Index (Classification-Explanations 4.3.1 API)

C D E F G I K L M N O P R S T W
All Classes and Interfaces|All Packages|Constant Field Values|Serialized Form

C

ColumnarExplainer<T extends org.tribuo.Output<T>> - Interface in org.tribuo.classification.explanations: An explainer for data using Tribuo's columnar data package.

D

DISTANCE_DELTA - Static variable in class org.tribuo.classification.explanations.lime.LIMEBase: Delta to consider two distances equal.

E

evaluator - Static variable in class org.tribuo.classification.explanations.lime.LIMEBase
explain(CommandInterpreter, String[]) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Explains a text classification.
explain(String) - Method in class org.tribuo.classification.explanations.lime.LIMEText
explain(String) - Method in interface org.tribuo.classification.explanations.TextExplainer: Converts the supplied text into an Example, and generates an explanation of the contained Model's prediction.
explain(Map<String, String>) - Method in interface org.tribuo.classification.explanations.ColumnarExplainer: Explains the supplied data.
explain(Map<String, String>) - Method in class org.tribuo.classification.explanations.lime.LIMEColumnar
explain(Example<Label>) - Method in class org.tribuo.classification.explanations.lime.LIMEBase
explain(Example<Label>) - Method in interface org.tribuo.classification.explanations.TabularExplainer: Explain why the supplied Example is classified a certain way.
explainWithSamples(Map<String, String>) - Method in class org.tribuo.classification.explanations.lime.LIMEColumnar
explainWithSamples(Example<Label>) - Method in class org.tribuo.classification.explanations.lime.LIMEBase
Explanation<T extends org.tribuo.Output<T>> - Interface in org.tribuo.classification.explanations: An explanation knows what features are used, what the explaining Model is and what the original Model's prediction is.
explanationTrainer - Variable in class org.tribuo.classification.explanations.lime.LIMEBase

F

featureInfo(CommandInterpreter, String) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Shows information on a particular feature.
fileCompleter() - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Completers for filenames.

G

generatesProbabilities(CommandInterpreter) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Does the model generate probabilities.
getActiveFeatures() - Method in interface org.tribuo.classification.explanations.Explanation: Returns the names of the active features in this explanation.
getActiveFeatures() - Method in class org.tribuo.classification.explanations.lime.LIMEExplanation
getDescription() - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI
getEvaluation() - Method in class org.tribuo.classification.explanations.lime.LIMEExplanation: Gets the evaluator which scores how close the sparse model's predictions are to the complex model's predictions.
getModel() - Method in interface org.tribuo.classification.explanations.Explanation: Returns the explanining model.
getModel() - Method in class org.tribuo.classification.explanations.lime.LIMEExplanation
getName() - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI
getPrediction() - Method in interface org.tribuo.classification.explanations.Explanation: The original model's prediction which is being explained.
getPrediction() - Method in class org.tribuo.classification.explanations.lime.LIMEExplanation
getRMSE(String) - Method in class org.tribuo.classification.explanations.lime.LIMEExplanation: Get the RMSE of a specific dimension of the explanation model.

I

innerModel - Variable in class org.tribuo.classification.explanations.lime.LIMEBase

K

kernelDist(double, double) - Static method in class org.tribuo.classification.explanations.lime.LIMEBase: Calculates an RBF kernel of a specific width.
kernelWidth - Variable in class org.tribuo.classification.explanations.lime.LIMEBase

L

LIMEBase - Class in org.tribuo.classification.explanations.lime: LIMEBase merges the lime_base.py and lime_tabular.py implementations, and deals with simple matrices of numerical or categorical data.
LIMEBase(SplittableRandom, Model<Label>, SparseTrainer<Regressor>, int) - Constructor for class org.tribuo.classification.explanations.lime.LIMEBase: Constructs a LIME explainer for a model which uses tabular data (i.e., no special treatment for text features).
LIMEColumnar - Class in org.tribuo.classification.explanations.lime: Uses the columnar data processing infrastructure to mix text and tabular data.
LIMEColumnar(SplittableRandom, Model<Label>, SparseTrainer<Regressor>, int, RowProcessor<Label>, Tokenizer) - Constructor for class org.tribuo.classification.explanations.lime.LIMEColumnar: Constructs a LIME explainer for a model which uses the columnar data processing system.
LIMEExplanation - Class in org.tribuo.classification.explanations.lime: An Explanation using LIME.
LIMEExplanation(SparseModel<Regressor>, Prediction<Label>, RegressionEvaluation) - Constructor for class org.tribuo.classification.explanations.lime.LIMEExplanation: Constructs a LIME explanation.
LIMEText - Class in org.tribuo.classification.explanations.lime: Uses a Tribuo TextFeatureExtractor to explain the prediction for a given piece of text.
LIMEText(SplittableRandom, Model<Label>, SparseTrainer<Regressor>, int, TextFeatureExtractor<Label>, Tokenizer) - Constructor for class org.tribuo.classification.explanations.lime.LIMEText: Constructs a LIME explainer for a model which uses text data.
LIMETextCLI - Class in org.tribuo.classification.explanations.lime: A CLI for interacting with LIMEText.
LIMETextCLI() - Constructor for class org.tribuo.classification.explanations.lime.LIMETextCLI: Constructs a LIME CLI.
LIMETextCLI.LIMETextCLIOptions - Class in org.tribuo.classification.explanations.lime: Command line options.
LIMETextCLIOptions() - Constructor for class org.tribuo.classification.explanations.lime.LIMETextCLI.LIMETextCLIOptions
loadModel(CommandInterpreter, File, boolean) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Loads a model in from disk.

