Presets API

Functions for accessing preset model configurations.

---

list_presets

from og_learn import list_presets

list_presets()

Print all available HV and LV presets.

---

get_hv_model

from og_learn.presets import get_hv_model

model = get_hv_model(name)

Get a High-Variance model instance.

Parameters:

Parameter	Type	Description
name	str	Preset name: 'lightgbm', 'biglightgbm', 'xgboost', 'catboost', 'random_forest', 'decision_tree', 'linear_regression'

Returns: Model instance with fit()/predict() methods

Example:

from og_learn.presets import get_hv_model

lgb = get_hv_model('lightgbm')
lgb.fit(X_train, y_train)
predictions = lgb.predict(X_test)

---

get_lv_model

from og_learn.presets import get_lv_model

model = get_lv_model(name, num_features, epochs=100)

Get a Low-Variance model instance.

Parameters:

Parameter	Type	Description
name	str	Preset name: 'mlp', 'bigmlp', 'resnet', 'transformer'
num_features	int	Number of input features
epochs	int	Training epochs (default: 100)

Returns: Model instance with fit()/predict() methods

Example:

from og_learn.presets import get_lv_model

mlp = get_lv_model('mlp', num_features=10, epochs=50)
mlp.fit(X_train, y_train)
predictions = mlp.predict(X_test)

---

Preset Configurations

HV_PRESETS

Dictionary of HV model configurations:

HV_PRESETS = {
    'lightgbm': {
        'n_estimators': 500,
        'num_leaves': 1200,
        'max_depth': 9,
        'learning_rate': 0.05,
        # ...
    },
    'biglightgbm': { ... },
    'xgboost': { ... },
    'catboost': { ... },
    'random_forest': { ... },
    'decision_tree': { ... },
    'linear_regression': { ... },
}

LV_PRESETS

Dictionary of LV model configurations:

LV_PRESETS = {
    'mlp': {
        'hidden_layers': [256, 128, 64],
        'dropout': 0.3,
        'batch_size': 256,
        'learning_rate': 0.001,
    },
    'bigmlp': { ... },
    'resnet': { ... },
    'transformer': { ... },
}