update state tracking documentation

Author: geraintpalmer

docs/Guides/state_trackers.rst

@@ -50,5 +50,9 @@ From this we can obtain the proportion of time the system spend in each state::
 
 So the system was in state :code:`0` (no individuals in the system) 55.4% of the time, in state :code:`1` (one individual in the system) 24.7% of the time, state :code:`2` (two individuals in the system) 13.1% of the time, and state :code:`3` (three individuals in the system) 6.8% of the time.
 
+If a warm up and cool down time is required when calculating the state probabilities, we can put in an observation period. For example, if we with to find the proportion of time the system spend in each state, between dates :code:`50` and :code:`200`, then we can use the following::
+
+    >>> Q.statetracker.state_probabilities(observation_period=(50, 200)) # doctest:+SKIP
+
+Note that different trackers represent different states in different ways, see :ref:`refs-statetrackers` for a list of implemented trackers.
 
-Note that different trackers represent different states in different ways, see :ref:`refs-statetrackers` for a list of implemented trackers.

docs/Reference/state_trackers.rst

@@ -8,6 +8,7 @@ Currently Ciw has the following state trackers:
 
 - :ref:`population`
 - :ref:`nodepop`
+- :ref:`nodepopsubset`
 - :ref:`nodeclssmatrix`
 - :ref:`naiveblock`
 - :ref:`matrixblock`
@@ -49,6 +50,24 @@ The Simulation object takes in the optional argument :code:`tracker` used as fol
     >>> Q = ciw.Simulation(N, tracker=ciw.trackers.NodePopulation()) # doctest:+SKIP
 
 
+.. _nodepopsubset:
+
+--------------------------------
+The NodePopulationSubset Tracker
+--------------------------------
+
+The NodePopulationSubset Tracker, similar to the NodePopulation Tracker, records the number of customers at each node. However this allows users to only track a subset of the nodes in the system.
+States take the form of list of numbers. An example of tracking a three node queueing network is shown below::
+
+    (2, 0, 5)
+
+This denotes that there are two customers at the first observed node, no customers at the second observed node, and five customers at the third observed node.
+
+The Simulation object takes in the optional argument :code:`tracker`, which takes an argument :code:`observed_nodes` a list of node numbers to observe, used as follows (observing the first, second, and fifth nodes)::
+
+    >>> Q = ciw.Simulation(N, tracker=ciw.trackers.NodePopulationSubset([0, 1, 4])) # doctest:+SKIP
+
+
 .. _nodeclssmatrix:
 
 ---------------------------