churnkit 0.76.0a3__py3-none-any.whl → 0.76.1a1__py3-none-any.whl

{churnkit-0.76.0a3.data → churnkit-0.76.1a1.data}/data/share/churnkit/exploration_notebooks/01_data_discovery.ipynb

@@ -137,10 +137,12 @@
     "# =============================================================================\n",
     "\n",
     "# DATA_PATH: Path to your data file (CSV, Parquet, or Delta)\n",
-    "DATA_PATH = \"../tests/fixtures/customer_retention_retail.csv\"\n",
+    "DATA_PATH = \"../tests/fixtures/customer_emails.csv\"\n",
+    "# or databricks /Volumes/churnkit/landing/datasets/customer_emails.csv\n",
+    "\n",
     "\n",
     "# TARGET_COLUMN: Your prediction target (set to None for auto-detection)\n",
-    "TARGET_COLUMN = \"unsubscribed\"\n",
+    "TARGET_COLUMN = None\n",
     "\n",
     "# ENTITY_COLUMN: Customer/user ID column (set to None for auto-detection)\n",
     "ENTITY_COLUMN = None\n",
@@ -169,7 +171,7 @@
     "#\n",
     "# EVENT-LEVEL (multiple rows per customer):\n",
     "# DATA_PATH = \"../tests/fixtures/customer_transactions.csv\"\n",
-    "DATA_PATH = \"../tests/fixtures/customer_emails.csv\"\n",
+    "# DATA_PATH = \"../tests/fixtures/customer_emails.csv\"\n",
     "# =============================================================================\n",
     "\n",
     "# OUTPUT_DIR: All outputs go here (gitignored)\n",

{churnkit-0.76.0a3.dist-info → churnkit-0.76.1a1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: churnkit
-Version: 0.76.0a3
+Version: 0.76.1a1
 Summary: Structured ML framework for customer churn prediction -- from exploration notebooks to production pipelines, locally or on Databricks.
 Project-URL: Homepage, https://github.com/aladjov/CR
 Project-URL: Documentation, https://github.com/aladjov/CR/wiki

