churnkit 0.76.1a1__py3-none-any.whl → 0.76.1a2__py3-none-any.whl

{churnkit-0.76.1a1.data → churnkit-0.76.1a2.data}/data/share/churnkit/exploration_notebooks/07_modeling_readiness.ipynb

@@ -82,18 +82,19 @@
    "outputs": [],
    "source": [
     "from customer_retention.analysis.notebook_progress import track_and_export_previous\n",
+    "\n",
     "track_and_export_previous(\"07_modeling_readiness.ipynb\")\n",
     "\n",
-    "from customer_retention.analysis.auto_explorer import ExplorationFindings\n",
-    "from customer_retention.analysis.visualization import ChartBuilder, display_figure, display_table\n",
-    "from customer_retention.core.config.column_config import ColumnType\n",
-    "from customer_retention.stages.modeling import ImbalanceRecommender, ImbalanceHandler, ImbalanceStrategy\n",
     "import pandas as pd\n",
-    "import numpy as np\n",
     "import plotly.graph_objects as go\n",
-    "import plotly.express as px\n",
-    "from plotly.subplots import make_subplots\n",
-    "from customer_retention.core.config.experiments import FINDINGS_DIR, EXPERIMENTS_DIR, OUTPUT_DIR, setup_experiments_structure"
+    "\n",
+    "from customer_retention.analysis.auto_explorer import ExplorationFindings\n",
+    "from customer_retention.analysis.visualization import ChartBuilder, display_figure\n",
+    "from customer_retention.core.config.column_config import ColumnType\n",
+    "from customer_retention.core.config.experiments import (\n",
+    "    FINDINGS_DIR,\n",
+    ")\n",
+    "from customer_retention.stages.modeling import ImbalanceRecommender"
    ]
   },
   {
@@ -150,7 +151,7 @@
     "findings = ExplorationFindings.load(FINDINGS_PATH)\n",
     "\n",
     "# Load data - handle aggregated vs standard paths\n",
-    "from customer_retention.stages.temporal import load_data_with_snapshot_preference, TEMPORAL_METADATA_COLS\n",
+    "from customer_retention.stages.temporal import TEMPORAL_METADATA_COLS, load_data_with_snapshot_preference\n",
     "\n",
     "# For aggregated data, load directly from the parquet source\n",
     "if \"_aggregated\" in FINDINGS_PATH and findings.source_path.endswith('.parquet'):\n",
@@ -217,11 +218,11 @@
     "has_target = findings.target_column is not None\n",
     "checklist.append({\"Check\": \"Target column identified\", \"Status\": \"Pass\" if has_target else \"Fail\"})\n",
     "\n",
-    "has_features = len([c for c in findings.columns.values() \n",
+    "has_features = len([c for c in findings.columns.values()\n",
     "                   if c.inferred_type not in [ColumnType.IDENTIFIER, ColumnType.TARGET]]) > 0\n",
     "checklist.append({\"Check\": \"Feature columns available\", \"Status\": \"Pass\" if has_features else \"Fail\"})\n",
     "\n",
-    "high_missing = any(c.universal_metrics.get(\"null_percentage\", 0) > 50 \n",
+    "high_missing = any(c.universal_metrics.get(\"null_percentage\", 0) > 50\n",
     "                   for c in findings.columns.values())\n",
     "checklist.append({\"Check\": \"No columns with >50% missing\", \"Status\": \"Fail\" if high_missing else \"Pass\"})\n",
     "\n",
@@ -289,31 +290,31 @@
     "if findings.target_column:\n",
     "    target = findings.target_column\n",
     "    target_series = df[target]\n",
-    "    \n",
+    "\n",
     "    print(\"=\" * 70)\n",
     "    print(\"CLASS IMBALANCE ANALYSIS\")\n",
     "    print(\"=\" * 70)\n",
-    "    \n",
+    "\n",
     "    print(f\"\\nTarget Column: {target}\")\n",
     "    print(f\"Target Type: {findings.target_type}\")\n",
