ins-pricing 0.3.0__tar.gz → 0.3.1__tar.gz

{ins_pricing-0.3.0 → ins_pricing-0.3.1}/PKG-INFO

@@ -1,162 +1,162 @@
-Metadata-Version: 2.4
-Name: ins_pricing
-Version: 0.3.0
-Summary: Reusable modelling, pricing, governance, and reporting utilities.
-Author: meishi125478
-License: Proprietary
-Keywords: pricing,insurance,bayesopt,ml
-Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3 :: Only
-Classifier: Programming Language :: Python :: 3.9
-Classifier: License :: Other/Proprietary License
-Classifier: Operating System :: OS Independent
-Classifier: Intended Audience :: Developers
-Requires-Python: >=3.9
-Description-Content-Type: text/markdown
-Requires-Dist: numpy>=1.20
-Requires-Dist: pandas>=1.4
-Provides-Extra: bayesopt
-Requires-Dist: torch>=1.13; extra == "bayesopt"
-Requires-Dist: optuna>=3.0; extra == "bayesopt"
-Requires-Dist: xgboost>=1.6; extra == "bayesopt"
-Requires-Dist: scikit-learn>=1.1; extra == "bayesopt"
-Requires-Dist: statsmodels>=0.13; extra == "bayesopt"
-Requires-Dist: joblib>=1.2; extra == "bayesopt"
-Requires-Dist: matplotlib>=3.5; extra == "bayesopt"
-Provides-Extra: plotting
-Requires-Dist: matplotlib>=3.5; extra == "plotting"
-Requires-Dist: scikit-learn>=1.1; extra == "plotting"
-Provides-Extra: explain
-Requires-Dist: torch>=1.13; extra == "explain"
-Requires-Dist: shap>=0.41; extra == "explain"
-Requires-Dist: scikit-learn>=1.1; extra == "explain"
-Provides-Extra: geo
-Requires-Dist: contextily>=1.3; extra == "geo"
-Requires-Dist: matplotlib>=3.5; extra == "geo"
-Provides-Extra: gnn
-Requires-Dist: torch>=1.13; extra == "gnn"
-Requires-Dist: pynndescent>=0.5; extra == "gnn"
-Requires-Dist: torch-geometric>=2.3; extra == "gnn"
-Provides-Extra: all
-Requires-Dist: torch>=1.13; extra == "all"
-Requires-Dist: optuna>=3.0; extra == "all"
-Requires-Dist: xgboost>=1.6; extra == "all"
-Requires-Dist: scikit-learn>=1.1; extra == "all"
-Requires-Dist: statsmodels>=0.13; extra == "all"
-Requires-Dist: joblib>=1.2; extra == "all"
-Requires-Dist: matplotlib>=3.5; extra == "all"
-Requires-Dist: shap>=0.41; extra == "all"
-Requires-Dist: contextily>=1.3; extra == "all"
-Requires-Dist: pynndescent>=0.5; extra == "all"
-Requires-Dist: torch-geometric>=2.3; extra == "all"
-
-# Insurance-Pricing
-
-A reusable toolkit for insurance modeling, pricing, governance, and reporting.
-
-## Overview
-
-Insurance-Pricing (ins_pricing) is an enterprise-grade Python library designed for machine learning model training, pricing calculations, and model governance workflows in the insurance industry.
-
-### Core Modules
-
-| Module | Description |
-|--------|-------------|
-| **modelling** | ML model training (GLM, XGBoost, ResNet, FT-Transformer, GNN) and model interpretability (SHAP, permutation importance) |
-| **pricing** | Factor table construction, numeric binning, premium calibration, exposure calculation, PSI monitoring |
-| **production** | Model prediction, batch scoring, data drift detection, production metrics monitoring |
-| **governance** | Model registry, version management, approval workflows, audit logging |
-| **reporting** | Report generation (Markdown format), report scheduling |
-| **utils** | Data validation, performance profiling, device management, logging configuration |
-
-### Quick Start
-
-```python
-# Model training with Bayesian optimization
-from ins_pricing import bayesopt as ropt
-
-model = ropt.BayesOptModel(
-    train_data, test_data,
-    model_name='my_model',
-    resp_nme='target',
-    weight_nme='weight',
-    factor_nmes=feature_list,
-    cate_list=categorical_features,
-)
-model.bayesopt_xgb(max_evals=100)      # Train XGBoost
-model.bayesopt_resnet(max_evals=50)    # Train ResNet
-model.bayesopt_ft(max_evals=50)        # Train FT-Transformer
-
-# Pricing: build factor table
-from ins_pricing.pricing import build_factor_table
-factors = build_factor_table(
-    df,
-    factor_col='age_band',
-    loss_col='claim_amount',
-    exposure_col='exposure',
-)
-
-# Production: batch scoring
-from ins_pricing.production import batch_score
-scores = batch_score(model.trainers['xgb'].predict, df)
-
