vitor-aignosi
cf5111e520
feat: integrate PluginStore and MinIO repository into model manager activities
...
- Added PluginStore integration for model management.
- Replaced StorageRepository with MinIORepository in Activities, Cleanup, and Training classes.
- Updated training logic to handle validation files and improved data management.
- Enhanced configuration for MinIO and PluginStore in connectors.
- Removed deprecated model repository and storage repository files.
- Updated environment variable handling for new configurations.
2026-03-11 17:35:05 -03:00
Kou Kinoshita
14f3fbadbe
SIENTIAPDE-1579: Switched from replace logic to validation + mapping
2026-02-18 18:57:41 -03:00
Kou Kinoshita
a4d94dd2ff
SIENTIAPDE-1579: Added date format and convert methods
2026-02-13 18:41:46 -03:00
Bruno Domingues
06571011f2
SIENTIAPDE-1430: Introduce static_threshold parameter for static window removal.
...
This parameter allows customizing the threshold (1-1000) used when rem_static_win is enabled, defaulting to 1 if null.
Updates include parameter definition, business rule validation, repository logic for passing the threshold, documentation in README.md and PIPELINE_PARAMS_CHANGELOG.md, and new unit and integration tests.
2025-12-19 15:44:18 -03:00
Bruno Domingues
7cbb7022e5
SIENTIAPDE-1430: Refactor DataPreprocessor date filtering, make rce_train radius optional, and introduce constants for model names.
2025-12-19 12:28:07 -03:00
Bruno Domingues
06fd08dc70
SIENTIAPDE-1430: Introduce comprehensive integration testing with JSON-based scenarios and detailed README documentation. Enhance training workflow to support advanced model configurations, including polynomial regression with mandatory scaler validation. Ensure robust prediction handling by calculating training predictions (y_train_pred) before denormalization and automatically configuring datetime indices for time-series operations.
2025-12-18 17:05:10 -03:00
Bruno Domingues
6e8f87b2a3
SIENTIAPDE-1430: Implement advanced model training capabilities and enhanced data preprocessing. This includes support for Polynomial Regression with configurable degree and interaction terms, flexible per-variable lag configurations, and new data filtering options by date range and removed intervals. Comprehensive business validations are now enforced for all parameters, and MLflow logging has been extended to capture these detailed configurations. Additionally, Reduced Coulomb Energy (RCE) metrics are added for drift detection, with a new changelog documenting all pipeline parameter updates.
2025-12-17 21:39:43 -03:00
Bruno Domingues
94697215aa
SIENTIAPDE-1241: Refactor: Improve documentation, exception handling, and configuration in model manager. This commit enhances clarity and robustness by adding detailed docstrings to methods, standardizing exception handling with custom types, and simplifying MLflow configuration.
2025-10-22 21:48:12 -03:00
Bruno Domingues
5789a13023
SIENTIAPDE-1241: refactor train_model workflow due to I/O errors.
2025-10-22 15:37:56 -03:00
Bruno Domingues
305a39b44c
SIENTIAPDE-1253: Implement business rule validation for training parameters. Adds a validate_business_rules method to the TrainModelParams class to enforce constraints on training parameters, improving data integrity and preventing errors. Also updates tests to include target variable in variable columns and adds tests for business rule validations.
2025-10-15 16:16:09 -03:00
Bruno Domingues
94e11df803
SIENTIAPDE-1252: Remove experiment_description from TrainModelParams and related tests.
2025-10-09 17:48:43 -03:00
Bruno Domingues
04c94efd98
SIENTIAPDE-1249: Refactor TrainModelParams to use dataclass and add from_dict method for validation, remove no-cache-dir from pip install in quality gate workflow, and rename X_train/X_test to x_train/x_test in TrainModelResult.
2025-10-06 19:11:02 -03:00
Bruno Domingues
d8583cdae7
SIENTIAPDE-1249: Implement data models for Model Manager and add unit tests. This commit introduces data transfer objects (DTOs) and model classes for experiment status, training parameters, and training results, along with corresponding unit tests to ensure their correct behavior.
2025-10-06 16:49:18 -03:00