     "    print(f\"Missing Values: {target_series.isnull().sum()}\")\n",
-    "    \n",
+    "\n",
     "    if findings.target_type == \"binary\":\n",
     "        value_counts = target_series.value_counts()\n",
     "        majority_class = value_counts.idxmax()\n",
     "        minority_class = value_counts.idxmin()\n",
     "        majority_count = value_counts.max()\n",
     "        minority_count = value_counts.min()\n",
-    "        \n",
-    "        print(f\"\\n📊 CLASS DISTRIBUTION:\")\n",
+    "\n",
+    "        print(\"\\n📊 CLASS DISTRIBUTION:\")\n",
     "        print(f\"  Majority Class ({majority_class}): {majority_count:,} ({majority_count/len(df)*100:.1f}%)\")\n",
     "        print(f\"  Minority Class ({minority_class}): {minority_count:,} ({minority_count/len(df)*100:.1f}%)\")\n",
-    "        \n",
+    "\n",
     "        # Use framework recommender for strategy recommendations\n",
     "        recommender = ImbalanceRecommender()\n",
     "        rec = recommender.recommend(target_series, n_samples=len(df))\n",
     "        rec.print_recommendation()\n",
-    "        \n",
+    "\n",
     "        # Visualize\n",
     "        severity_colors = {\"low\": \"#2ca02c\", \"moderate\": \"#ffbb00\", \"high\": \"#ff7f0e\", \"severe\": \"#d62728\"}\n",
     "        fig = go.Figure(go.Bar(\n",
@@ -330,14 +331,14 @@
     "            template='plotly_white', height=400\n",
     "        )\n",
     "        display_figure(fig)\n",
-    "        \n",
+    "\n",
     "        # Show sklearn class weights\n",
-    "        print(f\"\\n💡 SKLEARN CLASS WEIGHTS:\")\n",
+    "        print(\"\\n💡 SKLEARN CLASS WEIGHTS:\")\n",
     "        weight_minority = len(df) / (2 * minority_count)\n",
     "        weight_majority = len(df) / (2 * majority_count)\n",
     "        print(f\"  class_weight={{0: {weight_majority:.3f}, 1: {weight_minority:.3f}}}\")\n",
-    "        print(f\"  Or use class_weight='balanced'\")\n",
-    "        \n",
+    "        print(\"  Or use class_weight='balanced'\")\n",
+    "\n",
     "        # Store recommendation for later use\n",
     "        imbalance_recommendation = rec\n",
     "else:\n",
@@ -397,11 +398,11 @@
     "\n",
     "if findings.target_column:\n",
     "    target = findings.target_column\n",
-    "    \n",
+    "\n",
     "    for col_name, col_info in findings.columns.items():\n",
     "        if col_name == target or col_info.inferred_type == ColumnType.IDENTIFIER or col_name in TEMPORAL_METADATA_COLS:\n",
     "            continue\n",
-    "        \n",
+    "\n",
     "        if col_info.inferred_type in [ColumnType.NUMERIC_CONTINUOUS, ColumnType.NUMERIC_DISCRETE]:\n",
     "            corr = df[[col_name, target]].corr().iloc[0, 1]\n",
     "            if abs(corr) > 0.9:\n",
@@ -410,7 +411,7 @@
     "                    \"Risk\": \"High\",\n",
     "                    \"Reason\": f\"Very high correlation ({corr:.3f}) - potential leakage\"\n",
     "                })\n",
-    "        \n",
+    "\n",
     "        if any(kw in col_name.lower() for kw in ['future', 'outcome', 'result', 'after']):\n",
     "            leakage_risks.append({\n",
     "                \"Column\": col_name,\n",
@@ -473,7 +474,7 @@
     "for col_type, count in sorted(type_summary.items()):\n",
     "    print(f\"  {col_type}: {count}\")\n",
     "\n",
-    "usable_features = sum(1 for c in findings.columns.values() \n",
+    "usable_features = sum(1 for c in findings.columns.values()\n",
     "                      if c.inferred_type not in [ColumnType.IDENTIFIER, ColumnType.TARGET])\n",
