezazolanvari]

Requires more thought on how to implement it according to the best practices. It will be postponed to version 0.5.