diff --git a/Classification/Core/src/main/java/org/tribuo/classification/ensemble/ClassificationEnsembleOptions.java b/Classification/Core/src/main/java/org/tribuo/classification/ensemble/ClassificationEnsembleOptions.java
index 78539451a..a4ba507db 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/ensemble/ClassificationEnsembleOptions.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/ensemble/ClassificationEnsembleOptions.java
@@ -55,15 +55,25 @@ public enum EnsembleType {
         RF
     }
 
-    @Option(longName="ensemble-type",usage="Ensemble method, options are {ADABOOST, BAGGING, RF}.")
+    /**
+     * Ensemble method, options are {ADABOOST, BAGGING, RF}.
+     */
+    @Option(longName = "ensemble-type", usage = "Ensemble method, options are {ADABOOST, BAGGING, RF}.")
     public EnsembleType type = EnsembleType.BAGGING;
-    @Option(longName="ensemble-size",usage="Number of base learners in the ensemble.")
+    /**
+     * Number of base learners in the ensemble.
+     */
+    @Option(longName = "ensemble-size", usage = "Number of base learners in the ensemble.")
     public int ensembleSize = -1;
-    @Option(longName="ensemble-seed",usage="RNG seed.")
+    /**
+     * RNG seed.
+     */
+    @Option(longName = "ensemble-seed", usage = "RNG seed.")
     public long seed = Trainer.DEFAULT_SEED;
 
     /**
      * Wraps the supplied trainer using the ensemble trainer described by these options.
+     *
      * @param trainer The trainer to wrap.
      * @return An ensemble trainer.
      */
@@ -72,21 +82,21 @@ public Trainer<Label> wrapTrainer(Trainer<Label> trainer) {
             switch (type) {
                 case ADABOOST:
                     logger.info("Using Adaboost with " + ensembleSize + " members.");
-                    return new AdaBoostTrainer(trainer,ensembleSize,seed);
+                    return new AdaBoostTrainer(trainer, ensembleSize, seed);
                 case BAGGING:
                     logger.info("Using Bagging with " + ensembleSize + " members.");
-                    return new BaggingTrainer<>(trainer,new VotingCombiner(),ensembleSize,seed);
+                    return new BaggingTrainer<>(trainer, new VotingCombiner(), ensembleSize, seed);
                 case EXTRA_TREES:
                     if (trainer instanceof DecisionTreeTrainer) {
                         logger.info("Using Extra Trees with " + ensembleSize + " members.");
-                        return new ExtraTreesTrainer<>((DecisionTreeTrainer<Label>)trainer,new VotingCombiner(),ensembleSize,seed);
+                        return new ExtraTreesTrainer<>((DecisionTreeTrainer<Label>) trainer, new VotingCombiner(), ensembleSize, seed);
                     } else {
                         throw new IllegalArgumentException("ExtraTreesTrainer requires a DecisionTreeTrainer");
                     }
                 case RF:
                     if (trainer instanceof DecisionTreeTrainer) {
                         logger.info("Using Random Forests with " + ensembleSize + " members.");
-                        return new RandomForestTrainer<>((DecisionTreeTrainer<Label>)trainer,new VotingCombiner(),ensembleSize,seed);
+                        return new RandomForestTrainer<>((DecisionTreeTrainer<Label>) trainer, new VotingCombiner(), ensembleSize, seed);
                     } else {
                         throw new IllegalArgumentException("RandomForestTrainer requires a DecisionTreeTrainer");
                     }
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/evaluation/LabelEvaluationUtil.java b/Classification/Core/src/main/java/org/tribuo/classification/evaluation/LabelEvaluationUtil.java
index 08ed2034e..9859dab77 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/evaluation/LabelEvaluationUtil.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/evaluation/LabelEvaluationUtil.java
@@ -207,8 +207,17 @@ public TPFP(List<Integer> falsePos, List<Integer> truePos, List<Double> threshol
      * Not yet a record, but it will be one day.
      */
     public static class ROC {
+        /**
+         * The false positive rate at the corresponding threshold.
+         */
         public final double[] fpr;
+        /**
+         * The true positive rate at the corresponding threshold.
+         */
         public final double[] tpr;
+        /**
+         * The threshold values.
+         */
         public final double[] thresholds;
 
         /**
@@ -231,8 +240,17 @@ public ROC(double[] fpr, double[] tpr, double[] thresholds) {
      * Not yet a record, but it will be one day.
      */
     public static class PRCurve {
+        /**
+         * The precision at the corresponding threshold.
+         */
         public final double[] precision;
+        /**
+         * The recall at the corresponding threshold.
+         */
         public final double[] recall;
+        /**
+         * The threshold values.
+         */
         public final double[] thresholds;
 
         /**
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/evaluation/LabelMetric.java b/Classification/Core/src/main/java/org/tribuo/classification/evaluation/LabelMetric.java
index 20632b897..1730915c6 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/evaluation/LabelMetric.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/evaluation/LabelMetric.java
@@ -102,16 +102,30 @@ public static final class Context extends MetricContext<Label> {
 
         private final ConfusionMatrix<Label> cm;
 
+        /**
+         * Constructs a context and compute the confusion matrix using the specified model and predictions.
+         * @param model The model.
+         * @param predictions The predictions.
+         */
         public Context(Model<Label> model, List<Prediction<Label>> predictions) {
             super(model, predictions);
             this.cm = new LabelConfusionMatrix(model.getOutputIDInfo(), predictions);
         }
 
+        /**
+         * Constructs a context and compute the confusion matrix using the specified model and predictions.
+         * @param model The model.
+         * @param predictions The predictions.
+         */
         public Context(SequenceModel<Label> model, List<Prediction<Label>> predictions) {
             super(model, predictions);
             this.cm = new LabelConfusionMatrix(model.getOutputIDInfo(), predictions);
         }
 
+        /**
+         * Gets the confusion matrix.
+         * @return The confusion matrix.
+         */
         public ConfusionMatrix<Label> getCM() {
             return cm;
         }
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/example/GaussianLabelDataSource.java b/Classification/Core/src/main/java/org/tribuo/classification/example/GaussianLabelDataSource.java
index f020226de..317554cac 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/example/GaussianLabelDataSource.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/example/GaussianLabelDataSource.java
@@ -33,10 +33,10 @@
 public final class GaussianLabelDataSource extends DemoLabelDataSource {
 
     @Config(mandatory = true, description = "2d mean of the first Gaussian.")
-    public double[] firstMean;
+    private double[] firstMean;
 
     @Config(mandatory = true, description = "4 element covariance matrix of the first Gaussian.")
-    public double[] firstCovarianceMatrix;
+    private double[] firstCovarianceMatrix;
 
     @Config(mandatory = true, description = "2d mean of the second Gaussian.")
     private double[] secondMean;
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/example/LabelledDataGenerator.java b/Classification/Core/src/main/java/org/tribuo/classification/example/LabelledDataGenerator.java
index 04f8aeff1..045e6f7bb 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/example/LabelledDataGenerator.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/example/LabelledDataGenerator.java
@@ -40,6 +40,12 @@ private LabelledDataGenerator() {}
 
     private static final LabelFactory labelFactory = new LabelFactory();
 
+    /**
+     * Generates a train/test dataset pair which is dense in the features,
+     * each example has 4 features,{A,B,C,D}, and there are 4 classes,
+     * {Foo,Bar,Baz,Quux}.
+     * @return A pair of datasets.
+     */
     public static Pair<Dataset<Label>,Dataset<Label>> denseTrainTest() {
         return denseTrainTest(-1.0);
     }
@@ -99,6 +105,12 @@ public static Pair<Dataset<Label>,Dataset<Label>> denseTrainTest(double negate)
         return new Pair<>(train,test);
     }
 
+    /**
+     * Generates a pair of datasets, where the features are sparse,
+     * and unknown features appear in the test data. It has the same
+     * 4 classes {Foo,Bar,Baz,Quux}.
+     * @return A pair of train and test datasets.
+     */
     public static Pair<Dataset<Label>,Dataset<Label>> sparseTrainTest() {
         return sparseTrainTest(-1.0);
     }
@@ -173,6 +185,11 @@ public static Pair<Dataset<Label>,Dataset<Label>> sparseTrainTest(double negate)
         return new Pair<>(train,test);
     }
 
+    /**
+     * Generates a pair of datasets with sparse features and unknown features
+     * in the test data. Has binary labels {Foo,Bar}.
+     * @return A pair of train and test datasets.
+     */
     public static Pair<Dataset<Label>,Dataset<Label>> binarySparseTrainTest() {
         return binarySparseTrainTest(-1.0);
     }
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/sequence/ConfidencePredictingSequenceModel.java b/Classification/Core/src/main/java/org/tribuo/classification/sequence/ConfidencePredictingSequenceModel.java
index 288b3a77d..650067926 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/sequence/ConfidencePredictingSequenceModel.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/sequence/ConfidencePredictingSequenceModel.java
@@ -39,6 +39,13 @@
 public abstract class ConfidencePredictingSequenceModel extends SequenceModel<Label> {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * Constructs a ConfidencePredictingSequenceModel with the supplied parameters.
+     * @param name The model name.
+     * @param description The model provenance.
+     * @param featureIDMap The feature domain.
+     * @param labelIDMap The output domain.
+     */
     protected ConfidencePredictingSequenceModel(String name, ModelProvenance description, ImmutableFeatureMap featureIDMap, ImmutableOutputInfo<Label> labelIDMap) {
         super(name,description,featureIDMap,labelIDMap);
     }
@@ -81,7 +88,13 @@ private static <SUB extends Subsequence> Double multiplyWeights(List<Prediction<
      */
     public static class Subsequence implements Serializable {
         private static final long serialVersionUID = 1L;
+        /**
+         * The subsequence start index.
+         */
         public final int begin;
+        /**
+         * The subsequence end index.
+         */
         public final int end;
 
         /**
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/sequence/LabelSequenceEvaluation.java b/Classification/Core/src/main/java/org/tribuo/classification/sequence/LabelSequenceEvaluation.java
index 3696a66c0..dc9530fae 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/sequence/LabelSequenceEvaluation.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/sequence/LabelSequenceEvaluation.java
@@ -45,6 +45,12 @@ public class LabelSequenceEvaluation implements SequenceEvaluation<Label> {
     private final ConfusionMatrix<Label> cm;
     private final EvaluationProvenance provenance;
 
+    /**
+     * Constructs a LabelSequenceEvaluation using the supplied parameters.
+     * @param results The metric values.
+     * @param ctx The context.
+     * @param provenance The evaluation provenance.
+     */
     protected LabelSequenceEvaluation(Map<MetricID<Label>, Double> results,
                                       LabelMetric.Context ctx,
                                       EvaluationProvenance provenance) {
@@ -85,98 +91,204 @@ public double confusion(Label predictedLabel, Label trueLabel) {
         return cm.confusion(predictedLabel, trueLabel);
     }
 
+    /**
+     * Gets the true positive count for that label.
+     * @param label The label.
+     * @return The true positive count.
+     */
     public double tp(Label label) {
         return get(label, LabelMetrics.TP);
     }
 
+    /**
+     * Gets the micro averaged true positive count.
+     * @return The micro averaged true positive count.
+     */
     public double tp() {
         return get(EvaluationMetric.Average.MICRO, LabelMetrics.TP);
     }
 
+    /**
+     * Gets the macro averaged true positive count.
+     * @return The macro averaged true positive count.
+     */
     public double macroTP() {
         return get(EvaluationMetric.Average.MACRO, LabelMetrics.TP);
     }
 
+    /**
+     * The false positive count for this label.
+     * @param label The label.
+     * @return The false positive count.
+     */
     public double fp(Label label) {
         return get(label, LabelMetrics.FP);
     }
 
+    /**
+     * Gets the micro averaged false positive count.
+     * @return The micro averaged false positive count.
+     */
     public double fp() {
         return get(EvaluationMetric.Average.MICRO, LabelMetrics.FP);
     }
 
+    /**
+     * Gets the macro averaged false positive count.
+     * @return The macro averaged false positive count.
+     */
     public double macroFP() {
         return get(EvaluationMetric.Average.MACRO, LabelMetrics.FP);
     }
 
+    /**
+     * The true negative count for this label.
+     * @param label The label.
+     * @return The true negative count.
+     */
     public double tn(Label label) {
         return get(label, LabelMetrics.TN);
     }
 
+    /**
+     * Gets the micro averaged true negative count.
+     * @return The micro averaged true negative count.
+     */
     public double tn() {
         return get(EvaluationMetric.Average.MICRO, LabelMetrics.TN);
     }
 
+    /**
+     * Gets the macro averaged true negative count.
+     * @return The macro averaged true negative count.
+     */
     public double macroTN() {
         return get(EvaluationMetric.Average.MACRO, LabelMetrics.TN);
     }
 
+    /**
+     * The false negative count for this label.
+     * @param label The label.
+     * @return The false negative count.
+     */
     public double fn(Label label) {
         return get(label, LabelMetrics.FN);
     }
 
+    /**
+     * Gets the micro averaged false negative count.
+     * @return The micro averaged false negative count.
+     */
     public double fn() {
         return get(EvaluationMetric.Average.MICRO, LabelMetrics.FN);
     }
 
+    /**
+     * Gets the macro averaged false negative count.
+     * @return The macro averaged false negative count.
+     */
     public double macroFN() {
         return get(EvaluationMetric.Average.MACRO, LabelMetrics.FN);
     }
 
+    /**
+     * The precision for this label.
+     * @param label The label.
+     * @return The precision.
+     */
     public double precision(Label label) {
         return get(label, LabelMetrics.PRECISION);
     }
 
+    /**
+     * The micro averaged precision.
+     * @return The micro averaged precision.
+     */
     public double microAveragedPrecision() {
         return get(EvaluationMetric.Average.MICRO, LabelMetrics.PRECISION);
     }
 
+    /**
+     * The macro averaged precision.
+     * @return The macro averaged precision.
+     */
     public double macroAveragedPrecision() {
         return get(EvaluationMetric.Average.MACRO, LabelMetrics.PRECISION);
     }
 
+    /**
+     * The recall for this label.
+     * @param label The label.
+     * @return The recall.
+     */
     public double recall(Label label) {
         return get(label, LabelMetrics.RECALL);
     }
 
+    /**
+     * The micro averaged recall.
+     * @return The micro averaged recall.
+     */
     public double microAveragedRecall() {
         return get(EvaluationMetric.Average.MICRO, LabelMetrics.RECALL);
     }
 
+    /**
+     * The macro averaged recall.
+     * @return The macro averaged recall.
+     */
     public double macroAveragedRecall() {
         return get(EvaluationMetric.Average.MACRO, LabelMetrics.RECALL);
     }
 
+    /**
+     * The F1 for this label.
+     * @param label The label.
+     * @return The F1.
+     */
     public double f1(Label label) {
         return get(label, LabelMetrics.RECALL);
     }
 
+    /**
+     * The micro averaged F1.
+     * @return The micro averaged F1.
+     */
     public double microAveragedF1() {
         return get(EvaluationMetric.Average.MICRO, LabelMetrics.F1);
     }
 
+    /**
+     * The macro averaged F1.
+     * @return The macro averaged F1.
+     */
     public double macroAveragedF1() {
         return get(EvaluationMetric.Average.MACRO, LabelMetrics.F1);
     }
 
+    /**
+     * The accuracy.
+     * @return The accuracy.
+     */
     public double accuracy() {
         return get(EvaluationMetric.Average.MICRO, LabelMetrics.ACCURACY);
     }
 
+    /**
+     * Gets the accuracy for this label.
+     * @param label The label.
+     * @return The accuracy.
+     */
     public double accuracy(Label label) {
         return get(label, LabelMetrics.ACCURACY);
     }
 
+    /**
+     * Gets the balanced error rate.
+     * <p>
+     * Also known as 1 - the macro averaged recall.
+     * @return The balanced error rate.
+     */
     public double balancedErrorRate() {
         // Target doesn't matter for balanced error rate, so we just use Average.macro
         // as it's the macro averaged recall.
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/sequence/SeqTrainTest.java b/Classification/Core/src/main/java/org/tribuo/classification/sequence/SeqTrainTest.java
index 844465e02..d3bd0ce89 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/sequence/SeqTrainTest.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/sequence/SeqTrainTest.java
@@ -53,14 +53,29 @@ public String getOptionsDescription() {
             return "Trains and tests a sequence classification model on the specified dataset.";
         }
 
+        /**
+         * Name of the example dataset, options are {gorilla}.
+         */
         @Option(charName = 'd', longName = "dataset-name", usage = "Name of the example dataset, options are {gorilla}.")
         public String datasetName = "";
+        /**
+         * Path to serialize model to.
+         */
         @Option(charName = 'f', longName = "output-path", usage = "Path to serialize model to.")
         public Path outputPath;
+        /**
+         * Path to a serialised SequenceDataset used for training.
+         */
         @Option(charName = 'u', longName = "train-dataset", usage = "Path to a serialised SequenceDataset used for training.")
         public Path trainDataset = null;
+        /**
+         * Path to a serialised SequenceDataset used for testing.
+         */
         @Option(charName = 'v', longName = "test-dataset", usage = "Path to a serialised SequenceDataset used for testing.")
         public Path testDataset = null;
+        /**
+         * Name of the trainer in the configuration file.
+         */
         @Option(charName = 't', longName = "trainer-name", usage = "Name of the trainer in the configuration file.")
         public SequenceTrainer<Label> trainer;
     }
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/sequence/example/SequenceDataGenerator.java b/Classification/Core/src/main/java/org/tribuo/classification/sequence/example/SequenceDataGenerator.java
index c427aac16..5ab093b31 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/sequence/example/SequenceDataGenerator.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/sequence/example/SequenceDataGenerator.java
@@ -38,6 +38,11 @@ public final class SequenceDataGenerator {
 
     private SequenceDataGenerator() { }
 
+    /**
+     * Generates a simple dataset consisting of numCopies repeats of two sequences.
+     * @param numCopies The number of times to repeat the two sequence examples.
+     * @return The dataset.
+     */
     public static MutableSequenceDataset<Label> generateGorillaDataset(int numCopies) {
         List<SequenceExample<Label>> examples = new ArrayList<>();
 
@@ -49,6 +54,10 @@ public static MutableSequenceDataset<Label> generateGorillaDataset(int numCopies
         return new MutableSequenceDataset<>(examples, new SimpleDataSourceProvenance("ExampleSequenceDataset", OffsetDateTime.now(),labelFactory),labelFactory);
     }
 
+    /**
+     * Generates a sequence example with a mixture of features and three labels "O", "Status" and "Monkey".
+     * @return A sequence example.
+     */
     public static SequenceExample<Label> generateGorillaA() {
         //"The silverback gorilla is angry"
         List<Example<Label>> examples = new ArrayList<>();
@@ -85,6 +94,10 @@ public static SequenceExample<Label> generateGorillaA() {
         return new SequenceExample<>(examples);
     }
 
+    /**
+     * Generates a sequence example with a mixture of features and three labels "O", "Status" and "Monkey".
+     * @return A sequence example.
+     */
     public static SequenceExample<Label> generateGorillaB() {
         //"That is one angry looking gorilla"
         List<Example<Label>> examples = new ArrayList<>();
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/DefaultFeatureExtractor.java b/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/DefaultFeatureExtractor.java
index 28daebde5..d2b8a2468 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/DefaultFeatureExtractor.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/DefaultFeatureExtractor.java
@@ -70,10 +70,21 @@ public class DefaultFeatureExtractor implements LabelFeatureExtractor {
     @Config(mandatory = true, description = "Use 4-grams of the labels as features.")
     private boolean use4gram;
 
+    /**
+     * Constructs a default feature extractor for bigrams and trigrams using the past 3 outcomes.
+     */
     public DefaultFeatureExtractor() {
         this(1, 3, true, true, false);
     }
 
+    /**
+     * Constructs a default feature extractor using the supplied parameters.
+     * @param mostRecentOutcome The most recent outcome to include as a feature.
+     * @param leastRecentOutcome The least recent outcome to include as a feature.
+     * @param useBigram Use bigrams of the outcomes.
+     * @param useTrigram Use trigrams of the outcomes.
+     * @param use4gram Use 4-grams of the outcomes.
+     */
     public DefaultFeatureExtractor(int mostRecentOutcome, int leastRecentOutcome, boolean useBigram, boolean useTrigram, boolean use4gram) {
         this.mostRecentOutcome = mostRecentOutcome;
         this.leastRecentOutcome = leastRecentOutcome;
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/ViterbiModel.java b/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/ViterbiModel.java
index df222b244..0ac245882 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/ViterbiModel.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/ViterbiModel.java
@@ -47,7 +47,14 @@ public class ViterbiModel extends SequenceModel<Label> {
      * Types of label score aggregation.
      */
     public enum ScoreAggregation {
-        ADD, MULTIPLY
+        /**
+         * Adds the scores.
+         */
+        ADD,
+        /**
+         * Multiplies the scores.
+         */
+        MULTIPLY
     }
 
     private final Model<Label> model;
@@ -215,10 +222,18 @@ public int compareTo(Path that) {
 
     }
 
+    /**
+     * Gets the stack size of this model.
+     * @return The stack size.
+     */
     public int getStackSize() {
         return stackSize;
     }
 
+    /**
+     * Gets the score aggregation function.
+     * @return The score aggregation function.
+     */
     public ScoreAggregation getScoreAggregation() {
         return scoreAggregation;
     }
diff --git a/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/ViterbiTrainer.java b/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/ViterbiTrainer.java
index 6495d3990..394f8f437 100644
--- a/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/ViterbiTrainer.java
+++ b/Classification/Core/src/main/java/org/tribuo/classification/sequence/viterbi/ViterbiTrainer.java
@@ -60,11 +60,24 @@ public final class ViterbiTrainer implements SequenceTrainer<Label> {
 
     private int trainInvocationCounter = 0;
 
+    /**
+     * Constructs a ViterbiTrainer wrapping the specified trainer, with an unbounded stack size.
+     * @param trainer The trainer to wrap.
+     * @param labelFeatureExtractor The feature extraction function for labels.
+     * @param scoreAggregation The score aggregation function.
+     */
     public ViterbiTrainer(Trainer<Label> trainer, LabelFeatureExtractor labelFeatureExtractor,
                           ScoreAggregation scoreAggregation) {
         this(trainer, labelFeatureExtractor, -1, scoreAggregation);
     }
 
+    /**
+     * Constructs a ViterbiTrainer wrapping the specified trainer.
+     * @param trainer The trainer to wrap.
+     * @param labelFeatureExtractor The feature extraction function for labels.
+     * @param stackSize The stack size.
+     * @param scoreAggregation The score aggregation function.
+     */
     public ViterbiTrainer(Trainer<Label> trainer, LabelFeatureExtractor labelFeatureExtractor, int stackSize,
                           ScoreAggregation scoreAggregation) {
         this.trainer = trainer;
diff --git a/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/CARTClassificationOptions.java b/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/CARTClassificationOptions.java
index eed3b2831..7489c67b0 100644
--- a/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/CARTClassificationOptions.java
+++ b/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/CARTClassificationOptions.java
@@ -57,22 +57,49 @@ public enum ImpurityType {
         ENTROPY
     }
 
+    /**
+     * Maximum depth in the decision tree.
+     */
     @Option(longName = "cart-max-depth", usage = "Maximum depth in the decision tree.")
     public int cartMaxDepth = 6;
+    /**
+     * Minimum child weight.
+     */
     @Option(longName = "cart-min-child-weight", usage = "Minimum child weight.")
     public float cartMinChildWeight = 5.0f;
+    /**
+     * Minimum impurity decrease.
+     */
     @Option(longName = "cart-min-impurity-decrease", usage = "Minimum impurity decrease.")
     public float cartMinImpurityDecrease = 0.0f;
+    /**
+     * Fraction of features in split.
+     */
     @Option(longName = "cart-split-fraction", usage = "Fraction of features in split.")
     public float cartSplitFraction = 1.0f;
+    /**
+     * Whether to choose split points for features at random.
+     */
     @Option(longName = "cart-random-split", usage = "Whether to choose split points for features at random.")
     public boolean cartRandomSplit = false;
+    /**
+     * Impurity measure to use. Defaults to GINI.
+     */
     @Option(longName = "cart-impurity", usage = "Impurity measure to use. Defaults to GINI.")
     public ImpurityType cartImpurity = ImpurityType.GINI;
+    /**
+     * Prints the decision tree.
+     */
     @Option(longName = "cart-print-tree", usage = "Prints the decision tree.")
     public boolean cartPrintTree;
+    /**
+     * Tree algorithm to use (options are CART).
+     */
     @Option(longName = "cart-tree-algorithm", usage = "Tree algorithm to use (options are CART).")
     public TreeType cartTreeAlgorithm = TreeType.CART;
+    /**
+     * RNG seed.
+     */
     @Option(longName = "cart-seed", usage = "RNG seed.")
     public long cartSeed = Trainer.DEFAULT_SEED;
 
