datapruning 1.0.7__tar.gz → 1.0.8__tar.gz

{datapruning-1.0.7/datapruning.egg-info → datapruning-1.0.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datapruning
-Version: 1.0.7
+Version: 1.0.8
 Summary: Intelligent data pruning for ML datasets
 License: Proprietary
 Project-URL: Homepage, https://www.datapruning.com

datapruning-1.0.8/datapruning/engine/strategy_display.py

@@ -0,0 +1,111 @@
+STRATEGY_DISPLAY = {
+    "random": {
+        "display_name": "Fast Reduction",
+        "icon": "\u26a1",
+        "description": "Quickly reduces dataset size by randomly selecting a subset of samples.",
+        "best_for": "Large, balanced datasets where speed is the priority",
+        "advantages": [
+            "Extremely fast execution",
+            "Low memory usage",
+            "Stable and predictable",
+        ],
+        "tradeoffs": [
+            "Random selection may discard important samples",
+            "No class imbalance awareness",
+        ],
+        "recommended_models": ["LightGBM", "XGBoost", "Random Forest"],
+    },
+    "iht": {
+        "display_name": "Boundary Optimization",
+        "icon": "\ud83c\udfaf",
+        "description": "Preserves difficult-to-classify examples near decision boundaries.",
+        "best_for": "Fraud detection, rare event prediction, imbalanced classification",
+        "advantages": [
+            "Preserves difficult examples",
+            "Better decision boundaries",
+            "Model-informed selection",
+        ],
+        "tradeoffs": [
+            "Slightly slower than random methods",
+        ],
+        "recommended_models": ["XGBoost", "LightGBM", "CatBoost"],
+    },
+    "cluster": {
+        "display_name": "Representative Compression",
+        "icon": "\ud83d\udcca",
+        "description": "Creates a compact representative subset by clustering the dataset.",
+        "best_for": "Very large datasets needing maximum compression",
+        "advantages": [
+            "Keeps representative patterns",
+            "Good compression ratios",
+        ],
+        "tradeoffs": [
+            "May smooth out local structure",
+        ],
+        "recommended_models": ["LightGBM", "Random Forest"],
+    },
+    "density": {
+        "display_name": "Structural Optimization",
+        "icon": "\ud83e\udde9",
+        "description": "Removes redundant samples in high-density regions while preserving structure.",
+        "best_for": "General tabular datasets with clear structure",
+        "advantages": [
+            "Structure-aware selection",
+            "Stable results",
+        ],
+        "tradeoffs": [
+            "Experimental on highly imbalanced data",
+        ],
+        "recommended_models": ["XGBoost", "CatBoost", "LightGBM"],
+    },
+    "duplicate_aware": {
+        "display_name": "Duplicate Cleanup",
+        "icon": "\ud83e\uddf9",
+        "description": "Identifies and removes duplicate or near-duplicate records.",
+        "best_for": "Datasets with many repeated or near-identical records",
+        "advantages": [
+            "Removes exact and near duplicates",
+            "Very fast on most datasets",
+        ],
+        "tradeoffs": [
+            "Limited effect on already clean datasets",
+        ],
+        "recommended_models": ["Any \u2014 data quality improvement"],
+    },
+}
+
+
+def get_strategy_display(key: str) -> dict:
+    return dict(STRATEGY_DISPLAY.get(key, {
+        "display_name": key.replace("_", " ").title(),
+        "icon": "\ud83d\udd27",
+        "description": "",
+        "best_for": "",
+        "advantages": [],
+        "tradeoffs": [],
+        "recommended_models": [],
+    }))
+
+
+def enrich_algorithm_info(algo: dict) -> dict:
+    display = get_strategy_display(algo.get("key", ""))
+    return {**algo, **display}
+
+
+def enrich_recommendation(rec: dict) -> dict:
+    key = rec.get("algorithm_key", "")
+    display = get_strategy_display(key)
+    enriched = {
+        "friendly_name": display["display_name"],
+        "icon": display["icon"],
+        "description": display["description"],
+        "best_for": display["best_for"],
+        "advantages": display["advantages"],
+        "tradeoffs": display["tradeoffs"],
+        "recommended_models": display["recommended_models"],
+    }
+    return {**rec, **enriched}
+
+
+def enrich_alternatives(alternatives: list) -> list:
+    return [enrich_algorithm_info(alt) for alt in alternatives]

{datapruning-1.0.7 → datapruning-1.0.8/datapruning.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datapruning
-Version: 1.0.7
+Version: 1.0.8
 Summary: Intelligent data pruning for ML datasets
 License: Proprietary
 Project-URL: Homepage, https://www.datapruning.com

{datapruning-1.0.7 → datapruning-1.0.8}/datapruning.egg-info/SOURCES.txt

@@ -26,6 +26,7 @@ datapruning/engine/explainability.py
 datapruning/engine/intelligence.py
 datapruning/engine/optimizer.py
 datapruning/engine/scanner.py
+datapruning/engine/strategy_display.py
 datapruning/engine/strategy_selector.py
 datapruning/reports/__init__.py
 datapruning/reports/exporter.py

{datapruning-1.0.7 → datapruning-1.0.8}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "datapruning"
-version = "1.0.7"
+version = "1.0.8"
 description = "Intelligent data pruning for ML datasets"
 readme = "README.md"
 requires-python = ">=3.10"