     "print(f\"\\nUsable features for modeling: {usable_features}\")"
    ]
@@ -593,10 +594,10 @@
     "            print(f\"   • {col} ({info.availability_type}, {info.coverage_pct:.0f}% coverage)\")\n",
     "    if len(findings.problematic_availability_columns) > 10:\n",
     "        print(f\"   ... and {len(findings.problematic_availability_columns) - 10} more\")\n",
-    "    \n",
+    "\n",
     "    action = findings.metadata.get(\"availability_action\", \"exclude\")\n",
     "    print(f\"\\n📋 Action: {action.upper()}\")\n",
-    "    print(f\"   These features will be excluded in notebook 08.\")\n",
+    "    print(\"   These features will be excluded in notebook 08.\")\n",
     "else:\n",
     "    print(\"\\n✅ All features have full temporal coverage.\")"
    ]
@@ -700,15 +701,15 @@
     "if 'X' in dir() and 'y' in dir():\n",
     "    detector = LeakageDetector()\n",
     "    result = detector.run_all_checks(X, y, include_pit=False)\n",
-    "    \n",
+    "\n",
     "    print(\"=\" * 70)\n",
     "    print(\"FINAL LEAKAGE VALIDATION\")\n",
     "    print(\"=\" * 70)\n",
-    "    \n",
+    "\n",
     "    if result.passed:\n",
     "        print(\"\\n✅ PASSED: No critical leakage issues\")\n",
     "        print(f\"   Checks run: {len(result.checks)}\")\n",
-    "        \n",
+    "\n",
     "        # Show warnings for HIGH severity issues\n",
     "        high_issues = [c for c in result.checks if c.severity.value == 'high']\n",
     "        if high_issues:\n",

{churnkit-0.76.1a1.data → churnkit-0.76.1a2.data}/data/share/churnkit/exploration_notebooks/08_baseline_experiments.ipynb

@@ -83,25 +83,29 @@
    "outputs": [],
    "source": [
     "from customer_retention.analysis.notebook_progress import track_and_export_previous\n",
+    "\n",
     "track_and_export_previous(\"08_baseline_experiments.ipynb\")\n",
     "\n",
-    "from customer_retention.analysis.auto_explorer import ExplorationFindings\n",
-    "from customer_retention.analysis.visualization import ChartBuilder, display_figure, display_table\n",
-    "from customer_retention.core.config.column_config import ColumnType\n",
     "import pandas as pd\n",
-    "import numpy as np\n",
-    "\n",
-    "from sklearn.model_selection import train_test_split, cross_val_score\n",
-    "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
-    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
+    "from sklearn.ensemble import GradientBoostingClassifier, RandomForestClassifier\n",
     "from sklearn.linear_model import LogisticRegression\n",
-    "from sklearn.metrics import (roc_auc_score, classification_report, confusion_matrix,\n",
-    "                             roc_curve, precision_recall_curve, average_precision_score,\n",
-    "                             f1_score, precision_score, recall_score)\n",
+    "from sklearn.metrics import (\n",
+    "    average_precision_score,\n",
+    "    classification_report,\n",
+    "    f1_score,\n",
+    "    precision_score,\n",
+    "    recall_score,\n",
+    "    roc_auc_score,\n",
+    ")\n",
+    "from sklearn.model_selection import cross_val_score, train_test_split\n",
+    "from sklearn.preprocessing import LabelEncoder, StandardScaler\n",
     "\n",
-    "import plotly.graph_objects as go\n",
-    "from plotly.subplots import make_subplots\n",
-    "from customer_retention.core.config.experiments import FINDINGS_DIR, EXPERIMENTS_DIR, OUTPUT_DIR, setup_experiments_structure"
+    "from customer_retention.analysis.auto_explorer import ExplorationFindings\n",