@@ -93,8 +120,8 @@ public CARTClassificationTrainer getTrainer() {
         CARTClassificationTrainer trainer;
         switch (cartTreeAlgorithm) {
             case CART:
-                    trainer = new CARTClassificationTrainer(cartMaxDepth, cartMinChildWeight, cartMinImpurityDecrease
-                            ,cartSplitFraction, cartRandomSplit, impurity, cartSeed);
+                trainer = new CARTClassificationTrainer(cartMaxDepth, cartMinChildWeight, cartMinImpurityDecrease,
+                        cartSplitFraction, cartRandomSplit, impurity, cartSeed);
                 break;
             default:
                 throw new IllegalArgumentException("Unknown tree type " + cartTreeAlgorithm);
diff --git a/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/TrainTest.java b/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/TrainTest.java
index 6dff529a6..fe735911f 100644
--- a/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/TrainTest.java
+++ b/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/TrainTest.java
@@ -44,10 +44,21 @@ public String getOptionsDescription() {
             return "A simple train/test program for Classification trees.";
         }
 
+        /**
+         * The data loading options.
+         */
         public DataOptions generalOptions;
+        /**
+         * The CART trainer options.
+         */
         public CARTClassificationOptions cartOptions;
     }
 
+    /**
+     * Runs a TrainTest CLI.
+     * @param args the command line arguments
+     * @throws IOException if there is any error reading the examples.
+     */
     public static void main(String[] args) throws IOException {
         TrainTestOptions o = new TrainTestOptions();
         try (ConfigurationManager cm = new ConfigurationManager(args,o)){
diff --git a/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/impl/InvertedFeature.java b/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/impl/InvertedFeature.java
index 232b0510a..f577ce4b7 100644
--- a/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/impl/InvertedFeature.java
+++ b/Classification/DecisionTree/src/main/java/org/tribuo/classification/dtree/impl/InvertedFeature.java
@@ -39,6 +39,9 @@ class InvertedFeature implements Comparable<InvertedFeature>, Serializable {
 
     private static final int DEFAULT_SIZE = 8;
 
+    /**
+     * The value of this feature.
+     */
     public final double value;
 
     private final int numLabels;
diff --git a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/AllTrainerOptions.java b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/AllTrainerOptions.java
index 14a1a2a7f..11aa69628 100644
--- a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/AllTrainerOptions.java
+++ b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/AllTrainerOptions.java
@@ -78,19 +78,52 @@ public enum AlgorithmType {
         XGBOOST,
     }
 
+    /**
+     * Type of learner (or base learner). Defaults to SGD_LINEAR.
+     */
     @Option(longName = "algorithm", usage = "Type of learner (or base learner). Defaults to SGD_LINEAR.")
     public AlgorithmType algorithm = AlgorithmType.SGD_LINEAR;
 
+    /**
+     * Options for CART trainers.
+     */
     public CARTClassificationOptions cartOptions;
+    /**
+     * Options for K-NN trainers.
+     */
     public KNNClassifierOptions knnOptions;
+    /**
+     * Options for LibLinear trainers.
+     */
     public LibLinearOptions liblinearOptions;
+    /**
+     * Options for LibSVM trainers.
+     */
     public LibSVMOptions libsvmOptions;
+    /**
+     * Options for Multinomial Naive Bayes trainers.
+     */
     public MultinomialNaiveBayesOptions mnbOptions;
+    /**
+     * Options for Kernel SVM trainers.
+     */
     public KernelSVMOptions kernelSVMOptions;
+    /**
+     * Options for Linear SGD trainers.
+     */
     public LinearSGDOptions linearSGDOptions;
+    /**
+     * Options for XGBoost trainers.
+     */
     public XGBoostOptions xgBoostOptions;
 
+    /**
+     * Options for classifier ensembles.
+     */
     public ClassificationEnsembleOptions ensemble;
+    /**
+     * Options for hashing trainers.
+     */
     public HashingOptions hashingOptions;
 
     @Override
diff --git a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/ConfigurableTrainTest.java b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/ConfigurableTrainTest.java
index 443e92cdd..59525dd98 100644
--- a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/ConfigurableTrainTest.java
+++ b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/ConfigurableTrainTest.java
@@ -65,15 +65,27 @@ public String getOptionsDescription() {
             return "Loads a Trainer (and optionally a Datasource) from a config file, trains a Model, tests it and optionally saves it to disk.";
         }
 
+        /**
+         * Options for loading in data.
+         */
         public DataOptions general;
 
-        @Option(charName='t',longName="trainer",usage="Load a trainer from the config file.")
+        /**
+         * Load a trainer from the config file.
+         */
+        @Option(charName = 't', longName = "trainer", usage = "Load a trainer from the config file.")
         public Trainer<Label> trainer;
 
-        @Option(charName='w',longName="weights",usage="A list of weights to use in classification. Format = LABEL_NAME:weight,LABEL_NAME:weight...")
+        /**
+         * A list of weights to use in classification. Format = LABEL_NAME:weight,LABEL_NAME:weight...
+         */
+        @Option(charName = 'w', longName = "weights", usage = "A list of weights to use in classification. Format = LABEL_NAME:weight,LABEL_NAME:weight...")
         public List<String> weights;
 
-        @Option(charName='o',longName="predictions",usage="Path to write model predictions")
+        /**
+         * Path to write model predictions
+         */
+        @Option(charName = 'o', longName = "predictions", usage = "Path to write model predictions")
         public Path predictionPath;
     }
 
diff --git a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/RunAll.java b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/RunAll.java
index 8a1125e2f..6bfff3fe3 100644
--- a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/RunAll.java
+++ b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/RunAll.java
@@ -61,12 +61,24 @@ public static class RunAllOptions implements Options {
         public String getOptionsDescription() {
             return "Performs the same training and test experiment on all Trainers in the supplied configuration file.";
         }
+
+        /**
+         * Options for loading in data.
+         */
         public DataOptions general;
 
-        @Option(charName='d',longName="output-directory",usage="Directory to write out the models and test reports.")
+        /**
+         * Directory to write out the models and test reports.
+         */
+        @Option(charName = 'd', longName = "output-directory", usage = "Directory to write out the models and test reports.")
         public File directory;
     }
 
+    /**
+     * Runs the RunALL CLI.
+     * @param args The CLI arguments.
+     * @throws IOException If it failed to load the data.
+     */
     public static void main(String[] args) throws IOException {
         LabsLogFormatter.setAllLogFormatters();
 
diff --git a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/Test.java b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/Test.java
index e1300584d..3f57bad57 100644
--- a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/Test.java
+++ b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/Test.java
@@ -70,26 +70,60 @@ public static class ConfigurableTestOptions implements Options {
         public String getOptionsDescription() {
             return "Tests an already trained classifier on a dataset.";
         }
-        @Option(longName="hashing-dimension",usage="Hashing dimension used for standard text format.")
+
+        /**
+         * Hashing dimension used for standard text format.
+         */
+        @Option(longName = "hashing-dimension", usage = "Hashing dimension used for standard text format.")
         public int hashDim = 0;
-        @Option(longName="ngram",usage="Ngram size to generate when using standard text format. Defaults to 2.")
+        /**
+         * Ngram size to generate when using standard text format. Defaults to 2.
+         */
+        @Option(longName = "ngram", usage = "Ngram size to generate when using standard text format. Defaults to 2.")
         public int ngram = 2;
-        @Option(longName="term-counting",usage="Use term counts instead of boolean when using the standard text format.")
+        /**
+         * Use term counts instead of boolean when using the standard text format.
+         */
+        @Option(longName = "term-counting", usage = "Use term counts instead of boolean when using the standard text format.")
         public boolean termCounting;
-        @Option(longName="csv-response-name",usage="Response name in the csv file.")
+        /**
+         * Response name in the csv file.
+         */
+        @Option(longName = "csv-response-name", usage = "Response name in the csv file.")
         public String csvResponseName;
-        @Option(longName="libsvm-zero-indexed",usage="Is the libsvm file zero indexed.")
+        /**
+         * Is the libsvm file zero indexed.
+         */
+        @Option(longName = "libsvm-zero-indexed", usage = "Is the libsvm file zero indexed.")
         public boolean zeroIndexed = false;
-        @Option(charName='f',longName="model-path",usage="Load a trainer from the config file.")
+        /**
+         * Load a trainer from the config file.
+         */
+        @Option(charName = 'f', longName = "model-path", usage = "Load a trainer from the config file.")
         public Path modelPath;
-        @Option(charName='o',longName="predictions",usage="Path to write model predictions")
+        /**
+         * Path to write model predictions
+         */
+        @Option(charName = 'o', longName = "predictions", usage = "Path to write model predictions")
         public Path predictionPath;
-        @Option(charName='s',longName="input-format",usage="Loads the data using the specified format. Defaults to LIBSVM.")
+        /**
+         * Loads the data using the specified format. Defaults to LIBSVM.
+         */
+        @Option(charName = 's', longName = "input-format", usage = "Loads the data using the specified format. Defaults to LIBSVM.")
         public DataOptions.InputFormat inputFormat = DataOptions.InputFormat.LIBSVM;
-        @Option(charName='v',longName="testing-file",usage="Path to the testing file.")
+        /**
+         * Path to the testing file.
+         */
+        @Option(charName = 'v', longName = "testing-file", usage = "Path to the testing file.")
         public Path testingPath;
     }
 
+    /**
+     * Loads in the model and the dataset from the options.
+     * @param o The options.
+     * @return The model and the dataset.
+     * @throws IOException If either the model or dataset could not be read.
+     */
     @SuppressWarnings("unchecked") // deserialising generically typed datasets.
     public static Pair<Model<Label>,Dataset<Label>> load(ConfigurableTestOptions o) throws IOException {
         Path modelPath = o.modelPath;
@@ -160,6 +194,7 @@ public static Pair<Model<Label>,Dataset<Label>> load(ConfigurableTestOptions o)
     }
 
     /**
+     * Runs the Test CLI.
      * @param args the command line arguments
      */
     public static void main(String[] args) {
diff --git a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/TrainTest.java b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/TrainTest.java
index 693a20c1b..0b62469ff 100644
--- a/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/TrainTest.java
+++ b/Classification/Experiments/src/main/java/org/tribuo/classification/experiments/TrainTest.java
@@ -43,10 +43,22 @@ public String getOptionsDescription() {
             return "Trains and tests the specified classifier on the supplied datasets.";
         }
 
+        /**
+         * All the classification trainer options.
+         */
         public AllTrainerOptions trainerOptions;
+
+        /**
+         * Options for loading in data.
+         */
         public DataOptions general;
     }
 
+    /**
+     * Runs a TrainTest CLI.
+     * @param args the command line arguments
+     * @throws IOException if there is any error reading the examples.
+     */
     public static void main(String[] args) throws IOException {
         AllClassificationOptions o = new AllClassificationOptions();
         try (ConfigurationManager cm = new ConfigurationManager(args,o)){
diff --git a/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMEBase.java b/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMEBase.java
index 5932352ac..2fd4bef65 100644
--- a/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMEBase.java
+++ b/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMEBase.java
@@ -68,7 +68,13 @@
 public class LIMEBase implements TabularExplainer<Regressor> {
     private static final Logger logger = Logger.getLogger(LIMEBase.class.getName());
 
+    /**
+     * Width of the noise gaussian.
+     */
     public static final double WIDTH_CONSTANT = 0.75;
+    /**
+     * Delta to consider two distances equal.
+     */
     public static final double DISTANCE_DELTA = 1e-12;
 
     protected static final OutputFactory<Regressor> regressionFactory = new RegressionFactory();
diff --git a/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMEExplanation.java b/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMEExplanation.java
index 85b48c9bf..0aa0a0907 100644
--- a/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMEExplanation.java
+++ b/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMEExplanation.java
@@ -49,6 +49,12 @@ public class LIMEExplanation implements Explanation<Regressor> {
 
     private final RegressionEvaluation evaluation;
 
+    /**
+     * Constructs a LIME explanation.
+     * @param model The explanation model.
+     * @param prediction The prediction being explained.
+     * @param evaluation The evaluation of the explanation model on the sampled data.
+     */
     public LIMEExplanation(SparseModel<Regressor> model, Prediction<Label> prediction, RegressionEvaluation evaluation) {
         this.model = model;
         this.prediction = prediction;
diff --git a/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMETextCLI.java b/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMETextCLI.java
index 91cb87ee1..b9fa5774a 100644
--- a/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMETextCLI.java
+++ b/Classification/Explanations/src/main/java/org/tribuo/classification/explanations/lime/LIMETextCLI.java
@@ -57,7 +57,7 @@
 public class LIMETextCLI implements CommandGroup {
     private static final Logger logger = Logger.getLogger(LIMETextCLI.class.getName());
 
-    protected CommandInterpreter shell;
+    private final CommandInterpreter shell;
 
     private Model<Label> model;
 
@@ -74,6 +74,9 @@ public class LIMETextCLI implements CommandGroup {
 
     private LIMEText limeText = null;
 
+    /**
+     * Constructs a LIME CLI.
+     */
     public LIMETextCLI() {
         shell = new CommandInterpreter();
         shell.setPrompt("lime-text sh% ");
@@ -89,6 +92,10 @@ public String getDescription() {
         return "Commands for experimenting with LIME Text.";
     }
 
+    /**
+     * Completers for filenames.
+     * @return The filename completers.
+     */
     public Completer[] fileCompleter() {
         return new Completer[]{
                 new Completers.FileNameCompleter(),
@@ -104,6 +111,12 @@ public void startShell() {
         shell.start();
     }
 
+    /**
+     * Loads a model in from disk.
+     * @param ci The command interpreter.
+     * @param path The path to load the model from.
+     * @return A status message.
+     */
     @Command(usage = "<filename> - Load a model from disk.", completers="fileCompleter")
     public String loadModel(CommandInterpreter ci, File path) {
         try (ObjectInputStream ois = new ObjectInputStream(new BufferedInputStream(new FileInputStream(path)))) {
@@ -126,16 +139,32 @@ public String loadModel(CommandInterpreter ci, File path) {
         return "Loaded model from path " + path.toString();
     }
 
+    /**
+     * Does the model generate probabilities.
+     * @param ci The command interpreter.
+     * @return True if the model generates probabilities.
+     */
     @Command(usage="Does the model generate probabilities")
     public String generatesProbabilities(CommandInterpreter ci) {
         return ""+model.generatesProbabilities();
     }
 
+    /**
+     * Shows the model description.
+     * @param ci The command interpreter.
+     * @return The model description.
+     */
     @Command(usage="Shows the model description")
     public String modelDescription(CommandInterpreter ci) {
         return model.toString();
     }
 
+    /**
+     * Shows information on a particular feature.
+     * @param ci The command interpreter.
+     * @param featureName The feature to show.
+     * @return Feature information.
+     */
     @Command(usage="Shows the information on a particular feature")
     public String featureInfo(CommandInterpreter ci, String featureName) {
         VariableInfo f = model.getFeatureIDMap().get(featureName);
@@ -146,21 +175,33 @@ public String featureInfo(CommandInterpreter ci, String featureName) {
         }
     }
 
-    @Command(usage="Shows the output information.")
-    public String outputInfo(CommandInterpreter ci) {
-        return model.getOutputIDInfo().toReadableString();
-    }
-
+    /**
+     * Shows the top features of the loaded model.
+     * @param ci The command interpeter.
+     * @param numFeatures The number of features to show.
+     * @return The top features of the model.
+     */
     @Command(usage="<int> - Shows the top N features in the model")
     public String topFeatures(CommandInterpreter ci, int numFeatures) {
         return ""+ model.getTopFeatures(numFeatures);
     }
 
+    /**
+     * Shows the number of features.
+     * @param ci The command interpreter.
+     * @return The number of features in the model.
+     */
     @Command(usage="Shows the number of features in the model")
     public String numFeatures(CommandInterpreter ci) {
         return ""+ model.getFeatureIDMap().size();
     }
 
+    /**
+     * Shows the number of features that occurred more than minCount times.
+     * @param ci The command interpreter.
+     * @param minCount The minimum feature occurrence.
+     * @return The number of features with more than minCount occurrences.
+     */
     @Command(usage="<min count> - Shows the number of features that occurred more than min count times.")
     public String minCount(CommandInterpreter ci, int minCount) {
         int counter = 0;
@@ -172,17 +213,34 @@ public String minCount(CommandInterpreter ci, int minCount) {
         return counter + " features occurred more than " + minCount + " times.";
     }
 
+    /**
+     * Shows the output statistics.
+     * @param ci The command interpreter.
+     * @return The output statistics.
+     */
     @Command(usage="Shows the output statistics")
     public String showLabelStats(CommandInterpreter ci) {
         return "Label histogram : \n" + model.getOutputIDInfo().toReadableString();
     }
 
+    /**
+     * Sets the number of samples to use in LIME.
+     * @param ci The command interpreter.
+     * @param newNumSamples The number of samples to use in LIME.
+     * @return A status message.
+     */
     @Command(usage="Sets the number of samples to use in LIME")
     public String setNumSamples(CommandInterpreter ci, int newNumSamples) {
         numSamples = newNumSamples;
         return "Set number of samples to " + numSamples;
     }
 
+    /**
+     * Explains a text classification.
+     * @param ci The command interpreter.
+     * @param tokens A space separated token stream.
+     * @return An explanation.
+     */
     @Command(usage="Explain a text classification")
     public String explain(CommandInterpreter ci, String[] tokens) {
         String text = String.join(" ",tokens);
@@ -196,6 +254,12 @@ public String explain(CommandInterpreter ci, String[] tokens) {
         return "Explanation = " + explanation.toString();
     }
 
+    /**
+     * Sets the number of features LIME should use in an explanation.
+     * @param ci The command interpreter.
+     * @param newNumFeatures The number of features.
+     * @return A status message.
+     */
     @Command(usage="Sets the number of features LIME should use in an explanation")
     public String setNumFeatures(CommandInterpreter ci, int newNumFeatures) {
         numFeatures = newNumFeatures;
@@ -205,6 +269,12 @@ public String setNumFeatures(CommandInterpreter ci, int newNumFeatures) {
         return "Set the number of features in LIME to " + numFeatures;
     }
 
+    /**
+     * Makes a prediction using the loaded model.
+     * @param ci The command interpreter.
+     * @param tokens A space separated token stream.
+     * @return The prediction.
+     */
     @Command(usage="Make a prediction")
     public String predict(CommandInterpreter ci, String[] tokens) {
         String text = String.join(" ",tokens);
@@ -218,10 +288,17 @@ public String predict(CommandInterpreter ci, String[] tokens) {
      * Command line options.
      */
     public static class LIMETextCLIOptions implements Options {
-        @Option(charName='f',longName="filename",usage="Model file to load. Optional.")
+        /**
+         * Model file to load. Optional.
+         */
+        @Option(charName = 'f', longName = "filename", usage = "Model file to load. Optional.")
         public String modelFilename;
     }
 
+    /**
+     * Runs a LIMETextCLI.
+     * @param args The CLI arguments.
+     */
     public static void main(String[] args) {
         LIMETextCLI.LIMETextCLIOptions options = new LIMETextCLI.LIMETextCLIOptions();
         try {
diff --git a/Classification/LibLinear/src/main/java/org/tribuo/classification/liblinear/LinearClassificationType.java b/Classification/LibLinear/src/main/java/org/tribuo/classification/liblinear/LinearClassificationType.java
index 41ed3065c..b330d1aa2 100644
--- a/Classification/LibLinear/src/main/java/org/tribuo/classification/liblinear/LinearClassificationType.java
+++ b/Classification/LibLinear/src/main/java/org/tribuo/classification/liblinear/LinearClassificationType.java
@@ -76,6 +76,10 @@ public enum LinearType implements Serializable {
             this.type = type;
         }
 
+        /**
+         * Gets the LibLinear solver type.
+         * @return The solver type.
+         */
         public SolverType getSolverType() {
             return type;
         }
@@ -89,6 +93,10 @@ public SolverType getSolverType() {
      */
     private LinearClassificationType() {}
 
+    /**
+     * Constructs a LinearClassificationType using the supplied algorithm.
+     * @param type The liblinear algorithm.
+     */
     public LinearClassificationType(LinearType type) {
         this.type = type;
     }
diff --git a/Classification/LibLinear/src/main/java/org/tribuo/classification/liblinear/TrainTest.java b/Classification/LibLinear/src/main/java/org/tribuo/classification/liblinear/TrainTest.java
index 9993c5d84..3980df436 100644
--- a/Classification/LibLinear/src/main/java/org/tribuo/classification/liblinear/TrainTest.java
+++ b/Classification/LibLinear/src/main/java/org/tribuo/classification/liblinear/TrainTest.java
@@ -43,11 +43,25 @@ public String getOptionsDescription() {
             return "Trains and tests a LibLinear model on the specified datasets.";
         }
 
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
+        /**
+         * The liblinear options.
+         */
         public LibLinearOptions libLinearOptions;
+        /**
+         * The ensemble options.
+         */
         public ClassificationEnsembleOptions ensembleOptions;
     }
 
+    /**
+     * Runs a TrainTest CLI.
+     * @param args the command line arguments
+     * @throws IOException if there is any error reading the examples.
+     */
     public static void main(String[] args) throws IOException {
         TrainTestOptions o = new TrainTestOptions();
         try (ConfigurationManager cm = new ConfigurationManager(args, o)){
diff --git a/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/LibSVMOptions.java b/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/LibSVMOptions.java
index 71ba489a7..c161f25d5 100644
--- a/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/LibSVMOptions.java
+++ b/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/LibSVMOptions.java
@@ -33,14 +33,29 @@ public String getOptionsDescription() {
         return "Options for parameterising a LibSVM classification trainer.";
     }
 
+    /**
+     * Intercept in kernel function. Defaults to 0.0.
+     */
     @Option(longName = "svm-coefficient", usage = "Intercept in kernel function. Defaults to 0.0.")
     public double svmCoefficient = 0.0;  //TODO should this be 1.0?
+    /**
+     * Degree in polynomial kernel. Defaults to 3.
+     */
     @Option(longName = "svm-degree", usage = "Degree in polynomial kernel. Defaults to 3.")
     public int svmDegree = 3;
+    /**
+     * Gamma value in kernel function. Defaults to 0.0.
+     */
     @Option(longName = "svm-gamma", usage = "Gamma value in kernel function. Defaults to 0.0.")
     public double svmGamma = 0.0;  //TODO should the default be 0.1
+    /**
+     * Type of SVM kernel. Defaults to LINEAR.
+     */
     @Option(longName = "svm-kernel", usage = "Type of SVM kernel. Defaults to LINEAR.")
     public KernelType svmKernel = KernelType.LINEAR;
+    /**
+     * Type of SVM. Defaults to C_SVC.
+     */
     @Option(longName = "svm-type", usage = "Type of SVM. Defaults to C_SVC.")
     public SVMClassificationType.SVMMode svmType = SVMMode.C_SVC;
 
diff --git a/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/SVMClassificationType.java b/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/SVMClassificationType.java
index 8595304bc..2da01e73f 100644
--- a/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/SVMClassificationType.java
+++ b/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/SVMClassificationType.java
@@ -58,6 +58,10 @@ public enum SVMMode {
      */
     private SVMClassificationType() {}
 
+    /**
+     * Constructs an SVMClassificationType using the supplied SVM algorithm.
+     * @param type The SVM algorithm.
+     */
     public SVMClassificationType(SVMMode type) {
         this.type = type;
     }
diff --git a/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/TrainTest.java b/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/TrainTest.java
index a407fde93..6156e6f36 100644
--- a/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/TrainTest.java
+++ b/Classification/LibSVM/src/main/java/org/tribuo/classification/libsvm/TrainTest.java
@@ -46,11 +46,25 @@ public String getOptionsDescription() {
             return "Trains and tests a LibSVM model on the specified datasets.";
         }
 
