.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "gallery/print_train_and_save_agent.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        Click :ref:`here <sphx_glr_download_gallery_print_train_and_save_agent.py>`
        to download the full example code

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_gallery_print_train_and_save_agent.py:


==========================================
Train a PPO agent on Accenta's environment
==========================================

This example shows how to train a PPO agent.

.. GENERATED FROM PYTHON SOURCE LINES 13-14

Import required packages

.. GENERATED FROM PYTHON SOURCE LINES 14-22

.. code-block:: default


    from rlenv.envs.wall.core import AccentaEnv

    import torch as th
    from stable_baselines3 import PPO
    from rlagent.data import ppo_trained_model_example_path



.. GENERATED FROM PYTHON SOURCE LINES 23-24

Make the environment

.. GENERATED FROM PYTHON SOURCE LINES 24-28

.. code-block:: default


    env = AccentaEnv()



.. GENERATED FROM PYTHON SOURCE LINES 29-30

Make the agent's policy

.. GENERATED FROM PYTHON SOURCE LINES 30-38

.. code-block:: default


    # Custom actor (pi) and value function (vf) networks
    # of two layers of size 32 each with Relu activation function
    # https://stable-baselines3.readthedocs.io/en/master/guide/custom_policy.html#custom-network-architecture
    policy_kwargs = dict(activation_fn=th.nn.ReLU,
                         net_arch=[dict(pi=[32, 32], vf=[32, 32])])



.. GENERATED FROM PYTHON SOURCE LINES 39-40

Make the agent

.. GENERATED FROM PYTHON SOURCE LINES 40-44

.. code-block:: default


    model = PPO("MlpPolicy", env, policy_kwargs=policy_kwargs, verbose=1)



.. GENERATED FROM PYTHON SOURCE LINES 45-46

Train the agent

.. GENERATED FROM PYTHON SOURCE LINES 46-50

.. code-block:: default


    model.learn(total_timesteps=10000)



.. GENERATED FROM PYTHON SOURCE LINES 51-52

Save the trained agent

.. GENERATED FROM PYTHON SOURCE LINES 52-56

.. code-block:: default


    model.save(ppo_trained_model_example_path())



.. GENERATED FROM PYTHON SOURCE LINES 57-58

Reload the trained agent

.. GENERATED FROM PYTHON SOURCE LINES 58-61

.. code-block:: default


    del model
    model = PPO.load(ppo_trained_model_example_path(), env=env)


.. rst-class:: sphx-glr-timing

   **Total running time of the script:** ( 0 minutes  0.000 seconds)


.. _sphx_glr_download_gallery_print_train_and_save_agent.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example


    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: print_train_and_save_agent.py <print_train_and_save_agent.py>`

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: print_train_and_save_agent.ipynb <print_train_and_save_agent.ipynb>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_