.. _sphx_glr_auto_examples_DeepLearning:

Deep learning
=============================



.. raw:: html

    <div class="sphx-glr-thumbnails">

.. thumbnail-parent-div-open

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This script implements a model to predict a dummy dataset using MultiHeadAttention. The model  has a similar structure to modern large language models, but with way less parameters.">

.. only:: html

  .. image:: /auto_examples/DeepLearning/images/thumb/sphx_glr_Attention_thumb.png
    :alt:

  :ref:`sphx_glr_auto_examples_DeepLearning_Attention.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Deep learning with Attention</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This script implements a model to predict values of a simple sine function. It uses recurrent layers  to handle the sequential nature of the sine function.">

.. only:: html

  .. image:: /auto_examples/DeepLearning/images/thumb/sphx_glr_TimeSeriesAnalysis_thumb.png
    :alt:

  :ref:`sphx_glr_auto_examples_DeepLearning_TimeSeriesAnalysis.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Recurrent networks for time series analysis</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This script implements a model to classify the iris dataset. This model uses LSTM and  RNN layers with a Bidirectional wrapper for the predictions.">

.. only:: html

  .. image:: /auto_examples/DeepLearning/images/thumb/sphx_glr_BidirectionalClassification_thumb.png
    :alt:

  :ref:`sphx_glr_auto_examples_DeepLearning_BidirectionalClassification.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Bidirectional recurrent layers</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This script implements a model to predict values on a simple quadratic surface. It also  showcases some regularisation methods like Dropout and BatchNorm.">

.. only:: html

  .. image:: /auto_examples/DeepLearning/images/thumb/sphx_glr_Regression_thumb.png
    :alt:

  :ref:`sphx_glr_auto_examples_DeepLearning_Regression.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Regression with neural networks</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This script implements a model using Kolmogorov-Arnold networks. It fits to a simple  quadratic surface using only a few parameters.">

.. only:: html

  .. image:: /auto_examples/DeepLearning/images/thumb/sphx_glr_KANs_thumb.png
    :alt:

  :ref:`sphx_glr_auto_examples_DeepLearning_KANs.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Kolmogorov-Arnold Networks</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This script implements a model to classify the MNIST dataset. The model mainly consists of  convolutonal layers and pooling layers with a few dense layers at the end. As the script is  only for demonstration purposes, only 100 first datapoints are used to make the training faster.  For a full example, change the parameter n to 60000. If n is increased, more epochs may need  to be added and other hyperparameters tuned.">

.. only:: html

  .. image:: /auto_examples/DeepLearning/images/thumb/sphx_glr_ImageClassification_thumb.png
    :alt:

  :ref:`sphx_glr_auto_examples_DeepLearning_ImageClassification.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">MNIST Image classification</div>
    </div>


.. thumbnail-parent-div-close

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /auto_examples/DeepLearning/Attention
   /auto_examples/DeepLearning/TimeSeriesAnalysis
   /auto_examples/DeepLearning/BidirectionalClassification
   /auto_examples/DeepLearning/Regression
   /auto_examples/DeepLearning/KANs
   /auto_examples/DeepLearning/ImageClassification