Quick check on the current computer | Benchmark memory and manifests | Evidence shipped with this release | Reproducing a real-hardware run | What counts as validation

Quick check on the current computer | Evidence shipped with this release | Reproducing a real-hardware run | What counts as validation

Build contract | Tested combinations | Linux CPU-only installation | Optional Linux performance configuration | macOS CPU-only installation | Windows CPU-only installation | Linux CUDA and cuVS installation | Audit the installed capability | Common failures | Checks and licenses | Session information

Overview | Choosing a Model | Installation and Compilation | macOS | Ubuntu and Debian | Fedora | Windows x86-64 | Verify the Compiled Libraries | Backend and CPU Configuration | Supported Computational Routes | Public API and Data Inputs | Choosing a Starting Model | Classification Tasks | Fit And Predict A Classifier | Evaluate Classification Predictions | Classification Heads | Kernel PLS For Classification | Classification Score Plots | Regression Tasks | Univariate Regression | Float64 and Float32 | Evaluate Regression Predictions | Multivariate Regression | OPLS For Regression | Cross-Validation | Ordinary k-fold CV | Grouped k-fold CV with constrain | Leave-one-out CV and leave-one-group-out CV | Component and hyperparameter optimization CV | Interpreting R2Y, Q2Y, and RMSD | Permutation-test p-values | Nested or double CV | SVD Utility | CUDA and Metal Availability | Helper Functions | Methodological Scope | Method References | Session Information

Installation model | Standard installation | macOS | Ubuntu and Debian | Fedora | Windows x86-64 | Windows ARM64 | NVIDIA CUDA on Linux or Windows x86-64 | Verify the installed capabilities | Runtime backend and CPU-core selection | Build-time environment variables | Troubleshooting | The package is already in use | An x64 library is linked into an ARM64 build | CUDA or Metal is unavailable after installation | The vignette is not found after a GitHub installation | Record a reproducible installation | Session information

Introduction | Example data | Inspect execution capabilities | Run a nearest-neighbor search | Approximate search and observed recall | Automatic selection | Metrics and edge cases | Dense input contract | Reuse an index for prediction | Optional CUDA execution | Session information

Evaluation contract | Standard metrics reported | Dataset catalog | Load one real scenario | Example benchmark run | Reproducibility scripts

Functions | Refinement | Simulation | refine_spatial_labels() contract | Evaluation | Benchmarking | Scoring glossary

Purpose and input contract | Quick start | Refine irregular coordinates | refine_spatial_labels() | Binary specialization | Effective 2D and genuine 3D coordinates | Multiple specimens | Deterministic CPU use | Clean categorical masks | corrupt_categorical_mask() | clean_categorical_mask() | evaluate_mask_cleaning() | Build and evaluate benchmark objects | spatial_benchmark() | evaluate_spatial_refinement() | benchmark_spatial_refiners() | Simulation functions | simulate_spatial_clusters() | simulate_spatial_domains() | simulate_complex_spatial_domains() | simulate_gradient_regions() | simulate_volumetric_domains() | Bundled real benchmarks | available_spatial_benchmarks() | load_spatial_benchmark() | End-to-end analysis template | Reporting checklist

Evaluation contract | Standard metrics reported | Dataset catalog | Load one real scenario | Example benchmark run | Reproducibility scripts

Functions | Refinement | Simulation | refine_spatial_labels() contract | Evaluation | Benchmarking | Scoring glossary

Purpose and input contract | Quick start | Refine irregular coordinates | refine_spatial_labels() | Binary specialization | Effective 2D and genuine 3D coordinates | Multiple specimens | Deterministic CPU use | Clean categorical masks | corrupt_categorical_mask() | clean_categorical_mask() | evaluate_mask_cleaning() | Build and evaluate benchmark objects | spatial_benchmark() | evaluate_spatial_refinement() | benchmark_spatial_refiners() | Simulation functions | simulate_spatial_clusters() | simulate_spatial_domains() | simulate_complex_spatial_domains() | simulate_gradient_regions() | simulate_volumetric_domains() | Bundled real benchmarks | available_spatial_benchmarks() | load_spatial_benchmark() | End-to-end analysis template | Reporting checklist

Overview | Select the backend and CPU core count | A first analysis | A bundled Bioconductor expression dataset | Parameter philosophy and intended scope | Choosing between UMAP and t-SNE | Optional downstream graph clustering | PCA and reproducible t-SNE initialization | Reusing a nearest-neighbor graph | Independent UMAP initialization | UMAP graph modes | Float32 and memory-conscious analysis | Embedding new observations | Landmarking | Measuring embedding quality | CPU, Metal, and CUDA | Package boundary | UMAP implementation | Interpolation-based t-SNE implementation | Graph clustering and landmarking | Backend reporting and software acknowledgements | A practical analysis checklist | References

Overview | Select default execution resources | Basic workflow | Bioconductor objects | Migration from KODAMA 2.4 | Innovations | Renamed or deprecated interfaces | Session information

Introduction | 2 Installation | 2.1 Installation via CRAN | 2.3 Compatibility issues | 3 Getting Started | Prostate Data | Function: txtsummary | Example with the prostate dataset | Function: continuous.test | Comparing Continuous Variables Across Groups | Function Parameters | Example 1: Wilcoxon Rank-Sum Test (2 Groups, Non-Parametric) | Function: categorical.test | Comparing Categorical Variables Across Groups | Example 1: Unordered Categorical Variable (Fisher's Exact Test) | Example 2: Ordered Categorical Variable (Jonckheere–Terpstra Test) | Function: correlation.test | Function: multi_analysis | Function: intersect | Function: frequency_matching | Conclusion