This is the mlep Reference Manual, version 0.0.1, generated automatically by Declt version 4.0 beta 2 "William Riker" on Sun Dec 08 17:10:52 2024 GMT+0.
mlep/mlep.asd
mlep/package.lisp
mlep/macros/macros.lisp
mlep/datasets/iris.lisp
mlep/datasets/heights-weights.lisp
mlep/datasets/lenses.lisp
mlep/datasets/wages.lisp
mlep/utils/number-utils.lisp
mlep/utils/list-utils.lisp
mlep/utils/array-utils.lisp
mlep/utils/constants.lisp
mlep/utils/functions.lisp
mlep/utils/plot.lisp
mlep/core/generic.lisp
mlep/core/k-means.lisp
mlep/core/k-nearest.lisp
mlep/core/markov-chain.lisp
mlep/core/naive-bayes.lisp
mlep/core/perceptron.lisp
mlep/core/likelihood.lisp
mlep/core/neuronal.lisp
mlep/core/imputer.lisp
The main system appears first, followed by any subsystem dependency.
mlep
A Common Lisp machine learning library for educational purposes.
Frank Zalkow <frank_zalkow@web.de>
Frank Zalkow <frank_zalkow@web.de>
MIT License <http://opensource.org/licenses/MIT>
mlep is a Machine Learning library for Educational Purposes.
It aims at providing a collection of simple machine learning algorithms with the
following claims:
* to be implementation independent
* to be fairly easy to use so that even intermediate Common Lisp programmers should
be able to use this library instantly without pain
* to provide a tutorial-style documentation so that one should get to know this
library easily
0.0.1
package.lisp
(file).
macros
(module).
datasets
(module).
utils
(module).
core
(module).
Modules are listed depth-first from the system components tree.
mlep/macros
package.lisp
(file).
mlep
(system).
macros.lisp
(file).
mlep/datasets
macros
(module).
mlep
(system).
iris.lisp
(file).
heights-weights.lisp
(file).
lenses.lisp
(file).
wages.lisp
(file).
mlep/utils
datasets
(module).
mlep
(system).
number-utils.lisp
(file).
list-utils.lisp
(file).
array-utils.lisp
(file).
constants.lisp
(file).
functions.lisp
(file).
plot.lisp
(file).
mlep/core
utils
(module).
mlep
(system).
generic.lisp
(file).
k-means.lisp
(file).
k-nearest.lisp
(file).
markov-chain.lisp
(file).
naive-bayes.lisp
(file).
perceptron.lisp
(file).
likelihood.lisp
(file).
neuronal.lisp
(file).
imputer.lisp
(file).
Files are sorted by type and then listed depth-first from the systems components trees.
mlep/mlep.asd
mlep/package.lisp
mlep/macros/macros.lisp
mlep/datasets/iris.lisp
mlep/datasets/heights-weights.lisp
mlep/datasets/lenses.lisp
mlep/datasets/wages.lisp
mlep/utils/number-utils.lisp
mlep/utils/list-utils.lisp
mlep/utils/array-utils.lisp
mlep/utils/constants.lisp
mlep/utils/functions.lisp
mlep/utils/plot.lisp
mlep/core/generic.lisp
mlep/core/k-means.lisp
mlep/core/k-nearest.lisp
mlep/core/markov-chain.lisp
mlep/core/naive-bayes.lisp
mlep/core/perceptron.lisp
mlep/core/likelihood.lisp
mlep/core/neuronal.lisp
mlep/core/imputer.lisp
mlep/macros/macros.lisp
macros
(module).
do-lists
(macro).
dolist-with-index
(macro).
dotimes-fromto
(macro).
mlep/datasets/heights-weights.lisp
datasets
(module).
*heights-weights*
(special variable).
mlep/utils/number-utils.lisp
utils
(module).
random-from-to
(function).
euclidian-distance
(function).
range
(function).
square
(function).
mlep/utils/list-utils.lisp
utils
(module).
all-positions
(function).
get-column-ranges
(function).
get-row-ranges
(function).
get-row-ranges-list
(function).
max-arg
(function).
mean
(function).
min-arg
(function).
most-frequent-value
(function).
shuffle
(function).
transpose
(function).
transpose-list
(function).
mlep/utils/array-utils.lisp
utils
(module).
2d-col-to-vector
(function).
col-of-array
(function).