-# Model governance
-from ins_pricing.governance import ModelRegistry
-registry = ModelRegistry('models.json')
-registry.register(model_name, version, metrics=metrics)
-```
-
-### Project Structure
-
-```
-ins_pricing/
-├── cli/                    # Command-line entry points
-├── modelling/
-│   ├── core/bayesopt/     # ML model training core
-│   ├── explain/           # Model interpretability
-│   └── plotting/          # Model visualization
-├── pricing/               # Insurance pricing module
-├── production/            # Production deployment module
-├── governance/            # Model governance
-├── reporting/             # Report generation
-├── utils/                 # Utilities
-└── tests/                 # Test suite
-```
-
-### Installation
-
-```bash
-# Basic installation
-pip install ins_pricing
-
-# Full installation (all optional dependencies)
-pip install ins_pricing[all]
-
-# Install specific extras
-pip install ins_pricing[bayesopt]    # Model training
-pip install ins_pricing[explain]     # Model explanation
-pip install ins_pricing[plotting]    # Visualization
-pip install ins_pricing[gnn]         # Graph neural networks
-```
-
-#### Multi-platform & GPU installation notes
-
-- **PyTorch (CPU/GPU/MPS)**: Install the correct PyTorch build for your platform/GPU first (CUDA on
-  Linux/Windows, ROCm on supported AMD platforms, or MPS on Apple Silicon). Then install the
-  optional extras you need (e.g., `bayesopt`, `explain`, or `gnn`). This avoids pip pulling a
-  mismatched wheel.
-- **Torch Geometric (GNN)**: `torch-geometric` often requires platform-specific wheels (e.g.,
-  `torch-scatter`, `torch-sparse`). Follow the official PyG installation instructions for your
-  CUDA/ROCm/CPU environment, then install `ins_pricing[gnn]`.
-- **Multi-GPU**: Training code will use CUDA when available and can enable multi-GPU via
-  `torch.distributed`/`DataParallel` where supported. On Windows, CUDA DDP is not supported and will
-  fall back to single-GPU or DataParallel where possible.
-
-### Requirements
-
-- Python >= 3.9
-- Core dependencies: numpy >= 1.20, pandas >= 1.4
-
-### License
-
-Proprietary
+Metadata-Version: 2.4
+Name: ins_pricing
+Version: 0.3.1
+Summary: Reusable modelling, pricing, governance, and reporting utilities.
+Author: meishi125478
+License: Proprietary
+Keywords: pricing,insurance,bayesopt,ml
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.9
+Classifier: License :: Other/Proprietary License
+Classifier: Operating System :: OS Independent
+Classifier: Intended Audience :: Developers
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+Requires-Dist: numpy>=1.20
+Requires-Dist: pandas>=1.4
+Provides-Extra: bayesopt
+Requires-Dist: torch>=1.13; extra == "bayesopt"
+Requires-Dist: optuna>=3.0; extra == "bayesopt"
+Requires-Dist: xgboost>=1.6; extra == "bayesopt"
+Requires-Dist: scikit-learn>=1.1; extra == "bayesopt"
+Requires-Dist: statsmodels>=0.13; extra == "bayesopt"
+Requires-Dist: joblib>=1.2; extra == "bayesopt"
+Requires-Dist: matplotlib>=3.5; extra == "bayesopt"
+Provides-Extra: plotting
+Requires-Dist: matplotlib>=3.5; extra == "plotting"
+Requires-Dist: scikit-learn>=1.1; extra == "plotting"
+Provides-Extra: explain
+Requires-Dist: torch>=1.13; extra == "explain"
+Requires-Dist: shap>=0.41; extra == "explain"
+Requires-Dist: scikit-learn>=1.1; extra == "explain"
+Provides-Extra: geo
+Requires-Dist: contextily>=1.3; extra == "geo"
+Requires-Dist: matplotlib>=3.5; extra == "geo"
+Provides-Extra: gnn
+Requires-Dist: torch>=1.13; extra == "gnn"
+Requires-Dist: pynndescent>=0.5; extra == "gnn"
+Requires-Dist: torch-geometric>=2.3; extra == "gnn"
+Provides-Extra: all
+Requires-Dist: torch>=1.13; extra == "all"
+Requires-Dist: optuna>=3.0; extra == "all"
+Requires-Dist: xgboost>=1.6; extra == "all"
+Requires-Dist: scikit-learn>=1.1; extra == "all"
+Requires-Dist: statsmodels>=0.13; extra == "all"
+Requires-Dist: joblib>=1.2; extra == "all"
+Requires-Dist: matplotlib>=3.5; extra == "all"
+Requires-Dist: shap>=0.41; extra == "all"
+Requires-Dist: contextily>=1.3; extra == "all"
