Interface for machine learning modeling, testing and training 0.0.18
Released under the MIT
Interface for machine learning modeling, testing and training.
Installation
To install, add the following dependency to your project or build file:
[com.zensols.ml/model "0.0.18"]
Namespaces
zensols.model.classifier
A utility library that wraps Weka library. This library works with zensols.model.weka do the following: * Cross validate models * Manage and sort results (i.e. cross validations) * Train models * Read/write ARFF files
Public variables and functions:
- *arff-file*
- *best-result-criteria*
- *class-feature-meta*
- *classifier-class*
- *create-classifier-fn*
- *cross-fold-count*
- *cross-val-fns*
- *get-data-fn*
- *operation-write-instance-fns*
- *output-class-feature-meta*
- *rand-fn*
- analysis-report-resource
- classifier-name
- classify-instance
- compile-results
- cross-validate-tests
- excel-results
- excel-results-precision
- filter-attribute-data
- initialize
- model-exists?
- model-read-resource
- model-write-resource
- print-eval-results
- print-results
- read-arff
- read-model
- test-classifier
- train-classifier
- train-test-classifier
- write-arff
- write-model
zensols.model.eval-classifier
A client entry point library to help with evaluating machine learning models. This library not only wraps the Weka library but also provides additional functionality like a two pass cross validation (see with-two-pass).
Public variables and functions:
- *default-set-type*
- *throw-cross-validate*
- analysis-file
- compile-results
- create-model
- cross-fold-info
- display-features
- eval-and-write
- eval-and-write-results
- evaluations-file
- executing-two-pass?
- features-file
- print-best-results
- print-model-config
- read-arff
- read-model
- run-tests
- terse-results
- test-train-series-file
- train-model
- train-test-results
- train-test-series
- two-pass-model
- two-pass-test-instances
- two-pass-train-instances
- with-two-pass
- write-arff
- write-csv-train-test-series
- write-features
- write-model
zensols.model.execute-classifier
A client entry point library to help with executing a trained classifier. The classifier is tested and trained in the zensols.model.eval-classifier namespace.
Public variables and functions:
- classifier-file
- classify
- confusion-matrix-file
- cross-fold-instances
- display-predictions
- dump-model-info
- model-classifier-feature-types
- model-classifier-label
- model-config
- model-exists?
- predict
- predictions-file
- prime-model
- print-model-info
- read-model
- train-test-instances
- with-model-conf
- write-classifier
- write-confusion-matrix
- write-predictions
zensols.model.weka
Wraps the Weka Java API. This is probably the wrong library to use for most uses. Instead take a look at zensols.model.eval-classifier and zensols.model.execute-classifier.
Public variables and functions: