config

config

Module: config.py

This module defines the configuration for the Hybrid Bias Correction (LSEQM+DL) workflow. It loads settings from a YAML configuration file and provides them as module-level variables.

Configuration is loaded via initialize_config(), which should be called once at startup (typically in the notebook or main script). If not called, default values are used.

The configuration includes: - Directory settings for the main project, input data, and output locations. - Default input file paths for IMERG, CPC, and the land-sea mask. - Variable names for precipitation in each dataset. - Output directories and filename templates. - CF-compliant encoding settings for NetCDF outputs. - Statistical fitting and quantile mapping parameters (GPD, EQM). - Deep learning model training parameters. - Runtime settings (interactive mode, default actions for existing files).

Author: Benny Istanto Applied Climatology Study Program, Department of Geophysics and Meteorology, Bogor Agricultural University, Indonesia Email: bennyistanto@apps.ipb.ac.id

with supervision from Prof. Rizaldi Boer and Dr. I Putu Santikayasa

Update: 2026.03

Functions

Name	Description
find_project_root	Locate the project root using a strong fingerprint.
initialize_config	Load configuration from a YAML file and update module-level variables.
setup_logging	Configure logging so messages appear in Colab, Jupyter, and scripts.

find_project_root

config.find_project_root(extra_candidates=None)

Locate the project root using a strong fingerprint.

A directory is accepted as the project root only when it contains ALL of the following (prevents false positives from unrelated projects):

src/__init__.py
src/config.py
config.yml or config.yaml

Works across Google Colab, WSL, Linux, macOS, and Windows without any hardcoded paths. The search order is:

Directory containing this file (src/config.py → parent).
Current working directory and its parent.
extra_candidates supplied by the caller (if any).
Google Drive deep scan (Colab only, up to 4 levels).

Parameters

Name	Type	Description	Default
extra_candidates	list of str	Additional directory paths to consider.	`None`

Returns

Name	Type	Description
	str	Absolute path to the project root.

initialize_config

config.initialize_config(config_path=None)

Load configuration from a YAML file and update module-level variables.

This function should be called once at the start of your notebook or script. If config_path is None, it looks for config.yml (or config.yaml) in the project root directory.

Parameters:

config_path : str, optional Path to the YAML configuration file. If None, searches for config.yml then config.yaml in the project root (one level above src/).

Returns:

dict The loaded configuration dictionary.

setup_logging

config.setup_logging(level=logging.INFO)

Configure logging so messages appear in Colab, Jupyter, and scripts.

Google Colab’s IPython kernel installs its own root-logger handler before user code runs, which silently swallows logging.basicConfig() calls (because basicConfig is a no-op when handlers already exist).

This function works around that by:

Using force=True (Python 3.8+) to replace any pre-existing config.
Attaching a StreamHandler(sys.stdout) so output lands in the notebook cell, not in a hidden stderr stream.

Safe to call multiple times - subsequent calls just reset the level.

Parameters

Name	Type	Description	Default
level	int	Logging level (default `logging.INFO`).	`logging.INFO`

# config { #src.config } `config` Module: config.py This module defines the configuration for the Hybrid Bias Correction (LSEQM+DL) workflow. It loads settings from a YAML configuration file and provides them as module-level variables. Configuration is loaded via initialize_config(), which should be called once at startup (typically in the notebook or main script). If not called, default values are used. The configuration includes: - Directory settings for the main project, input data, and output locations. - Default input file paths for IMERG, CPC, and the land-sea mask. - Variable names for precipitation in each dataset. - Output directories and filename templates. - CF-compliant encoding settings for NetCDF outputs. - Statistical fitting and quantile mapping parameters (GPD, EQM). - Deep learning model training parameters. - Runtime settings (interactive mode, default actions for existing files). **Author**: Benny Istanto Applied Climatology Study Program, Department of Geophysics and Meteorology, Bogor Agricultural University, Indonesia Email: bennyistanto@apps.ipb.ac.id with supervision from Prof. Rizaldi Boer and Dr. I Putu Santikayasa Update: 2026.03 ## Functions | Name | Description | | --- | --- | | [find_project_root](#src.config.find_project_root) | Locate the project root using a **strong fingerprint**. | | [initialize_config](#src.config.initialize_config) | Load configuration from a YAML file and update module-level variables. | | [setup_logging](#src.config.setup_logging) | Configure logging so messages appear in Colab, Jupyter, and scripts. | ### find_project_root { #src.config.find_project_root } ```python config.find_project_root(extra_candidates=None) ``` Locate the project root using a **strong fingerprint**. A directory is accepted as the project root only when it contains ALL of the following (prevents false positives from unrelated projects): * ``src/__init__.py`` * ``src/config.py`` * ``config.yml`` **or** ``config.yaml`` Works across Google Colab, WSL, Linux, macOS, and Windows without any hardcoded paths. The search order is: 1. Directory containing *this* file (``src/config.py`` → parent). 2. Current working directory and its parent. 3. ``extra_candidates`` supplied by the caller (if any). 4. Google Drive deep scan (Colab only, up to 4 levels). #### Parameters {.doc-section .doc-section-parameters} | Name | Type | Description | Default | |------------------|-------------|-----------------------------------------|-----------| | extra_candidates | list of str | Additional directory paths to consider. | `None` | #### Returns {.doc-section .doc-section-returns} | Name | Type | Description | |--------|--------------|------------------------------------| | | [str](`str`) | Absolute path to the project root. | ### initialize_config { #src.config.initialize_config } ```python config.initialize_config(config_path=None) ``` Load configuration from a YAML file and update module-level variables. This function should be called once at the start of your notebook or script. If config_path is None, it looks for config.yml (or config.yaml) in the project root directory. #### Parameters: {.doc-section .doc-section-parameters} config_path : str, optional Path to the YAML configuration file. If None, searches for config.yml then config.yaml in the project root (one level above src/). #### Returns: {.doc-section .doc-section-returns} dict The loaded configuration dictionary. ### setup_logging { #src.config.setup_logging } ```python config.setup_logging(level=logging.INFO) ``` Configure logging so messages appear in Colab, Jupyter, and scripts. Google Colab's IPython kernel installs its own root-logger handler before user code runs, which silently swallows ``logging.basicConfig()`` calls (because ``basicConfig`` is a no-op when handlers already exist). This function works around that by: * Using ``force=True`` (Python 3.8+) to replace any pre-existing config. * Attaching a ``StreamHandler(sys.stdout)`` so output lands in the notebook cell, not in a hidden stderr stream. Safe to call multiple times - subsequent calls just reset the level. #### Parameters {.doc-section .doc-section-parameters} | Name | Type | Description | Default | |--------|--------------|-------------------------------------------|----------------| | level | [int](`int`) | Logging level (default ``logging.INFO``). | `logging.INFO` |