+Requires-Dist: pynndescent>=0.5; extra == "all"
+Requires-Dist: torch-geometric>=2.3; extra == "all"
+
+# Insurance-Pricing
+
+A reusable toolkit for insurance modeling, pricing, governance, and reporting.
+
+## Overview
+
+Insurance-Pricing (ins_pricing) is an enterprise-grade Python library designed for machine learning model training, pricing calculations, and model governance workflows in the insurance industry.
+
+### Core Modules
+
+| Module | Description |
+|--------|-------------|
+| **modelling** | ML model training (GLM, XGBoost, ResNet, FT-Transformer, GNN) and model interpretability (SHAP, permutation importance) |
+| **pricing** | Factor table construction, numeric binning, premium calibration, exposure calculation, PSI monitoring |
+| **production** | Model prediction, batch scoring, data drift detection, production metrics monitoring |
+| **governance** | Model registry, version management, approval workflows, audit logging |
+| **reporting** | Report generation (Markdown format), report scheduling |
+| **utils** | Data validation, performance profiling, device management, logging configuration |
+
+### Quick Start
+
+```python
+# Model training with Bayesian optimization
+from ins_pricing import bayesopt as ropt
+
+model = ropt.BayesOptModel(
+    train_data, test_data,
+    model_name='my_model',
+    resp_nme='target',
+    weight_nme='weight',
+    factor_nmes=feature_list,
+    cate_list=categorical_features,
+)
+model.bayesopt_xgb(max_evals=100)      # Train XGBoost
+model.bayesopt_resnet(max_evals=50)    # Train ResNet
+model.bayesopt_ft(max_evals=50)        # Train FT-Transformer
+
+# Pricing: build factor table
+from ins_pricing.pricing import build_factor_table
+factors = build_factor_table(
+    df,
+    factor_col='age_band',
+    loss_col='claim_amount',
+    exposure_col='exposure',
+)
+
+# Production: batch scoring
+from ins_pricing.production import batch_score
+scores = batch_score(model.trainers['xgb'].predict, df)
+
+# Model governance
+from ins_pricing.governance import ModelRegistry
+registry = ModelRegistry('models.json')
+registry.register(model_name, version, metrics=metrics)
+```
+
+### Project Structure
+
+```
+ins_pricing/
+├── cli/                    # Command-line entry points
+├── modelling/
+│   ├── core/bayesopt/     # ML model training core
+│   ├── explain/           # Model interpretability
+│   └── plotting/          # Model visualization
+├── pricing/               # Insurance pricing module
+├── production/            # Production deployment module
+├── governance/            # Model governance
+├── reporting/             # Report generation
+├── utils/                 # Utilities
+└── tests/                 # Test suite
+```
+
+### Installation
+
+```bash
+# Basic installation
+pip install ins_pricing
+
+# Full installation (all optional dependencies)
+pip install ins_pricing[all]
+
+# Install specific extras
+pip install ins_pricing[bayesopt]    # Model training
+pip install ins_pricing[explain]     # Model explanation
+pip install ins_pricing[plotting]    # Visualization
+pip install ins_pricing[gnn]         # Graph neural networks
+```
+
+#### Multi-platform & GPU installation notes
+
+- **PyTorch (CPU/GPU/MPS)**: Install the correct PyTorch build for your platform/GPU first (CUDA on
+  Linux/Windows, ROCm on supported AMD platforms, or MPS on Apple Silicon). Then install the
+  optional extras you need (e.g., `bayesopt`, `explain`, or `gnn`). This avoids pip pulling a
+  mismatched wheel.
+- **Torch Geometric (GNN)**: `torch-geometric` often requires platform-specific wheels (e.g.,
+  `torch-scatter`, `torch-sparse`). Follow the official PyG installation instructions for your
+  CUDA/ROCm/CPU environment, then install `ins_pricing[gnn]`.
+- **Multi-GPU**: Training code will use CUDA when available and can enable multi-GPU via
+  `torch.distributed`/`DataParallel` where supported. On Windows, CUDA DDP is not supported and will
+  fall back to single-GPU or DataParallel where possible.
+
+### Requirements
+
+- Python >= 3.9
+- Core dependencies: numpy >= 1.20, pandas >= 1.4
+
+### License
+
+Proprietary

{ins_pricing-0.3.0 → ins_pricing-0.3.1}/ins_pricing/cli/Explain_entry.py

@@ -491,54 +491,56 @@ def explain_from_config(args: argparse.Namespace) -> None:
         categorical_features = cfg.get("categorical_features")
         plot_path_style = runtime_cfg["plot_path_style"]
 
-        model = ropt.BayesOptModel(
-            train_df,
-            test_df,
-            model_name,
-            cfg["target"],
-            cfg["weight"],
-            feature_list,
-            task_type=str(cfg.get("task_type", "regression")),
-            binary_resp_nme=binary_target,
-            cate_list=categorical_features,
-            prop_test=prop_test,
-            rand_seed=rand_seed,
-            epochs=int(runtime_cfg["epochs"]),
-            use_gpu=bool(cfg.get("use_gpu", True)),
-            output_dir=output_dir,
-            xgb_max_depth_max=runtime_cfg["xgb_max_depth_max"],
-            xgb_n_estimators_max=runtime_cfg["xgb_n_estimators_max"],
-            resn_weight_decay=cfg.get("resn_weight_decay"),
-            final_ensemble=bool(cfg.get("final_ensemble", False)),
-            final_ensemble_k=int(cfg.get("final_ensemble_k", 3)),
-            final_refit=bool(cfg.get("final_refit", True)),
-            optuna_storage=runtime_cfg["optuna_storage"],
-            optuna_study_prefix=runtime_cfg["optuna_study_prefix"],
-            best_params_files=runtime_cfg["best_params_files"],
-            gnn_use_approx_knn=cfg.get("gnn_use_approx_knn", True),
-            gnn_approx_knn_threshold=cfg.get("gnn_approx_knn_threshold", 50000),
-            gnn_graph_cache=cfg.get("gnn_graph_cache"),
-            gnn_max_gpu_knn_nodes=cfg.get("gnn_max_gpu_knn_nodes", 200000),
-            gnn_knn_gpu_mem_ratio=cfg.get("gnn_knn_gpu_mem_ratio", 0.9),
-            gnn_knn_gpu_mem_overhead=cfg.get("gnn_knn_gpu_mem_overhead", 2.0),
-            region_province_col=cfg.get("region_province_col"),
-            region_city_col=cfg.get("region_city_col"),
-            region_effect_alpha=cfg.get("region_effect_alpha"),
-            geo_feature_nmes=cfg.get("geo_feature_nmes"),
-            geo_token_hidden_dim=cfg.get("geo_token_hidden_dim"),
-            geo_token_layers=cfg.get("geo_token_layers"),
-            geo_token_dropout=cfg.get("geo_token_dropout"),
-            geo_token_k_neighbors=cfg.get("geo_token_k_neighbors"),
-            geo_token_learning_rate=cfg.get("geo_token_learning_rate"),
-            geo_token_epochs=cfg.get("geo_token_epochs"),
-            ft_role=str(cfg.get("ft_role", "model")),
-            ft_feature_prefix=str(cfg.get("ft_feature_prefix", "ft_emb")),
-            ft_num_numeric_tokens=cfg.get("ft_num_numeric_tokens"),
-            infer_categorical_max_unique=int(cfg.get("infer_categorical_max_unique", 50)),
-            infer_categorical_max_ratio=float(cfg.get("infer_categorical_max_ratio", 0.05)),
-            reuse_best_params=runtime_cfg["reuse_best_params"],
-            plot_path_style=plot_path_style,
-        )
+        config_fields = getattr(ropt.BayesOptConfig, "__dataclass_fields__", {})
+        allowed_config_keys = set(config_fields.keys())
+        config_payload = {k: v for k, v in cfg.items() if k in allowed_config_keys}
+        config_payload.update({
+            "model_nme": model_name,
+            "resp_nme": cfg["target"],
+            "weight_nme": cfg["weight"],
+            "factor_nmes": feature_list,
+            "task_type": str(cfg.get("task_type", "regression")),
+            "binary_resp_nme": binary_target,
+            "cate_list": categorical_features,
+            "prop_test": prop_test,
+            "rand_seed": rand_seed,
+            "epochs": int(runtime_cfg["epochs"]),
+            "use_gpu": bool(cfg.get("use_gpu", True)),
+            "output_dir": output_dir,
+            "xgb_max_depth_max": runtime_cfg["xgb_max_depth_max"],
+            "xgb_n_estimators_max": runtime_cfg["xgb_n_estimators_max"],
+            "resn_weight_decay": cfg.get("resn_weight_decay"),
+            "final_ensemble": bool(cfg.get("final_ensemble", False)),
+            "final_ensemble_k": int(cfg.get("final_ensemble_k", 3)),
+            "final_refit": bool(cfg.get("final_refit", True)),
+            "optuna_storage": runtime_cfg["optuna_storage"],
+            "optuna_study_prefix": runtime_cfg["optuna_study_prefix"],
+            "best_params_files": runtime_cfg["best_params_files"],
+            "gnn_use_approx_knn": cfg.get("gnn_use_approx_knn", True),
+            "gnn_approx_knn_threshold": cfg.get("gnn_approx_knn_threshold", 50000),
+            "gnn_graph_cache": cfg.get("gnn_graph_cache"),
+            "gnn_max_gpu_knn_nodes": cfg.get("gnn_max_gpu_knn_nodes", 200000),
+            "gnn_knn_gpu_mem_ratio": cfg.get("gnn_knn_gpu_mem_ratio", 0.9),
+            "gnn_knn_gpu_mem_overhead": cfg.get("gnn_knn_gpu_mem_overhead", 2.0),
+            "region_province_col": cfg.get("region_province_col"),
+            "region_city_col": cfg.get("region_city_col"),
+            "region_effect_alpha": cfg.get("region_effect_alpha"),
+            "geo_feature_nmes": cfg.get("geo_feature_nmes"),
+            "geo_token_hidden_dim": cfg.get("geo_token_hidden_dim"),
+            "geo_token_layers": cfg.get("geo_token_layers"),
+            "geo_token_dropout": cfg.get("geo_token_dropout"),
+            "geo_token_k_neighbors": cfg.get("geo_token_k_neighbors"),
+            "geo_token_learning_rate": cfg.get("geo_token_learning_rate"),
+            "geo_token_epochs": cfg.get("geo_token_epochs"),
+            "ft_role": str(cfg.get("ft_role", "model")),
+            "ft_feature_prefix": str(cfg.get("ft_feature_prefix", "ft_emb")),
+            "ft_num_numeric_tokens": cfg.get("ft_num_numeric_tokens"),
+            "reuse_best_params": runtime_cfg["reuse_best_params"],
+            "plot_path_style": plot_path_style or "nested",
+        })
+        config_payload = {k: v for k, v in config_payload.items() if v is not None}
+        config = ropt.BayesOptConfig(**config_payload)
+        model = ropt.BayesOptModel(train_df, test_df, config=config)
 
         output_overrides = resolve_explain_output_overrides(
             explain_cfg,

{ins_pricing-0.3.0 → ins_pricing-0.3.1}/ins_pricing/cli/bayesopt_entry_runner.py

@@ -1223,76 +1223,79 @@ def train_from_config(args: argparse.Namespace) -> None:
             cfg.get("ft_feature_prefix", args.ft_feature_prefix))
         ft_num_numeric_tokens = cfg.get("ft_num_numeric_tokens")
 
