skyulf-core 0.3.2

The core machine learning library for Skyulf.

Skyulf Core

Skyulf Core (skyulf-core) is the standalone machine learning library that powers the Skyulf MLOps platform. It provides a robust, type-safe, and modular set of tools for:

• Data Preprocessing: A comprehensive suite of transformers for cleaning, scaling, encoding, and feature engineering.
• Text & NLP: Count / TF-IDF / Hashing vectorizers, a configurable Tokenizer, Multinomial & Bernoulli Naive Bayes, and optional dense Sentence Embeddings (skyulf-core[nlp]).
• Modeling: Unified interfaces for classification and regression models, wrapping Scikit-Learn and other libraries.
• Pipeline Management: Tools to build, serialize, and execute complex ML pipelines.
• Tuning: Advanced hyperparameter tuning capabilities with support for Grid Search, Random Search, and Optuna.
• Evaluation: Standardized metrics and evaluation schemas for model performance tracking.

Website & Documentation

Visit the full documentation and project site for guides, API reference, and examples:

• Project site / docs: https://www.skyulf.com
• Repository: https://github.com/flyingriverhorse/Skyulf
• PyPI package: https://pypi.org/project/skyulf-core

Installation

pip install skyulf-core

# EDA-focused install (core EDA + optional advanced EDA + visualization)
pip install skyulf-core[eda,viz]

# For visualization support (Rich dashboard + Matplotlib plots)
pip install skyulf-core[viz]

# For advanced EDA add-ons (sentiment + causal discovery)
pip install skyulf-core[eda]

# For hyperparameter tuning engines
pip install skyulf-core[tuning]

# For imbalance-aware preprocessing (e.g., SMOTE)
pip install skyulf-core[preprocessing-imbalanced]

# For XGBoost modeling nodes
pip install skyulf-core[modeling-xgboost]

Quick Start: Automated EDA

Skyulf Core includes a powerful automated Exploratory Data Analysis (EDA) module.

import polars as pl
from skyulf.profiling.analyzer import EDAAnalyzer
from skyulf.profiling.visualizer import EDAVisualizer

# 1. Load Data
df = pl.read_csv("data.csv")

# 2. Analyze
analyzer = EDAAnalyzer(df)
# Optional: Manually specify special columns if auto-detection fails
profile = analyzer.analyze(
    target_col="target",
    date_col="timestamp",  # Optional
    lat_col="latitude",    # Optional
    lon_col="longitude"    # Optional
)

# 3. Visualize
viz = EDAVisualizer(profile, df)
viz.summary()  # Prints rich terminal dashboard
viz.plot()     # Opens interactive plots

Features

• Automated EDA: One-line profiling with Data Quality, Outliers, Time Series, and Geospatial analysis.
• Type-Safe: Built with modern Python type hints and Pydantic models.
• Modular: Use only the components you need.
• Serializable: All components are designed to be easily serialized for storage and deployment.
• Extensible: Easy to extend with your own custom transformers and models.

License

This project is licensed under the terms of the Apache 2.0 license.