mlr3hf | Hugging Face Objects | Data | Arguments | Active Bindings | hfdt() and htsk() | Working with hfdt() | Datasets with Predefined Splits | Using a Validation Split | Using tsk() | Listing Datasets | API Authentication | Caching | Supported Data Formats

Step 1: Search for a Dataset | Step 2: Create an HFData Object | Step 3: Select a Configuration | Step 4: Create an mlr3 Task | Shortcut: Specify Everything at Once

Core Concepts | Basic Example | Permutation Feature Importance (PFI) | Leave-One-Covariate-Out (LOCO) | Feature Samplers | Detailed Scoring Information | Observation-wise losses and importances | Statistical Inference | Using Pre-trained Learners | Parallelization | Example with future | Example with mirai

Ordering Features | Accessing Internals From $state | Default vs. Initial Parameters | Complex Defaults

Defining the Parameter Set of a Learner | Train function | Predict function | Optional Extractors | Testing the learner | Autotest | Skipping tests for optional dependencies | Special considerations for Python learners | Checking Parameters | Documenting the Learner | Example Templates | When to Use Custom Templates | Creating a Custom Template | Using Templates in Learner Files | Contributing to mlr3extralearners

Preparations | Benchmark

Setup | Base Class: FeatureSampler | Key Properties | Permutation Sampler | Marginal Reference Sampler | Conditional Samplers | Gaussian Conditional Sampler | ARF Sampler | Ctree Conditional Sampler | kNN Conditional Sampler | Example 1: All-numeric conditioning (Euclidean distance) | Example 2: Mixed-type conditioning (Gower distance) | Knockoff Samplers | Gaussian Knockoffs | Summary

Adding new Tuners | Adding a new Tuner | Template | Optimize method | Writing a custom iteration | Calling an external optimization function | Assign result method | Transform optimizer to tuner | Add unit tests

Introduction | Overview of Simulation Settings | 1. Correlated Features DGP | Mathematical Model | Causal Structure | Usage Example | 2. Mediated Effects DGP | 3. Confounding DGP | 4. Interaction Effects DGP | 5. Independent Features DGP (Baseline) | Expected Behavior | 6. Ewald et al. (2024) DGP

Parameters (using paradox) | Reference Based Objects | Defining a Parameter Space | Domain Representing Single Parameters | Type / Range Checking | Parameter Sets | Values in a ParamSet | Dependencies | Vector Parameters | Parameter Sampling | Parameter Designs | Grid Design | Random Sampling | Generalized Sampling: The Sampler Class | 1D-Samplers | Hierarchical Sampler | Joint Sampler | SamplerUnif | Parameter Transformation | Transformation between Types | Defining a Tuning Spaces | Creating ParamSets | Transformations (trafo) | Automatic Factor Level Transformation | Parameter Dependencies (depends) | Creating Tuning ParamSets from other ParamSets

Getting Started | Construction and Parameterisation | Parameters in distr6 | Representing a distribution | d/p/q/r | Mathematical and Statistical Results | Listing in distr6 | Further Documentation

3D Visualization (spatiotemporal) | Examples of spatial partitioning (2D)

Introduction | Creating a spatial Task | Contributed reflections by | Task Type | Task Column Roles | Resampling Methods | Examples Tasks | Upstream Packages and Scientific References | References

General Case Example: PipeOpCopy | First Steps: Inheriting from PipeOp | Channel Definitions | Train and Predict | Putting it Together | Special Case: Preprocessing | Example: PipeOpDropNA | Example: PipeOpScaleAlways | Special Case: Preprocessing with Simple Train | Example: PipeOpDropConst | Example: PipeOpScaleAlwaysSimple | Hyperparameters | Hyperparameter Example: PipeOpScale

Constructing a Set | The different kind of sets | Comparisons and Containedness | Algebra of Sets | Union of Sets | Relative Complement | Cartesian Product | Intersection | Wrappers | Going Forward