Description Usage Arguments Details Value Functions
Effects of Testing on un-encoded features and thresholds
1 2 3 4 5 |
... |
Arugments passed to approriate functions based on type-checking. See details. |
mem |
Matrix of memory strengths. |
nFeatures |
Number of features the memory for each item has |
thresh |
Matrix of memory thresholds. Threshold is the number of active features needed to support pattern completetion |
acc |
Logical matrix of performance for previous test |
LR |
Learning Rate. Gives the probability of learning a new features through practice |
TR |
Threshold Reduction rate. Describes the probability of removing the need for a remembering a particular feature following successfull recall |
FR |
Forgetting Rate. Gives the probability of forgetting a feature |
thresh |
Matrix of memory thresholds. Threshold is the number of active features needed to support pattern completetion. Must hold integer values. |
acc |
Logical matrix of performance for previous test |
TR |
Threshold Reduction rate. Describes the probability of removing the need for a remembering a particular feature following successfull recall |
test
is a conveince function, and will attempt to type-check
the codemem matrix to call the appropriate function (test_beta for doubles,
test_binomial for integers).
test_beta
and test_binomial
can (and should) be called directly though,
especially when fitting models and computational time matters. See Functions section
for more details.
List of two matrices. mem is a matrix of memory strengths following succcesful recall. thresh is a matrix of feature thresholds following succesful recall.
test_beta
: test_beta
is designed to be used with a matrix of
continous feature and threhsold values (represented as doubles) drawn from
a beta distribution
test_binomial
: test_binomial
is designed to be used with a matrix of
discrete feature and threshold values (represented as integers) drawn from
a binomial distribution
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.