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
+        /**
+         * The LibSVM options.
+         */
         public LibSVMOptions libsvmOptions;
+        /**
+         * The ensemble options.
+         */
         public ClassificationEnsembleOptions ensembleOptions;
     }
 
+    /**
+     * Runs a TrainTest CLI.
+     * @param args the command line arguments
+     * @throws IOException if there is any error reading the examples.
+     */
     public static void main(String[] args) throws IOException {
         TrainTestOptions o = new TrainTestOptions();
         try (ConfigurationManager cm = new ConfigurationManager(args, o)) {
diff --git a/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/MultinomialNaiveBayesOptions.java b/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/MultinomialNaiveBayesOptions.java
index 24fa5d30c..83c279ab0 100644
--- a/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/MultinomialNaiveBayesOptions.java
+++ b/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/MultinomialNaiveBayesOptions.java
@@ -24,6 +24,9 @@
  */
 public class MultinomialNaiveBayesOptions implements ClassificationOptions<MultinomialNaiveBayesTrainer> {
 
+    /**
+     * Smoothing parameter for the conditional probabilities.
+     */
     @Option(charName = 'a', longName = "mnb-alpha", usage = "Smoothing parameter for the conditional probabilities.")
     public double mnbAlpha;
 
diff --git a/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/MultinomialNaiveBayesTrainer.java b/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/MultinomialNaiveBayesTrainer.java
index e954eea23..e4d4a34bb 100644
--- a/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/MultinomialNaiveBayesTrainer.java
+++ b/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/MultinomialNaiveBayesTrainer.java
@@ -57,10 +57,17 @@ public class MultinomialNaiveBayesTrainer implements Trainer<Label>, WeightedExa
 
     private int invocationCount = 0;
 
+    /**
+     * Constructs a multinomial naive bayes trainer using a smoothing value of 1.0.
+     */
     public MultinomialNaiveBayesTrainer() {
         this(1.0);
     }
 
+    /**
+     * Constructs a multinomial naive bayes trainer with the specified smoothing value.
+     * @param alpha The smoothing value.
+     */
     //TODO support different alphas for different features?
     public MultinomialNaiveBayesTrainer(double alpha) {
         if(alpha <= 0.0) {
diff --git a/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/TrainTest.java b/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/TrainTest.java
index f06716fa5..e3737ae42 100644
--- a/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/TrainTest.java
+++ b/Classification/MultinomialNaiveBayes/src/main/java/org/tribuo/classification/mnb/TrainTest.java
@@ -39,8 +39,17 @@ public class TrainTest {
      * Command line options.
      */
     public static class TrainTestOptions implements Options {
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
+        /**
+         * The multinomial naive bayes options.
+         */
         public MultinomialNaiveBayesOptions mnbOptions;
+        /**
+         * The ensemble options.
+         */
         public ClassificationEnsembleOptions ensembleOptions;
 
         @Override
@@ -50,6 +59,11 @@ public String getOptionsDescription() {
 
     }
 
+    /**
+     * Runs a TrainTest CLI.
+     * @param args the command line arguments
+     * @throws IOException if there is any error reading the examples.
+     */
     public static void main(String[] args) throws IOException {
         TrainTestOptions o = new TrainTestOptions();
         try (ConfigurationManager cm = new ConfigurationManager(args,o)){
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/TrainTest.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/TrainTest.java
index 9c51fc664..adf096801 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/TrainTest.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/TrainTest.java
@@ -45,11 +45,25 @@ public String getOptionsDescription() {
             return "Trains and tests a linear SGD model on the specified datasets.";
         }
 
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
+        /**
+         * The linear SGD trainer options.
+         */
         public LinearSGDOptions trainerOptions;
+        /**
+         * The ensemble options.
+         */
         public ClassificationEnsembleOptions ensembleOptions;
     }
 
+    /**
+     * Runs a TrainTest CLI.
+     * @param args the command line arguments
+     * @throws IOException if there is any error reading the examples.
+     */
     public static void main(String[] args) throws IOException {
         TrainTestOptions o = new TrainTestOptions();
         try (ConfigurationManager cm = new ConfigurationManager(args,o)){
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/Util.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/Util.java
index 996575e4b..8b6239eed 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/Util.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/Util.java
@@ -127,10 +127,25 @@ public static ExampleArray shuffle(SparseVector[] features, int[] labels, double
      * A nominal tuple. One day it'll be a record, but not today.
      */
     public static class ExampleArray {
+        /**
+         * The examples encoded as sparse vectors.
+         */
         public final SparseVector[] features;
+        /**
+         * The label indices.
+         */
         public final int[] labels;
+        /**
+         * The example weights.
+         */
         public final double[] weights;
 
+        /**
+         * Constructs an example array.
+         * @param features The examples' features.
+         * @param labels The label indices.
+         * @param weights The example weights.
+         */
         public ExampleArray(SparseVector[] features, int[] labels, double[] weights) {
             this.features = features;
             this.labels = labels;
@@ -206,8 +221,17 @@ public static SequenceExampleArray shuffle(SGDVector[][] features, int[][] label
      * A nominal tuple. One day it'll be a record, but not today.
      */
     public static class SequenceExampleArray {
+        /**
+         * The array of sequence example features.
+         */
         public final SGDVector[][] features;
+        /**
+         * The sequence example label indices.
+         */
         public final int[][] labels;
+        /**
+         * The sequence example weights.
+         */
         public final double[] weights;
 
         SequenceExampleArray(SGDVector[][] features, int[][] labels, double[] weights) {
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/CRFOptions.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/CRFOptions.java
index db8e38176..40be2a63c 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/CRFOptions.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/CRFOptions.java
@@ -26,6 +26,9 @@
  */
 public class CRFOptions implements Options {
 
+    /**
+     * Options for the gradient optimiser.
+     */
     public GradientOptimiserOptions sgoOptions;
 
     @Option(longName = "crf-epochs", usage = "Epochs of SGD.")
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/ChainHelper.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/ChainHelper.java
index 8a5d7c92f..33ff7fa74 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/ChainHelper.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/ChainHelper.java
@@ -251,9 +251,21 @@ public static double sumLogProbs(double[] input) {
      * Belief Propagation results. One day it'll be a record, but not today.
      */
     public static final class ChainBPResults {
+        /**
+         * The normalization constant.
+         */
         public final double logZ;
+        /**
+         * The alpha values array from forward propagation.
+         */
         public final DenseVector[] alphas;
+        /**
+         * The beta values array from backward propagation.
+         */
         public final DenseVector[] betas;
+        /**
+         * The local clique scores (i.e., the linear model scores).
+         */
         public final ChainCliqueValues scores;
 
         ChainBPResults(double logZ, DenseVector[] alphas, DenseVector[] betas, ChainCliqueValues scores) {
@@ -268,7 +280,13 @@ public static final class ChainBPResults {
      * Clique scores within a chain. One day it'll be a record, but not today.
      */
     public static final class ChainCliqueValues {
+        /**
+         * The per element values.
+         */
         public final DenseVector[] localValues;
+        /**
+         * The label-label transition matrix.
+         */
         public final DenseMatrix transitionValues;
 
         ChainCliqueValues(DenseVector[] localValues, DenseMatrix transitionValues) {
@@ -281,8 +299,17 @@ public static final class ChainCliqueValues {
      * Viterbi output from a linear chain. One day it'll be a record, but not today.
      */
     public static final class ChainViterbiResults {
+        /**
+         * The score for this result.
+         */
         public final double mapScore;
+        /**
+         * The viterbi states.
+         */
         public final int[] mapValues;
+        /**
+         * The pre-viterbi scores.
+         */
         public final ChainCliqueValues scores;
 
         ChainViterbiResults(double mapScore, int[] mapValues, ChainCliqueValues scores) {
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/Chunk.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/Chunk.java
index 0dc9da08e..4575eeb20 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/Chunk.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/Chunk.java
@@ -20,14 +20,29 @@
  * Chunk class used for chunk level confidence prediction in the {@link CRFModel}.
  */
 public class Chunk {
+    /**
+     * The starting point of this chunk.
+     */
     public final int begin;
+    /**
+     * The labels for the next labels.length values.
+     */
     public final int[] labels;
 
+    /**
+     * Constructs a chunk.
+     * @param begin The starting point of this chunk.
+     * @param labels The chunk labels.
+     */
     public Chunk(int begin, int[] labels) {
         this.begin = begin;
         this.labels = labels;
     }
 
+    /**
+     * Copies the labels into the constraints array starting at {@link #begin}.
+     * @param constraints The constraints array.
+     */
     public void unpack(int[] constraints) {
         System.arraycopy(labels, 0, constraints, begin, labels.length);
     }
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/SeqTest.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/SeqTest.java
index d4666e5ba..ad59803df 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/SeqTest.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/crf/SeqTest.java
@@ -61,27 +61,63 @@ public String getOptionsDescription() {
             return "Tests a linear chain CRF model on the specified dataset.";
         }
 
+        /**
+         * Gradient descent options.
+         */
         public GradientOptimiserOptions gradientOptions;
+        /**
+         * Name of the example dataset, options are {gorilla}.
+         */
         @Option(charName = 'd', longName = "dataset-name", usage = "Name of the example dataset, options are {gorilla}.")
         public String datasetName = "";
+        /**
+         * Path to serialize model to.
+         */
         @Option(charName = 'f', longName = "output-path", usage = "Path to serialize model to.")
         public Path outputPath;
+        /**
+         * Number of SGD epochs.
+         */
         @Option(charName = 'i', longName = "epochs", usage = "Number of SGD epochs.")
         public int epochs = 5;
+        /**
+         * Print out feature, label and other model details.
+         */
         @Option(charName = 'o', longName = "print-model", usage = "Print out feature, label and other model details.")
         public boolean logModel = false;
+        /**
+         * Log the objective after n examples.
+         */
         @Option(charName = 'p', longName = "logging-interval", usage = "Log the objective after <int> examples.")
         public int loggingInterval = 100;
+        /**
+         * RNG seed.
+         */
         @Option(charName = 'r', longName = "seed", usage = "RNG seed.")
         public long seed = 1;
+        /**
+         * Shuffle the data each epoch (default: true).
+         */
         @Option(longName = "shuffle", usage = "Shuffle the data each epoch (default: true).")
         public boolean shuffle = true;
+        /**
+         * Path to a serialised SequenceDataset used for training.
+         */
         @Option(charName = 'u', longName = "train-dataset", usage = "Path to a serialised SequenceDataset used for training.")
         public Path trainDataset = null;
+        /**
+         * Path to a serialised SequenceDataset used for testing.
+         */
         @Option(charName = 'v', longName = "test-dataset", usage = "Path to a serialised SequenceDataset used for testing.")
         public Path testDataset = null;
+        /**
+         * Hash the model during training. Defaults to no hashing.
+         */
         @Option(longName = "model-hashing-algorithm", usage = "Hash the model during training. Defaults to no hashing.")
         public ModelHashingType modelHashingAlgorithm = ModelHashingType.NONE;
+        /**
+         * Salt for hashing the model.
+         */
         @Option(longName = "model-hashing-salt", usage = "Salt for hashing the model.")
         public String modelHashingSalt = "";
     }
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/kernel/KernelSVMOptions.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/kernel/KernelSVMOptions.java
index d2756f650..c321d81cb 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/kernel/KernelSVMOptions.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/kernel/KernelSVMOptions.java
@@ -63,22 +63,46 @@ public enum KernelEnum {
         RBF
     }
 
+    /**
+     * Intercept in kernel function. Defaults to 1.0.
+     */
     @Option(longName = "kernel-intercept", usage = "Intercept in kernel function. Defaults to 1.0.")
     public double kernelIntercept = 1.0;
+    /**
+     * Degree in polynomial kernel function. Defaults to 1.0.
+     */
     @Option(longName = "kernel-degree", usage = "Degree in polynomial kernel function. Defaults to 1.0.")
     public double kernelDegree = 1.0;
+    /**
+     * Gamma value in kernel function. Defaults to 1.0.
+     */
     @Option(longName = "kernel-gamma", usage = "Gamma value in kernel function. Defaults to 1.0.")
     public double kernelGamma = 1.0;
+    /**
+     * Number of SGD epochs. Defaults to 5.
+     */
     @Option(longName = "kernel-epochs", usage = "Number of SGD epochs. Defaults to 5.")
     public int kernelEpochs = 5;
+    /**
+     * Kernel function. Defaults to LINEAR.
+     */
     @Option(longName = "kernel-kernel", usage = "Kernel function. Defaults to LINEAR.")
     public KernelEnum kernelKernel = KernelEnum.LINEAR; //TODO should the default be KernelEnum.RBF?
+    /**
+     * Lambda value in gradient optimisation. Defaults to 0.01.
+     */
     @Option(longName = "kernel-lambda", usage = "Lambda value in gradient optimisation. Defaults to 0.01.")
     public double kernelLambda = 0.01;
+    /**
+     * Log the objective after n examples. Defaults to 100.
+     */
     @Option(longName = "kernel-logging-interval", usage = "Log the objective after <int> examples. Defaults to 100.")
     public int kernelLoggingInterval = 100;
+    /**
+     * Sets the random seed for the Kernel SVM.
+     */
     @Option(longName = "kernel-seed", usage = "Sets the random seed for the Kernel SVM.")
-    private long kernelSeed = Trainer.DEFAULT_SEED;
+    public long kernelSeed = Trainer.DEFAULT_SEED;
 
     @Override
     public KernelSVMTrainer getTrainer() {
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/kernel/TrainTest.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/kernel/TrainTest.java
index c5ade46ba..3b5af0a6e 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/kernel/TrainTest.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/kernel/TrainTest.java
@@ -43,11 +43,18 @@ public String getOptionsDescription() {
             return "Trains and tests a Kernel SVM model using SGD on the specified datasets.";
         }
 
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
+        /**
+         * The Kernel SVM SGD options.
+         */
         public KernelSVMOptions trainerOptions;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
      * @throws IOException if there is any error reading the examples.
      */
diff --git a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/linear/LinearSGDOptions.java b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/linear/LinearSGDOptions.java
index df77e4e35..7d37be672 100644
--- a/Classification/SGD/src/main/java/org/tribuo/classification/sgd/linear/LinearSGDOptions.java
+++ b/Classification/SGD/src/main/java/org/tribuo/classification/sgd/linear/LinearSGDOptions.java
@@ -36,23 +36,51 @@ public class LinearSGDOptions implements ClassificationOptions<LinearSGDTrainer>
     /**
      * Available loss types.
      */
-    public enum LossEnum {HINGE, LOG}
+    public enum LossEnum {
+        /**
+         * Hinge loss.
+         */
+        HINGE,
+        /**
+         * Log loss, i.e., cross-entropy.
+         */
+        LOG
+    }
 
+    /**
+     * The gradient descent optimiser options.
+     */
     public GradientOptimiserOptions sgoOptions;
 
+    /**
+     * Number of SGD epochs. Defaults to 5.
+     */
     @Option(longName = "sgd-epochs", usage = "Number of SGD epochs. Defaults to 5.")
     public int sgdEpochs = 5;
+    /**
+     * Loss function. Defaults to LOG.
+     */
     @Option(longName = "sgd-objective", usage = "Loss function. Defaults to LOG.")
     public LossEnum sgdObjective = LossEnum.LOG;
+    /**
+     * Log the objective after n examples. Defaults to 100.
+     */
     @Option(longName = "sgd-logging-interval", usage = "Log the objective after <int> examples. Defaults to 100.")
     public int sgdLoggingInterval = 100;
+    /**
+     * Minibatch size. Defaults to 1.
+     */
     @Option(longName = "sgd-minibatch-size", usage = "Minibatch size. Defaults to 1.")
     public int sgdMinibatchSize = 1;
+    /**
+     * Sets the random seed for the LinearSGDTrainer.
+     */
     @Option(longName = "sgd-seed", usage = "Sets the random seed for the LinearSGDTrainer.")
-    private long sgdSeed = Trainer.DEFAULT_SEED;
+    public long sgdSeed = Trainer.DEFAULT_SEED;
 
     /**
      * Returns the loss function specified in the arguments.
+     *
      * @return The loss function.
      */
     public LabelObjective getLoss() {
diff --git a/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/TrainTest.java b/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/TrainTest.java
index 6938a6006..8382ddf7b 100644
--- a/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/TrainTest.java
+++ b/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/TrainTest.java
@@ -44,13 +44,20 @@ public String getOptionsDescription() {
             return "Trains and tests an XGBoost classification model on the specified datasets.";
         }
 
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
+        /**
+         * The XGBoost trainer options.
+         */
         public XGBoostOptions xgboostOptions;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
-     * @throws java.io.IOException if there is any error reading the examples.
+     * @throws IOException if there is any error reading the examples.
      */
     public static void main(String[] args) throws IOException {
         TrainTestOptions o = new TrainTestOptions();
diff --git a/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostClassificationConverter.java b/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostClassificationConverter.java
index 862f2cc3e..1abc94786 100644
--- a/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostClassificationConverter.java
+++ b/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostClassificationConverter.java
@@ -32,6 +32,9 @@
 public final class XGBoostClassificationConverter implements XGBoostOutputConverter<Label> {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * Constructs an XGBoostClassificationConverter.
+     */
     public XGBoostClassificationConverter() {}
 
     @Override
diff --git a/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostClassificationTrainer.java b/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostClassificationTrainer.java
index 69c57b9b2..e24700ae2 100644
--- a/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostClassificationTrainer.java
+++ b/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostClassificationTrainer.java
@@ -73,10 +73,22 @@ public final class XGBoostClassificationTrainer extends XGBoostTrainer<Label> {
     @Config(description="Evaluation metric to use. The default value is set based on the objective function, so this can be usually left blank.")
     private String evalMetric = "";
 
+    /**
+     * Create an XGBoost trainer.
+     *
+     * @param numTrees Number of trees to boost.
+     */
     public XGBoostClassificationTrainer(int numTrees) {
         this(numTrees, 0.3, 0, 6, 1, 1, 1, 1, 0, 4, true, Trainer.DEFAULT_SEED);
     }
 
+    /**
+     * Create an XGBoost trainer.
+     *
+     * @param numTrees Number of trees to boost.
+     * @param numThreads Number of threads to use.
+     * @param silent Silence the training output text.
+     */
     public XGBoostClassificationTrainer(int numTrees, int numThreads, boolean silent) {
         this(numTrees, 0.3, 0, 6, 1, 1, 1, 1, 0, numThreads, silent, Trainer.DEFAULT_SEED);
     }
diff --git a/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostOptions.java b/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostOptions.java
index 79661c22e..e250337d6 100644
--- a/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostOptions.java
+++ b/Classification/XGBoost/src/main/java/org/tribuo/classification/xgboost/XGBoostOptions.java
@@ -30,32 +30,74 @@
 public class XGBoostOptions implements ClassificationOptions<XGBoostClassificationTrainer> {
     private static final Logger logger = Logger.getLogger(XGBoostOptions.class.getName());
 
+    /**
+     * Weak learning algorithm.
+     */
     @Option(longName = "xgb-booster-type", usage = "Weak learning algorithm.")
     public XGBoostTrainer.BoosterType xgbBoosterType = XGBoostTrainer.BoosterType.GBTREE;
+    /**
+     * Tree building algorithm.
+     */
     @Option(longName = "xgb-tree-method", usage = "Tree building algorithm.")
     public XGBoostTrainer.TreeMethod xgbTreeMethod = XGBoostTrainer.TreeMethod.AUTO;
+    /**
+     * Number of trees in the ensemble.
+     */
     @Option(longName = "xgb-ensemble-size", usage = "Number of trees in the ensemble.")
     public int xgbEnsembleSize = -1;
+    /**
+     * L1 regularization term for weights.
+     */
     @Option(longName = "xgb-alpha", usage = "L1 regularization term for weights.")
     public float xbgAlpha = 0.0f;
+    /**
+     * Minimum sum of instance weights needed in a leaf (range [0,Infinity]).
+     */
     @Option(longName = "xgb-min-weight", usage = "Minimum sum of instance weights needed in a leaf (range [0,Infinity]).")
     public float xgbMinWeight = 1;
+    /**
+     * Max tree depth (range (0,Integer.MAX_VALUE]).
+     */
     @Option(longName = "xgb-max-depth", usage = "Max tree depth (range (0,Integer.MAX_VALUE]).")
     public int xgbMaxDepth = 6;
+    /**
+     * Step size shrinkage parameter (range [0,1]).
+     */
     @Option(longName = "xgb-eta", usage = "Step size shrinkage parameter (range [0,1]).")
     public float xgbEta = 0.3f;
+    /**
+     * Subsample features for each tree (range (0,1]).
+     */
     @Option(longName = "xgb-subsample-features", usage = "Subsample features for each tree (range (0,1]).")
     public float xgbSubsampleFeatures = 0.0f;
+    /**
+     * Minimum loss reduction to make a split (range [0,Infinity]).
+     */
     @Option(longName = "xgb-gamma", usage = "Minimum loss reduction to make a split (range [0,Infinity]).")
     public float xgbGamma = 0.0f;
+    /**
+     * L2 regularization term for weights.
+     */
     @Option(longName = "xgb-lambda", usage = "L2 regularization term for weights.")
     public float xgbLambda = 1.0f;
+    /**
+     * Deprecated, use xgb-loglevel.
+     */
     @Option(longName = "xgb-quiet", usage = "Deprecated, use xgb-loglevel.")
     public boolean xgbQuiet;
+    /**
+     * Make the XGBoost training procedure quiet.
+     */
     @Option(longName = "xgb-loglevel", usage = "Make the XGBoost training procedure quiet.")
     public XGBoostTrainer.LoggingVerbosity xgbLogLevel = XGBoostTrainer.LoggingVerbosity.WARNING;
+    /**
+     * Subsample size for each tree (range (0,1]).
+     */
     @Option(longName = "xgb-subsample", usage = "Subsample size for each tree (range (0,1]).")
     public float xgbSubsample = 1.0f;
+    /**
+     * Number of threads to use (range (1, num hw threads)). The default of 0 means use all hw threads.
+     */
     @Option(longName = "xgb-num-threads", usage = "Number of threads to use (range (1, num hw threads)). The default of 0 means use all hw threads.")
     public int xgbNumThreads = 0;
     @Option(longName = "xgb-seed", usage = "Sets the random seed for XGBoost.")
@@ -67,7 +109,7 @@ public XGBoostClassificationTrainer getTrainer() {
             throw new IllegalArgumentException("Please supply the number of trees.");
         }
         if (xgbQuiet) {
-            logger.log(Level.WARNING,"Silencing XGBoost, overriding logging verbosity. Please switch to the 'xgb-loglevel' argument.");
+            logger.log(Level.WARNING, "Silencing XGBoost, overriding logging verbosity. Please switch to the 'xgb-loglevel' argument.");
             xgbLogLevel = XGBoostTrainer.LoggingVerbosity.SILENT;
         }
         return new XGBoostClassificationTrainer(xgbBoosterType, xgbTreeMethod, xgbEnsembleSize, xgbEta, xgbGamma, xgbMaxDepth, xgbMinWeight, xgbSubsample, xgbSubsampleFeatures, xgbLambda, xbgAlpha, xgbNumThreads, xgbLogLevel, xgbSeed);
diff --git a/Clustering/Core/src/main/java/org/tribuo/clustering/ClusterID.java b/Clustering/Core/src/main/java/org/tribuo/clustering/ClusterID.java
index cff14990a..1189ec713 100644
--- a/Clustering/Core/src/main/java/org/tribuo/clustering/ClusterID.java
+++ b/Clustering/Core/src/main/java/org/tribuo/clustering/ClusterID.java
@@ -34,6 +34,9 @@
 public class ClusterID implements Output<ClusterID> {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * The sentinel unassigned cluster ID.
+     */
     public static final int UNASSIGNED = -1;
 
     private final int id;
diff --git a/Clustering/Core/src/main/java/org/tribuo/clustering/ClusteringFactory.java b/Clustering/Core/src/main/java/org/tribuo/clustering/ClusteringFactory.java
index 8b46424b7..1ed819208 100644
--- a/Clustering/Core/src/main/java/org/tribuo/clustering/ClusteringFactory.java
+++ b/Clustering/Core/src/main/java/org/tribuo/clustering/ClusteringFactory.java
@@ -37,6 +37,9 @@
 public final class ClusteringFactory implements OutputFactory<ClusterID> {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * The sentinel unassigned cluster id, used when there is no ground truth clustering.
+     */
     public static final ClusterID UNASSIGNED_CLUSTER_ID = new ClusterID(ClusterID.UNASSIGNED);
 
     private static final ClusteringFactoryProvenance provenance = new ClusteringFactoryProvenance();
diff --git a/Clustering/Core/src/main/java/org/tribuo/clustering/ImmutableClusteringInfo.java b/Clustering/Core/src/main/java/org/tribuo/clustering/ImmutableClusteringInfo.java
index 082f2524f..c03e7e4fc 100644
--- a/Clustering/Core/src/main/java/org/tribuo/clustering/ImmutableClusteringInfo.java
+++ b/Clustering/Core/src/main/java/org/tribuo/clustering/ImmutableClusteringInfo.java
@@ -38,6 +38,10 @@ public class ImmutableClusteringInfo extends ClusteringInfo implements Immutable
 
     private final Set<ClusterID> domain;
 
+    /**
+     * Constructs an immutable clustering info from the supplied cluster counts.
+     * @param counts The cluster counts.
+     */
     public ImmutableClusteringInfo(Map<Integer,MutableLong> counts) {
         super();
         clusterCounts.putAll(MutableNumber.copyMap(counts));
@@ -49,6 +53,10 @@ public ImmutableClusteringInfo(Map<Integer,MutableLong> counts) {
         domain = Collections.unmodifiableSet(outputs);
     }
 
+    /**
+     * Copies the supplied clustering info, generating id numbers.
+     * @param other The clustering info to copy.
+     */
     public ImmutableClusteringInfo(ClusteringInfo other) {
         super(other);
         Set<ClusterID> outputs = new HashSet<>();
diff --git a/Clustering/Core/src/main/java/org/tribuo/clustering/evaluation/ClusteringMetric.java b/Clustering/Core/src/main/java/org/tribuo/clustering/evaluation/ClusteringMetric.java
index 684491b0d..996405c07 100644
--- a/Clustering/Core/src/main/java/org/tribuo/clustering/evaluation/ClusteringMetric.java
+++ b/Clustering/Core/src/main/java/org/tribuo/clustering/evaluation/ClusteringMetric.java
@@ -38,6 +38,12 @@ public class ClusteringMetric implements EvaluationMetric<ClusterID, ClusteringM
     private final String name;
     private final BiFunction<MetricTarget<ClusterID>, Context, Double> impl;
 
+    /**
+     * Constructs a clustering metric using the supplied parameters.
+     * @param target The metric target.
+     * @param name The metric name.
+     * @param impl The implementation function.
+     */
     public ClusteringMetric(MetricTarget<ClusterID> target, String name, BiFunction<MetricTarget<ClusterID>, Context, Double> impl) {
         this.target = target;
         this.name = name;
@@ -92,10 +98,18 @@ static final class Context extends MetricContext<ClusterID> {
             }
         }
 
+        /**
+         * Gets the predicted cluster ids.
+         * @return The predicted cluster ids.
+         */
         public ArrayList<Integer> getPredictedIDs() {
             return predictedIDs;
         }
 
+        /**
+         * Gets the ground truth cluster ids.
+         * @return The ground truth cluster ids.
+         */
         public ArrayList<Integer> getTrueIDs() {
             return trueIDs;
         }
diff --git a/Clustering/Core/src/main/java/org/tribuo/clustering/evaluation/ClusteringMetrics.java b/Clustering/Core/src/main/java/org/tribuo/clustering/evaluation/ClusteringMetrics.java
index ffcfbb7fc..a452dc970 100644
--- a/Clustering/Core/src/main/java/org/tribuo/clustering/evaluation/ClusteringMetrics.java
+++ b/Clustering/Core/src/main/java/org/tribuo/clustering/evaluation/ClusteringMetrics.java
@@ -47,10 +47,19 @@ public enum ClusteringMetrics {
         this.impl = impl;
     }
 
+    /**
+     * Gets the implementing function for this metric.
+     * @return The implementing function.
+     */
     public BiFunction<MetricTarget<ClusterID>, ClusteringMetric.Context, Double> getImpl() {
         return impl;
     }
 
+    /**
+     * Constructs the metric for the specified metric target.
+     * @param tgt The metric target.
+     * @return The clustering metric for that target.
+     */
     public ClusteringMetric forTarget(MetricTarget<ClusterID> tgt) {
         return new ClusteringMetric(tgt, this.name(), this.getImpl());
     }
diff --git a/Clustering/Core/src/main/java/org/tribuo/clustering/example/ClusteringDataGenerator.java b/Clustering/Core/src/main/java/org/tribuo/clustering/example/ClusteringDataGenerator.java
index 015a69c21..a6b59d4bc 100644
--- a/Clustering/Core/src/main/java/org/tribuo/clustering/example/ClusteringDataGenerator.java
+++ b/Clustering/Core/src/main/java/org/tribuo/clustering/example/ClusteringDataGenerator.java
@@ -91,6 +91,12 @@ public static Dataset<ClusterID> gaussianClusters(long size, long seed) {
         return new MutableDataset<>(new ListDataSource<>(trainingData,clusteringFactory,trainingProvenance));
     }
 
+    /**
+     * Generates a train/test dataset pair which is dense in the features,
+     * each example has 4 features,{A,B,C,D}, and there are 4 clusters,
+     * {0,1,2,3}.
+     * @return A pair of datasets.
+     */
     public static Pair<Dataset<ClusterID>,Dataset<ClusterID>> denseTrainTest() {
         return denseTrainTest(-1.0);
     }
@@ -148,6 +154,12 @@ public static Pair<Dataset<ClusterID>,Dataset<ClusterID>> denseTrainTest(double
         return new Pair<>(train,test);
     }
 
+    /**
+     * Generates a pair of datasets, where the features are sparse,
+     * and unknown features appear in the test data. It has the same
+     * 4 clusters {0,1,2,3}.
+     * @return A pair of datasets.
+     */
     public static Pair<Dataset<ClusterID>,Dataset<ClusterID>> sparseTrainTest() {
         return sparseTrainTest(-1.0);
     }
diff --git a/Clustering/KMeans/src/main/java/org/tribuo/clustering/kmeans/KMeansOptions.java b/Clustering/KMeans/src/main/java/org/tribuo/clustering/kmeans/KMeansOptions.java
index 6807bf7ba..977e9fe84 100644
--- a/Clustering/KMeans/src/main/java/org/tribuo/clustering/kmeans/KMeansOptions.java
+++ b/Clustering/KMeans/src/main/java/org/tribuo/clustering/kmeans/KMeansOptions.java
@@ -30,22 +30,41 @@
 public class KMeansOptions implements Options {
     private static final Logger logger = Logger.getLogger(KMeansOptions.class.getName());
 
-    @Option(longName="kmeans-interations",usage="Iterations of the k-means algorithm. Defaults to 10.")
+    /**
+     * Iterations of the k-means algorithm. Defaults to 10.
+     */
+    @Option(longName = "kmeans-interations", usage = "Iterations of the k-means algorithm. Defaults to 10.")
     public int iterations = 10;
-    @Option(longName="kmeans-num-centroids",usage="Number of centroids in K-Means. Defaults to 10.")
+    /**
+     * Number of centroids in K-Means. Defaults to 10.
+     */
+    @Option(longName = "kmeans-num-centroids", usage = "Number of centroids in K-Means. Defaults to 10.")
     public int centroids = 10;
-    @Option(longName="kmeans-distance",usage="Distance function in K-Means. Defaults to EUCLIDEAN.")
+    /**
+     * Distance function in K-Means. Defaults to EUCLIDEAN.
+     */
+    @Option(longName = "kmeans-distance", usage = "Distance function in K-Means. Defaults to EUCLIDEAN.")
     public Distance distance = Distance.EUCLIDEAN;
-    @Option(longName="kmeans-initialisation",usage="Initialisation function in K-Means. Defaults to RANDOM.")
+    /**
+     * Initialisation function in K-Means. Defaults to RANDOM.
+     */
+    @Option(longName = "kmeans-initialisation", usage = "Initialisation function in K-Means. Defaults to RANDOM.")
     public Initialisation initialisation = Initialisation.RANDOM;
-    @Option(longName="kmeans-num-threads",usage="Number of computation threads in K-Means. Defaults to 4.")
+    /**
+     * Number of computation threads in K-Means. Defaults to 4.
+     */
+    @Option(longName = "kmeans-num-threads", usage = "Number of computation threads in K-Means. Defaults to 4.")
     public int numThreads = 4;
-    @Option(longName="kmeans-seed", usage = "Sets the random seed for K-Means.")
+    @Option(longName = "kmeans-seed", usage = "Sets the random seed for K-Means.")
     private long seed = Trainer.DEFAULT_SEED;
 
+    /**
+     * Gets the configured KMeansTrainer using the options in this object.
+     * @return A KMeansTrainer.
+     */
     public KMeansTrainer getTrainer() {
         logger.info("Configuring K-Means Trainer");
         //public KMeansTrainer(int centroids, int iterations, Distance distanceType, int numThreads, int seed) {
-        return new KMeansTrainer(centroids,iterations,distance,initialisation,numThreads,seed);
+        return new KMeansTrainer(centroids, iterations, distance, initialisation, numThreads, seed);
     }
 }
diff --git a/Clustering/KMeans/src/main/java/org/tribuo/clustering/kmeans/TrainTest.java b/Clustering/KMeans/src/main/java/org/tribuo/clustering/kmeans/TrainTest.java
index 4358895e0..df557af43 100644
--- a/Clustering/KMeans/src/main/java/org/tribuo/clustering/kmeans/TrainTest.java
+++ b/Clustering/KMeans/src/main/java/org/tribuo/clustering/kmeans/TrainTest.java
@@ -50,21 +50,41 @@ public static class KMeansOptions implements Options {
         public String getOptionsDescription() {
             return "Trains and evaluates a K-Means model on the specified dataset.";
         }
+
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
 
-        @Option(charName='n',longName="num-clusters",usage="Number of clusters to infer.")
+        /**
+         * Number of clusters to infer.
+         */
+        @Option(charName = 'n', longName = "num-clusters", usage = "Number of clusters to infer.")
         public int centroids = 5;
-        @Option(charName='i',longName="iterations",usage="Maximum number of iterations.")
+        /**
+         * Maximum number of iterations.
+         */
+        @Option(charName = 'i', longName = "iterations", usage = "Maximum number of iterations.")
         public int iterations = 10;
-        @Option(charName='d',longName="distance",usage="Distance function to use in the e step.")
+        /**
+         * Distance function to use in the e step.
+         */
+        @Option(charName = 'd', longName = "distance", usage = "Distance function to use in the e step.")
         public Distance distance = Distance.EUCLIDEAN;
-        @Option(charName='s',longName="initialisation",usage="Type of initialisation to use for centroids.")
+        /**
+         * Type of initialisation to use for centroids.
+         */
+        @Option(charName = 's', longName = "initialisation", usage = "Type of initialisation to use for centroids.")
         public Initialisation initialisation = Initialisation.RANDOM;
-        @Option(charName='t',longName="num-threads",usage="Number of threads to use (range (1, num hw threads)).")
+        /**
+         * Number of threads to use (range (1, num hw threads)).
+         */
+        @Option(charName = 't', longName = "num-threads", usage = "Number of threads to use (range (1, num hw threads)).")
         public int numThreads = 4;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
      * @throws IOException if there is any error reading the examples.
      */
diff --git a/Common/LibLinear/src/main/java/org/tribuo/common/liblinear/LibLinearModel.java b/Common/LibLinear/src/main/java/org/tribuo/common/liblinear/LibLinearModel.java
index 68eca9778..8b9a6e004 100644
--- a/Common/LibLinear/src/main/java/org/tribuo/common/liblinear/LibLinearModel.java
+++ b/Common/LibLinear/src/main/java/org/tribuo/common/liblinear/LibLinearModel.java
@@ -100,7 +100,6 @@ public List<de.bwaldvogel.liblinear.Model> getInnerModels() {
      * LibLinear model.
      * <p>
      * Prefer {@link LibLinearModel#getExcuses} to get multiple excuses.
-     * <p>
      * @param e The example to excuse.
      * @return An {@link Excuse} for this example.
      */
diff --git a/Common/LibSVM/src/main/java/org/tribuo/common/libsvm/LibSVMModel.java b/Common/LibSVM/src/main/java/org/tribuo/common/libsvm/LibSVMModel.java
index ba83b0cea..9601fa20f 100644
--- a/Common/LibSVM/src/main/java/org/tribuo/common/libsvm/LibSVMModel.java
+++ b/Common/LibSVM/src/main/java/org/tribuo/common/libsvm/LibSVMModel.java
@@ -85,7 +85,6 @@ protected LibSVMModel(String name, ModelProvenance description, ImmutableFeature
 
     /**
      * Returns an unmodifiable copy of the underlying list of libsvm models.
-     * <p>
      * @deprecated Deprecated to unify the names across LibLinear, LibSVM and XGBoost.
      * @return The underlying model list.
      */
diff --git a/Common/LibSVM/src/main/java/org/tribuo/common/libsvm/SVMParameters.java b/Common/LibSVM/src/main/java/org/tribuo/common/libsvm/SVMParameters.java
index 36283c707..6956a6566 100644
--- a/Common/LibSVM/src/main/java/org/tribuo/common/libsvm/SVMParameters.java
+++ b/Common/LibSVM/src/main/java/org/tribuo/common/libsvm/SVMParameters.java
@@ -37,6 +37,11 @@ public class SVMParameters<T extends Output<T>> implements Serializable {
     
     protected final svm_parameter parameters = new svm_parameter();
 
+    /**
+     * Constructs the default SVMParameters using the specified SVMType and KernelType.
+     * @param svmType The SVM algorithm.
+     * @param kernelType The kernel.
+     */
     public SVMParameters(SVMType<T> svmType, KernelType kernelType) {
         this.svmType = svmType;
         this.kernelType = kernelType;
@@ -60,14 +65,26 @@ public SVMParameters(SVMType<T> svmType, KernelType kernelType) {
         parameters.weight = new double[0];
     }
 
+    /**
+     * Gets the SVM type.
+     * @return The SVM type.
+     */
     public SVMType<T> getSvmType() {
         return svmType;
     }
 
+    /**
+     * Gets the kernel type.
+     * @return The kernel type.
+     */
     public KernelType getKernelType() {
         return kernelType;
     }
 
+    /**
+     * Gets the underlying SVM parameter object.
+     * @return The SVM parameters.
+     */
     public svm_parameter getParameters() {
         return parameters;
     }
@@ -83,14 +100,22 @@ public String toString() {
     public void setProbability() {
         parameters.probability = 1;
     }
-    
+
+    /**
+     * Sets the cost for C_SVC.
+     * @param c The cost.
+     */
     public void setCost(double c) {
         if(svmType.isNu() || !svmType.isClassification()) {
             logger.warning(String.format("Setting cost %f for non-C_SVC model", c));
         }
         parameters.C = c;
     }
-    
+
+    /**
+     * Sets the value of nu for NU_SVM.
+     * @param nu The nu.
+     */
     public void setNu(double nu) {
         if(!svmType.isNu()) {
             logger.warning(String.format("Setting nu %f for non-NU_SVM model", nu));
@@ -98,26 +123,50 @@ public void setNu(double nu) {
         parameters.nu = nu;
     }
 
+    /**
+     * Sets the coefficient.
+     * @param coeff The coefficient.
+     */
     public void setCoeff(double coeff) {
         parameters.coef0 = coeff;
     }
 
+    /**
+     * Sets the termination closeness.
+     * @param epsilon The termination criterion.
+     */
     public void setEpsilon(double epsilon) {
         parameters.p = epsilon;
     }
 
+    /**
+     * Sets the degree of the polynomial kernel.
+     * @param degree The polynomial degree.
+     */
     public void setDegree(int degree) {
         parameters.degree = degree;
     }
 
+    /**
+     * Sets gamma in the RBF kernel.
+     * @param gamma The gamma.
+     */
     public void setGamma(double gamma) {
         parameters.gamma = gamma;
     }
-    
+
+    /**
+     * Gets the gamma value.
+     * @return The gamma value.
+     */
     public double getGamma() {
         return parameters.gamma;
     }
-    
+
+    /**
+     * Sets the cache size.
+     * @param cacheMB The cache size.
+     */
     public void setCacheSize(double cacheMB) {
         if(cacheMB <= 0) {
             throw new IllegalArgumentException("Cache must be larger than 0MB");
diff --git a/Common/NearestNeighbour/src/main/java/org/tribuo/common/nearest/KNNClassifierOptions.java b/Common/NearestNeighbour/src/main/java/org/tribuo/common/nearest/KNNClassifierOptions.java
index 812677e08..6949283b8 100644
--- a/Common/NearestNeighbour/src/main/java/org/tribuo/common/nearest/KNNClassifierOptions.java
+++ b/Common/NearestNeighbour/src/main/java/org/tribuo/common/nearest/KNNClassifierOptions.java
@@ -34,17 +34,41 @@ public class KNNClassifierOptions implements ClassificationOptions<KNNTrainer<La
     /**
      * The type of combination function.
      */
-    public enum EnsembleCombinerType { VOTING, FULLY_WEIGHTED_VOTING}
+    public enum EnsembleCombinerType {
+        /**
+         * Use a {@link VotingCombiner}.
+         */
+        VOTING,
+        /**
+         * Use a {@link FullyWeightedVotingCombiner}.
+         */
+        FULLY_WEIGHTED_VOTING
+    }
 
-    @Option(longName="knn-k",usage="K nearest neighbours to use. Defaults to 1.")
+    /**
+     * K nearest neighbours to use. Defaults to 1.
+     */
+    @Option(longName = "knn-k", usage = "K nearest neighbours to use.")
     public int knnK = 1;
-    @Option(longName="knn-num-threads",usage="Number of threads to use. Defaults to 1.")
+    /**
+     * Number of threads to use. Defaults to 1.
+     */
+    @Option(longName = "knn-num-threads", usage = "Number of threads to use.")
     public int knnNumThreads = 1;
-    @Option(longName="knn-distance",usage="Distance metric to use. Defaults to L2.")
+    /**
+     * Distance metric to use. Defaults to L2.
+     */
+    @Option(longName = "knn-distance", usage = "Distance metric to use.")
     public Distance knnDistance = Distance.L2;
-    @Option(longName="knn-backend",usage="Parallel backend to use.")
+    /**
+     * Parallel backend to use.
+     */
+    @Option(longName = "knn-backend", usage = "Parallel backend to use.")
     public Backend knnBackend = Backend.STREAMS;
-    @Option(longName="knn-voting",usage="Parallel backend to use.")
+    /**
+     * Parallel backend to use.
+     */
+    @Option(longName = "knn-voting", usage = "Parallel backend to use.")
     public EnsembleCombinerType knnEnsembleCombiner = EnsembleCombinerType.VOTING;
 
     @Override
@@ -65,6 +89,6 @@ private EnsembleCombiner<Label> getEnsembleCombiner() {
 
     @Override
     public KNNTrainer<Label> getTrainer() {
-    	return new KNNTrainer<>(knnK, knnDistance, knnNumThreads, getEnsembleCombiner(), knnBackend);
+        return new KNNTrainer<>(knnK, knnDistance, knnNumThreads, getEnsembleCombiner(), knnBackend);
     }
 }
diff --git a/Common/Trees/src/main/java/org/tribuo/common/tree/AbstractCARTTrainer.java b/Common/Trees/src/main/java/org/tribuo/common/tree/AbstractCARTTrainer.java
index d5847a863..5ea49c3fc 100644
--- a/Common/Trees/src/main/java/org/tribuo/common/tree/AbstractCARTTrainer.java
+++ b/Common/Trees/src/main/java/org/tribuo/common/tree/AbstractCARTTrainer.java
@@ -91,7 +91,6 @@ public abstract class AbstractCARTTrainer<T extends Output<T>> implements Decisi
 
     protected int trainInvocationCounter;
 
-
     /**
      * After calls to this superconstructor subclasses must call postConfig().
      * @param maxDepth The maximum depth of the tree.
@@ -223,6 +222,12 @@ public TreeModel<T> train(Dataset<T> examples, Map<String, Provenance> runProven
         return new TreeModel<>("cart-tree", provenance, featureIDMap, outputIDInfo, false, root.convertTree());
     }
 
+    /**
+     * Makes the initial training node.
+     * @param examples The dataset to use.
+     * @param leafDeterminer The leaf determination function.
+     * @return The initial training node.
+     */
     protected abstract AbstractTrainingNode<T> mkTrainingNode(Dataset<T> examples,
                                                               AbstractTrainingNode.LeafDeterminer leafDeterminer);
 
@@ -233,10 +238,19 @@ protected abstract AbstractTrainingNode<T> mkTrainingNode(Dataset<T> examples,
     protected static abstract class AbstractCARTTrainerProvenance extends SkeletalTrainerProvenance {
         private static final long serialVersionUID = 1L;
 
+        /**
+         * Constructs a provenance for the host AbstractCARTTrainer.
+         * @param host The host trainer.
+         * @param <T> The trainer type.
+         */
         protected <T extends Output<T>> AbstractCARTTrainerProvenance(AbstractCARTTrainer<T> host) {
             super(host);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenance map.
+         */
         protected AbstractCARTTrainerProvenance(Map<String,Provenance> map) {
             super(map);
         }
diff --git a/Common/Trees/src/main/java/org/tribuo/common/tree/AbstractTrainingNode.java b/Common/Trees/src/main/java/org/tribuo/common/tree/AbstractTrainingNode.java
index 7e81d3e36..e766ac91b 100644
--- a/Common/Trees/src/main/java/org/tribuo/common/tree/AbstractTrainingNode.java
+++ b/Common/Trees/src/main/java/org/tribuo/common/tree/AbstractTrainingNode.java
@@ -160,25 +160,44 @@ public Node<T> copy() {
     /**
      * Contains parameters needed to determine whether a node is a leaf.
      */
+    // Will be a record one day.
     public static class LeafDeterminer {
         private final int maxDepth;
         private final float minChildWeight;
         private final float scaledMinImpurityDecrease;
 
+        /**
+         * Constructs a leaf determiner using the supplied parameters.
+         * @param maxDepth The maximum tree depth.
+         * @param minChildWeight The minimum example weight of each child node.
+         * @param scaledMinImpurityDecrease  The scaled minimum impurity decrease necessary to split a node.
+         */
         public LeafDeterminer(int maxDepth, float minChildWeight, float scaledMinImpurityDecrease) {
             this.maxDepth = maxDepth;
             this.minChildWeight = minChildWeight;
             this.scaledMinImpurityDecrease = scaledMinImpurityDecrease;
         }
 
+        /**
+         * Gets the maximum tree depth.
+         * @return The maximum tree depth.
+         */
         public int getMaxDepth() {
             return maxDepth;
         }
 
+        /**
+         * Gets the minimum example weight of a child node.
+         * @return The mimimum weight of a child node.
+         */
         public float getMinChildWeight() {
             return minChildWeight;
         }
 
+        /**
+         * Gets the minimum impurity decrease necessary to split a node.
+         * @return The minimum impurity decrease to split a node.
+         */
         public float getScaledMinImpurityDecrease() {
             return scaledMinImpurityDecrease;
         }
diff --git a/Common/Trees/src/main/java/org/tribuo/common/tree/impl/IntArrayContainer.java b/Common/Trees/src/main/java/org/tribuo/common/tree/impl/IntArrayContainer.java
index 49413b25e..a286e785e 100644
--- a/Common/Trees/src/main/java/org/tribuo/common/tree/impl/IntArrayContainer.java
+++ b/Common/Trees/src/main/java/org/tribuo/common/tree/impl/IntArrayContainer.java
@@ -27,7 +27,13 @@
 public class IntArrayContainer {
     private static final Logger logger = Logger.getLogger(IntArrayContainer.class.getName());
 
+    /**
+     * The array of ints.
+     */
     public int[] array;
+    /**
+     * The number of elements in the array.
+     */
     public int size;
 
     /**
diff --git a/Common/XGBoost/src/main/java/org/tribuo/common/xgboost/XGBoostTrainer.java b/Common/XGBoost/src/main/java/org/tribuo/common/xgboost/XGBoostTrainer.java
index 60a8c9062..a7158456e 100644
--- a/Common/XGBoost/src/main/java/org/tribuo/common/xgboost/XGBoostTrainer.java
+++ b/Common/XGBoost/src/main/java/org/tribuo/common/xgboost/XGBoostTrainer.java
@@ -108,6 +108,9 @@ public enum TreeMethod {
          */
         GPU_HIST("gpu_hist");
 
