{% import macros as m with context  %}

.. _w2_optimal_inventory:

Week 2: Optimal planning in the Inventory-environment
=============================================================================================================

{{ m.embed_game('week2_inventory_solved') }}

.. topic:: Controls
    :class: margin

    :kbd:`Space`
        Take an (optimal) action according to the DP algorithm
    :kbd:`0`, :kbd:`1`, :kbd:`2`
        Buy the given number of items.
    :kbd:`p`
        Automatically take optimal actions
    :kbd:`r`
        Reset the game

    .. rubric:: Run locally

    :gitref:`../irlc/lectures/lec02/lecture_02_optimal_inventory.py`




What you see
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This example shows what happens when you apply the DP algorithm to the inventory control problem. The optimal
policy is in this case very simple: buy one item if the inventory is empty.


How it works
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
The planning occurs using the DP algorithm that you implement during week 2 and to make the visualization work
I use the DP Agent you will implement during week 2.