(setf col-of-array)
(function).
determinant
(function).
determinant-helper
(function).
get-row-ranges-array
(function).
list-to-2d-array
(function).
list-to-vector
(function).
map-pointwise
(function).
matrix-expt
(function).
matrix-identity
(function).
multiply-matrices
(function).
outer-product
(function).
row-of-array
(function).
(setf row-of-array)
(function).
scalar-product
(function).
transpose-array
(function).
vector-to-2d-col
(function).
mlep/utils/functions.lisp
utils
(module).
*diff-function-dict*
(special variable).
d-sigmoid
(function).
d-step-function
(function).
d-tanh
(function).
diff
(function).
sigmoid
(function).
step-function
(function).
mlep/utils/plot.lisp
utils
(module).
plot-points
(function).
plot-values
(function).
mlep/core/generic.lisp
core
(module).
analyze
(generic function).
classify
(generic function).
data-set
(generic reader).
distance
(generic reader).
forward
(generic function).
k
(generic reader).
learning-rate
(generic reader).
means
(generic reader).
order
(generic reader).
probabilities
(generic reader).
run
(generic function).
set-labels
(generic reader).
synthesize
(generic function).
test-set
(generic reader).
transform
(generic function).
unique
(generic reader).
backprop
(generic function).
initialize
(generic function).
mlep/core/k-means.lisp
core
(module).
classify
(method).
data-set
(reader method).
(setf data-set)
(writer method).
distance
(reader method).
(setf distance)
(writer method).
initialize-instance
(method).
k
(reader method).
(setf k)
(method).
k-means
(class).
means
(reader method).
(setf means)
(writer method).
run
(method).
mlep/core/k-nearest.lisp
core
(module).
data-set
(reader method).
(setf data-set)
(writer method).
distance
(reader method).
(setf distance)
(writer method).
k
(reader method).
(setf k)
(writer method).
k-nearest-neighbors
(class).
run
(method).
set-labels
(reader method).
(setf set-labels)
(writer method).
test-set
(reader method).
(setf test-set)
(writer method).
histogram
(function).
k-maxmin-idx
(function).
mlep/core/markov-chain.lisp
core
(module).
analyze
(method).
data-set
(reader method).
(setf data-set)
(writer method).
markov-chain
(class).
order
(reader method).
(setf order)
(writer method).
probabilities
(reader method).
(setf probabilities)
(writer method).
run
(method).
synthesize
(method).
unique
(reader method).
altern
(function).
normalize-tensor
(function).
mlep/core/naive-bayes.lisp
core
(module).
data-set
(reader method).
(setf data-set)
(writer method).
(setf data-set)
(method).
initialize-instance
(method).
naive-bayes
(class).
run
(method).
set-labels
(reader method).
(setf set-labels)
(writer method).
(setf set-labels)
(method).
test-set
(reader method).
(setf test-set)
(writer method).
all-labels
(reader method).
(setf all-labels)
(writer method).
compute-likelihoods
(function).
compute-possible-data-values
(function).
compute-prior-probabilities
(function).
initialize
(method).
label-count
(reader method).
(setf label-count)
(writer method).
likelihoods
(reader method).
(setf likelihoods)
(writer method).
possible-data-values
(reader method).
(setf possible-data-values)
(writer method).
prior-probabilities
(reader method).
(setf prior-probabilities)
(writer method).
mlep/core/perceptron.lisp
core
(module).
classify
(method).
data-set
(reader method).
(setf data-set)
(writer method).
(setf data-set)
(method).
initialize-instance
(method).
learning-rate
(reader method).
(setf learning-rate)
(writer method).
perceptron
(class).
run
(method).
set-labels
(reader method).
(setf set-labels)
(writer method).
(setf set-labels)
(method).
activation-function
(reader method).
(setf activation-function)
(writer method).
get-actual-output
(function).
get-overall-error
(function).
initialize
(method).
max-weight-init-value
(reader method).
(setf max-weight-init-value)
(writer method).
(setf max-weight-init-value)
(method).
weights
(reader method).
(setf weights)
(writer method).
mlep/core/likelihood.lisp
core
(module).
data-set
(reader method).
(setf data-set)
(writer method).
max-likelihood
(class).
run
(method).
degrees-of-freedom
(reader method).
(setf degrees-of-freedom)
(writer method).
mlep/core/neuronal.lisp
core
(module).
classify
(method).
data-set
(reader method).
(setf data-set)
(writer method).
(setf data-set)
(method).
forward
(method).
initialize-instance
(method).
learning-rate
(reader method).
(setf learning-rate)
(writer method).
neuronal-network
(class).
run
(method).
set-labels
(reader method).