+        /**
+         * The parameter name used by the XGBoost native library.
+         */
         public final String paramName;
 
         TreeMethod(String paramName) {
@@ -119,11 +122,27 @@ public enum TreeMethod {
      * The logging verbosity of the native library.
      */
     public enum LoggingVerbosity {
+        /**
+         * No logging.
+         */
         SILENT(0),
+        /**
+         * Only warnings are logged.
+         */
         WARNING(1),
+        /**
+         * Tree building info is logged as well as warnings.
+         */
         INFO(2),
+        /**
+         * All the logging.
+         */
+        // You've never seen so much logging.
         DEBUG(3);
 
+        /**
+         * The log value used by the XGBoost native library.
+         */
         public final int value;
 
         LoggingVerbosity(int value) {
@@ -148,6 +167,9 @@ public enum BoosterType {
          */
         DART("dart");
 
+        /**
+         * The parameter value used by the XGBoost native library.
+         */
         public final String paramName;
 
         BoosterType(String paramName) {
@@ -208,10 +230,20 @@ public enum BoosterType {
 
     protected int trainInvocationCounter = 0;
 
+    /**
+     * Constructs an XGBoost trainer using the specified number of trees.
+     * @param numTrees The number of trees.
+     */
     protected XGBoostTrainer(int numTrees) {
         this(numTrees, 0.3, 0, 6, 1, 1, 1, 1, 0, 4, true, Trainer.DEFAULT_SEED);
     }
 
+    /**
+     * Constructs an XGBoost trainer using the specified number of trees.
+     * @param numTrees The number of trees.
+     * @param numThreads The number of training threads.
+     * @param silent Should the logging be silenced?
+     */
     protected XGBoostTrainer(int numTrees, int numThreads, boolean silent) {
         this(numTrees, 0.3, 0, 6, 1, 1, 1, 1, 0, numThreads, silent, Trainer.DEFAULT_SEED);
     }
diff --git a/Core/src/main/java/org/tribuo/Dataset.java b/Core/src/main/java/org/tribuo/Dataset.java
index 7ee45c2ec..f3a4b0e89 100644
--- a/Core/src/main/java/org/tribuo/Dataset.java
+++ b/Core/src/main/java/org/tribuo/Dataset.java
@@ -48,7 +48,6 @@
  * A class for sets of data, which are used to train and evaluate classifiers.
  * <p>
  * Subclass {@link MutableDataset} rather than this class.
- * <p>
  * @param <T> the type of the features in the data set.
  */
 public abstract class Dataset<T extends Output<T>> implements Iterable<Example<T>>, Provenancable<DatasetProvenance>, Serializable {
diff --git a/Core/src/main/java/org/tribuo/Model.java b/Core/src/main/java/org/tribuo/Model.java
index fb5ad0b68..ea14efe85 100644
--- a/Core/src/main/java/org/tribuo/Model.java
+++ b/Core/src/main/java/org/tribuo/Model.java
@@ -35,7 +35,6 @@
  * <p>
  * If two features map to the same id in the featureIDMap, then
  * occurrences of those features will be combined at prediction time.
- * <p>
  * @param <T> the type of prediction produced by the model.
  */
 public abstract class Model<T extends Output<T>> implements Provenancable<ModelProvenance>, Serializable {
diff --git a/Core/src/main/java/org/tribuo/ModelExplorer.java b/Core/src/main/java/org/tribuo/ModelExplorer.java
index 8644164f8..3d9e1eead 100644
--- a/Core/src/main/java/org/tribuo/ModelExplorer.java
+++ b/Core/src/main/java/org/tribuo/ModelExplorer.java
@@ -44,7 +44,7 @@ public class ModelExplorer implements CommandGroup {
     /**
      * The command shell instance.
      */
-    protected CommandInterpreter shell;
+    private final CommandInterpreter shell;
 
     private Model<?> model;
 
@@ -198,9 +198,9 @@ public String minCount(CommandInterpreter ci, int minCount) {
      */
     public static class ModelExplorerOptions implements Options {
         /**
-         * Model file to load.
+         * Model file to load. Optional.
          */
-        @Option(charName='f',longName="filename",usage="Model file to load. Optional.")
+        @Option(charName = 'f', longName = "filename", usage = "Model file to load. Optional.")
         public String modelFilename;
     }
 
diff --git a/Core/src/main/java/org/tribuo/dataset/DatasetView.java b/Core/src/main/java/org/tribuo/dataset/DatasetView.java
index d2f2f722e..d1b437782 100644
--- a/Core/src/main/java/org/tribuo/dataset/DatasetView.java
+++ b/Core/src/main/java/org/tribuo/dataset/DatasetView.java
@@ -395,6 +395,10 @@ <T extends Output<T>> DatasetViewProvenance(DatasetView<T> dataset, boolean stor
             this.indices = storeIndices ? dataset.indices : new int[0];
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public DatasetViewProvenance(Map<String,Provenance> map) {
             super(map);
             this.size = ObjectProvenance.checkAndExtractProvenance(map,SIZE,IntProvenance.class, DatasetViewProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/dataset/MinimumCardinalityDataset.java b/Core/src/main/java/org/tribuo/dataset/MinimumCardinalityDataset.java
index d03d081da..25f9e8553 100644
--- a/Core/src/main/java/org/tribuo/dataset/MinimumCardinalityDataset.java
+++ b/Core/src/main/java/org/tribuo/dataset/MinimumCardinalityDataset.java
@@ -166,6 +166,10 @@ <T extends Output<T>> MinimumCardinalityDatasetProvenance(MinimumCardinalityData
             this.minCardinality = new IntProvenance(MIN_CARDINALITY,dataset.minCardinality);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public MinimumCardinalityDatasetProvenance(Map<String,Provenance> map) {
             super(map);
             this.minCardinality = ObjectProvenance.checkAndExtractProvenance(map,MIN_CARDINALITY,IntProvenance.class,MinimumCardinalityDatasetProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/datasource/AggregateConfigurableDataSource.java b/Core/src/main/java/org/tribuo/datasource/AggregateConfigurableDataSource.java
index 07708e0f6..81b279ff4 100644
--- a/Core/src/main/java/org/tribuo/datasource/AggregateConfigurableDataSource.java
+++ b/Core/src/main/java/org/tribuo/datasource/AggregateConfigurableDataSource.java
@@ -101,6 +101,11 @@ public DataSourceProvenance getProvenance() {
     public static class AggregateConfigurableDataSourceProvenance extends SkeletalConfiguredObjectProvenance implements DataSourceProvenance {
         private static final long serialVersionUID = 1L;
 
+        /**
+         * Creates a AggregateConfigurableDataSourceProvenance from the host object.
+         * @param host The data source to provenance.
+         * @param <T> The type of the data source.
+         */
         <T extends Output<T>> AggregateConfigurableDataSourceProvenance(AggregateConfigurableDataSource<T> host) {
             super(host, "DataSource");
         }
@@ -113,10 +118,19 @@ public AggregateConfigurableDataSourceProvenance(Map<String, Provenance> map) {
             this(extractProvenanceInfo(map));
         }
 
+        /**
+         * Inner deserialization constructor.
+         * @param info The extracted information.
+         */
         private AggregateConfigurableDataSourceProvenance(ExtractedInfo info) {
             super(info);
         }
 
+        /**
+         * Extracts the class name and host type fields from the provenance map.
+         * @param map The provenance map.
+         * @return The extracted information.
+         */
         protected static ExtractedInfo extractProvenanceInfo(Map<String, Provenance> map) {
             Map<String, Provenance> configuredParameters = new HashMap<>(map);
             String className = ObjectProvenance.checkAndExtractProvenance(configuredParameters,CLASS_NAME, StringProvenance.class, AggregateConfigurableDataSourceProvenance.class.getSimpleName()).getValue();
diff --git a/Core/src/main/java/org/tribuo/datasource/AggregateDataSource.java b/Core/src/main/java/org/tribuo/datasource/AggregateDataSource.java
index dbd87b464..606f58eee 100644
--- a/Core/src/main/java/org/tribuo/datasource/AggregateDataSource.java
+++ b/Core/src/main/java/org/tribuo/datasource/AggregateDataSource.java
@@ -197,12 +197,21 @@ public static class AggregateDataSourceProvenance implements DataSourceProvenanc
         private final ListProvenance<DataSourceProvenance> provenances;
         private EnumProvenance<IterationOrder> orderProvenance;
 
+        /**
+         * Constructs an AggregateDataSourceProvenance from the host object.
+         * @param host The host data source.
+         * @param <T> The data source type.
+         */
         <T extends Output<T>> AggregateDataSourceProvenance(AggregateDataSource<T> host) {
             this.className = new StringProvenance(CLASS_NAME,host.getClass().getName());
             this.provenances = ListProvenance.createListProvenance(host.sources);
             this.orderProvenance = new EnumProvenance<>(ORDER,host.order);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenance map.
+         */
         @SuppressWarnings({"unchecked","rawtypes"}) //ListProvenance cast, EnumProvenance cast
         public AggregateDataSourceProvenance(Map<String,Provenance> map) {
             this.className = ObjectProvenance.checkAndExtractProvenance(map,CLASS_NAME, StringProvenance.class,AggregateDataSourceProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/datasource/IDXDataSource.java b/Core/src/main/java/org/tribuo/datasource/IDXDataSource.java
index fd7831324..003a33ccd 100644
--- a/Core/src/main/java/org/tribuo/datasource/IDXDataSource.java
+++ b/Core/src/main/java/org/tribuo/datasource/IDXDataSource.java
@@ -70,11 +70,29 @@ public final class IDXDataSource<T extends Output<T>> implements ConfigurableDat
      * The possible IDX input formats.
      */
     public enum IDXType {
+        /**
+         * An unsigned byte.
+         */
         UBYTE((byte) 0x08),
+        /**
+         * A signed byte.
+         */
         BYTE((byte) 0x09),
+        /**
+         * A 16-bit integer.
+         */
         SHORT((byte) 0x0B),
+        /**
+         * A 32-bit integer.
+         */
         INT((byte) 0x0C),
+        /**
+         * A 32-bit float.
+         */
         FLOAT((byte) 0x0D),
+        /**
+         * A 64-bit float.
+         */
         DOUBLE((byte) 0x0E);
 
         /**
@@ -493,10 +511,25 @@ private static DataOutputStream makeStream(Path outputPath, boolean gzip) throws
     public static final class IDXDataSourceProvenance extends SkeletalConfiguredObjectProvenance implements DataSourceProvenance {
         private static final long serialVersionUID = 1L;
 
+        /**
+         * The name of the output file modified time provenance field.
+         */
         public static final String OUTPUT_FILE_MODIFIED_TIME = "output-file-modified-time";
+        /**
+         * The name of the features file modified time provenance field.
+         */
         public static final String FEATURES_FILE_MODIFIED_TIME = "features-file-modified-time";
+        /**
+         * The name of the provenance field for the feature file hash.
+         */
         public static final String FEATURES_RESOURCE_HASH = "features-resource-hash";
+        /**
+         * The name of the provenance field for the output file hash.
+         */
         public static final String OUTPUT_RESOURCE_HASH = "output-resource-hash";
+        /**
+         * The name of the provenance field for the idx feature type.
+         */
         public static final String FEATURE_TYPE = "idx-feature-type";
 
         private final DateTimeProvenance featuresFileModifiedTime;
@@ -516,6 +549,10 @@ <T extends Output<T>> IDXDataSourceProvenance(IDXDataSource<T> host) {
             this.featureType = new EnumProvenance<>(FEATURE_TYPE, host.dataType);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public IDXDataSourceProvenance(Map<String, Provenance> map) {
             this(extractProvenanceInfo(map));
         }
@@ -532,6 +569,11 @@ private IDXDataSourceProvenance(ExtractedInfo info) {
             this.featureType = (EnumProvenance<IDXType>) info.instanceValues.get(FEATURE_TYPE);
         }
 
+        /**
+         * Separates out the configured and non-configured provenance values.
+         * @param map The provenances to separate.
+         * @return The extracted provenance information.
+         */
         protected static ExtractedInfo extractProvenanceInfo(Map<String, Provenance> map) {
             Map<String, Provenance> configuredParameters = new HashMap<>(map);
             String className = ObjectProvenance.checkAndExtractProvenance(configuredParameters, CLASS_NAME, StringProvenance.class, IDXDataSourceProvenance.class.getSimpleName()).getValue();
diff --git a/Core/src/main/java/org/tribuo/datasource/ListDataSource.java b/Core/src/main/java/org/tribuo/datasource/ListDataSource.java
index 3f1dafd1a..1a9846917 100644
--- a/Core/src/main/java/org/tribuo/datasource/ListDataSource.java
+++ b/Core/src/main/java/org/tribuo/datasource/ListDataSource.java
@@ -42,6 +42,12 @@ public class ListDataSource<T extends Output<T>> implements DataSource<T> {
 
     private final DataSourceProvenance provenance;
 
+    /**
+     * Constructs an in-memory data source wrapping the supplied examples.
+     * @param list The examples.
+     * @param factory The output factory.
+     * @param provenance The data source provenance.
+     */
     public ListDataSource(List<Example<T>> list, OutputFactory<T> factory, DataSourceProvenance provenance) {
         this.data = Collections.unmodifiableList(new ArrayList<>(list));
         this.factory = factory;
diff --git a/Core/src/main/java/org/tribuo/ensemble/BaggingTrainer.java b/Core/src/main/java/org/tribuo/ensemble/BaggingTrainer.java
index fb5d1801f..a1d3065de 100644
--- a/Core/src/main/java/org/tribuo/ensemble/BaggingTrainer.java
+++ b/Core/src/main/java/org/tribuo/ensemble/BaggingTrainer.java
@@ -74,10 +74,23 @@ public class BaggingTrainer<T extends Output<T>> implements Trainer<T> {
      */
     protected BaggingTrainer() { }
 
+    /**
+     * Constructs a bagging trainer with the supplied parameters using {@link Trainer#DEFAULT_SEED} as the RNG seed.
+     * @param trainer The ensemble member trainer.
+     * @param combiner The combination function.
+     * @param numMembers The number of ensemble members to train.
+     */
     public BaggingTrainer(Trainer<T> trainer, EnsembleCombiner<T> combiner, int numMembers) {
         this(trainer, combiner, numMembers, Trainer.DEFAULT_SEED);
     }
 
+    /**
+     * Constructs a bagging trainer with the supplied parameters.
+     * @param trainer The ensemble member trainer.
+     * @param combiner The combination function.
+     * @param numMembers The number of ensemble members to train.
+     * @param seed The RNG seed used to bootstrap the datasets.
+     */
     public BaggingTrainer(Trainer<T> trainer, EnsembleCombiner<T> combiner, int numMembers, long seed) {
         this.innerTrainer = trainer;
         this.combiner = combiner;
@@ -94,6 +107,10 @@ public synchronized void postConfig() {
         this.rng = new SplittableRandom(seed);
     }
 
+    /**
+     * Default name of the generated ensemble.
+     * @return The default ensemble name.
+     */
     protected String ensembleName() {
         return "bagging-ensemble";
     }
@@ -137,6 +154,15 @@ public Model<T> train(Dataset<T> examples, Map<String, Provenance> runProvenance
         return new WeightedEnsembleModel<>(ensembleName(),provenance,featureIDs,labelIDs,models,combiner);
     }
 
+    /**
+     * Trains a single model.
+     * @param examples The training dataset.
+     * @param featureIDs The feature domain.
+     * @param labelIDs The output domain.
+     * @param localRNG The local RNG instance.
+     * @param runProvenance Provenance for this instance.
+     * @return The trained ensemble member.
+     */
     protected Model<T> trainSingleModel(Dataset<T> examples, ImmutableFeatureMap featureIDs, ImmutableOutputInfo<T> labelIDs, SplittableRandom localRNG, Map<String,Provenance> runProvenance) {
         DatasetView<T> bag = DatasetView.createBootstrapView(examples,examples.size(),localRNG.nextInt(),featureIDs,labelIDs);
         Model<T> newModel = innerTrainer.train(bag,runProvenance);
diff --git a/Core/src/main/java/org/tribuo/ensemble/EnsembleExcuse.java b/Core/src/main/java/org/tribuo/ensemble/EnsembleExcuse.java
index 60d0a806a..fcae10cd3 100644
--- a/Core/src/main/java/org/tribuo/ensemble/EnsembleExcuse.java
+++ b/Core/src/main/java/org/tribuo/ensemble/EnsembleExcuse.java
@@ -32,6 +32,13 @@ public class EnsembleExcuse<T extends Output<T>> extends Excuse<T> {
 
     private final List<Excuse<T>> innerExcuses;
 
+    /**
+     * Constructs an ensemble excuse, comprising the excuses from each ensemble member, along with the feature weights.
+     * @param example The example.
+     * @param prediction The prediction to excuse.
+     * @param weights The weights.
+     * @param innerExcuses The ensemble member excuses.
+     */
     public EnsembleExcuse(Example<T> example, Prediction<T> prediction, Map<String,List<Pair<String,Double>>> weights, List<Excuse<T>> innerExcuses) {
         super(example,prediction,weights);
         this.innerExcuses = innerExcuses;
diff --git a/Core/src/main/java/org/tribuo/evaluation/DescriptiveStats.java b/Core/src/main/java/org/tribuo/evaluation/DescriptiveStats.java
index 5758cd086..f106eb288 100644
--- a/Core/src/main/java/org/tribuo/evaluation/DescriptiveStats.java
+++ b/Core/src/main/java/org/tribuo/evaluation/DescriptiveStats.java
@@ -31,8 +31,15 @@ public final class DescriptiveStats {
 
     private final List<Double> samples = new ArrayList<>();
 
+    /**
+     * Create an empty DescriptiveStats.
+     */
     public DescriptiveStats() {}
 
+    /**
+     * Create a DescriptiveStats initialized with the supplied values.
+     * @param values The initial values.
+     */
     public DescriptiveStats(List<Double> values) {
         this.samples.addAll(values);
     }
diff --git a/Core/src/main/java/org/tribuo/evaluation/KFoldSplitter.java b/Core/src/main/java/org/tribuo/evaluation/KFoldSplitter.java
index 6ab98c333..b89de03d8 100644
--- a/Core/src/main/java/org/tribuo/evaluation/KFoldSplitter.java
+++ b/Core/src/main/java/org/tribuo/evaluation/KFoldSplitter.java
@@ -122,20 +122,30 @@ public TrainTestFold<T> next() {
 
     /**
      * Stores a train/test split for a dataset.
-     *
+     * <p>
+     * Will be a record one day.
      * @see KFoldSplitter#split
      *
      * @param <T> the type of the examples that make up the data we've split.
      */
     public static class TrainTestFold<T extends Output<T>> {
-
+        /**
+         * The training fold.
+         */
         public final DatasetView<T> train;
+        /**
+         * The testing fold.
+         */
         public final DatasetView<T> test;
 
+        /**
+         * Constructs a train test fold.
+         * @param train The training fold.
+         * @param test The testing fold.
+         */
         TrainTestFold(DatasetView<T> train, DatasetView<T> test) {
             this.train = train;
             this.test = test;
         }
-
     }
 }
diff --git a/Core/src/main/java/org/tribuo/evaluation/TrainTestSplitter.java b/Core/src/main/java/org/tribuo/evaluation/TrainTestSplitter.java
index e48792183..489d9d262 100644
--- a/Core/src/main/java/org/tribuo/evaluation/TrainTestSplitter.java
+++ b/Core/src/main/java/org/tribuo/evaluation/TrainTestSplitter.java
@@ -154,6 +154,10 @@ <T extends Output<T>> SplitDataSourceProvenance(TrainTestSplitter<T> host, boole
             this.isTrain = new BooleanProvenance(IS_TRAIN,isTrain);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public SplitDataSourceProvenance(Map<String, Provenance> map) {
             this.className = ObjectProvenance.checkAndExtractProvenance(map,CLASS_NAME,StringProvenance.class,SplitDataSourceProvenance.class.getSimpleName());
             this.innerSourceProvenance = ObjectProvenance.checkAndExtractProvenance(map,SOURCE,DataSourceProvenance.class,SplitDataSourceProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/evaluation/metrics/EvaluationMetric.java b/Core/src/main/java/org/tribuo/evaluation/metrics/EvaluationMetric.java
index 76f08644b..1a1150a1b 100644
--- a/Core/src/main/java/org/tribuo/evaluation/metrics/EvaluationMetric.java
+++ b/Core/src/main/java/org/tribuo/evaluation/metrics/EvaluationMetric.java
@@ -82,7 +82,13 @@ public default C createContext(Model<T> model, Dataset<T> dataset) {
      */
     // Note, if we extend this enum, update MetricTarget with new singletons.
     enum Average {
+        /**
+         * The macro average.
+         */
         MACRO,
+        /**
+         * The micro average.
+         */
         MICRO
     }
 }
\ No newline at end of file
diff --git a/Core/src/main/java/org/tribuo/evaluation/metrics/MetricID.java b/Core/src/main/java/org/tribuo/evaluation/metrics/MetricID.java
index 438f8ffbc..e213f2551 100644
--- a/Core/src/main/java/org/tribuo/evaluation/metrics/MetricID.java
+++ b/Core/src/main/java/org/tribuo/evaluation/metrics/MetricID.java
@@ -24,12 +24,17 @@
  * Just an easier-to-read alias for {@code Pair<MetricTarget<T>, String>}.
  * <p>
  * Used as a key in metric result maps. See {@link Evaluation#asMap()}.
- * <p>
  * @param <T> The type of output.
  */
 public final class MetricID<T extends Output<T>> extends Pair<MetricTarget<T>, String> {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * Constructs a metric id.
+     * @param target The metric target.
+     * @param metricName The name of the metric.
+     */
+    // Should be a record one day.
     public MetricID(MetricTarget<T> target, String metricName) {
         super(target, metricName);
     }
diff --git a/Core/src/main/java/org/tribuo/hash/HashCodeHasher.java b/Core/src/main/java/org/tribuo/hash/HashCodeHasher.java
index 65c356436..cfbc797b6 100644
--- a/Core/src/main/java/org/tribuo/hash/HashCodeHasher.java
+++ b/Core/src/main/java/org/tribuo/hash/HashCodeHasher.java
@@ -32,7 +32,7 @@
 public final class HashCodeHasher extends Hasher {
     private static final long serialVersionUID = 2L;
 
-    @Config(mandatory = true,description="Salt used in the hash.")
+    @Config(mandatory = true, redact = true, description="Salt used in the hash.")
     private transient String salt = null;
 
     private static final HashCodeHasherProvenance provenance = new HashCodeHasherProvenance();
@@ -42,6 +42,10 @@ public final class HashCodeHasher extends Hasher {
      */
     private HashCodeHasher() { }
 
+    /**
+     * Constructs a HashCodeHasher using the specified salt value.
+     * @param salt The salt value.
+     */
     public HashCodeHasher(String salt) {
         this.salt = salt;
     }
@@ -87,6 +91,10 @@ public final static class HashCodeHasherProvenance implements ConfiguredObjectPr
 
         HashCodeHasherProvenance() {}
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public HashCodeHasherProvenance(Map<String, Provenance> map) { }
 
         @Override
diff --git a/Core/src/main/java/org/tribuo/hash/HashingOptions.java b/Core/src/main/java/org/tribuo/hash/HashingOptions.java
index d54bc840c..89a5f348d 100644
--- a/Core/src/main/java/org/tribuo/hash/HashingOptions.java
+++ b/Core/src/main/java/org/tribuo/hash/HashingOptions.java
@@ -33,15 +33,43 @@ public class HashingOptions implements Options {
     /**
      * Supported types of hashes in CLI programs.
      */
-    public enum ModelHashingType { NONE, MOD, HC, SHA1, SHA256 }
+    public enum ModelHashingType {
+        /**
+         * No hashing applied.
+         */
+        NONE,
+        /**
+         * Takes the String hash code mod some value.
+         */
+        MOD,
+        /**
+         * Uses the String hash code.
+         */
+        HC,
+        /**
+         * Uses SHA-1.
+         */
+        SHA1,
+        /**
+         * Uses SHA-256.
+         */
+        SHA256
+    }
 