{churnkit-0.76.0a3.dist-info → churnkit-0.76.1a1.dist-info}/RECORD

@@ -1,9 +1,9 @@
-customer_retention/__init__.py,sha256=1f6U04yIo69jlaQhX87DCNUo8P9P2tRzDJCik5ELPE0,1406
+customer_retention/__init__.py,sha256=fNqb1iL6Q95kXsIVQBbVJtV_NmNSWBtgbGZxRy7vuYM,1406
 customer_retention/cli.py,sha256=Wdl540cZgu_9mV-hWmTV9jD3S8QTDR8Ik-5hQXYCvmg,2466
 customer_retention/analysis/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 customer_retention/analysis/jupyter_save_hook.py,sha256=iiNFIL83yOPX8BGUjCE6Pt5Kc8X-2adtE1_NZTMUaZQ,947
 customer_retention/analysis/notebook_html_exporter.py,sha256=AMOTcD6nZncM4MPdVS1Kn4WF2YoaOoODMI2X48oEZ24,4491
-customer_retention/analysis/notebook_progress.py,sha256=aauU-0AXHraFf0M_91aSmZwI-HmZslaGnmu-Msw4xPM,2298
+customer_retention/analysis/notebook_progress.py,sha256=Tm98ihaFQMXRAlEHTH89T23gLNZardzrCMC4svcoAUw,2463
 customer_retention/analysis/plotly_preprocessor.py,sha256=Bdd_9-AmfmJdrmm030wzgpLflbiszp9KhXPbw_F5Id0,5300
 customer_retention/analysis/auto_explorer/__init__.py,sha256=0isViyt62QvDkYc2oxOhsDQ9RNMqBq1ihvwEZgoLb_s,1572
 customer_retention/analysis/auto_explorer/exploration_manager.py,sha256=60ObVRhYwAWqHnLrkeJ6_oQjPvXOl4gkLutE66_k8uc,18028
@@ -84,7 +84,7 @@ customer_retention/core/components/components/transformer.py,sha256=saEO6cRzKitU
 customer_retention/core/components/components/validator.py,sha256=5IbUqPYhsvZBTRx0X3MKV2dvZrgTcI19MM9c5_9t2CU,1405
 customer_retention/core/config/__init__.py,sha256=VXNmwSFG3wY6Budh82WRj26X07WCQKgl-M9sVwx8eds,1587
 customer_retention/core/config/column_config.py,sha256=rmMJFV4wK66q-DDQAJXe0EuXdrWd_6bg8s81NQQ54_A,3051
-customer_retention/core/config/experiments.py,sha256=CCxFgzO4sHpuNu3W2UkAaOAqsoPkXTmdHupMaMs8PvM,3234
+customer_retention/core/config/experiments.py,sha256=PQJwO1l8NjS9k_F6qOu0u0fTFcf_k--e6NIUMFBK1p4,4703
 customer_retention/core/config/pipeline_config.py,sha256=jriAcP-_UAlVTT_vVlWUPF97ieIguqlE5hrl9Ny0UiI,3675
 customer_retention/core/config/source_config.py,sha256=NnZUytq4NVvRVmp1ZtoFO_SiaIvSoJwkhw5WXy4Wi_c,2534
 customer_retention/core/utils/__init__.py,sha256=9b8SwZGiLP-glYwzcp-1aWCeTGIploAPokwITbUCneA,971
@@ -131,7 +131,7 @@ customer_retention/generators/spec_generator/generic_generator.py,sha256=I_glnOO
 customer_retention/generators/spec_generator/mlflow_pipeline_generator.py,sha256=LME87sjzP_MjOMA3NTxqRfOhCroUJAb40BAnSH4-I74,29866
 customer_retention/generators/spec_generator/pipeline_spec.py,sha256=c8v1SWgTdeGmNs96l1hOS0qx1B1ua0iwPhw1I5w9OIo,10705
 customer_retention/integrations/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-customer_retention/integrations/databricks_init.py,sha256=Qte81tiF4MjDzX7o5dPNOKsY_IoVjVXtJoPUDtZ4kx0,5641
+customer_retention/integrations/databricks_init.py,sha256=xQgCBhzx0DoGGGeocYX9Iu3H2dEl--bvA36T7_uF6TM,6526
 customer_retention/integrations/adapters/__init__.py,sha256=Fgdp0ESROTUHnOb2RN9Ubo0A4BdfoenOGuUz61lHz8g,583
 customer_retention/integrations/adapters/base.py,sha256=z6dVAowDKGogKsYGR7VMcLkS6VhcB9h4zgN1tilNYRg,254
 customer_retention/integrations/adapters/factory.py,sha256=CMsqOeDozADbWnk8fzktZvAyL1FEmUjDMvfDCpLDVaU,1202
@@ -277,27 +277,27 @@ customer_retention/transforms/artifact_store.py,sha256=FYLpDcv2N6-dUTX5RPEIK3aCW
