Utilities

Transition matrix validation

There are currently two helper-functions to validate a transition matrix: validate_matrix_sum() and validate_matrix_positive(). Both functions return a boolean indicating whether a transition matrix is valid or not.

First we make the necessary imports:

>>> from chain_simulator.utilities import validate_matrix_sum, validate_matrix_negative
>>> import numpy as np

Then we create an array and validate it:

>>> valid_matrix = np.array(
...    [
...       [0.0, 1.0, 0.0],
...       [0.0, 0.5, 0.5],
...       [0.0, 0.0, 1.0],
...    ]
... )
>>> validate_matrix_sum(valid_matrix)
True
>>> validate_matrix_positive(valid_matrix)
True

Logging errors

It may be helpful to know which row doesn’t sum to 1 or which probability is not positive. This information is logged to the console using the logging module from the Python standard library. It is disabled by default but can be enabled by adding the following statements to your code:

>>> import logging
>>> logging.basicConfig()

If we were to validate a faulty transition matrix:

>>> faulty_sum_matrix = np.array([[-1, 1, 0], [1, 0, -1], [0, -1, 1]])
>>> validate_matrix_sum(faulty_sum_matrix)
False

Then errors that have been found will be logged:

WARNING:chain_simulator.utilities:Row 0 sums to 0 instead of 1!
WARNING:chain_simulator.utilities:Row 1 sums to 0 instead of 1!
WARNING:chain_simulator.utilities:Row 2 sums to 0 instead of 1!