(setf set-labels)
(writer method).
(setf set-labels)
(method).
activation-function
(reader method).
(setf activation-function)
(writer method).
backprop
(method).
initialize
(method).
net-structure
(reader method).
(setf net-structure)
(writer method).
(setf net-structure)
(method).
net-structure-to-array
(function).
net-structure-to-weights
(function).
output-net
(reader method).
(setf output-net)
(writer method).
output-net-before-activation
(reader method).
(setf output-net-before-activation)
(writer method).
weight-init-range
(reader method).
(setf weight-init-range)
(writer method).
(setf weight-init-range)
(method).
weights
(reader method).
(setf weights)
(writer method).
mlep/core/imputer.lisp
core
(module).
data-set
(reader method).
(setf data-set)
(writer method).
imputer
(class).
run
(method).
transform
(method).
missing-value
(reader method).
(setf missing-value)
(writer method).
missing-value-test
(reader method).
(setf missing-value-test)
(writer method).
replacers
(reader method).
(setf replacers)
(writer method).
Packages are listed by definition order.
mlep
@code{mlep} is a Machine Learning library for Educational Purposes.
It aims at providing a collection of simple machine learning algorithms with the
following claims:
@itemize{
@item{to use only ANSI Common Lisp (thus to be implementation independent)}
@item{to be fairly easy to use so that even intermediate Common Lisp programmers should
be able to use this library instantly without pain}
@item{to provide a tutorial-style documentation so that one should get to know this
library easily}}
common-lisp
.
*heights-weights*
(special variable).
*iris*
(special variable).
*lenses*
(special variable).
*wages*
(special variable).
analyze
(generic function).
classify
(generic function).
data-set
(generic reader).
(setf data-set)
(generic function).
distance
(generic reader).
(setf distance)
(generic writer).
forward
(generic function).
imputer
(class).
k
(generic reader).
(setf k)
(generic function).
k-means
(class).
k-nearest-neighbors
(class).
learning-rate
(generic reader).
(setf learning-rate)
(generic writer).
markov-chain
(class).
max-likelihood
(class).
means
(generic reader).
(setf means)
(generic writer).
naive-bayes
(class).
neuronal-network
(class).
order
(generic reader).
(setf order)
(generic writer).
perceptron
(class).
plot-points
(function).
plot-values
(function).
probabilities
(generic reader).
(setf probabilities)
(generic writer).
random-from-to
(function).
run
(generic function).
set-labels
(generic reader).
(setf set-labels)
(generic function).
synthesize
(generic function).
test-set
(generic reader).
(setf test-set)
(generic writer).
transform
(generic function).
unique
(generic reader).
*diff-function-dict*
(special variable).
+e+
(constant).
2d-col-to-vector
(function).
activation-function
(generic reader).
(setf activation-function)
(generic writer).
all-labels
(generic reader).
(setf all-labels)
(generic writer).
all-positions
(function).
altern
(function).
backprop
(generic function).
col-of-array
(function).
(setf col-of-array)
(function).
compute-likelihoods
(function).
compute-possible-data-values
(function).
compute-prior-probabilities
(function).
d-sigmoid
(function).
d-step-function
(function).
d-tanh
(function).
degrees-of-freedom
(generic reader).
(setf degrees-of-freedom)
(generic writer).
determinant
(function).
determinant-helper
(function).
diff
(function).
do-lists
(macro).
dolist-with-index
(macro).
dotimes-fromto
(macro).
euclidian-distance
(function).
get-actual-output
(function).
get-column-ranges
(function).
get-overall-error
(function).
get-row-ranges
(function).
get-row-ranges-array
(function).
get-row-ranges-list
(function).
histogram
(function).
initialize
(generic function).
k-maxmin-idx
(function).
label-count
(generic reader).
(setf label-count)
(generic writer).
likelihoods
(generic reader).
(setf likelihoods)
(generic writer).
list-to-2d-array
(function).
list-to-vector
(function).
map-pointwise
(function).
matrix-expt
(function).
matrix-identity
(function).
max-arg
(function).
max-weight-init-value
(generic reader).
(setf max-weight-init-value)
(generic function).
mean
(function).
min-arg
(function).
missing-value
(generic reader).
(setf missing-value)
(generic writer).
missing-value-test
(generic reader).
(setf missing-value-test)
(generic writer).
most-frequent-value
(function).
multiply-matrices
(function).
net-structure
(generic reader).