-    @Option(longName="model-hashing-algorithm",usage="Hash the model during training, options are {NONE,MOD,HC,SHA1,SHA256}")
+    /**
+     * Hash the model during training, options are {NONE,MOD,HC,SHA1,SHA256}
+     */
+    @Option(longName = "model-hashing-algorithm", usage = "Hash the model during training, options are {NONE,MOD,HC,SHA1,SHA256}")
     public ModelHashingType modelHashingAlgorithm = ModelHashingType.NONE;
-    @Option(longName="model-hashing-salt",usage="Salt for hashing the model")
+    /**
+     * Salt for hashing the model
+     */
+    @Option(longName = "model-hashing-salt", usage = "Salt for hashing the model")
     public String modelHashingSalt = "";
 
     /**
      * Get the specified hasher.
+     *
      * @return The configured hasher.
      */
     public Optional<Hasher> getHasher() {
@@ -69,14 +97,15 @@ public Optional<Hasher> getHasher() {
 
     /**
      * Gets the trainer wrapped in a hashing trainer.
+     *
      * @param innerTrainer The inner trainer.
-     * @param <T> The output type.
+     * @param <T>          The output type.
      * @return The hashing trainer.
      */
     public <T extends Output<T>> Trainer<T> getHashedTrainer(Trainer<T> innerTrainer) {
         Optional<Hasher> hasherOpt = getHasher();
         if (hasherOpt.isPresent()) {
-            return new HashingTrainer<>(innerTrainer,hasherOpt.get());
+            return new HashingTrainer<>(innerTrainer, hasherOpt.get());
         } else {
             throw new IllegalArgumentException("Invalid Hasher");
         }
diff --git a/Core/src/main/java/org/tribuo/hash/HashingTrainer.java b/Core/src/main/java/org/tribuo/hash/HashingTrainer.java
index cd2faa24a..c7f48439f 100644
--- a/Core/src/main/java/org/tribuo/hash/HashingTrainer.java
+++ b/Core/src/main/java/org/tribuo/hash/HashingTrainer.java
@@ -52,6 +52,11 @@ public final class HashingTrainer<T extends Output<T>> implements Trainer<T> {
      */
     private HashingTrainer() {}
 
+    /**
+     * Constructs a hashing trainer using the supplied parameters.
+     * @param trainer The trainer to use.
+     * @param hasher The feature hasher to apply.
+     */
     public HashingTrainer(Trainer<T> trainer, Hasher hasher) {
         this.innerTrainer = trainer;
         this.hasher = hasher;
diff --git a/Core/src/main/java/org/tribuo/hash/MessageDigestHasher.java b/Core/src/main/java/org/tribuo/hash/MessageDigestHasher.java
index a2d9537ed..fcf79a336 100644
--- a/Core/src/main/java/org/tribuo/hash/MessageDigestHasher.java
+++ b/Core/src/main/java/org/tribuo/hash/MessageDigestHasher.java
@@ -41,6 +41,9 @@
 public final class MessageDigestHasher extends Hasher {
     private static final long serialVersionUID = 3L;
 
+    /**
+     * Alias for {@link StandardCharsets#UTF_8}.
+     */
     public static final Charset utf8Charset = StandardCharsets.UTF_8;
 
     static final String HASH_TYPE = "hashType";
@@ -65,6 +68,11 @@ public final class MessageDigestHasher extends Hasher {
      */
     private MessageDigestHasher() {}
 
+    /**
+     * Constructs a message digest hasher.
+     * @param hashType The hash function to use.
+     * @param salt The salt value.
+     */
     public MessageDigestHasher(String hashType, String salt) {
         this.hashType = hashType;
         this.salt = salt.getBytes(utf8Charset);
@@ -151,6 +159,10 @@ public final static class MessageDigestHasherProvenance implements ConfiguredObj
             this.hashType = new StringProvenance(HASH_TYPE,hashType);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public MessageDigestHasherProvenance(Map<String, Provenance> map) {
             hashType = ObjectProvenance.checkAndExtractProvenance(map,HASH_TYPE,StringProvenance.class,MessageDigestHasherProvenance.class.getSimpleName());
         }
diff --git a/Core/src/main/java/org/tribuo/hash/ModHashCodeHasher.java b/Core/src/main/java/org/tribuo/hash/ModHashCodeHasher.java
index ac2ae4811..0f0104053 100644
--- a/Core/src/main/java/org/tribuo/hash/ModHashCodeHasher.java
+++ b/Core/src/main/java/org/tribuo/hash/ModHashCodeHasher.java
@@ -37,7 +37,7 @@ public final class ModHashCodeHasher extends Hasher {
 
     static final String DIMENSION = "dimension";
 
-    @Config(mandatory = true,description="Salt used in the hash.")
+    @Config(mandatory = true,redact = true,description="Salt used in the hash.")
     private transient String salt = null;
 
     @Config(mandatory = true,description="Range of the hashing function.")
@@ -50,10 +50,19 @@ public final class ModHashCodeHasher extends Hasher {
      */
     private ModHashCodeHasher() { }
 
+    /**
+     * Constructs a ModHashCodeHasher with a fixed dimensionality of 100.
+     * @param salt The salt value.
+     */
     public ModHashCodeHasher(String salt) {
         this(100,salt);
     }
 
+    /**
+     * Constructs a ModHashCodeHasher with the supplied parameters.
+     * @param dimension The dimensionality.
+     * @param salt The salt value.
+     */
     public ModHashCodeHasher(int dimension, String salt) {
         this.dimension = dimension;
         this.salt = salt;
@@ -113,6 +122,10 @@ public final static class ModHashCodeHasherProvenance implements ConfiguredObjec
             this.dimension = new IntProvenance(DIMENSION,dimension);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public ModHashCodeHasherProvenance(Map<String, Provenance> map) {
             dimension = ObjectProvenance.checkAndExtractProvenance(map,DIMENSION,IntProvenance.class,ModHashCodeHasherProvenance.class.getSimpleName());
         }
diff --git a/Core/src/main/java/org/tribuo/impl/ArrayExample.java b/Core/src/main/java/org/tribuo/impl/ArrayExample.java
index e5fdbfdb8..05dcb0181 100644
--- a/Core/src/main/java/org/tribuo/impl/ArrayExample.java
+++ b/Core/src/main/java/org/tribuo/impl/ArrayExample.java
@@ -48,12 +48,24 @@ public class ArrayExample<T extends Output<T>> extends Example<T> {
 
     private static final Logger logger = Logger.getLogger(ArrayExample.class.getName());
 
+    /**
+     * Default initial size of the backing arrays.
+     */
     public static final int DEFAULT_SIZE = 10;
 
+    /**
+     * Feature names array.
+     */
     protected String[] featureNames;
 
+    /**
+     * Feature values array.
+     */
     protected double[] featureValues;
 
+    /**
+     * Number of valid features in this example.
+     */
     protected int size = 0;
 
     /**
diff --git a/Core/src/main/java/org/tribuo/impl/BinaryFeaturesExample.java b/Core/src/main/java/org/tribuo/impl/BinaryFeaturesExample.java
index eca52463e..cfa60bc7f 100644
--- a/Core/src/main/java/org/tribuo/impl/BinaryFeaturesExample.java
+++ b/Core/src/main/java/org/tribuo/impl/BinaryFeaturesExample.java
@@ -62,10 +62,19 @@ public final class BinaryFeaturesExample<T extends Output<T>> extends Example<T>
 
     private static final Logger logger = Logger.getLogger(BinaryFeaturesExample.class.getName());
 
+    /**
+     * Default initial size of the backing arrays.
+     */
     public static final int DEFAULT_SIZE = 10;
 
+    /**
+     * Feature names array.
+     */
     protected String[] featureNames;
 
+    /**
+     * Number of valid features in this example.
+     */
     protected int size = 0;
 
     /**
@@ -212,10 +221,19 @@ public void add(String name) {
         sort();
     }
 
+    /**
+     * Is the supplied feature binary (i.e., does it have a value of 1.0)?
+     * @param feature The feature to check.
+     * @return True if it's a binary feature.
+     */
     public static boolean isBinary(Feature feature) {
         return feature.getValue() == 1.0;
     }
 
+    /**
+     * Checks if the supplied feature is binary, if not throw an {@link IllegalArgumentException}.
+     * @param feature The feature to check.
+     */
     public static void checkIsBinary(Feature feature) {
         if(!isBinary(feature)) {
             throw new IllegalArgumentException("non-binary features are not allowed in BinaryFeaturesExample: value="+feature.getValue());
diff --git a/Core/src/main/java/org/tribuo/impl/IndexedArrayExample.java b/Core/src/main/java/org/tribuo/impl/IndexedArrayExample.java
index 80d3a9b90..6e55f73db 100644
--- a/Core/src/main/java/org/tribuo/impl/IndexedArrayExample.java
+++ b/Core/src/main/java/org/tribuo/impl/IndexedArrayExample.java
@@ -44,8 +44,15 @@
 public class IndexedArrayExample<T extends Output<T>> extends ArrayExample<T> {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * Feature id numbers from the internal featureMap.
+     */
     protected int[] featureIDs;
 
+    /**
+     * Output id from the internal output map. Note this only works for single
+     * dimensional outputs, in multi-dimensional cases it is likely to be -1.
+     */
     protected final int outputID;
 
     private final ImmutableFeatureMap featureMap;
@@ -327,12 +334,32 @@ public Iterable<FeatureTuple> idIterator() {
      * A tuple of the feature name, id and value.
      */
     public static class FeatureTuple {
+        /**
+         * The feature name.
+         */
         public String name;
+        /**
+         * The feature id number.
+         */
         public int id;
+        /**
+         * The feature value.
+         */
         public double value;
 
+        /**
+         * Constructs an empty feature tuple.
+         */
+        // TODO this should be made package private.
         public FeatureTuple() { }
 
+        /**
+         * Constructs a feature tuple using the specified values.
+         * @param name The feature name.
+         * @param id The feature id number.
+         * @param value The feature value.
+         */
+        // TODO this should be made package private.
         public FeatureTuple(String name, int id, double value) {
             this.name = name;
             this.id = id;
diff --git a/Core/src/main/java/org/tribuo/impl/ListExample.java b/Core/src/main/java/org/tribuo/impl/ListExample.java
index 5e196821c..709bdc1d0 100644
--- a/Core/src/main/java/org/tribuo/impl/ListExample.java
+++ b/Core/src/main/java/org/tribuo/impl/ListExample.java
@@ -49,14 +49,27 @@ public class ListExample<T extends Output<T>> extends Example<T> implements Seri
 
     private final List<Feature> features = new ArrayList<>();
 
+    /**
+     * Constructs a ListExample for the specified output and weight.
+     * @param output The output.
+     * @param weight The example weight.
+     */
     public ListExample(T output, float weight) {
         super(output,weight);
     }
-     
+
+    /**
+     * Constructs a ListExample for the specified output with a weight of {@link Example#DEFAULT_WEIGHT}.
+     * @param output The output.
+     */
     public ListExample(T output) {
         super(output);
     }
 
+    /**
+     * Copies the supplied example's features, weight, output and metadata into this example.
+     * @param other The example to copy.
+     */
     public ListExample(Example<T> other) {
         super(other);
         for (Feature f : other) {
@@ -65,6 +78,11 @@ public ListExample(Example<T> other) {
         sort();
     }
 
+    /**
+     * Constructs a ListExample using the specified output and feature list.
+     * @param output The output.
+     * @param features The features.
+     */
     public ListExample(T output, List<? extends Feature> features) {
         super(output);
         for (Feature f : features) {
@@ -73,6 +91,12 @@ public ListExample(T output, List<? extends Feature> features) {
         sort();
     }
 
+    /**
+     * Constructs a ListExample from the specified output, feature names and feature values.
+     * @param output The output.
+     * @param featureNames The feature names.
+     * @param featureValues The feature values.
+     */
     public ListExample(T output, String[] featureNames, double[] featureValues) {
         super(output);
         if (featureNames.length != featureValues.length) {
@@ -103,7 +127,10 @@ public void addAll(Collection<? extends Feature> features) {
         this.features.addAll(features);
         sort();
     }
-    
+
+    /**
+     * Clears the features from this example.
+     */
     public void clear() {
         features.clear();
     }
diff --git a/Core/src/main/java/org/tribuo/onnx/ONNXOperators.java b/Core/src/main/java/org/tribuo/onnx/ONNXOperators.java
index 795500f95..e7878a497 100644
--- a/Core/src/main/java/org/tribuo/onnx/ONNXOperators.java
+++ b/Core/src/main/java/org/tribuo/onnx/ONNXOperators.java
@@ -42,7 +42,6 @@ public enum ONNXOperators {
     SIGMOID("Sigmoid",1,1),
     /**
      * Softmax.
-     * <p>
      * <ul>
      *     <li>{@code axis} defaults to -1.</li>
      * </ul>
@@ -68,7 +67,6 @@ public enum ONNXOperators {
     DIV("Div",2,1),
     /**
      * Compute the minimum along the specified axes of the tensor.
-     * <p>
      * <ul>
      *     <li>{@code axes} defaults to all dimensions.</li>
      *     <li>{@code keepdims} defaults to 1 which means keep.</li>
@@ -80,7 +78,6 @@ public enum ONNXOperators {
     )),
     /**
      * Compute the sum along the specified axes of the tensor.
-     * <p>
      * <ul>
      *     <li>{@code axes} defaults to all dimensions.</li>
      *     <li>{@code keepdims} defaults to 1 which means keep.</li>
@@ -92,7 +89,6 @@ public enum ONNXOperators {
     )),
     /**
      * General Matrix Multiply: alpha*AB + beta*C.
-     * <p>
      * <ul>
      *     <li>{@code alpha} defaults to 1.0</li>
      *     <li>{@code beta} defaults to 1.0</li>
diff --git a/Core/src/main/java/org/tribuo/provenance/DataSourceProvenance.java b/Core/src/main/java/org/tribuo/provenance/DataSourceProvenance.java
index deb654850..ef70b1dcd 100644
--- a/Core/src/main/java/org/tribuo/provenance/DataSourceProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/DataSourceProvenance.java
@@ -23,12 +23,24 @@
  */
 public interface DataSourceProvenance extends DataProvenance, ObjectProvenance {
 
+    /**
+     * The name of the provenance field for the resource hash.
+     */
     public static final String RESOURCE_HASH = "resource-hash";
 
+    /**
+     * The name of the provenance field for the file timestamp.
+     */
     public static final String FILE_MODIFIED_TIME = "file-modified-time";
 
+    /**
+     * The name of the provenance field for the datasource timestamp.
+     */
     public static final String DATASOURCE_CREATION_TIME = "datasource-creation-time";
 
+    /**
+     * The name of the provenance field for the output factory.
+     */
     public static final String OUTPUT_FACTORY = "outputFactory";
 
 }
diff --git a/Core/src/main/java/org/tribuo/provenance/DatasetProvenance.java b/Core/src/main/java/org/tribuo/provenance/DatasetProvenance.java
index 32ba7760e..0110d6af7 100644
--- a/Core/src/main/java/org/tribuo/provenance/DatasetProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/DatasetProvenance.java
@@ -73,14 +73,39 @@ public class DatasetProvenance implements DataProvenance, ObjectProvenance {
 
     private final String versionString;
 
+    /**
+     * Creates a dataset provenance from the supplied dataset.
+     * @param sourceProvenance The data source provenance.
+     * @param transformationProvenance The transformations applied to this dataset.
+     * @param dataset The dataset itself.
+     * @param <T> The output type of the dataset.
+     */
     public <T extends Output<T>> DatasetProvenance(DataProvenance sourceProvenance, ListProvenance<ObjectProvenance> transformationProvenance, Dataset<T> dataset) {
         this(sourceProvenance,transformationProvenance,dataset.getClass().getName(),dataset instanceof MutableDataset && ((MutableDataset<T>) dataset).isDense(),false,dataset.size(),dataset.getFeatureMap().size(),dataset.getOutputInfo().size());
     }
 
+    /**
+     * Creates a dataset provenance from the supplied sequence dataset.
+     * @param sourceProvenance The data source provenance.
+     * @param transformationProvenance The transformations applied to this sequence dataset.
+     * @param dataset The sequence dataset itself.
+     * @param <T> The output type of the sequence dataset.
+     */
     public <T extends Output<T>> DatasetProvenance(DataProvenance sourceProvenance, ListProvenance<ObjectProvenance> transformationProvenance, SequenceDataset<T> dataset) {
         this(sourceProvenance,transformationProvenance,dataset.getClass().getName(),dataset instanceof MutableSequenceDataset && ((MutableSequenceDataset<T>) dataset).isDense(),true,dataset.size(),dataset.getFeatureMap().size(),dataset.getOutputInfo().size());
     }
 
+    /**
+     * Constructs a dataset provenance using the supplied information.
+     * @param sourceProvenance The data source provenance.
+     * @param transformationProvenance The transformations applied to the dataset.
+     * @param datasetClassName The dataset class name.
+     * @param isDense Is the dataset dense?
+     * @param isSequence Is it a sequence dataset?
+     * @param numExamples The number of examples in this dataset.
+     * @param numFeatures The number of features in this dataset.
+     * @param numOutputs The output dimensionality.
+     */
     protected DatasetProvenance(DataProvenance sourceProvenance, ListProvenance<ObjectProvenance> transformationProvenance, String datasetClassName, boolean isDense, boolean isSequence, int numExamples, int numFeatures, int numOutputs) {
         this.className = datasetClassName;
         this.sourceProvenance = sourceProvenance;
@@ -93,6 +118,10 @@ protected DatasetProvenance(DataProvenance sourceProvenance, ListProvenance<Obje
         this.versionString = Tribuo.VERSION;
     }
 
+    /**
+     * Deserialization constructor.
+     * @param map The provenances.
+     */
     @SuppressWarnings("unchecked") //ListProvenance assignment
     public DatasetProvenance(Map<String,Provenance> map) {
         this.className = ObjectProvenance.checkAndExtractProvenance(map,CLASS_NAME,StringProvenance.class, DatasetProvenance.class.getSimpleName()).getValue();
@@ -181,6 +210,10 @@ public Iterator<Pair<String, Provenance>> iterator() {
         return iterable.iterator();
     }
 
+    /**
+     * Returns a list of all the provenances.
+     * @return The provenances.
+     */
     protected List<Pair<String,Provenance>> allProvenances() {
         ArrayList<Pair<String,Provenance>> provenances = new ArrayList<>();
         provenances.add(new Pair<>(CLASS_NAME,new StringProvenance(CLASS_NAME,className)));
diff --git a/Core/src/main/java/org/tribuo/provenance/EnsembleModelProvenance.java b/Core/src/main/java/org/tribuo/provenance/EnsembleModelProvenance.java
index f9efa8bdb..1cd4a9722 100644
--- a/Core/src/main/java/org/tribuo/provenance/EnsembleModelProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/EnsembleModelProvenance.java
@@ -32,6 +32,9 @@
 public class EnsembleModelProvenance extends ModelProvenance {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * The name of the provenance field where the member provenances are stored.
+     */
     protected static final String MEMBERS = "member-provenance";
 
     private final ListProvenance<? extends ModelProvenance> memberProvenance;
diff --git a/Core/src/main/java/org/tribuo/provenance/EvaluationProvenance.java b/Core/src/main/java/org/tribuo/provenance/EvaluationProvenance.java
index 1b7f05303..ab75df193 100644
--- a/Core/src/main/java/org/tribuo/provenance/EvaluationProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/EvaluationProvenance.java
@@ -42,6 +42,11 @@ public final class EvaluationProvenance implements ObjectProvenance {
     private final DataProvenance datasetProvenance;
     private final StringProvenance versionString;
 
+    /**
+     * Constructs an evaluation provenance from the supplied provenances.
+     * @param modelProvenance The model provenance.
+     * @param datasetProvenance The test data provenance.
+     */
     public EvaluationProvenance(ModelProvenance modelProvenance, DataProvenance datasetProvenance) {
         this.className = new StringProvenance(CLASS_NAME, EvaluationProvenance.class.getName());
         this.modelProvenance = modelProvenance;
@@ -49,6 +54,10 @@ public EvaluationProvenance(ModelProvenance modelProvenance, DataProvenance data
         this.versionString = new StringProvenance(TRIBUO_VERSION_STRING,Tribuo.VERSION);
     }
 
+    /**
+     * Deserialization constructor.
+     * @param map The provenances.
+     */
     public EvaluationProvenance(Map<String,Provenance> map) {
         this.className = ObjectProvenance.checkAndExtractProvenance(map,CLASS_NAME,StringProvenance.class, EvaluationProvenance.class.getSimpleName());
         this.modelProvenance = ObjectProvenance.checkAndExtractProvenance(map,MODEL_PROVENANCE_NAME,ModelProvenance.class, EvaluationProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/provenance/SimpleDataSourceProvenance.java b/Core/src/main/java/org/tribuo/provenance/SimpleDataSourceProvenance.java
index 60633cf8c..c7aaf78b1 100644
--- a/Core/src/main/java/org/tribuo/provenance/SimpleDataSourceProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/SimpleDataSourceProvenance.java
@@ -41,6 +41,9 @@
 public class SimpleDataSourceProvenance implements DataSourceProvenance {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * The description field in the provenance.
+     */
     public static final String DESCRIPTION = "description";
 
     private final String className = DataSource.class.getName();
diff --git a/Core/src/main/java/org/tribuo/provenance/TrainerProvenance.java b/Core/src/main/java/org/tribuo/provenance/TrainerProvenance.java
index d1a5bfbdb..a0f3e6a86 100644
--- a/Core/src/main/java/org/tribuo/provenance/TrainerProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/TrainerProvenance.java
@@ -23,10 +23,19 @@
  */
 public interface TrainerProvenance extends ConfiguredObjectProvenance {
 
+    /**
+     * The name of the provenance field recording the train invocation count.
+     */
     public static final String TRAIN_INVOCATION_COUNT = "train-invocation-count";
 
+    /**
+     * The name of the provenance field recording if this is a sequence trainer.
+     */
     public static final String IS_SEQUENCE = DatasetProvenance.IS_SEQUENCE;
 
+    /**
+     * The name of the provenance field recording the Tribuo version used in training.
+     */
     public static final String TRIBUO_VERSION_STRING = "tribuo-version";
 
 }
diff --git a/Core/src/main/java/org/tribuo/provenance/impl/EmptyDataSourceProvenance.java b/Core/src/main/java/org/tribuo/provenance/impl/EmptyDataSourceProvenance.java
index 2f86e7680..ef8bd08b9 100644
--- a/Core/src/main/java/org/tribuo/provenance/impl/EmptyDataSourceProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/impl/EmptyDataSourceProvenance.java
@@ -32,8 +32,15 @@
 public final class EmptyDataSourceProvenance implements DataSourceProvenance {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * Constructs an empty data source provenance.
+     */
     EmptyDataSourceProvenance() { }
 
+    /**
+     * Deserialization constructor.
+     * @param map The provenances, which are ignored as this provenance is empty.
+     */
     public EmptyDataSourceProvenance(Map<String,Provenance> map) {}
 
     @Override
diff --git a/Core/src/main/java/org/tribuo/provenance/impl/EmptyDatasetProvenance.java b/Core/src/main/java/org/tribuo/provenance/impl/EmptyDatasetProvenance.java
index afcaf167b..800a7a130 100644
--- a/Core/src/main/java/org/tribuo/provenance/impl/EmptyDatasetProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/impl/EmptyDatasetProvenance.java
@@ -29,10 +29,17 @@
 public final class EmptyDatasetProvenance extends DatasetProvenance {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * An empty dataset provenance.
+     */
     public EmptyDatasetProvenance() {
         super(new EmptyDataSourceProvenance(), new ListProvenance<>(), Dataset.class.getName(), false, false, -1, -1, -1);
     }
 
+    /**
+     * Deserialization constructor.
+     * @param map The provenance map, which is ignored as this provenance is empty.
+     */
     public EmptyDatasetProvenance(Map<String, Provenance> map) {
         this();
     }
diff --git a/Core/src/main/java/org/tribuo/provenance/impl/EmptyTrainerProvenance.java b/Core/src/main/java/org/tribuo/provenance/impl/EmptyTrainerProvenance.java
index 837e23dc7..cd1c81892 100644
--- a/Core/src/main/java/org/tribuo/provenance/impl/EmptyTrainerProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/impl/EmptyTrainerProvenance.java
@@ -29,8 +29,15 @@
 public final class EmptyTrainerProvenance implements TrainerProvenance {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * Constructs an empty trainer provenance.
+     */
     public EmptyTrainerProvenance() { }
 
