tessl/pypi-flaml

A fast library for automated machine learning and tuning

—

Pending

Quality

Pending

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

Overview

Eval results

Files

Default Estimators and Hyperparameter Suggestions

Name: tessl/pypi-flaml
Author: tessl

Enhanced versions of popular machine learning estimators with optimized hyperparameters and intelligent hyperparameter suggestion functions. The default module provides pre-tuned estimators and utilities for automatic hyperparameter configuration.

Capabilities

Enhanced Estimators

Pre-configured versions of popular scikit-learn estimators with optimized hyperparameters based on FLAML's research and extensive testing.

class ExtraTreesClassifier:
    """Enhanced ExtraTreesClassifier with optimized hyperparameters."""
    def __init__(self, n_estimators=100, **kwargs): ...
    def fit(self, X, y, **kwargs): ...
    def predict(self, X): ...
    def predict_proba(self, X): ...

class ExtraTreesRegressor:
    """Enhanced ExtraTreesRegressor with optimized hyperparameters."""
    def __init__(self, n_estimators=100, **kwargs): ...
    def fit(self, X, y, **kwargs): ...
    def predict(self, X): ...

class LGBMClassifier:
    """Enhanced LightGBM classifier with optimized hyperparameters."""
    def __init__(self, n_estimators=100, **kwargs): ...
    def fit(self, X, y, **kwargs): ...
    def predict(self, X): ...
    def predict_proba(self, X): ...

class LGBMRegressor:
    """Enhanced LightGBM regressor with optimized hyperparameters."""
    def __init__(self, n_estimators=100, **kwargs): ...
    def fit(self, X, y, **kwargs): ...
    def predict(self, X): ...

class RandomForestClassifier:
    """Enhanced RandomForest classifier with optimized hyperparameters."""
    def __init__(self, n_estimators=100, **kwargs): ...
    def fit(self, X, y, **kwargs): ...
    def predict(self, X): ...
    def predict_proba(self, X): ...

class RandomForestRegressor:
    """Enhanced RandomForest regressor with optimized hyperparameters."""
    def __init__(self, n_estimators=100, **kwargs): ...
    def fit(self, X, y, **kwargs): ...
    def predict(self, X): ...

class XGBClassifier:
    """Enhanced XGBoost classifier with optimized hyperparameters."""
    def __init__(self, n_estimators=100, **kwargs): ...
    def fit(self, X, y, **kwargs): ...
    def predict(self, X): ...
    def predict_proba(self, X): ...

class XGBRegressor:
    """Enhanced XGBoost regressor with optimized hyperparameters."""
    def __init__(self, n_estimators=100, **kwargs): ...
    def fit(self, X, y, **kwargs): ...
    def predict(self, X): ...

Estimator Enhancement

Function for enhancing existing estimators with FLAML's optimized hyperparameters.

def flamlize_estimator(estimator_class, task="classification", **kwargs):
    """
    Enhance an estimator class with FLAML's optimized hyperparameters.
    
    Args:
        estimator_class: Scikit-learn compatible estimator class
        task (str): Task type - 'classification' or 'regression'
        **kwargs: Additional parameters
        
    Returns:
        Enhanced estimator class with optimized hyperparameters
    """

Hyperparameter Suggestion Functions

Intelligent hyperparameter suggestion based on dataset characteristics and meta-learning.

def suggest_hyperparams(estimator_name, X, y, task="classification"):
    """
    Suggest hyperparameters for an estimator based on dataset characteristics.
    
    Args:
        estimator_name (str): Name of the estimator
        X: Training features
        y: Training target
        task (str): Task type - 'classification' or 'regression'
        
    Returns:
        dict: Suggested hyperparameters
    """

def suggest_config(estimator_name, X, y, task="classification", time_budget=60):
    """
    Suggest configuration including hyperparameters and training settings.
    
    Args:
        estimator_name (str): Name of the estimator
        X: Training features  
        y: Training target
        task (str): Task type
        time_budget (float): Available time budget in seconds
        
    Returns:
        dict: Suggested configuration
    """

def suggest_learner(X, y, task="classification"):
    """
    Suggest the best learner for given dataset characteristics.
    
    Args:
        X: Training features
        y: Training target
        task (str): Task type - 'classification' or 'regression'
        
    Returns:
        str: Recommended learner name
    """

def preprocess_and_suggest_hyperparams(X, y, task, estimator_name, time_budget=60):
    """
    Preprocess data and suggest hyperparameters.
    
    Args:
        X: Training features
        y: Training target
        task (str): Task type
        estimator_name (str): Name of the estimator
        time_budget (float): Available time budget
        
    Returns:
        tuple: (preprocessed_X, preprocessed_y, suggested_hyperparams)
    """

def meta_feature(X, y, task="classification"):
    """
    Extract meta-features from dataset for hyperparameter suggestion.
    
    Args:
        X: Training features
        y: Training target
        task (str): Task type
        
    Returns:
        dict: Meta-features of the dataset
    """

Usage Examples

Using Enhanced Estimators

from flaml.default import LGBMRegressor, XGBClassifier
import pandas as pd
from sklearn.model_selection import train_test_split

# Load your data
X, y = load_data()  # your dataset
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# Use enhanced LightGBM regressor with optimized hyperparameters
regressor = LGBMRegressor()
regressor.fit(X_train, y_train)
predictions = regressor.predict(X_test)

# Use enhanced XGBoost classifier
classifier = XGBClassifier()
classifier.fit(X_train, y_train)
probabilities = classifier.predict_proba(X_test)

Hyperparameter Suggestions

from flaml.default import suggest_hyperparams, suggest_learner

# Get hyperparameter suggestions
suggested_params = suggest_hyperparams("lgbm", X_train, y_train, task="regression")
print(f"Suggested hyperparameters: {suggested_params}")

# Get learner recommendation
best_learner = suggest_learner(X_train, y_train, task="classification")
print(f"Recommended learner: {best_learner}")

Install with Tessl CLI