+    "from customer_retention.analysis.visualization import ChartBuilder, display_figure, display_table\n",
+    "from customer_retention.core.config.column_config import ColumnType\n",
+    "from customer_retention.core.config.experiments import (\n",
+    "    FINDINGS_DIR,\n",
+    ")"
    ]
   },
   {
@@ -158,7 +162,7 @@
     "findings = ExplorationFindings.load(FINDINGS_PATH)\n",
     "\n",
     "# Load data - handle aggregated vs standard paths\n",
-    "from customer_retention.stages.temporal import load_data_with_snapshot_preference, TEMPORAL_METADATA_COLS\n",
+    "from customer_retention.stages.temporal import TEMPORAL_METADATA_COLS, load_data_with_snapshot_preference\n",
     "\n",
     "# For aggregated data, load directly from the parquet source\n",
     "if \"_aggregated\" in FINDINGS_PATH and findings.source_path.endswith('.parquet'):\n",
@@ -269,7 +273,7 @@
     "    print(f\"   {col_type}: {count}\")\n",
     "\n",
     "# Show excluded columns\n",
-    "excluded = [name for name, col in findings.columns.items() \n",
+    "excluded = [name for name, col in findings.columns.items()\n",
     "            if col.inferred_type in [ColumnType.IDENTIFIER, ColumnType.TARGET, ColumnType.TEXT]]\n",
     "if excluded:\n",
     "    print(f\"\\n⛔ Excluded Columns: {', '.join(excluded)}\")"
@@ -309,19 +313,19 @@
     "    selector = FeatureSelector(target_column=findings.target_column)\n",
     "    availability_recs = selector.get_availability_recommendations(findings.feature_availability)\n",
     "    unavailable_features = [rec.column for rec in availability_recs]\n",
-    "    \n",
+    "\n",
     "    print(f\"\\n⚠️  {len(availability_recs)} feature(s) have availability issues:\\n\")\n",
     "    for rec in availability_recs:\n",
     "        print(f\"   • {rec.column} ({rec.issue_type}, {rec.coverage_pct:.0f}% coverage)\")\n",
-    "    \n",
+    "\n",
     "    print(\"\\n📋 Alternative approaches (for investigation):\")\n",
     "    print(\"   • segment_by_cohort: Train separate models per availability period\")\n",
     "    print(\"   • add_indicator: Create availability flags and impute missing\")\n",
     "    print(\"   • filter_window: Restrict data to feature's available period\")\n",
-    "    \n",
+    "\n",
     "    original_count = len(feature_cols)\n",
     "    feature_cols = [f for f in feature_cols if f not in unavailable_features]\n",
-    "    \n",
+    "\n",
     "    print(f\"\\n🗑️  Removed {original_count - len(feature_cols)} unavailable features\")\n",
     "    print(f\"📊 Features remaining: {len(feature_cols)}\")\n",
     "else:\n",
@@ -377,7 +381,7 @@
     "\n",
     "print(f\"Train size: {len(X_train):,} ({len(X_train)/len(X)*100:.0f}%)\")\n",
     "print(f\"Test size: {len(X_test):,} ({len(X_test)/len(X)*100:.0f}%)\")\n",
-    "print(f\"\\nTrain class distribution:\")\n",
+    "print(\"\\nTrain class distribution:\")\n",
     "print(f\"  Retained (1): {(y_train == 1).sum():,} ({(y_train == 1).sum()/len(y_train)*100:.1f}%)\")\n",
     "print(f\"  Churned (0): {(y_train == 0).sum():,} ({(y_train == 0).sum()/len(y_train)*100:.1f}%)\")"
    ]
@@ -438,7 +442,7 @@
     "\n",
     "for name, model in models.items():\n",
     "    print(f\"Training {name}...\")\n",
-    "    \n",
+    "\n",
     "    # Use scaled data for Logistic Regression, unscaled for tree-based\n",
     "    if \"Logistic\" in name:\n",
     "        model.fit(X_train_scaled, y_train)\n",
@@ -448,20 +452,20 @@
     "        model.fit(X_train, y_train)\n",
     "        y_pred = model.predict(X_test)\n",