 customer_retention/transforms/executor.py,sha256=oML5dCidxbW_q6YUkAwWcutYP6bIFB6IdD3BvemK45A,6304
 customer_retention/transforms/fitted.py,sha256=3pNvnae-P3t3bKMeZz1Bl0xww-feapIYdoeTY6aUtI8,3278
 customer_retention/transforms/ops.py,sha256=Xg2g9UOOudq_y9Hf3oWsjpqw3dEoykQR5pDSoyW8GX0,4294
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/00_start_here.ipynb,sha256=zCyhftEd3v9fc0Ta6wvA6b-9LcoGzRi8bS1tMZ3iu9w,21911
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/01_data_discovery.ipynb,sha256=up0X3oDJ5sAo1-tbqMyZj_f1h6D542G2uAxjVmtYCOI,46430
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/01a_a_temporal_text_deep_dive.ipynb,sha256=uai8T3iJSqOrabBQnVi8Z0k8zZGVgs_VVQWRHyXN8QU,33690
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/01a_temporal_deep_dive.ipynb,sha256=fC1ASNtvI8X1lAe-Lzcw3oX2cptDC-ymPeEtKKWhg20,67326
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/01b_temporal_quality.ipynb,sha256=RU5hxgrTVMZs1ytChVv1t49WpTO0Oj6B_Fu8g0xS0To,23039
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/01c_temporal_patterns.ipynb,sha256=ZGYfztP6JhOEwPmTYdC0l7w579fKXcNEJXq-PnCLc2I,153167
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/01d_event_aggregation.ipynb,sha256=-FT3SoBU0fhaZxGeTo-_UQl6riCrtoJaFnUg31opk64,63244
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/02_column_deep_dive.ipynb,sha256=mbP2LQWsXDyTsWg0bhrCBHEfHsEer_XOXRYV9f8JxAk,60250
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/02a_text_columns_deep_dive.ipynb,sha256=M9YN8yAjjuC6ZaUlc-rVqVLEkWd7Rc_GNILHS9qO3PU,29704
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/03_quality_assessment.ipynb,sha256=H49LLmn1PHbcbAvSQfteESRGk125QwkPI5qbLk3yZgc,68595
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/04_relationship_analysis.ipynb,sha256=Rr-B4-xg0ILuAIgztlZkiGJdTzLuNjOqBFxO8W4o9iU,78624
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/05_multi_dataset.ipynb,sha256=bBxkuZyTl1yZg4kMXO87WRjgZMhj_6hwLGX6m3XC270,62664
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/06_feature_opportunities.ipynb,sha256=cBJF5o4z3Z-dustQ4CVklnfTcQ8saG97tlgswWK9uWE,67409
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/07_modeling_readiness.ipynb,sha256=IiA04fyb-l097Glp3MtR03vPjQsZlS1Icg-hjEHa_Dg,28376
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/08_baseline_experiments.ipynb,sha256=KmjhnDf1JdpEiIcdfQ-ZFo_at6t9JRC30B6NmmvMBmg,34226
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/09_business_alignment.ipynb,sha256=tMNfGM7AH50N1ONzHhGW2HZLpQwraIxVzOiVnI-10X8,17214
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/10_spec_generation.ipynb,sha256=KeUdfL9Mvdi6023XpnfZ6oLEDNZaWiIHUfsAWig24mE,42847
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/11_scoring_validation.ipynb,sha256=5fi3eHMm03ZKZgdFAXMgydtZ3qX2TtR3L9bZS2MpWPE,49937
-churnkit-0.76.0a3.data/data/share/churnkit/exploration_notebooks/12_view_documentation.ipynb,sha256=aQF7CG8HxckqUKOKqnmZgMkSvfVzyO2LlYPrymLYjBY,4405
-churnkit-0.76.0a3.dist-info/METADATA,sha256=MYmaYdzvcZWtkAL6AtSWEBBAhluczHz_im33olePY3k,13005
-churnkit-0.76.0a3.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
-churnkit-0.76.0a3.dist-info/entry_points.txt,sha256=swQFVe-jjgQSBJQNO2Ulkz2F5odaE-TsnlTor3HQBjw,70