(setf net-structure)
(generic function).
net-structure-to-array
(function).
net-structure-to-weights
(function).
normalize-tensor
(function).
outer-product
(function).
output-net
(generic reader).
(setf output-net)
(generic writer).
output-net-before-activation
(generic reader).
(setf output-net-before-activation)
(generic writer).
possible-data-values
(generic reader).
(setf possible-data-values)
(generic writer).
prior-probabilities
(generic reader).
(setf prior-probabilities)
(generic writer).
range
(function).
replacers
(generic reader).
(setf replacers)
(generic writer).
row-of-array
(function).
(setf row-of-array)
(function).
scalar-product
(function).
shuffle
(function).
sigmoid
(function).
square
(function).
step-function
(function).
transpose
(function).
transpose-array
(function).
transpose-list
(function).
vector-to-2d-col
(function).
weight-init-range
(generic reader).
(setf weight-init-range)
(generic function).
weights
(generic reader).
(setf weights)
(generic writer).
Definitions are sorted by export status, category, package, and then by lexicographic order.
SOCR Data Dinov 020108 HeightsWeights
Human Height and Weight are mostly hereditable, but lifestyles, diet, health and environmental factors also play a role in determining individual’s physical characteristics. The dataset below contains 25,000 records of human heights and weights. These data were obtained in 1993 by a Growth Survey of 25,000 children from birth to 18 years of age recruited from Maternal and Child Health Centres (MCHC) and schools and were used to develop Hong Kong’s current growth charts for weight, height, weight-for-age, weight-for-height and body mass index (BMI). See also the Major League Baseball Players Height and Weight dataset.
@a[http://wiki.stat.ucla.edu/socr/index.php/SOCR_Data_Dinov_020108_HeightsWeights]{http://wiki.stat.ucla.edu/socr/index.php/SOCR_Data_Dinov_020108_HeightsWeights}
Attribute Information:
@itemize{@item{1. Height (Inches)}
@item{2. Weight (Pounds)}}
mlep
.
Iris flower data set by Sir Ronald Fisher (1936)
@a[http://archive.ics.uci.edu/ml/datasets/Iris]{http://archive.ics.uci.edu/ml/datasets/Iris}
Attribute Information:
@itemize{
@item{1. sepal length in cm}
@item{2. sepal width in cm}
@item{3. petal length in cm}
@item{4. petal width in cm}
@item{5. class (Iris Setosa, Iris Versicolour, Iris Virginica)}}
Lenses Data Set by J. Cendrowska (1987)
@a[https://archive.ics.uci.edu/ml/datasets/Lenses]{https://archive.ics.uci.edu/ml/datasets/Lenses}
Attribute Information:
@itemize{
@item{1. age of the patient (1 = young, 2 = pre-presbyopic, 3 = presbyopic)}
@item{2. spectacle prescription (1 = myope, 2 = hypermetrope)}
@item{3. astigmatic (1 = no, 2 = yes)}
@item{4. tear production rate (1 = reduced, 2 = normal)}
@item{5. class @itemize{@item{1 = the patient should be fitted with hard contact lenses,}
@item{2 = the patient should be fitted with soft contact lenses,}
@item{3 = the patient should not be fitted with contact lenses.}}}}
mlep
.
Determinants of Wages from the 1985 Current Population Survey
Therese Stukel
The datafile contains 534 observations on 11 variables
sampled from the Current Population Survey of 1985
@a[http://lib.stat.cmu.edu/datasets/CPS_85_Wages]{http://lib.stat.cmu.edu/datasets/CPS_85_Wages}
Attribute Information:
@itemize{
@item{ 1. EDUCATION: Number of years of education.}
@item{ 2. SOUTH: Indicator variable for Southern Region (1=Person lives in South, 0=Person lives elsewhere).}
@item{ 3. SEX: Indicator variable for sex (1=Female, 0=Male).}
@item{ 4. EXPERIENCE: Number of years of work experience.}
@item{ 5. UNION: Indicator variable for union membership (1=Union member, 0=Not union member).}
@item{ 6. WAGE: Wage (dollars per hour).}
@item{ 7. AGE: Age (years).}
@item{ 8. RACE: Race (1=Other, 2=Hispanic, 3=White).}
@item{ 9. OCCUPATION: Occupational category (1=Management, 2=Sales, 3=Clerical, 4=Service, 5=Professional, 6=Other).}
@item{ 10. SECTOR: Sector (0=Other, 1=Manufacturing, 2=Construction).}
@item{ 11. MARR: Marital Status (0=Unmarried, 1=Married)}}
mlep
.
@arg[vals]{a list of list with x/y-points or a 2d-array – @code{((x1 y1) ... (xn yn))} or @code{#2a((x1 y1) ... (xn yn))}}
@arg[height]{the height in characters used for the plot}
@arg[width]{the width in characters used for the plot}
@arg[char]{the character used for printing}
@return{nothing}
Plotting points with x/y-coordinates.
@arg[vals]{a sequence of numbers to be plotted}
@arg[height]{the height in characters used for the plot}
@arg[char]{the character used for printing}
@return{nothing}
Plot the values of @code{vals} successively.
@arg[from]{the lower bound (inclusive)}
@arg[to]{the upper bound (exclusive)}
@arg[state]{a random state object containing information used by the pseudo-random number generator}
@return{a random number}
Gives a random number in certain range.
mlep
.
@arg[instance]{an instance of @code{markov-chain}}
@arg[input]{some input data}
@return{the probability of @code{input}}
Check the probability of @code{input} being generated by @code{instance}.
mlep
.
markov-chain
) input &key) ¶@arg[instance]{an instance of @code{k-means}, @code{perceptron} or @code{neuronal-network}} @arg[new-data-set]{use @code{new-data-set} instead of the internal @code{data-set}} @arg[verbose]{print some more information (only taken into account for @code{neuronal-network})} @return{a list with a classification number according to each sample in the classified data-set} Classifying some data-set.
mlep
.
neuronal-network
) &key new-data-set verbose) ¶perceptron
) &key new-data-set) ¶@arg[instance]{an instance of any @code{mlep} learning algorithm}
@return{the data-set of @code{instance}}
Get the data-set of @code{instance}.
mlep
.
neuronal-network
)) ¶The data-set to be analyzed.
max-likelihood
)) ¶The data-set to be analyzed.
perceptron
)) ¶The data-set to be analyzed.
naive-bayes
)) ¶The data-set that is already known. (@code{set-labels} go hand in hand with it.)
markov-chain
)) ¶The data-set to be analyzed.
k-nearest-neighbors
)) ¶The data-set that is already known. (@code{set-labels} go hand in hand with it.)
mlep
.
imputer
)) ¶The data-set to be analyzed.
neuronal-network
)) ¶neuronal-network
)) ¶The data-set to be analyzed.
max-likelihood
)) ¶The data-set to be analyzed.
perceptron
)) ¶perceptron
)) ¶The data-set to be analyzed.
naive-bayes
)) ¶naive-bayes
)) ¶The data-set that is already known. (@code{set-labels} go hand in hand with it.)
markov-chain
)) ¶The data-set to be analyzed.
k-nearest-neighbors
)) ¶The data-set that is already known. (@code{set-labels} go hand in hand with it.)
@arg[instance]{an instance of @code{k-means} or @code{k-nearest-neighbors}}
@return{the function for calculating the distance for @code{instance}}
Get the function for calculating the distance for @code{instance}, e.g. @code{#’euclidian-distance}.
mlep
.
k-nearest-neighbors
)) ¶A distance measuring function.
mlep
.
k-nearest-neighbors
)) ¶A distance measuring function.
@arg[instance]{an instance of @code{neuronal-network}}
@arg[input]{the input data to be considered}
@return{the output of the @code{neuronal-network} given the @code{input}}
Computes a forward path through the network and gives its output.
mlep
.
neuronal-network
) &key input) ¶@arg[instance]{an instance of @code{k-means} or @code{k-nearest-neighbors}}
@return{the parameter @code{k}}
@code{k} determines how many means are assumed (for @code{k-means}) resp. how many neighbors are considered (for @code{k-nearest-neighbors}).
@arg[instance]{an instance of @code{neuronal-network} or @code{perceptron}}
@return{the learning-rate}
The learning rate controls the size of change during updating the weights.
mlep
.
neuronal-network
)) ¶The learning rate.
perceptron
)) ¶The learning rate.
mlep
.
neuronal-network
)) ¶The learning rate.
perceptron
)) ¶The learning rate.
@arg[instance]{an instance of @code{k-means} or @code{principal-component-analysis}}
@return{the current means}
Get the current means.
@arg[instance]{an instance of @code{markov-chain}}
@return{the order of the markov chain}
The order of a markov chain determines how much past events are considered for producing a current event.
mlep
.
markov-chain
)) ¶The order of the markov chain.
mlep
.
markov-chain
)) ¶The order of the markov chain.
@arg[instance]{an instance of @code{markov-chain}}
@return{the probabilities of the markov chain}
Get the probability matrix (or tensor) – it’s rank is @code{order+1}.
mlep
.
markov-chain
)) ¶A matrix/tensor with probabilities.
mlep
.
markov-chain
)) ¶A matrix/tensor with probabilities.
@arg[instance]{an instance of any @code{mlep} learning algorithm}
@arg[threshold]{a threshold that is the minimum global error to be achieved – iterative training runs until threshold is reached (supported by @code{perceptron})}
@arg[epochs]{number of how often a iterative algorithm should be performed (supported by @code{k-means} and @code{neuronal-network})}
@return{depends on the learning algorithm:
@itemize{
@item{@code{k-means}: the computed means}
@item{@code{k-nearest-neighbors}: the classes assigned to @code{test-set}}
@item{@code{max-likelihood}: a list with the first item being the mean and the second one being the co-variance matrix of the normal distribution}
@item{@code{markov-chain}: the probability matrix (or tensor)}
@item{@code{naive-bayes}: the classes assigned to @code{test-set}}
@item{@code{neuronal-network}: nothing}
@item{@code{perceptron}: nothing}
@item{@code{principal-component-analysis}: list with three matrices - @code{unitary-matrix1 U} (orthogonal matrix), @code{unitary-matrix2 Vt} (orthogonal matrix) and @code{singular-values} (a diagonal matrix with the diagonal elements being the singular values called @code{D}; @code{UxDxVt} should be a reconstruction of the input matrix)}
@item{@code{imputer}: The default replace values for each column.}
}}
A general interface for ’running’ a learning algorithm.
mlep
.
neuronal-network
) &key epochs) ¶max-likelihood
) &key) ¶perceptron
) &key threshold) ¶naive-bayes
) &key) ¶markov-chain
) &key) ¶k-nearest-neighbors
) &key) ¶@arg[instance]{an instance of @code{k-nearest-neighbors}, @code{naive-bayes}, @code{neuronal-network} or @code{perceptron}}
@return{the target labels for @code{data-set} of a supervised learning algorithm.}
Get the target labels.
mlep
.
neuronal-network
)) ¶The output-values for @code{data-set}.
perceptron
)) ¶The output-values for @code{data-set}.
naive-bayes
)) ¶The labels for @code{data-set}.
k-nearest-neighbors
)) ¶The labels for @code{data-set}.
mlep
.
neuronal-network
)) ¶neuronal-network
)) ¶The output-values for @code{data-set}.
perceptron
)) ¶perceptron
)) ¶The output-values for @code{data-set}.
naive-bayes
)) ¶naive-bayes
)) ¶The labels for @code{data-set}.
k-nearest-neighbors
)) ¶The labels for @code{data-set}.
@arg[instance]{an instance of @code{markov-chain}}
@arg[start]{1) the symbol @code{random} – a new random sequence as beginning is generated; 2) @code{nil} (default) – a literal subsequence with random starting index is taken as beginning, 3) a list with an user given starting sequence}
@arg[howmany]{number of elements to be synthesized}
Synthesize some data.
mlep
.
markov-chain
) &key start howmany) ¶@arg[instance]{an instance of @code{k-nearest-neighbors} or @code{naive-bayes}} @return{the test, i.e. a set that has no target labels and needs to be classified.} Get the test-set.
mlep
.
naive-bayes
)) ¶The data-set that has no labels and needs to be classified.
k-nearest-neighbors
)) ¶The data-set that has no labels and needs to be classified.
mlep
.
naive-bayes
)) ¶The data-set that has no labels and needs to be classified.
k-nearest-neighbors
)) ¶The data-set that has no labels and needs to be classified.
@arg[instance]{an instance of @code{principal-component-analysis} or @code{imputer}}
@arg[components]{a number that states how many dimensions should be used for a transformation (default is @code{nil} which means that it should use all dimensions of @code{data-set}}
@arg[inverse]{do an inverse transformation (@code{t} or @code{nil}, default: @code{nil})}
@arg[new-data]{do the transformation on this data-set (default is @code{nil} which means, that it should use @code{data-set})}
@return{the transformed data-set}
Project some data on its principal components. / Fit missing values.
@arg[instance]{an instance of @code{markov-chain}} @return{all unique elements in @code{data-set}} Get all unique values are considered by the chain.
mlep
.
markov-chain
)) ¶Unique values of data-set
neuronal-network
) &key) ¶perceptron
) &key) ¶naive-bayes
) &key) ¶Replace missing value by the mean (for numercial data) or the mode (for categorical data).
mlep
.
The data-set to be analyzed.
:data-set
The value that is recognized as a missing value.
:missing-value
The test-function for comparing each item in a data-set with this missing value.
(function eq)
:missing-value-test
The values that are used for replacing missing values per column.
k-means is a simple unsupervised clustering algorithm for a known number of clusters.
mlep
.
The number of groups/clusters to be determined.
2
:k
k
.
This slot is read-only.
The data-set to be analyzed.
:data-set
A distance measuring function.
(function mlep::euclidian-distance)
:distance
The means of the data points.
:means
k-nearest-neighbors is a simple supervised clustering algorithm for a known number of clusters.
mlep
.
The number of neighbors to be taken into account.
The data-set that is already known. (@code{set-labels} go hand in hand with it.)
:data-set
The labels for @code{data-set}.
:set-labels
A distance measuring function.
(function mlep::euclidian-distance)
:distance
The data-set that has no labels and needs to be classified.
:test-set
A Markov-Chain.
mlep
.
The data-set to be analyzed.
:data-set
The order of the markov chain.
1
:order
Unique values of data-set
:unique
This slot is read-only.
A matrix/tensor with probabilities.
:probabilities
With max-likelihood one can estimate the parameters of the normal distributed probability density function that fits the data-set.
Naive-bayes takes an probabilistic approach for a simple supervised clustering.
mlep
.
(setf all-labels)
.
all-labels
.
(setf data-set)
.
(setf data-set)
.
data-set
.
initialize
.
initialize-instance
.
(setf label-count)
.
label-count
.
(setf likelihoods)
.
likelihoods
.
(setf possible-data-values)
.
possible-data-values
.
(setf prior-probabilities)
.
prior-probabilities
.
run
.
(setf set-labels)
.
(setf set-labels)
.
set-labels
.
(setf test-set)
.
test-set
.
The data-set that is already known. (@code{set-labels} go hand in hand with it.)
:data-set
The labels for @code{data-set}.
:set-labels
The data-set that has no labels and needs to be classified.
:test-set
All possible values occurring in each attribute of @code{data-set}. To be pre-computed.
:possible-data-values
All given data-labels. To be pre-computed.
:all-labels
Counting of each item of a label. To be pre-computed.
:label-count
The prior-probabilities of each class. To be pre-computed.
:prior-probabilities
The likelihoods of a feature attribute given a class. To be pre-computed.
mlep::likelihoods
A fully-connected Feed-Forward Multi-Layer Perceptron.
mlep
.
(setf activation-function)
.
activation-function
.
backprop
.
classify
.
(setf data-set)
.
(setf data-set)
.
data-set
.
forward
.
initialize
.
initialize-instance
.
(setf learning-rate)
.
learning-rate
.
(setf net-structure)
.
(setf net-structure)
.
net-structure
.
(setf output-net)
.
output-net
.
(setf output-net-before-activation)
.
output-net-before-activation
.
run
.
(setf set-labels)
.
(setf set-labels)
.
set-labels
.
(setf weight-init-range)
.
(setf weight-init-range)
.
weight-init-range
.
(setf weights)
.
weights
.
The data-set to be analyzed.
:data-set
The output-values for @code{data-set}.
:set-labels
The net-structure of the net. (Neurons per layer.)
:net-structure
The weights from input to output values.
:weights
The output of all neurons of the network.
:output-net
The output of all neurons of the network before the activation function was applied.
:output-net-before-activation
The maximum range for initializing the weights.
(quote (-0.1d0 0.1d0))
:weight-init-range
The activation function, usually a Heaviside step function.
(function tanh)
:activation-function
The learning rate.
0.2d0
:learning-rate
A perceptron is a very simple neuron model and turns out to be a linear classificator.
mlep
.
(setf activation-function)
.
activation-function
.
classify
.
(setf data-set)
.
(setf data-set)
.
data-set
.
initialize
.
initialize-instance
.
(setf learning-rate)
.
learning-rate
.
(setf max-weight-init-value)
.
(setf max-weight-init-value)
.
max-weight-init-value
.
run
.
(setf set-labels)
.
(setf set-labels)
.
set-labels
.
(setf weights)
.
weights
.
The data-set to be analyzed.
:data-set
The output-values for @code{data-set}.
:set-labels
The weights from input to output values.
:weights
The maximum value for initializing the weights.
0.1d0
:max-weight-init-value
The activation function, usually a Heaviside step function.
(function mlep::step-function)
:activation-function
The learning rate.
0.2d0
:learning-rate
mlep
.
Gets the indices of all elements in LIST that safisfy the TEST.
mlep
.
mlep
.
Computes the determinant of ‘matrix’ - which must a square matrix of rank 2..
mlep
.
Gives the euclidian distance between to points of arbitrary dimension.
mlep
.
SET is a two-dimensional list or array. Gets the minmal and maximal value of each column of SET. Column means the corresponding nth values of each sublist.
mlep
.
SET is a two-dimensional list or array. Gets the minimal and maximal value of each row (i.e. sublist) of SET.
mlep
.
Count all occurrences in vector. Returns list of all unique items in vector and their frequencies.
mlep
.
Get the index of k max/min values in vector.
mlep
.
Converts a list of list to a 2-dimensional array.
mlep
.
Converts a list to a vector.
mlep
.
Calls ‘function’ to one or more ‘arrays’ pointwise.
mlep
.
Takes the first argument (a matrix) and multiplies it by itself exp times
mlep
.
Creates a new identity matrix of size dim*dim
mlep
.
Get the most frequent value of a sequence x.
mlep
.
Takes two 2D arrays and returns their product, or an error if they cannot be multiplied
mlep
.
mlep
.
Computes the outer product (aka tensor product) of two vectors/lists ‘x1’ and ‘x2’.
mlep
.
Gives a list with all numbers from ‘FROM’ to (exclusive) ‘TO’.
mlep
.
Computes the scalar product (aka dot product aka inner product) of two vectors/lists.
mlep
.
mlep
.
neuronal-network
)) ¶neuronal-network
)) ¶The activation function, usually a Heaviside step function.
perceptron
)) ¶perceptron
)) ¶The activation function, usually a Heaviside step function.
mlep
.
naive-bayes
)) ¶naive-bayes
)) ¶All given data-labels. To be pre-computed.
Adjust weights with backpropagation.
mlep
.
neuronal-network
) input wanted) ¶mlep
.
max-likelihood
)) ¶max-likelihood
)) ¶Delta Degrees of Freedom. Divisor is length of @code{data-set} minus @code{degrees-of-freedom}.
Initialize the arguments of INSTANCE.
mlep
.
neuronal-network
) &key) ¶perceptron
) &key) ¶naive-bayes
) &key) ¶Pre-computes: ‘all-labels’, ‘label-count’, ‘possible-data-values’, ‘prior-probabilities’, ‘likelihoods’
mlep
.
naive-bayes
)) ¶naive-bayes
)) ¶Counting of each item of a label. To be pre-computed.
mlep
.
naive-bayes
)) ¶naive-bayes
)) ¶The likelihoods of a feature attribute given a class. To be pre-computed.
mlep
.
perceptron
)) ¶The maximum value for initializing the weights.
mlep
.
perceptron
)) ¶perceptron
)) ¶The maximum value for initializing the weights.
mlep
.
neuronal-network
)) ¶The net-structure of the net. (Neurons per layer.)
mlep
.
neuronal-network
)) ¶neuronal-network
)) ¶The net-structure of the net. (Neurons per layer.)
mlep
.
neuronal-network
)) ¶neuronal-network
)) ¶The output of all neurons of the network.
mlep
.
neuronal-network
)) ¶neuronal-network
)) ¶The output of all neurons of the network before the activation function was applied.
mlep
.
naive-bayes
)) ¶naive-bayes
)) ¶All possible values occurring in each attribute of @code{data-set}. To be pre-computed.
mlep
.
naive-bayes
)) ¶naive-bayes
)) ¶The prior-probabilities of each class. To be pre-computed.
mlep
.
neuronal-network
)) ¶The maximum range for initializing the weights.
mlep
.
neuronal-network
)) ¶neuronal-network
)) ¶The maximum range for initializing the weights.
mlep
.
neuronal-network
)) ¶neuronal-network
)) ¶The weights from input to output values.
perceptron
)) ¶perceptron
)) ¶The weights from input to output values.
Jump to: | (
2
A B C D E F G H I K L M N O P R S T U V W |
---|
Jump to: | (
2
A B C D E F G H I K L M N O P R S T U V W |
---|
Jump to: | *
+
A C D K L M N O P R S T U W |
---|
Jump to: | *
+
A C D K L M N O P R S T U W |
---|
Jump to: | A C D F G H I K L M N P S U W |
---|
Jump to: | A C D F G H I K L M N P S U W |
---|