0.3.0

What's Changed

• Fix: Calculate ROC-AUC using predicted probabilities instead of class labels by @shankarpandala in #533
• feat: Add K-fold cross-validation and built-in predict function by @shankarpandala in #534
• Fix #324: Prevent custom_metric crash with array length mismatch by @shankarpandala in #535
• Add Precision and Recall metrics to LazyClassifier by @shankarpandala in #536
• Fix #438: Improve verbosity control with tqdm disable parameter by @shankarpandala in #537
• Fix #455: Add PerpetualBooster support by @shankarpandala in #538
• Fix #440: Add timeout parameter to stop slow algorithms by @shankarpandala in #539
• Fix #436: Handle boolean DataFrame features correctly by @shankarpandala in #540
• Add Intel Extension for Scikit-learn support (#382) by @shankarpandala in #541
• feat: Add categorical_encoder parameter to LazyClassifier and LazyRegressor by @shankarpandala in #542
• Refactor Supervised.py: add type hints, logging, input validation, and improve code organization by @shankarpandala in #547
• feat: Add LazyForecaster time series forecasting — v0.3.0a1 by @shankarpandala in #548
• feat: Enable GPU support for models that support GPU acceleration by @shankarpandala in #550
• 0.3.0a2 by @shankarpandala in #551
• Merge pull request #551 from shankarpandala/dev by @shankarpandala in #552
• fix: Resolve all GitHub Actions CI failures (lint, docs, conda-build) by @shankarpandala in #553
• fix: Resolve release blockers for 0.3.0a2 PyPI publish by @shankarpandala in #556
• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #558
• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #560
• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #561
• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #562
• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #563
• Add time series visualization and diagnostics modules by @shankarpandala in #564
• Use lazy imports for torch/timesfm to avoid OpenMP deadlocks by @shankarpandala in #565
• Bump version to 0.3.0 release by @shankarpandala in #566

Full Changelog: v0.2.16...v0.3.0

0.3.0a5

What's Changed

• Add time series visualization and diagnostics modules by @shankarpandala in #564

Full Changelog: v0.3.0a4...v0.3.0a5

v0.3.0a4

What's Changed

• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #563

Full Changelog: v0.3.0a3...v0.3.0a4

v0.3.0a3

What's Changed

• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #560
• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #561
• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #562

Full Changelog: v0.3.0a2...v0.3.0a3

v0.3.0a2

What's Changed

• feat: Enable GPU support for models that support GPU acceleration by @shankarpandala in #550
• 0.3.0a2 by @shankarpandala in #551
• Merge pull request #551 from shankarpandala/dev by @shankarpandala in #552
• fix: Resolve all GitHub Actions CI failures (lint, docs, conda-build) by @shankarpandala in #553
• fix: Resolve release blockers for 0.3.0a2 PyPI publish by @shankarpandala in #556
• Claude/release blockers 0.3.0a2 4 i4 dp by @shankarpandala in #558

Full Changelog: v0.3.0a1...v0.3.0a2

v0.3.0a1 — LazyForecaster (Time Series)

What's New

Time Series Forecasting — LazyForecaster

Benchmark 26 forecasting models in a single call:

• Statistical: Naive, SeasonalNaive, SimpleExpSmoothing, Holt, HoltWinters (add/mul), Theta, SARIMAX, AutoARIMA
• Machine Learning: LinearRegression, Ridge, Lasso, ElasticNet, KNN, SVR, DecisionTree, RandomForest, GradientBoosting, AdaBoost, Bagging, ExtraTrees, XGBoost, LightGBM
• Deep Learning: LSTM, GRU (via PyTorch)
• Foundation Model: Google TimesFM 2.5 (200M-parameter zero-shot pretrained transformer)

Features

• Automatic seasonal period detection via autocorrelation (ACF)
• Exogenous variable support for SARIMAX, AutoARIMA, and ML models
• Cross-validation with expanding window (TimeSeriesSplit)
• New forecasting metrics: MAPE, SMAPE, MASE
• New install extras: pip install lazypredict[timeseries], [deeplearning], [foundation]
• Added categorical_encoder parameter to LazyClassifier and LazyRegressor
• Refactored Supervised.py with type hints, logging, and input validation

Install

pip install lazypredict==0.3.0a1

v0.2.16

What's Changed

• add requirements.txt for documentation dependencies by @shankarpandala in #494
• Add documentation structure and examples for Lazy Predict by @shankarpandala in #495
• 0.2.16 by @shankarpandala in #496

Full Changelog: v0.2.15...v0.2.16

v0.2.15

What's Changed

• changed way of releasing package by @shankarpandala in #471
• Updated way of releasing by @shankarpandala in #472
• Mlflow integration by @shankarpandala in #491
• bump version to 2.13 in setup.py and init.py; add .bumpversion.cf… by @shankarpandala in #492
• bump version to 2.15 in setup.py, init.py, and .bumpversion.cfg by @shankarpandala in #493

Full Changelog: v0.2.13...v0.2.15

v0.2.14-alpha.1

Full Changelog: v0.2.14-alpha...v0.2.14-beta

Full Changelog: v0.2.14-alpha...v0.2.14-alpha.1

v0.2.14-alpha

Full Changelog: v0.2.13...v0.2.14-alpha