M

main(String[]) - Static method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Runs a LIMETextCLI.
measureDistance(ImmutableFeatureMap, long, SparseVector, SparseVector) - Static method in class org.tribuo.classification.explanations.lime.LIMEBase: Measures the distance between an input point and a sampled point.
minCount(CommandInterpreter, int) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Shows the number of features that occurred more than minCount times.
modelDescription(CommandInterpreter) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Shows the model description.
modelFilename - Variable in class org.tribuo.classification.explanations.lime.LIMETextCLI.LIMETextCLIOptions: Model file to load.

N

nameFeature(String, int) - Method in class org.tribuo.classification.explanations.lime.LIMEText: Generate the feature name by combining the word and index.
nameFeature(String, String, int) - Method in class org.tribuo.classification.explanations.lime.LIMEColumnar: Generate the feature name by combining the word and index.
numFeatures(CommandInterpreter) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Shows the number of features.
numSamples - Variable in class org.tribuo.classification.explanations.lime.LIMEBase
numTrainingExamples - Variable in class org.tribuo.classification.explanations.lime.LIMEBase

O

org.tribuo.classification.explanations - package org.tribuo.classification.explanations: Provides core infrastructure for local model based explanations.
org.tribuo.classification.explanations.lime - package org.tribuo.classification.explanations.lime: Provides an implementation of LIME (Locally Interpretable Model Explanations).

P

predict(CommandInterpreter, String[]) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Makes a prediction using the loaded model.
protobufFormat - Variable in class org.tribuo.classification.explanations.lime.LIMETextCLI.LIMETextCLIOptions: Load the model from a protobuf.

R

regressionFactory - Static variable in class org.tribuo.classification.explanations.lime.LIMEBase
rng - Variable in class org.tribuo.classification.explanations.lime.LIMEBase

S

sampleData(String, List<Token>) - Method in class org.tribuo.classification.explanations.lime.LIMEText: Samples a new dataset from the input text.
samplePoint(Random, ImmutableFeatureMap, long, SparseVector) - Static method in class org.tribuo.classification.explanations.lime.LIMEBase: Samples a single example from the supplied feature map and input vector.
setNumFeatures(CommandInterpreter, int) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Sets the number of features LIME should use in an explanation.
setNumSamples(CommandInterpreter, int) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Sets the number of samples to use in LIME.
showLabelStats(CommandInterpreter) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Shows the output statistics.
startShell() - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Start the command shell

T

TabularExplainer<T extends org.tribuo.Output<T>> - Interface in org.tribuo.classification.explanations: An explainer for tabular data.
TextExplainer<T extends org.tribuo.Output<T>> - Interface in org.tribuo.classification.explanations: An explainer for text data.
topFeatures(CommandInterpreter, int) - Method in class org.tribuo.classification.explanations.lime.LIMETextCLI: Shows the top features of the loaded model.
toString() - Method in class org.tribuo.classification.explanations.lime.LIMEExplanation
trainExplainer(Example<Regressor>, List<Example<Regressor>>) - Method in class org.tribuo.classification.explanations.lime.LIMEBase: Trains the explanation model using the supplied sampled data and the input example.
transformOutput(Prediction<Label>) - Static method in class org.tribuo.classification.explanations.lime.LIMEBase: Transforms a Prediction for a multiclass problem into a Regressor output which represents the probability for each class.

W

WIDTH_CONSTANT - Static variable in class org.tribuo.classification.explanations.lime.LIMEBase: Width of the noise gaussian.

C D E F G I K L M N O P R S T W
All Classes and Interfaces|All Packages|Constant Field Values|Serialized Form