-        model = ropt.BayesOptModel(
-            train_df,
-            test_df,
-            model_name,
-            cfg["target"],
-            cfg["weight"],
-            feature_list,
-            task_type=task_type,
-            binary_resp_nme=binary_target,
-            cate_list=categorical_features,
-            prop_test=val_ratio,
-            rand_seed=rand_seed,
-            epochs=epochs,
-            use_gpu=use_gpu,
-            use_resn_data_parallel=use_resn_dp,
-            use_ft_data_parallel=use_ft_dp,
-            use_resn_ddp=use_resn_ddp,
-            use_ft_ddp=use_ft_ddp,
-            use_gnn_data_parallel=use_gnn_dp,
-            use_gnn_ddp=use_gnn_ddp,
-            output_dir=output_dir,
-            xgb_max_depth_max=xgb_max_depth_max,
-            xgb_n_estimators_max=xgb_n_estimators_max,
-            resn_weight_decay=cfg.get("resn_weight_decay"),
-            final_ensemble=bool(cfg.get("final_ensemble", False)),
-            final_ensemble_k=int(cfg.get("final_ensemble_k", 3)),
-            final_refit=bool(cfg.get("final_refit", True)),
-            optuna_storage=optuna_storage,
-            optuna_study_prefix=optuna_study_prefix,
-            best_params_files=best_params_files,
-            gnn_use_approx_knn=gnn_use_ann,
-            gnn_approx_knn_threshold=gnn_threshold,
-            gnn_graph_cache=gnn_graph_cache,
-            gnn_max_gpu_knn_nodes=gnn_max_gpu_nodes,
-            gnn_knn_gpu_mem_ratio=gnn_gpu_mem_ratio,
-            gnn_knn_gpu_mem_overhead=gnn_gpu_mem_overhead,
-            region_province_col=region_province_col,
-            region_city_col=region_city_col,
-            region_effect_alpha=region_effect_alpha,
-            geo_feature_nmes=geo_feature_nmes,
-            geo_token_hidden_dim=geo_token_hidden_dim,
-            geo_token_layers=geo_token_layers,
-            geo_token_dropout=geo_token_dropout,
-            geo_token_k_neighbors=geo_token_k_neighbors,
-            geo_token_learning_rate=geo_token_learning_rate,
-            geo_token_epochs=geo_token_epochs,
-            ft_role=ft_role,
-            ft_feature_prefix=ft_feature_prefix,
-            ft_num_numeric_tokens=ft_num_numeric_tokens,
-            infer_categorical_max_unique=int(
-                cfg.get("infer_categorical_max_unique", 50)),
-            infer_categorical_max_ratio=float(
-                cfg.get("infer_categorical_max_ratio", 0.05)),
-            reuse_best_params=reuse_best_params,
-            bo_sample_limit=bo_sample_limit,
-            cache_predictions=cache_predictions,
-            prediction_cache_dir=prediction_cache_dir,
-            prediction_cache_format=prediction_cache_format,
-            cv_strategy=cv_strategy or split_strategy,
-            cv_group_col=cv_group_col or split_group_col,
-            cv_time_col=cv_time_col or split_time_col,
-            cv_time_ascending=cv_time_ascending,
-            cv_splits=cv_splits,
-            ft_oof_folds=ft_oof_folds,
-            ft_oof_strategy=ft_oof_strategy,
-            ft_oof_shuffle=ft_oof_shuffle,
-            save_preprocess=save_preprocess,
-            preprocess_artifact_path=preprocess_artifact_path,
-            plot_path_style=plot_path_style,
-        )
+        config_fields = getattr(ropt.BayesOptConfig,
+                                "__dataclass_fields__", {})
+        allowed_config_keys = set(config_fields.keys())
+        config_payload = {k: v for k,
