Refactor repository

README.md

@@ -1,6 +1,6 @@
-# MicroImpute
+# Microimpute
 
-MicroImpute enables variable imputation through different statistical methods. It facilitates comparison and benchmarking across methods through quantile loss calculations.
+Microimpute enables variable imputation through different statistical methods. It facilitates comparison and benchmarking across methods through quantile loss calculations.
 
 To install, run pip install microimpute.
 

changelog_entry.yaml

@@ -0,0 +1,7 @@
+- bump: patch
+  changes:
+    changed:
+    - Refactor the Imputer base class.
+    - Refactor cross-validation, comparison, visualization, and autoimputation functions.
+    added:
+    - Improved test coverage.

microimpute/__init__.py

@@ -1,11 +1,28 @@
-"""MicroImpute Package
+"""Microimpute: advanced statistical imputation for microdata
 
-A package for benchmarking different imputation methods using microdata.
+Microimpute is a comprehensive Python package for benchmarking and applying
+various statistical imputation methods to microdata. It provides tools for
+automated model selection, cross-validation, and visualization of imputation results.
+
+Key features:
+    - Multiple imputation models (OLS, QRF, QuantReg, Matching)
+    - Automated model selection with cross-validation
+    - Quantile-based imputation and evaluation
+    - Comprehensive visualization tools
+    - Support for weighted imputation
+    - Hyperparameter tuning capabilities
+
+Main components:
+    - autoimpute: automated imputation with model selection
+    - Models: OLS, QRF, QuantReg, Matching (optional)
+    - Evaluation: cross-validation and quantile loss metrics
+    - Visualization: performance and comparison plots
 """
 
 __version__ = "1.1.2"
 
-from microimpute.comparisons.autoimpute import autoimpute
+# Import automated imputation
+from microimpute.comparisons.autoimpute import AutoImputeResult, autoimpute
 from microimpute.comparisons.imputations import get_imputations
 
 # Import comparison utilities
@@ -15,6 +32,14 @@
     quantile_loss,
 )
 
+# Import validation utilities
+from microimpute.comparisons.validation import (
+    validate_columns_exist,
+    validate_dataframe_compatibility,
+    validate_imputation_inputs,
+    validate_quantiles,
+)
+
 # Main configuration
 from microimpute.config import (
     PLOT_CONFIG,
@@ -30,15 +55,17 @@
 from microimpute.models import OLS, QRF, Imputer, ImputerResults, QuantReg
 
 # Import data handling functions
-from microimpute.utils.data import preprocess_data
+from microimpute.utils.data import preprocess_data, unnormalize_predictions
 
 try:
     from microimpute.models.matching import Matching
 except ImportError:
     pass
 
 # Import visualization modules
-from microimpute.visualizations.plotting import (
+from microimpute.visualizations import (
+    MethodComparisonResults,
+    PerformanceResults,
     method_comparison_results,
     model_performance_results,
 )

microimpute/comparisons/__init__.py

@@ -1,17 +1,34 @@
-"""Data Comparison Utilities
+"""Imputation method comparison and evaluation utilities
 
-This module contains utilities for comparing different imputation methods.
+This module provides comprehensive tools for comparing and evaluating different
+imputation methods. It includes automated model selection, quantile loss metrics,
+and validation utilities for ensuring data integrity.
+
+Key components:
+    - autoimpute: automated imputation method selection and application
+    - get_imputations: generate imputations using multiple model classes
+    - quantile_loss: calculate quantile-based loss metrics
+    - compare_quantile_loss: compare performance across imputation methods
+    - Validation utilities for data and parameter validation
 """
 
 # Import automated imputation utilities
-from .autoimpute import autoimpute
+from microimpute.comparisons.autoimpute import AutoImputeResult, autoimpute
 
 # Import imputation utilities
-from .imputations import get_imputations
+from microimpute.comparisons.imputations import get_imputations
 
 # Import loss functions
-from .quantile_loss import (
+from microimpute.comparisons.quantile_loss import (
     compare_quantile_loss,
     compute_quantile_loss,
     quantile_loss,
 )
+
+# Import validation utilities
+from microimpute.comparisons.validation import (
+    validate_columns_exist,
+    validate_dataframe_compatibility,
+    validate_imputation_inputs,
+    validate_quantiles,
+)