+    /**
+     * Deserialization constructor.
+     * @param map The provenances, which are ignored as this provenance is empty.
+     */
     public EmptyTrainerProvenance(Map<String,Provenance> map) {}
 
     @Override
diff --git a/Core/src/main/java/org/tribuo/provenance/impl/TimestampedTrainerProvenance.java b/Core/src/main/java/org/tribuo/provenance/impl/TimestampedTrainerProvenance.java
index 78ccafd7f..84b144925 100644
--- a/Core/src/main/java/org/tribuo/provenance/impl/TimestampedTrainerProvenance.java
+++ b/Core/src/main/java/org/tribuo/provenance/impl/TimestampedTrainerProvenance.java
@@ -33,11 +33,15 @@
 
 /**
  * A TrainerProvenance with a timestamp, used when there was no trainer
- * involved in model construction (e.g., creating an EnsembleModel from existing models).
+ * involved in model construction (e.g., creating an {@link org.tribuo.ensemble.EnsembleModel}
+ * from existing models).
  */
 public final class TimestampedTrainerProvenance implements TrainerProvenance {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * The name of the provenance field storing the model creation time.
+     */
     public static final String CREATION_TIME = "creation-time";
 
     private final DateTimeProvenance creationTime;
diff --git a/Core/src/main/java/org/tribuo/sequence/HashingSequenceTrainer.java b/Core/src/main/java/org/tribuo/sequence/HashingSequenceTrainer.java
index dd8e8d5f2..2112a9418 100644
--- a/Core/src/main/java/org/tribuo/sequence/HashingSequenceTrainer.java
+++ b/Core/src/main/java/org/tribuo/sequence/HashingSequenceTrainer.java
@@ -48,6 +48,11 @@ public final class HashingSequenceTrainer<T extends Output<T>> implements Sequen
      */
     private HashingSequenceTrainer() {}
 
+    /**
+     * Constructs a hashing sequence trainer using the supplied parameters.
+     * @param trainer The sequence trainer to use.
+     * @param hasher The feature hasher to apply.
+     */
     public HashingSequenceTrainer(SequenceTrainer<T> trainer, Hasher hasher) {
         this.innerTrainer = trainer;
         this.hasher = hasher;
@@ -103,6 +108,10 @@ <T extends Output<T>> HashingSequenceTrainerProvenance(HashingSequenceTrainer<T>
             super(host);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public HashingSequenceTrainerProvenance(Map<String, Provenance> map) {
             super(extractProvenanceInfo(map));
         }
diff --git a/Core/src/main/java/org/tribuo/sequence/ImmutableSequenceDataset.java b/Core/src/main/java/org/tribuo/sequence/ImmutableSequenceDataset.java
index 32e96dd9e..cd1921764 100644
--- a/Core/src/main/java/org/tribuo/sequence/ImmutableSequenceDataset.java
+++ b/Core/src/main/java/org/tribuo/sequence/ImmutableSequenceDataset.java
@@ -64,10 +64,21 @@ protected ImmutableSequenceDataset(DataProvenance sourceProvenance, OutputFactor
         super(sourceProvenance,outputFactory);
     }
 
+    /**
+     * Creates a dataset from a data source, taking the output and feature domains from the supplied model.
+     * @param dataSource The input data.
+     * @param model The model to use for the feature and output domains.
+     */
     public ImmutableSequenceDataset(SequenceDataSource<T> dataSource, SequenceModel<T> model) {
         this(dataSource,dataSource.getProvenance(),model.getFeatureIDMap(),model.getOutputIDInfo(),dataSource.getOutputFactory());
     }
 
+    /**
+     * Creates a dataset from a data source, using the specified output and feature domains.
+     * @param dataSource The input data.
+     * @param featureIDMap The feature domain.
+     * @param outputIDInfo The output domain.
+     */
     public ImmutableSequenceDataset(SequenceDataSource<T> dataSource, FeatureMap featureIDMap, OutputInfo<T> outputIDInfo) {
         this(dataSource,dataSource.getProvenance(),featureIDMap,outputIDInfo,dataSource.getOutputFactory());
     }
diff --git a/Core/src/main/java/org/tribuo/sequence/MinimumCardinalitySequenceDataset.java b/Core/src/main/java/org/tribuo/sequence/MinimumCardinalitySequenceDataset.java
index f387a9cad..bad142085 100644
--- a/Core/src/main/java/org/tribuo/sequence/MinimumCardinalitySequenceDataset.java
+++ b/Core/src/main/java/org/tribuo/sequence/MinimumCardinalitySequenceDataset.java
@@ -188,6 +188,10 @@ <T extends Output<T>> MinimumCardinalitySequenceDatasetProvenance(
             this.minCardinality = new IntProvenance(MIN_CARDINALITY, dataset.minCardinality);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public MinimumCardinalitySequenceDatasetProvenance(Map<String, Provenance> map) {
             super(map);
             this.minCardinality = ObjectProvenance.checkAndExtractProvenance(map, MIN_CARDINALITY, IntProvenance.class,
diff --git a/Core/src/main/java/org/tribuo/sequence/MutableSequenceDataset.java b/Core/src/main/java/org/tribuo/sequence/MutableSequenceDataset.java
index 7c14465d5..efb7a2e47 100644
--- a/Core/src/main/java/org/tribuo/sequence/MutableSequenceDataset.java
+++ b/Core/src/main/java/org/tribuo/sequence/MutableSequenceDataset.java
@@ -49,6 +49,9 @@ public class MutableSequenceDataset<T extends Output<T>> extends SequenceDataset
      */
     protected final MutableFeatureMap featureMap;
 
+    /**
+     * Does this dataset have a dense feature space.
+     */
     protected boolean dense = true;
 
     /**
@@ -78,10 +81,20 @@ public MutableSequenceDataset(Iterable<SequenceExample<T>> dataSource, DataProve
         }
     }
 
+    /**
+     * Builds a dataset from the supplied data source.
+     * @param dataSource The data source.
+     */
     public MutableSequenceDataset(SequenceDataSource<T> dataSource) {
         this(dataSource,dataSource.getProvenance(),dataSource.getOutputFactory());
     }
 
+    /**
+     * Copies the immutable dataset into a mutable dataset.
+     * <p>
+     * This should be infrequently used and mostly exists for the ViterbiTrainer.
+     * @param dataset The dataset to copy.
+     */
     //special purpose constructor created for ViterbiTrainer
     public MutableSequenceDataset(ImmutableSequenceDataset<T> dataset) {
         super(dataset.getProvenance(),dataset.getOutputFactory());
diff --git a/Core/src/main/java/org/tribuo/sequence/SequenceExample.java b/Core/src/main/java/org/tribuo/sequence/SequenceExample.java
index 6235cc14d..0c1befce4 100644
--- a/Core/src/main/java/org/tribuo/sequence/SequenceExample.java
+++ b/Core/src/main/java/org/tribuo/sequence/SequenceExample.java
@@ -40,6 +40,9 @@ public class SequenceExample<T extends Output<T>> implements Iterable<Example<T>
 
     private static final Logger logger = Logger.getLogger(SequenceExample.class.getName());
 
+    /**
+     * The default sequence example weight.
+     */
     public static final float DEFAULT_WEIGHT = 1.0f;
 
     private final List<Example<T>> examples;
@@ -99,10 +102,29 @@ public SequenceExample(List<T> outputs, List<? extends List<? extends Feature>>
         this(outputs, features, weight, false);
     }
 
+    /**
+     * Creates a sequence example from the supplied weight, outputs and list of list of features.
+     * <p>
+     * The features are copied out by this constructor. The outputs and features lists
+     * must be of the same length. Sets the weight to {@link SequenceExample#DEFAULT_WEIGHT}.
+     * @param outputs The outputs for each sequence element.
+     * @param features The features for each sequence element.
+     * @param attemptBinaryFeatures Attempt to use {@link BinaryFeaturesExample} as the inner examples.
+     */
     public SequenceExample(List<T> outputs, List<? extends List<? extends Feature>> features, boolean attemptBinaryFeatures) {
         this(outputs, features, DEFAULT_WEIGHT, attemptBinaryFeatures);
     }
-    
+
+    /**
+     * Creates a sequence example from the supplied weight, outputs and list of list of features.
+     * <p>
+     * The features are copied out by this constructor. The outputs and features lists
+     * must be of the same length.
+     * @param outputs The outputs for each sequence element.
+     * @param features The features for each sequence element.
+     * @param weight The weight for this sequence example.
+     * @param attemptBinaryFeatures Attempt to use {@link BinaryFeaturesExample} as the inner examples.
+     */
     public SequenceExample(List<T> outputs, List<? extends List<? extends Feature>> features, float weight, boolean attemptBinaryFeatures) {
         if (outputs.size() != features.size()) {
             throw new IllegalArgumentException("outputs.size() = " + outputs.size() + ", features.size() = " + features.size());
diff --git a/Core/src/main/java/org/tribuo/sequence/SequenceModel.java b/Core/src/main/java/org/tribuo/sequence/SequenceModel.java
index 314cb5ab1..13bb492f8 100644
--- a/Core/src/main/java/org/tribuo/sequence/SequenceModel.java
+++ b/Core/src/main/java/org/tribuo/sequence/SequenceModel.java
@@ -48,6 +48,13 @@ public abstract class SequenceModel<T extends Output<T>> implements Provenancabl
 
     protected final ImmutableOutputInfo<T> outputIDMap;
 
+    /**
+     * Builds a SequenceModel.
+     * @param name The model name.
+     * @param provenance The model provenance.
+     * @param featureIDMap The feature domain.
+     * @param outputIDMap The output domain.
+     */
     public SequenceModel(String name, ModelProvenance provenance, ImmutableFeatureMap featureIDMap, ImmutableOutputInfo<T> outputIDMap) {
         this.name = name;
         this.provenance = provenance;
@@ -174,6 +181,12 @@ public List<List<Prediction<T>>> predict(SequenceDataset<T> examples) {
      */
     public abstract Map<String, List<Pair<String, Double>>> getTopFeatures(int n);
 
+    /**
+     * Extracts a list of the predicted outputs from the list of prediction objects.
+     * @param predictions The predictions.
+     * @param <T> The prediction type.
+     * @return A list of predicted outputs.
+     */
     public static <T extends Output<T>> List<T> toMaxLabels(List<Prediction<T>> predictions) {
         return predictions.stream().map(Prediction::getOutput).collect(Collectors.toList());
     }
diff --git a/Core/src/main/java/org/tribuo/sequence/SequenceModelExplorer.java b/Core/src/main/java/org/tribuo/sequence/SequenceModelExplorer.java
index 20f7ca219..dfb163d1e 100644
--- a/Core/src/main/java/org/tribuo/sequence/SequenceModelExplorer.java
+++ b/Core/src/main/java/org/tribuo/sequence/SequenceModelExplorer.java
@@ -42,10 +42,13 @@
 public class SequenceModelExplorer implements CommandGroup {
     private static final Logger logger = Logger.getLogger(SequenceModelExplorer.class.getName());
 
-    protected CommandInterpreter shell;
+    private final CommandInterpreter shell;
 
     private SequenceModel<?> model;
 
+    /**
+     * Builds a sequence model explorer shell.
+     */
     public SequenceModelExplorer() {
         shell = new CommandInterpreter();
         shell.setPrompt("model sh% ");
@@ -61,6 +64,10 @@ public String getDescription() {
         return "Commands for inspecting a SequenceModel.";
     }
 
+    /**
+     * Completers for files.
+     * @return The completers for file commands.
+     */
     public Completer[] fileCompleter() {
         return new Completer[]{
                 new Completers.FileNameCompleter(),
@@ -76,6 +83,12 @@ public void startShell() {
         shell.start();
     }
 
+    /**
+     * Loads a model.
+     * @param ci The shell instance.
+     * @param path The path to load.
+     * @return A status string.
+     */
     @Command(usage = "<filename> - Load a model from disk.", completers="fileCompleter")
     public String loadModel(CommandInterpreter ci, File path) {
         try (ObjectInputStream ois = new ObjectInputStream(new BufferedInputStream(new FileInputStream(path)))) {
@@ -94,11 +107,22 @@ public String loadModel(CommandInterpreter ci, File path) {
         return "Loaded model from path " + path.toString();
     }
 
+    /**
+     * Shows the model description.
+     * @param ci The command interpreter.
+     * @return The model description.
+     */
     @Command(usage="Shows the model description")
     public String modelDescription(CommandInterpreter ci) {
         return model.toString();
     }
 
+    /**
+     * Shows information on a particular feature.
+     * @param ci The command interpreter.
+     * @param featureName The feature name.
+     * @return The feature information.
+     */
     @Command(usage="Shows the information on a particular feature")
     public String featureInfo(CommandInterpreter ci, String featureName) {
         VariableInfo f = model.getFeatureIDMap().get(featureName);
@@ -109,21 +133,43 @@ public String featureInfo(CommandInterpreter ci, String featureName) {
         }
     }
 
+    /**
+     * Shows the output information.
+     * @param ci The command interpreter.
+     * @return The output information.
+     */
     @Command(usage="Shows the output information.")
     public String outputInfo(CommandInterpreter ci) {
         return model.getOutputIDInfo().toReadableString();
     }
 
+    /**
+     * Shows the top n features in this model.
+     * @param ci The command interpreter.
+     * @param numFeatures The number of features to display.
+     * @return The top features.
+     */
     @Command(usage="<int> - Shows the top N features in the model")
     public String topFeatures(CommandInterpreter ci, int numFeatures) {
         return ""+ model.getTopFeatures(numFeatures);
     }
 
+    /**
+     * Shows the number of features in this model.
+     * @param ci The command interpreter.
+     * @return The number of features.
+     */
     @Command(usage="Shows the number of features in the model")
     public String numFeatures(CommandInterpreter ci) {
         return ""+ model.getFeatureIDMap().size();
     }
 
+    /**
+     * Shows the number of features which occurred more than minCount times in the training data.
+     * @param ci The command interpreter.
+     * @param minCount The minimum occurrence count.
+     * @return The number of features which occurred more than minCount times.
+     */
     @Command(usage="<min count> - Shows the number of features that occurred more than min count times.")
     public String minCount(CommandInterpreter ci, int minCount) {
         int counter = 0;
@@ -135,25 +181,21 @@ public String minCount(CommandInterpreter ci, int minCount) {
         return counter + " features occurred more than " + minCount + " times.";
     }
 
-    public static String usage() {
-        StringBuilder string = new StringBuilder();
-        string.append("Usage: ModelExplorer\n");
-
-        string.append("Optional parameters\n");
-        string.append("     -f <model-filename>\n");
-        string.append("         Load in a model from file.\n");
-
-        return string.toString();
-    }
-
     /**
      * Command line options.
      */
     public static class SequenceModelExplorerOptions implements Options {
-        @Option(charName='f',longName="filename",usage="Model file to load. Optional.")
+        /**
+         * Model file to load. Optional.
+         */
+        @Option(charName = 'f', longName = "filename", usage = "Model file to load. Optional.")
         public String modelFilename;
     }
 
+    /**
+     * Runs the sequence model explorer.
+     * @param args CLI arguments.
+     */
     public static void main(String[] args) {
         SequenceModelExplorerOptions options = new SequenceModelExplorerOptions();
         ConfigurationManager cm = new ConfigurationManager(args,options,false);
diff --git a/Core/src/main/java/org/tribuo/transform/TransformationMap.java b/Core/src/main/java/org/tribuo/transform/TransformationMap.java
index aca46b4e3..53a9a525c 100644
--- a/Core/src/main/java/org/tribuo/transform/TransformationMap.java
+++ b/Core/src/main/java/org/tribuo/transform/TransformationMap.java
@@ -67,6 +67,11 @@ public class TransformationMap implements Configurable, Provenancable<Configured
      */
     private TransformationMap() {}
 
+    /**
+     * Creates a transformation map from the supplied global transformation list, and per feature transformations.
+     * @param globalTransformations The global transformations.
+     * @param featureTransformations The per feature transformations.
+     */
     public TransformationMap(List<Transformation> globalTransformations, Map<String,List<Transformation>> featureTransformations) {
         this.globalTransformations = new ArrayList<>(globalTransformations);
         this.featureTransformations.putAll(featureTransformations);
@@ -78,10 +83,18 @@ public TransformationMap(List<Transformation> globalTransformations, Map<String,
         
     }
 
+    /**
+     * Creates a TransformationMap with only global transformations.
+     * @param globalTransformations The global transformations.
+     */
     public TransformationMap(List<Transformation> globalTransformations) {
         this(globalTransformations, Collections.emptyMap());
     }
 
+    /**
+     * Creates a TransformationMap with only per feature transformations.
+     * @param featureTransformations The per feature transformations.
+     */
     public TransformationMap(Map<String,List<Transformation>> featureTransformations) {
         this(Collections.emptyList(),featureTransformations);
     }
@@ -175,11 +188,18 @@ private ConfiguredObjectProvenanceImpl cacheProvenance() {
      * Will be deprecated if/when OLCUT supports this.
      */
     public final static class TransformationList implements Configurable, Provenancable<ConfiguredObjectProvenance> {
+        /**
+         * The list of transformations.
+         */
         @Config(description="A list of transformations to apply.")
         public List<Transformation> list;
 
         private TransformationList() {}
 
+        /**
+         * Constructs a transformation list.
+         * @param list The list of transformations.
+         */
         public TransformationList(List<Transformation> list) {
             this.list = list;
         }
diff --git a/Core/src/main/java/org/tribuo/transform/TransformerMap.java b/Core/src/main/java/org/tribuo/transform/TransformerMap.java
index 5618b9829..624df19cb 100644
--- a/Core/src/main/java/org/tribuo/transform/TransformerMap.java
+++ b/Core/src/main/java/org/tribuo/transform/TransformerMap.java
@@ -207,6 +207,10 @@ public final static class TransformerMapProvenance implements ObjectProvenance {
             this.datasetProvenance = host.datasetProvenance;
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public TransformerMapProvenance(Map<String,Provenance> map) {
             this.className = ObjectProvenance.checkAndExtractProvenance(map,CLASS_NAME,StringProvenance.class,TransformerMapProvenance.class.getSimpleName()).getValue();
             this.transformationMapProvenance = ObjectProvenance.checkAndExtractProvenance(map,TRANSFORMATION_MAP,ConfiguredObjectProvenance.class,TransformerMapProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/transform/transformations/BinningTransformation.java b/Core/src/main/java/org/tribuo/transform/transformations/BinningTransformation.java
index d53d659dc..e8d31ffa4 100644
--- a/Core/src/main/java/org/tribuo/transform/transformations/BinningTransformation.java
+++ b/Core/src/main/java/org/tribuo/transform/transformations/BinningTransformation.java
@@ -134,6 +134,10 @@ public final static class BinningTransformationProvenance implements Transformat
             this.type = new EnumProvenance<>(TYPE,host.type);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenance map.
+         */
         @SuppressWarnings("unchecked") // Enum cast
         public BinningTransformationProvenance(Map<String,Provenance> map) {
             numBins = ObjectProvenance.checkAndExtractProvenance(map,NUM_BINS,IntProvenance.class,BinningTransformationProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/transform/transformations/IDFTransformation.java b/Core/src/main/java/org/tribuo/transform/transformations/IDFTransformation.java
index 0d9d9eae0..566441bab 100644
--- a/Core/src/main/java/org/tribuo/transform/transformations/IDFTransformation.java
+++ b/Core/src/main/java/org/tribuo/transform/transformations/IDFTransformation.java
@@ -16,6 +16,11 @@ public class IDFTransformation implements Transformation {
     
     private TransformationProvenance provenance;
 
+    /**
+     * Constructs an IDFTransformation.
+     */
+    public IDFTransformation() {}
+
     @Override
     public TransformStatistics createStats() {
         return new IDFStatistics();
@@ -72,10 +77,15 @@ public Transformer generateTransformer() {
     private static class IDFTransformer implements Transformer {
         private static final long serialVersionUID = 1L;
 
-        private double df;
-        
-        private double N;
+        private final double df;
         
+        private final double N;
+
+        /**
+         * Constructs an IDFTransformer using the supplied parameters.
+         * @param df The document frequency.
+         * @param N The number of documents.
+         */
         public IDFTransformer(int df, int N) {
             this.df = df;
             this.N = N;
@@ -96,6 +106,10 @@ public final static class IDFTransformationProvenance implements TransformationP
 
         IDFTransformationProvenance() { }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         // IDFTransformation has no state to record.
         public IDFTransformationProvenance(Map<String,Provenance> map) { }
 
diff --git a/Core/src/main/java/org/tribuo/transform/transformations/LinearScalingTransformation.java b/Core/src/main/java/org/tribuo/transform/transformations/LinearScalingTransformation.java
index 516b86f4f..1b66a2fc6 100644
--- a/Core/src/main/java/org/tribuo/transform/transformations/LinearScalingTransformation.java
+++ b/Core/src/main/java/org/tribuo/transform/transformations/LinearScalingTransformation.java
@@ -56,6 +56,11 @@ public final class LinearScalingTransformation implements Transformation {
      */
     public LinearScalingTransformation() { }
 
+    /**
+     * Constructs a LinearScalingTransformation which puts feature values into the specified range.
+     * @param targetMin The new minimum feature value.
+     * @param targetMax The new maximum feature value.
+     */
     public LinearScalingTransformation(double targetMin, double targetMax) {
         this.targetMin = targetMin;
         this.targetMax = targetMax;
@@ -99,6 +104,10 @@ public final static class LinearScalingTransformationProvenance implements Trans
             this.targetMax = new DoubleProvenance(TARGET_MAX,host.targetMax);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public LinearScalingTransformationProvenance(Map<String,Provenance> map) {
             targetMin = ObjectProvenance.checkAndExtractProvenance(map,TARGET_MIN,DoubleProvenance.class,LinearScalingTransformationProvenance.class.getSimpleName());
             targetMax = ObjectProvenance.checkAndExtractProvenance(map,TARGET_MAX,DoubleProvenance.class,LinearScalingTransformationProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/transform/transformations/MeanStdDevTransformation.java b/Core/src/main/java/org/tribuo/transform/transformations/MeanStdDevTransformation.java
index bfd5aa943..cd84a8eb0 100644
--- a/Core/src/main/java/org/tribuo/transform/transformations/MeanStdDevTransformation.java
+++ b/Core/src/main/java/org/tribuo/transform/transformations/MeanStdDevTransformation.java
@@ -56,6 +56,11 @@ public final class MeanStdDevTransformation implements Transformation {
      */
     public MeanStdDevTransformation() { }
 
+    /**
+     * Constructs a MeanStdDevTransformation targetting the specified mean and standard deviation.
+     * @param targetMean The target mean.
+     * @param targetStdDev The target standard deviation.
+     */
     public MeanStdDevTransformation(double targetMean, double targetStdDev) {
         this.targetMean = targetMean;
         this.targetStdDev = targetStdDev;
@@ -99,6 +104,10 @@ public final static class MeanStdDevTransformationProvenance implements Transfor
             this.targetStdDev = new DoubleProvenance(TARGET_STDDEV, host.targetStdDev);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public MeanStdDevTransformationProvenance(Map<String, Provenance> map) {
             targetMean = ObjectProvenance.checkAndExtractProvenance(map, TARGET_MEAN, DoubleProvenance.class, MeanStdDevTransformationProvenance.class.getSimpleName());
             targetStdDev = ObjectProvenance.checkAndExtractProvenance(map, TARGET_STDDEV, DoubleProvenance.class, MeanStdDevTransformationProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/transform/transformations/SimpleTransform.java b/Core/src/main/java/org/tribuo/transform/transformations/SimpleTransform.java
index 9ee566174..3133220b5 100644
--- a/Core/src/main/java/org/tribuo/transform/transformations/SimpleTransform.java
+++ b/Core/src/main/java/org/tribuo/transform/transformations/SimpleTransform.java
@@ -50,6 +50,9 @@ public final class SimpleTransform implements Transformer, Transformation, Trans
     private static final String OPERAND = "operand";
     private static final String SECOND_OPERAND = "secondOperand";
 
+    /**
+     * Epsilon for determining when two double values are the same.
+     */
     public static final double EPSILON = 1e-12;
 
