.. _examples:

========
Examples
========

.. seealso::

   To explore the trusted and explainable methods we use under the hood, explore our free `knowledge base <https://www.digilab.co.uk/posts>`_.


Tutorials 
~~~~~~~~~

.. list-table::
   :header-rows: 1

   * - Name
     - Description
     - GitHub Link
   * - Quickstart
     - Get started with twinLab. 
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Quickstart.ipynb>`__
   * - Prediction
     - Make predictions with your model. 
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Prediction.ipynb>`__
   * - Recommend: Explore
     - Figure out how to add the most informative new data points to your training set. 
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Recommend_explore.ipynb>`__
   * - Recommend: Optimize
     - Solve an optimization problem by finding data points with your trained emulator. 
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Recommend_optimize.ipynb>`__
   * - Noisy data  
     - Learn how to handle noisy observations in your data using twinLab.
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Noisy_Data.ipynb>`__
   * - Noiseless data
     - Train and use a model that assumes no noise in the observations.
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Noiseless_data.ipynb>`__
   * - Model selection
     - How twinLab can automate the process of kernel selection and composition. 
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/model-selection.ipynb>`__
   * - Bayesian optimisation
     - Finding the maxima of an unknown function using twinLab. 
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/bayesian_optimisation.ipynb>`__
   * - Functional decomposition 
     - Effectively processing the training dataset.
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/analyse-dataset.ipynb>`__
   * - Dimensionality reduction
     - Truncating the number of features to accelerate your emulator's training.
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Dimensionality_reduction.ipynb>`__
   * - Initial design
     - Recommendations of points for sampling before training an emulator.
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/initial_design.ipynb>`__
   * - Mixture of experts model
     - Train and use a model that uses a combination of multiple GP experts.
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Mixture_of_experts.ipynb>`__
   * - Classification
     - Train and use a model for binary classification problems.
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/Classification.ipynb>`__ 
   * - Solving inverse problems
     - Determining the underlying conditions (inputs) that led to observed outcomes.
     - `View Jupyter Notebook <https://github.com/digiLab-ai/twinLab-Tutorials/blob/main/notebooks/solve_inverse_linear.ipynb>`__

.. seealso::

   Now you've completed our tutorials, start constructing emulators for your own problems with help from our :ref:`Python documentation <api>`.

   Let us know if you get stuck - we're always `happy to help <https://www.digilab.co.uk/contact>`_!