ogboost 0.5.5__py3-none-any.whl

ogboost/__init__.py

@@ -0,0 +1,8 @@
+# ogboost/__init__.py
+
+# Re-export key components from submodules
+from .data import load_wine_quality
+from .main import GradientBoostingOrdinal, concordance_index, LinkFunctions
+
+# Define the public API
+__all__ = ["load_wine_quality", "GradientBoostingOrdinal", "concordance_index", "LinkFunctions"]

ogboost/data.py

@@ -0,0 +1,103 @@
+"""
+Module for loading the Wine Quality Dataset.
+
+This module provides functionality to download and load the Wine Quality Dataset from the 
+UCI Machine Learning Repository. The dataset includes information about physicochemical 
+tests (e.g., pH, alcohol content) and quality scores for red and white wines.
+
+The module contains the following functionality:
+- Automatically downloads the datasets if not found locally.
+- Loads the red and white wine datasets into pandas DataFrames.
+- Rescales the 'quality' column to start from 0 and ensures it is of integer type.
+
+Functions
+---------
+load_wine_quality():
+    Loads the wine quality dataset. Automatically downloads the datasets 
+    if they are not found locally and returns them as pandas DataFrames.
+
+_download_wine_quality_datasets(destination: str):
+    Downloads the wine quality datasets (red and white) from the UCI repository 
+    to the specified destination folder.
+
+Examples
+--------
+>>> from mymodule import load_wine_quality
+>>> red_wine, white_wine = load_wine_quality()
+>>> print(red_wine.head())
+>>> print(white_wine.head())
+"""
+import os
+import urllib.request
+import pandas as pd
+
+def load_wine_quality():
+    """
+    Loads the wine quality dataset from the UCI repository. 
+    If the datasets are not already downloaded,
+    they will be downloaded automatically.
+
+    Returns
+    -------
+    tuple of pandas.DataFrame
+        A tuple containing two DataFrames:
+        - The first for red wine data.
+        - The second for white wine data.
+    """
+    # Define the local paths for the datasets
+    dataset_folder = os.path.join(os.path.dirname(__file__), "data")
+    red_wine_path = os.path.join(dataset_folder, "winequality-red.csv")
+    white_wine_path = os.path.join(dataset_folder, "winequality-white.csv")
+
+    # Check if the datasets exist locally; if not, download them
+    if not os.path.exists(red_wine_path) or not os.path.exists(white_wine_path):
+        print("Datasets not found locally. Downloading from UCI repository...")
+        os.makedirs(dataset_folder, exist_ok=True)
+        _download_wine_quality_datasets(dataset_folder)
+
+    # Load the datasets into pandas DataFrames
+    red_wine_df = pd.read_csv(red_wine_path, sep=";")
+    white_wine_df = pd.read_csv(white_wine_path, sep=";")
+
+    # rescale quality to start from 0
+    red_wine_df["quality"] -= red_wine_df["quality"].min()
+    white_wine_df["quality"] -= white_wine_df["quality"].min()
+
+    # ensure data type for response is integer
+    red_wine_df["quality"] = red_wine_df["quality"].astype(int)
+    white_wine_df["quality"] = white_wine_df["quality"].astype(int)
+
+    return red_wine_df, white_wine_df
+
+
+def _download_wine_quality_datasets(destination: str):
+    """
+    Downloads the wine quality datasets from the UCI repository.
+
+    Parameters
+    ----------
+    destination : str
+        Directory where the datasets will be saved.
+    """
+    # UCI repository URLs
+    red_wine_url = (
+        "https://archive.ics.uci.edu/ml/machine-learning-databases/"
+        "wine-quality/winequality-red.csv"
+    )
+    white_wine_url = (
+        "https://archive.ics.uci.edu/ml/machine-learning-databases/"
+        "wine-quality/winequality-white.csv"
+    )
+
+    # Define file paths
+    red_wine_path = os.path.join(destination, "winequality-red.csv")
+    white_wine_path = os.path.join(destination, "winequality-white.csv")
+
+    # Download the datasets
+    print(f"Downloading red wine dataset to {red_wine_path}...")
+    urllib.request.urlretrieve(red_wine_url, red_wine_path)
+    print("Red wine dataset downloaded.")
+
+    print(f"Downloading white wine dataset to {white_wine_path}...")
+    urllib.request.urlretrieve(white_wine_url, white_wine_path)
+    print("White wine dataset downloaded.")