-churnkit-0.76.0a3.dist-info/licenses/LICENSE,sha256=Bud8Oj25tnpoIuXCWW0xcSfmGPeEZAAHrDRoKdSYtZY,11344
-churnkit-0.76.0a3.dist-info/RECORD,,
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/00_start_here.ipynb,sha256=zCyhftEd3v9fc0Ta6wvA6b-9LcoGzRi8bS1tMZ3iu9w,21911
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/01_data_discovery.ipynb,sha256=EDhWyKM2stHtWg2bS8HdRMItZ6WtokWRyZ90_MjTPFQ,46500
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/01a_a_temporal_text_deep_dive.ipynb,sha256=uai8T3iJSqOrabBQnVi8Z0k8zZGVgs_VVQWRHyXN8QU,33690
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/01a_temporal_deep_dive.ipynb,sha256=fC1ASNtvI8X1lAe-Lzcw3oX2cptDC-ymPeEtKKWhg20,67326
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/01b_temporal_quality.ipynb,sha256=RU5hxgrTVMZs1ytChVv1t49WpTO0Oj6B_Fu8g0xS0To,23039
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/01c_temporal_patterns.ipynb,sha256=ZGYfztP6JhOEwPmTYdC0l7w579fKXcNEJXq-PnCLc2I,153167
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/01d_event_aggregation.ipynb,sha256=-FT3SoBU0fhaZxGeTo-_UQl6riCrtoJaFnUg31opk64,63244
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/02_column_deep_dive.ipynb,sha256=mbP2LQWsXDyTsWg0bhrCBHEfHsEer_XOXRYV9f8JxAk,60250
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/02a_text_columns_deep_dive.ipynb,sha256=M9YN8yAjjuC6ZaUlc-rVqVLEkWd7Rc_GNILHS9qO3PU,29704
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/03_quality_assessment.ipynb,sha256=H49LLmn1PHbcbAvSQfteESRGk125QwkPI5qbLk3yZgc,68595
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/04_relationship_analysis.ipynb,sha256=Rr-B4-xg0ILuAIgztlZkiGJdTzLuNjOqBFxO8W4o9iU,78624
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/05_multi_dataset.ipynb,sha256=bBxkuZyTl1yZg4kMXO87WRjgZMhj_6hwLGX6m3XC270,62664
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/06_feature_opportunities.ipynb,sha256=cBJF5o4z3Z-dustQ4CVklnfTcQ8saG97tlgswWK9uWE,67409
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/07_modeling_readiness.ipynb,sha256=IiA04fyb-l097Glp3MtR03vPjQsZlS1Icg-hjEHa_Dg,28376
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/08_baseline_experiments.ipynb,sha256=KmjhnDf1JdpEiIcdfQ-ZFo_at6t9JRC30B6NmmvMBmg,34226
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/09_business_alignment.ipynb,sha256=tMNfGM7AH50N1ONzHhGW2HZLpQwraIxVzOiVnI-10X8,17214
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/10_spec_generation.ipynb,sha256=KeUdfL9Mvdi6023XpnfZ6oLEDNZaWiIHUfsAWig24mE,42847
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/11_scoring_validation.ipynb,sha256=5fi3eHMm03ZKZgdFAXMgydtZ3qX2TtR3L9bZS2MpWPE,49937
+churnkit-0.76.1a1.data/data/share/churnkit/exploration_notebooks/12_view_documentation.ipynb,sha256=aQF7CG8HxckqUKOKqnmZgMkSvfVzyO2LlYPrymLYjBY,4405
+churnkit-0.76.1a1.dist-info/METADATA,sha256=CkjX1u5qlqFbTuNbtFC93u25moHVAf2TaxYJwFAMWZY,13005
+churnkit-0.76.1a1.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
+churnkit-0.76.1a1.dist-info/entry_points.txt,sha256=swQFVe-jjgQSBJQNO2Ulkz2F5odaE-TsnlTor3HQBjw,70
+churnkit-0.76.1a1.dist-info/licenses/LICENSE,sha256=Bud8Oj25tnpoIuXCWW0xcSfmGPeEZAAHrDRoKdSYtZY,11344
+churnkit-0.76.1a1.dist-info/RECORD,,