+                          v in cfg.items() if k in allowed_config_keys}
+        config_payload.update({
+            "model_nme": model_name,
+            "resp_nme": cfg["target"],
+            "weight_nme": cfg["weight"],
+            "factor_nmes": feature_list,
+            "task_type": task_type,
+            "binary_resp_nme": binary_target,
+            "cate_list": categorical_features,
+            "prop_test": val_ratio,
+            "rand_seed": rand_seed,
+            "epochs": epochs,
+            "use_gpu": use_gpu,
+            "use_resn_data_parallel": use_resn_dp,
+            "use_ft_data_parallel": use_ft_dp,
+            "use_gnn_data_parallel": use_gnn_dp,
+            "use_resn_ddp": use_resn_ddp,
+            "use_ft_ddp": use_ft_ddp,
+            "use_gnn_ddp": use_gnn_ddp,
+            "output_dir": output_dir,
+            "xgb_max_depth_max": xgb_max_depth_max,
+            "xgb_n_estimators_max": xgb_n_estimators_max,
+            "resn_weight_decay": cfg.get("resn_weight_decay"),
+            "final_ensemble": bool(cfg.get("final_ensemble", False)),
+            "final_ensemble_k": int(cfg.get("final_ensemble_k", 3)),
+            "final_refit": bool(cfg.get("final_refit", True)),
+            "optuna_storage": optuna_storage,
+            "optuna_study_prefix": optuna_study_prefix,
+            "best_params_files": best_params_files,
+            "gnn_use_approx_knn": gnn_use_ann,
+            "gnn_approx_knn_threshold": gnn_threshold,
+            "gnn_graph_cache": gnn_graph_cache,
+            "gnn_max_gpu_knn_nodes": gnn_max_gpu_nodes,
+            "gnn_knn_gpu_mem_ratio": gnn_gpu_mem_ratio,
+            "gnn_knn_gpu_mem_overhead": gnn_gpu_mem_overhead,
+            "region_province_col": region_province_col,
+            "region_city_col": region_city_col,
+            "region_effect_alpha": region_effect_alpha,
+            "geo_feature_nmes": geo_feature_nmes,
+            "geo_token_hidden_dim": geo_token_hidden_dim,
+            "geo_token_layers": geo_token_layers,
+            "geo_token_dropout": geo_token_dropout,
+            "geo_token_k_neighbors": geo_token_k_neighbors,
+            "geo_token_learning_rate": geo_token_learning_rate,
+            "geo_token_epochs": geo_token_epochs,
+            "ft_role": ft_role,
+            "ft_feature_prefix": ft_feature_prefix,
+            "ft_num_numeric_tokens": ft_num_numeric_tokens,
+            "reuse_best_params": reuse_best_params,
+            "bo_sample_limit": bo_sample_limit,
+            "cache_predictions": cache_predictions,
+            "prediction_cache_dir": prediction_cache_dir,
+            "prediction_cache_format": prediction_cache_format,
+            "cv_strategy": cv_strategy or split_strategy,
+            "cv_group_col": cv_group_col or split_group_col,
+            "cv_time_col": cv_time_col or split_time_col,
+            "cv_time_ascending": cv_time_ascending,
+            "cv_splits": cv_splits,
+            "ft_oof_folds": ft_oof_folds,
+            "ft_oof_strategy": ft_oof_strategy,
+            "ft_oof_shuffle": ft_oof_shuffle,
+            "save_preprocess": save_preprocess,
+            "preprocess_artifact_path": preprocess_artifact_path,
+            "plot_path_style": plot_path_style or "nested",
+        })
+        config_payload = {
+            k: v for k, v in config_payload.items() if v is not None}
+        config = ropt.BayesOptConfig(**config_payload)
+        model = ropt.BayesOptModel(train_df, test_df, config=config)
 