     "        y_pred_proba = model.predict_proba(X_test)[:, 1]\n",
-    "    \n",
+    "\n",
     "    # Calculate metrics\n",
     "    auc = roc_auc_score(y_test, y_pred_proba)\n",
     "    pr_auc = average_precision_score(y_test, y_pred_proba)\n",
     "    f1 = f1_score(y_test, y_pred)\n",
     "    precision = precision_score(y_test, y_pred)\n",
     "    recall = recall_score(y_test, y_pred)\n",
-    "    \n",
+    "\n",
     "    # Cross-validation\n",
     "    if \"Logistic\" in name:\n",
     "        cv_scores = cross_val_score(model, X_train_scaled, y_train, cv=5, scoring='roc_auc')\n",
     "    else:\n",
     "        cv_scores = cross_val_score(model, X_train, y_train, cv=5, scoring='roc_auc')\n",
-    "    \n",
+    "\n",
     "    results.append({\n",
     "        \"Model\": name,\n",
     "        \"Test AUC\": auc,\n",
@@ -472,7 +476,7 @@
     "        \"CV AUC Mean\": cv_scores.mean(),\n",
     "        \"CV AUC Std\": cv_scores.std()\n",
     "    })\n",
-    "    \n",
+    "\n",
     "    model_predictions[name] = {\n",
     "        'y_pred': y_pred,\n",
     "        'y_pred_proba': y_pred_proba,\n",
@@ -799,12 +803,12 @@
     "print(f\"   PR-AUC: {best_model['PR-AUC']:.4f}\")\n",
     "print(f\"   F1-Score: {best_model['F1-Score']:.4f}\")\n",
     "\n",
-    "print(f\"\\n📊 TOP 3 IMPORTANT FEATURES:\")\n",
+    "print(\"\\n📊 TOP 3 IMPORTANT FEATURES:\")\n",
     "for i, feat in enumerate(importance_df.head(3)['Feature'].tolist(), 1):\n",
     "    imp = importance_df[importance_df['Feature'] == feat]['Importance'].values[0]\n",
     "    print(f\"   {i}. {feat} ({imp:.3f})\")\n",
     "\n",
-    "print(f\"\\n📈 MODEL PERFORMANCE ASSESSMENT:\")\n",
+    "print(\"\\n📈 MODEL PERFORMANCE ASSESSMENT:\")\n",
     "if best_model['Test AUC'] > 0.90:\n",
     "    print(\"   Excellent predictive signal - likely production-ready with tuning\")\n",
     "elif best_model['Test AUC'] > 0.80:\n",
@@ -814,7 +818,7 @@
     "else:\n",
     "    print(\"   Weak signal - may need more data or different features\")\n",
     "\n",
-    "print(f\"\\n💡 NEXT STEPS:\")\n",
+    "print(\"\\n💡 NEXT STEPS:\")\n",
     "print(\"   1. Feature engineering with derived features (notebook 05)\")\n",
     "print(\"   2. Hyperparameter tuning (GridSearchCV)\")\n",
     "print(\"   3. Threshold optimization for business metrics\")\n",

{churnkit-0.76.1a1.data → churnkit-0.76.1a2.data}/data/share/churnkit/exploration_notebooks/09_business_alignment.ipynb

@@ -66,13 +66,15 @@
    "outputs": [],
    "source": [
     "from customer_retention.analysis.notebook_progress import track_and_export_previous\n",
+    "\n",
     "track_and_export_previous(\"09_business_alignment.ipynb\")\n",
     "\n",
-    "from customer_retention.analysis.auto_explorer import ExplorationFindings\n",
-    "from customer_retention.analysis.visualization import display_table\n",
     "import pandas as pd\n",
-    "from customer_retention.core.config.experiments import FINDINGS_DIR, EXPERIMENTS_DIR, OUTPUT_DIR, setup_experiments_structure\n",
-    "from customer_retention.stages.temporal import TEMPORAL_METADATA_COLS\n"
+    "\n",
+    "from customer_retention.analysis.auto_explorer import ExplorationFindings\n",
+    "from customer_retention.core.config.experiments import (\n",
+    "    FINDINGS_DIR,\n",
+    ")\n"
    ]
   },
   {
@@ -98,7 +100,6 @@
    "outputs": [],
    "source": [
     "# === CONFIGURATION ===\n",
-    "from pathlib import Path\n",
     "\n",
     "# FINDINGS_DIR imported from customer_retention.core.config.experiments\n",
     "\n",