customer_retention/__init__.py

@@ -17,7 +17,7 @@ Main module categories:
                 llm_context, iteration)
 """
 
-__version__ = "0.76.0a3"
+__version__ = "0.76.1a1"
 
 # Environment utilities (always available)
 from .core.compat import (

customer_retention/analysis/notebook_progress.py

@@ -5,7 +5,16 @@ from pathlib import Path
 from typing import Optional
 
 from customer_retention.core.compat import is_databricks
-from customer_retention.core.config.experiments import get_notebook_experiments_dir
+from customer_retention.core.config.experiments import get_notebook_experiments_dir, reload_config
+
+
+def _ensure_databricks_config_loaded() -> None:
+    if not is_databricks():
+        return
+    reload_config()
+
+
+_ensure_databricks_config_loaded()
 
 
 def track_and_export_previous(current_notebook: str) -> None:

customer_retention/core/config/experiments.py

@@ -1,7 +1,49 @@
+import json
 import os
 from pathlib import Path
 from typing import Optional
 
+_DATABRICKS_CONFIG_FILENAME = ".churnkit_config.json"
+
+
+def _workspace_config_path(workspace_path: str) -> Path:
+    return Path(f"/Workspace/{workspace_path}") / _DATABRICKS_CONFIG_FILENAME
+
+
+def _read_config_file(path: Path) -> dict | None:
+    try:
+        return json.loads(path.read_text()) if path.exists() else None
+    except (json.JSONDecodeError, OSError):
+        return None
+
+
+def _load_persisted_databricks_config() -> dict | None:
+    if not os.environ.get("DATABRICKS_RUNTIME_VERSION"):
+        return None
+    workspace_path = os.environ.get("CR_WORKSPACE_PATH")
+    if workspace_path:
+        return _read_config_file(_workspace_config_path(workspace_path))
+    cwd = Path.cwd()
+    for _ in range(5):
+        result = _read_config_file(cwd / _DATABRICKS_CONFIG_FILENAME)
+        if result:
+            return result
+        if cwd.parent == cwd:
+            break
+        cwd = cwd.parent
+    return None
+
+
+def persist_databricks_config(experiments_dir: str, catalog: str, schema: str, workspace_path: str | None = None) -> None:
+    if not workspace_path:
+        return
+    try:
+        _workspace_config_path(workspace_path).write_text(json.dumps({
+            "experiments_dir": experiments_dir, "catalog": catalog, "schema": schema,
+        }))
+    except OSError:
+        pass
+
 
 def _find_project_root() -> Path:
     path = Path(__file__).parent
@@ -17,6 +59,9 @@ def get_experiments_dir(default: Optional[str] = None) -> Path:
         return Path(os.environ["CR_EXPERIMENTS_DIR"])
     if default:
         return Path(default)
+    persisted = _load_persisted_databricks_config()
+    if persisted and "experiments_dir" in persisted:
+        return Path(persisted["experiments_dir"])
     return _find_project_root() / "experiments"
 
 

customer_retention/integrations/databricks_init.py

@@ -40,11 +40,14 @@ def databricks_init(
     _validate_databricks_environment()
     if workspace_path:
         workspace_path = _normalize_workspace_path(workspace_path)
+        _ensure_workspace_directory(workspace_path)
     _set_environment_variables(catalog, schema, workspace_path)
+    _persist_config(catalog, schema, workspace_path)
     resolved_experiment_name = experiment_name or _resolve_experiment_name_from_notebook_path()
     resolved_experiment_name = _make_absolute_experiment_path(resolved_experiment_name, workspace_path)
     _set_experiment_name_env_var(resolved_experiment_name)
     _reload_config_constants()
+    _setup_experiment_directories()
     _configure_mlflow_experiment(resolved_experiment_name)
     notebooks_copied: list[str] = []
     if copy_notebooks and workspace_path:
@@ -81,12 +84,27 @@ def _set_experiment_name_env_var(experiment_name: str) -> None:
     os.environ["CR_EXPERIMENT_NAME"] = experiment_name
 
 
+def _persist_config(catalog: str, schema: str, workspace_path: str | None) -> None:
+    from customer_retention.core.config.experiments import persist_databricks_config
+
+    persist_databricks_config(f"/Volumes/{catalog}/{schema}/experiments", catalog, schema, workspace_path)
+
+
 def _reload_config_constants() -> None:
     from customer_retention.core.config.experiments import reload_config
 
     reload_config()
 
 
+def _setup_experiment_directories() -> None:
+    from customer_retention.core.config.experiments import setup_experiments_structure
+
+    try:
+        setup_experiments_structure()
+    except OSError:
+        pass
+
+
 def _resolve_experiment_name_from_notebook_path() -> str:
     try:
         dbutils = _get_dbutils()
@@ -133,6 +151,13 @@ def _configure_mlflow_experiment(experiment_name: str) -> None:
         pass
 
 
+def _ensure_workspace_directory(workspace_path: str) -> None:
+    try:
+        Path(f"/Workspace/{workspace_path}").mkdir(parents=True, exist_ok=True)
+    except OSError:
+        pass
+
+
 def _copy_exploration_notebooks(workspace_path: str) -> list[str]:
     from customer_retention.generators.notebook_generator.project_init import ProjectInitializer
 
@@ -154,7 +179,9 @@ def _copy_exploration_notebooks(workspace_path: str) -> list[str]:
 
 
 def _display_init_summary(result: DatabricksInitResult) -> None:
-    print("ChurnKit Databricks Initialization Complete")
+    from customer_retention import __version__
+
+    print(f"ChurnKit v{__version__} Databricks Initialization Complete")
     print("=" * 45)
     print(f"  Catalog:          {result.catalog}")
     print(f"  Schema:           {result.schema}")