         if plot_requested:
             plot_cfg = cfg.get("plot", {})

{ins_pricing-0.3.0 → ins_pricing-0.3.1}/ins_pricing/modelling/core/bayesopt/models/model_ft_trainer.py

@@ -626,6 +626,7 @@ class FTTransformerSklearn(TorchTrainerMixin, nn.Module):
         best_state = None
         patience_counter = 0
         is_ddp_model = isinstance(self.ft, DDP)
+        use_collectives = dist.is_initialized() and is_ddp_model
 
         clip_fn = None
         if self.device.type == 'cuda':
@@ -669,7 +670,7 @@ class FTTransformerSklearn(TorchTrainerMixin, nn.Module):
                             device=X_num_b.device)
                         local_bad = 0 if bool(torch.isfinite(batch_loss)) else 1
                         global_bad = local_bad
-                        if dist.is_initialized():
+                        if use_collectives:
                             bad = torch.tensor(
                                 [local_bad],
                                 device=batch_loss.device,
@@ -774,7 +775,7 @@ class FTTransformerSklearn(TorchTrainerMixin, nn.Module):
                             total_n += float(end - start)
                     val_loss_tensor[0] = total_val / max(total_n, 1.0)
 
-                if dist.is_initialized():
+                if use_collectives:
                     dist.broadcast(val_loss_tensor, src=0)
                 val_loss_value = float(val_loss_tensor.item())
                 prune_now = False
@@ -806,7 +807,7 @@ class FTTransformerSklearn(TorchTrainerMixin, nn.Module):
                     if trial.should_prune():
                         prune_now = True
 
-                if dist.is_initialized():
+                if use_collectives:
                     flag = torch.tensor(
                         [1 if prune_now else 0],
                         device=loss_tensor_device,