     /**
@@ -202,6 +205,10 @@ public final static class SimpleTransformProvenance implements TransformationPro
             this.secondOperand = new DoubleProvenance(SECOND_OPERAND,host.secondOperand);
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         @SuppressWarnings("unchecked") // Enum cast
         public SimpleTransformProvenance(Map<String,Provenance> map) {
             op = ObjectProvenance.checkAndExtractProvenance(map,OP,EnumProvenance.class, SimpleTransformProvenance.class.getSimpleName());
diff --git a/Core/src/main/java/org/tribuo/util/MurmurHash3.java b/Core/src/main/java/org/tribuo/util/MurmurHash3.java
index 086acbcf7..67b8e53f2 100644
--- a/Core/src/main/java/org/tribuo/util/MurmurHash3.java
+++ b/Core/src/main/java/org/tribuo/util/MurmurHash3.java
@@ -27,10 +27,21 @@ public final class MurmurHash3 {
 
     /** 128 bits of state */
     public static final class LongPair {
+        /**
+         * First value.
+         */
         public long val1;
+        /**
+         * Second value.
+         */
         public long val2;
     }
 
+    /**
+     * 32-bit mixing function.
+     * @param h Value to mix.
+     * @return Mixed value.
+     */
     public static final int fmix32(int h) {
         h ^= h >>> 16;
         h *= 0x85ebca6b;
@@ -40,6 +51,11 @@ public static final int fmix32(int h) {
         return h;
     }
 
+    /**
+     * 64-bit mixing function.
+     * @param k Value to mix.
+     * @return Mixed value.
+     */
     public static final long fmix64(long k) {
         k ^= k >>> 33;
         k *= 0xff51afd7ed558ccdL;
@@ -126,7 +142,6 @@ public static int murmurhash3_x86_32(byte[] data, int offset, int len, int seed)
         return h1;
     }
 
-
     /**
      * Returns the MurmurHash3_x86_32 hash of the UTF-8 bytes of the String without actually encoding
      * the string to a temporary buffer.  This is more than 2x faster than hashing the result
@@ -255,7 +270,6 @@ else if (code < 0xD800 || code > 0xDFFF || pos>=end) {
         return h1;
     }
 
-
     /**
      * Returns the MurmurHash3_x64_128 hash, placing the result in "out".
      * @param key The data to hash.
diff --git a/Core/src/main/java/org/tribuo/util/Util.java b/Core/src/main/java/org/tribuo/util/Util.java
index dc2f67e5f..a4444a03b 100644
--- a/Core/src/main/java/org/tribuo/util/Util.java
+++ b/Core/src/main/java/org/tribuo/util/Util.java
@@ -515,6 +515,12 @@ public static int sampleFromCDF(double[] cdf, SplittableRandom rng) {
         }
     }
 
+    /**
+     * Generates an array of the specified length filled with the specified value.
+     * @param length The array length.
+     * @param value The array value.
+     * @return A filled array.
+     */
     public static double[] generateUniformVector(int length, double value) {
         double[] output = new double[length];
 
@@ -523,6 +529,12 @@ public static double[] generateUniformVector(int length, double value) {
         return output;
     }
 
+    /**
+     * Generates an array of the specified length filled with the specified value.
+     * @param length The array length.
+     * @param value The array value.
+     * @return A filled array.
+     */
     public static float[] generateUniformVector(int length, float value) {
         float[] output = new float[length];
 
@@ -531,6 +543,13 @@ public static float[] generateUniformVector(int length, float value) {
         return output;
     }
 
+    /**
+     * Generates a normalized version of the input array.
+     * <p>
+     * Normalization in this case means the array sums to 1.0.
+     * @param input The input array.
+     * @return A normalized probability distribution.
+     */
     public static double[] normalizeToDistribution(double[] input) {
         double[] output = new double[input.length];
         double sum = 0.0;
@@ -547,6 +566,13 @@ public static double[] normalizeToDistribution(double[] input) {
         return output;
     }
 
+    /**
+     * Generates a normalized version of the input array.
+     * <p>
+     * Normalization in this case means the array sums to 1.0.
+     * @param input The input array.
+     * @return A normalized probability distribution.
+     */
     public static double[] normalizeToDistribution(float[] input) {
         double[] output = new double[input.length];
         double sum = 0.0;
@@ -563,6 +589,13 @@ public static double[] normalizeToDistribution(float[] input) {
         return output;
     }
 
+    /**
+     * Normalizes the input array in place.
+     * <p>
+     * Normalization in this case means the array sums to 1.0.
+     * @param input The input array.
+     * @return The input array after it has been normalized into a probability distribution.
+     */
     public static double[] inplaceNormalizeToDistribution(double[] input) {
         double sum = 0.0;
 
@@ -577,6 +610,13 @@ public static double[] inplaceNormalizeToDistribution(double[] input) {
         return input;
     }
 
+    /**
+     * Normalizes the input array in place.
+     * <p>
+     * Normalization in this case means the array sums to 1.0.
+     * @param input The input array.
+     */
+    // TODO why does this not return the array and the one for doubles does?
     public static void inplaceNormalizeToDistribution(float[] input) {
         float sum = 0.0f;
 
@@ -590,6 +630,14 @@ public static void inplaceNormalizeToDistribution(float[] input) {
 
     }
 
+    /**
+     * Logs the supplied array to the supplied logger at the specified level.
+     * <p>
+     * Prints the array as tuples of index and value.
+     * @param otherLogger The logger.
+     * @param level The log level.
+     * @param input The input array to log.
+     */
     public static void logVector(Logger otherLogger, Level level, double[] input) {
         StringBuilder buffer = new StringBuilder();
 
@@ -604,6 +652,14 @@ public static void logVector(Logger otherLogger, Level level, double[] input) {
         otherLogger.log(level, buffer.toString());
     }
 
+    /**
+     * Logs the supplied array to the supplied logger at the specified level.
+     * <p>
+     * Prints the array as tuples of index and value.
+     * @param otherLogger The logger.
+     * @param level The log level.
+     * @param input The input array to log.
+     */
     public static void logVector(Logger otherLogger, Level level, float[] input) {
         StringBuilder buffer = new StringBuilder();
 
@@ -618,6 +674,11 @@ public static void logVector(Logger otherLogger, Level level, float[] input) {
         otherLogger.log(level, buffer.toString());
     }
 
+    /**
+     * Converts a boxed list of integers into an array of primitive doubles.
+     * @param input The input integer list.
+     * @return An array of doubles.
+     */
     public static double[] toPrimitiveDoubleFromInteger(List<Integer> input) {
         double[] output = new double[input.size()];
 
@@ -628,6 +689,11 @@ public static double[] toPrimitiveDoubleFromInteger(List<Integer> input) {
         return output;
     }
 
+    /**
+     * Converts a boxed list of doubles into an array of primitive doubles.
+     * @param input The input double list.
+     * @return An array of doubles.
+     */
     public static double[] toPrimitiveDouble(List<Double> input) {
         double[] output = new double[input.size()];
 
@@ -638,6 +704,11 @@ public static double[] toPrimitiveDouble(List<Double> input) {
         return output;
     }
 
+    /**
+     * Converts a boxed list of floats into an array of primitive floats.
+     * @param input The input float list.
+     * @return An array of floats.
+     */
     public static float[] toPrimitiveFloat(List<Float> input) {
         float[] output = new float[input.size()];
 
@@ -648,6 +719,11 @@ public static float[] toPrimitiveFloat(List<Float> input) {
         return output;
     }
 
+    /**
+     * Converts a boxed list of integers into an array of primitive ints.
+     * @param input The input integer list.
+     * @return An array of ints.
+     */
     public static int[] toPrimitiveInt(List<Integer> input) {
         int[] output = new int[input.size()];
 
@@ -658,6 +734,11 @@ public static int[] toPrimitiveInt(List<Integer> input) {
         return output;
     }
 
+    /**
+     * Converts a boxed list of longs into an array of primitive longs.
+     * @param input The input long list.
+     * @return An array of longs.
+     */
     public static long[] toPrimitiveLong(List<Long> input) {
         long[] output = new long[input.size()];
 
@@ -668,6 +749,13 @@ public static long[] toPrimitiveLong(List<Long> input) {
         return output;
     }
 
+    /**
+     * Samples an array of ints from the supplied rng in [0,range).
+     * @param rng The RNG to sample from.
+     * @param size The number of samples to draw.
+     * @param range The range of the samples.
+     * @return A sampled int array.
+     */
     public static int[] sampleInts(Random rng, int size, int range) {
         int[] output = new int[size];
 
@@ -678,38 +766,68 @@ public static int[] sampleInts(Random rng, int size, int range) {
         return output;
     }
 
+    /**
+     * Adds update to input in place.
+     * @param input The input array.
+     * @param update The array to add.
+     */
     public static void inPlaceAdd(double[] input, double[] update) {
         for (int i = 0; i < input.length; i++) {
             input[i] += update[i];
         }
     }
 
+    /**
+     * Subtracts update from input in place.
+     * @param input The input array.
+     * @param update The array to subtract.
+     */
     public static void inPlaceSubtract(double[] input, double[] update) {
         for (int i = 0; i < input.length; i++) {
             input[i] -= update[i];
         }
     }
 
+    /**
+     * Adds update to input in place.
+     * @param input The input array.
+     * @param update The array to add.
+     */
     public static void inPlaceAdd(float[] input, float[] update) {
         for (int i = 0; i < input.length; i++) {
             input[i] += update[i];
         }
     }
 
+    /**
+     * Subtracts update from input in place.
+     * @param input The input array.
+     * @param update The array to subtract.
+     */
     public static void inPlaceSubtract(float[] input, float[] update) {
         for (int i = 0; i < input.length; i++) {
             input[i] -= update[i];
         }
     }
 
+    /**
+     * Computes the vector two-norm.
+     * @param input The input vector.
+     * @return The euclidean norm of the vector.
+     */
     public static double vectorNorm(double[] input) {
         double norm = 0.0;
         for (double d : input) {
             norm += d * d;
         }
-        return norm;
+        return Math.sqrt(norm);
     }
 
+    /**
+     * Computes the sum of the input vector.
+     * @param input The vector to sum.
+     * @return The vector sum.
+     */
     public static double sum(double[] input) {
         double sum = 0.0;
         for (double d : input) {
@@ -718,6 +836,11 @@ public static double sum(double[] input) {
         return sum;
     }
 
+    /**
+     * Computes the sum of the input vector.
+     * @param input The vector to sum.
+     * @return The vector sum.
+     */
     public static float sum(float[] input) {
         float sum = 0.0f;
         for (float d : input) {
@@ -726,6 +849,12 @@ public static float sum(float[] input) {
         return sum;
     }
 
+    /**
+     * Computes the sum of the input vector up to length elements.
+     * @param array The vector to sum.
+     * @param length The number of elements to sum.
+     * @return The vector sum.
+     */
     public static double sum(double[] array, int length) {
         double sum = 0.0;
         for (int i = 0; i < length; i++) {
@@ -734,6 +863,12 @@ public static double sum(double[] array, int length) {
         return sum;
     }
 
+    /**
+     * Computes the sum of the input vector up to length elements.
+     * @param array The vector to sum.
+     * @param length The number of elements to sum.
+     * @return The vector sum.
+     */
     public static float sum(float[] array, int length) {
         float sum = 0.0f;
         for (int i = 0; i < length; i++) {
@@ -742,6 +877,13 @@ public static float sum(float[] array, int length) {
         return sum;
     }
 
+    /**
+     * Computes the sum of the specified indices in the input array.
+     * @param indices The indices to sum.
+     * @param indicesLength The number of elements of indices to sum over.
+     * @param input The input array to sum the values of.
+     * @return The sum of the input array.
+     */
     public static float sum(int[] indices, int indicesLength, float[] input) {
         float sum = 0.0f;
         for (int i = 0; i < indicesLength; i++) {
@@ -750,10 +892,22 @@ public static float sum(int[] indices, int indicesLength, float[] input) {
         return sum;
     }
 
+    /**
+     * Computes the sum of the specified indices in the input array.
+     * @param indices The indices to sum.
+     * @param input The input array to sum the values of.
+     * @return The sum of the input array.
+     */
     public static float sum(int[] indices, float[] input) {
         return sum(indices,indices.length,input);
     }
 
+    /**
+     * Generates a float vector of the specified length filled with the specified value.
+     * @param length The vector length.
+     * @param value The initial value.
+     * @return A float vector initialised with the value.
+     */
     public static float[] generateUniformFloatVector(int length, float value) {
         float[] output = new float[length];
 
@@ -936,11 +1090,23 @@ public static double mean(double[] inputs) {
         return output / inputs.length;
     }
 
+    /**
+     * Computes the mean of the first length elements of array.
+     * @param array The array to take the mean of.
+     * @param length The number of elements to take the mean over.
+     * @return The mean.
+     */
     public static double mean(double[] array, int length) {
         double sum = sum(array,length);
         return sum / length;
     }
 
+    /**
+     * Computes the mean of the collection.
+     * @param values The collection.
+     * @param <V> The collection type.
+     * @return The mean of the collection.
+     */
     public static <V extends Number> double mean(Collection<V> values) {
         double total = 0d;
         for (V v : values) {
@@ -949,6 +1115,12 @@ public static <V extends Number> double mean(Collection<V> values) {
         return total / values.size();
     }
 
+    /**
+     * Computes the sample variance of the collection.
+     * @param values The collection.
+     * @param <V> The collection type.
+     * @return The sample variance of the collection.
+     */
     public static <V extends Number> double sampleVariance(Collection<V> values) {
         double mean = mean(values);
         double total = 0d;
@@ -958,15 +1130,35 @@ public static <V extends Number> double sampleVariance(Collection<V> values) {
         return total / (values.size() - 1);
     }
 
+    /**
+     * Computes the sample standard deviation of the collection.
+     * @param values The collection.
+     * @param <V> The collection type.
+     * @return The sample standard deviation.
+     */
     public static <V extends Number> double sampleStandardDeviation(Collection<V> values) {
         return Math.sqrt(sampleVariance(values));
     }
 
+    /**
+     * Computes the weighted mean of the first length elements of the array.
+     * @param array The array to mean.
+     * @param weights The element weights.
+     * @param length The number of elements to use.
+     * @return The weighted mean.
+     */
     public static double weightedMean(double[] array, float[] weights, int length) {
         double sum = weightedSum(array,weights,length);
         return sum / sum(weights,length);
     }
 
+    /**
+     * Computes the weighted sum of the first length elements of the array.
+     * @param array The array to sum.
+     * @param weights The element weights.
+     * @param length The number of elements to use.
+     * @return The weighted sum.
+     */
     public static double weightedSum(double[] array, float[] weights, int length) {
         if (array.length != weights.length) {
             throw new IllegalArgumentException("array and weights must be the same length, array.length = " + array.length + ", weights.length = " + weights.length);
diff --git a/Data/src/main/java/org/tribuo/data/CompletelyConfigurableTrainTest.java b/Data/src/main/java/org/tribuo/data/CompletelyConfigurableTrainTest.java
index f96127cdb..536c7b41d 100644
--- a/Data/src/main/java/org/tribuo/data/CompletelyConfigurableTrainTest.java
+++ b/Data/src/main/java/org/tribuo/data/CompletelyConfigurableTrainTest.java
@@ -60,22 +60,40 @@ public String getOptionsDescription() {
             return "Loads a Trainer and two DataSources from a config file, trains a Model, tests it and optionally saves it to disk.";
         }
 
-        @Option(charName='f',longName="model-output-path",usage="Path to serialize model to.")
+        /**
+         * Path to serialize model to.
+         */
+        @Option(charName = 'f', longName = "model-output-path", usage = "Path to serialize model to.")
         public Path outputPath;
 
-        @Option(charName='u',longName="train-source",usage="Load the training DataSource from the config file.")
+        /**
+         * Load the training DataSource from the config file.
+         */
+        @Option(charName = 'u', longName = "train-source", usage = "Load the training DataSource from the config file.")
         public ConfigurableDataSource<?> trainSource;
 
-        @Option(charName='v',longName="test-source",usage="Load the testing DataSource from the config file.")
+        /**
+         * Load the testing DataSource from the config file.
+         */
+        @Option(charName = 'v', longName = "test-source", usage = "Load the testing DataSource from the config file.")
         public ConfigurableDataSource<?> testSource;
 
-        @Option(charName='t',longName="trainer",usage="Load a trainer from the config file.")
+        /**
+         * Load a trainer from the config file.
+         */
+        @Option(charName = 't', longName = "trainer", usage = "Load a trainer from the config file.")
         public Trainer<?> trainer;
 
-        @Option(longName="transformer",usage="Load a transformation map from the config file.")
+        /**
+         * Load a transformation map from the config file.
+         */
+        @Option(longName = "transformer", usage = "Load a transformation map from the config file.")
         public TransformationMap transformationMap;
 
-        @Option(charName='m',longName="minimum-count",usage="Remove features which occur fewer than <int> times.")
+        /**
+         * Remove features which occur fewer than m times.
+         */
+        @Option(charName = 'm', longName = "minimum-count", usage = "Remove features which occur fewer than <int> times.")
         public int minCount = -1;
     }
 
diff --git a/Data/src/main/java/org/tribuo/data/ConfigurableTrainTest.java b/Data/src/main/java/org/tribuo/data/ConfigurableTrainTest.java
index e9cc0bbba..cbe51a02f 100644
--- a/Data/src/main/java/org/tribuo/data/ConfigurableTrainTest.java
+++ b/Data/src/main/java/org/tribuo/data/ConfigurableTrainTest.java
@@ -64,21 +64,39 @@ public String getOptionsDescription() {
             return "Loads a Trainer from a config file, trains a Model (optionally with cross-validation), tests it and optionally saves it to disk.";
         }
 
+        /**
+         * Data loading options.
+         */
         public DataOptions general;
 
-        @Option(charName='t',longName="trainer",usage="Load a trainer from the config file.")
+        /**
+         * Load a trainer from the config file.
+         */
+        @Option(charName = 't', longName = "trainer", usage = "Load a trainer from the config file.")
         public Trainer<?> trainer;
 
-        @Option(longName="transformer",usage="Load a transformation map from the config file.")
+        /**
+         * Load a transformation map from the config file.
+         */
+        @Option(longName = "transformer", usage = "Load a transformation map from the config file.")
         public TransformationMap transformationMap;
 
-        @Option(charName='a',longName="output-factory",usage="The output factory to construct.")
+        /**
+         * The output factory to construct.
+         */
+        @Option(charName = 'a', longName = "output-factory", usage = "The output factory to construct.")
         public OutputFactory<?> outputFactory;
 
-        @Option(charName='x',longName="cross-validate",usage="Cross-validate the output metrics.")
+        /**
+         * Cross-validate the output metrics.
+         */
+        @Option(charName = 'x', longName = "cross-validate", usage = "Cross-validate the output metrics.")
         public boolean crossValidation;
 
-        @Option(charName='n',longName="num-folds",usage="The number of cross validation folds.")
+        /**
+         * The number of cross validation folds.
+         */
+        @Option(charName = 'n', longName = "num-folds", usage = "The number of cross validation folds.")
         public int numFolds = 5;
     }
 
diff --git a/Data/src/main/java/org/tribuo/data/DataOptions.java b/Data/src/main/java/org/tribuo/data/DataOptions.java
index b917dcd27..a897b54f0 100644
--- a/Data/src/main/java/org/tribuo/data/DataOptions.java
+++ b/Data/src/main/java/org/tribuo/data/DataOptions.java
@@ -58,16 +58,50 @@ public final class DataOptions implements Options {
      * The input formats supported by this options object.
      */
     public enum InputFormat {
-        SERIALIZED, LIBSVM, TEXT, CSV, COLUMNAR
+        /**
+         * Serialized Tribuo datasets.
+         */
+        SERIALIZED,
+        /**
+         * LibSVM format data.
+         */
+        LIBSVM,
+        /**
+         * Text data in Tribuo's default text format "output ## text".
+         */
+        TEXT,
+        /**
+         * Simple numeric CSV file.
+         */
+        CSV,
+        /**
+         * A CSV file using a {@link RowProcessor}.
+         */
+        COLUMNAR
     }
 
     /**
      * The delimiters supported by CSV files in this options object.
      */
     public enum Delimiter {
-        COMMA(','), TAB('\t'), SEMICOLON(';');
+        /**
+         * Comma separator.
+         */
+        COMMA(','),
+        /**
+         * Tab separator.
+         */
+        TAB('\t'),
+        /**
+         * Semicolon separator.
+         */
+        SEMICOLON(';');
 
+        /**
+         * The delimiter character.
+         */
         public final char value;
+
         Delimiter(char value) {
             this.value = value;
         }
@@ -78,34 +112,80 @@ public String getOptionsDescription() {
         return "Options for loading and processing train and test data.";
     }
 
-    @Option(longName="hashing-dimension",usage="Hashing dimension used for standard text format.")
+    /**
+     * Hashing dimension used for standard text format.
+     */
+    @Option(longName = "hashing-dimension", usage = "Hashing dimension used for standard text format.")
     public int hashDim = 0;
-    @Option(longName="ngram",usage="Ngram size to generate when using standard text format.")
+    /**
+     * Ngram size to generate when using standard text format.
+     */
+    @Option(longName = "ngram", usage = "Ngram size to generate when using standard text format.")
     public int ngram = 2;
-    @Option(longName="term-counting",usage="Use term counts instead of boolean when using the standard text format.")
+    /**
+     * Use term counts instead of boolean when using the standard text format.
+     */
+    @Option(longName = "term-counting", usage = "Use term counts instead of boolean when using the standard text format.")
     public boolean termCounting;
-    @Option(charName='f',longName="model-output-path",usage="Path to serialize model to.")
+    /**
+     * Path to serialize model to.
+     */
+    @Option(charName = 'f', longName = "model-output-path", usage = "Path to serialize model to.")
     public Path outputPath;
-    @Option(charName='r',longName="seed",usage="RNG seed.")
+    /**
+     * RNG seed.
+     */
+    @Option(charName = 'r', longName = "seed", usage = "RNG seed.")
     public long seed = Trainer.DEFAULT_SEED;
-    @Option(charName='s',longName="input-format",usage="Loads the data using the specified format.")
+    /**
+     * Loads the data using the specified format.
+     */
+    @Option(charName = 's', longName = "input-format", usage = "Loads the data using the specified format.")
     public InputFormat inputFormat = InputFormat.LIBSVM;
-    @Option(longName="csv-response-name",usage="Response name in the csv file.")
+    /**
+     * Response name in the csv file.
+     */
+    @Option(longName = "csv-response-name", usage = "Response name in the csv file.")
     public String csvResponseName;
-    @Option(longName="csv-delimiter",usage="Delimiter")
+    /**
+     * Delimiter
+     */
+    @Option(longName = "csv-delimiter", usage = "Delimiter")
     public Delimiter delimiter = Delimiter.COMMA;
-    @Option(longName="csv-quote-char",usage="Quote character in the CSV file.")
+    /**
+     * Quote character in the CSV file.
+     */
+    @Option(longName = "csv-quote-char", usage = "Quote character in the CSV file.")
     public char csvQuoteChar = '"';
-    @Option(longName="columnar-row-processor",usage="The name of the row processor from the config file.")
+    /**
+     * The name of the row processor from the config file.
+     */
+    @Option(longName = "columnar-row-processor", usage = "The name of the row processor from the config file.")
     public RowProcessor<?> rowProcessor;
-    @Option(longName="min-count",usage="Minimum cardinality of the features.")
+    /**
+     * Minimum cardinality of the features.
+     */
+    @Option(longName = "min-count", usage = "Minimum cardinality of the features.")
     public int minCount = 0;
-    @Option(charName='u',longName="training-file",usage="Path to the training file.")
+    /**
+     * Path to the training file.
+     */
+    @Option(charName = 'u', longName = "training-file", usage = "Path to the training file.")
     public Path trainingPath;
-    @Option(charName='v',longName="testing-file",usage="Path to the testing file.")
+    /**
+     * Path to the testing file.
+     */
+    @Option(charName = 'v', longName = "testing-file", usage = "Path to the testing file.")
     public Path testingPath;
 
-    public <T extends Output<T>> Pair<Dataset<T>,Dataset<T>> load(OutputFactory<T> outputFactory) throws IOException {
+    /**
+     * Loads the datasets specified in this options.
+     * @param outputFactory The output factory to use.
+     * @param <T> The type of the dataset.
+     * @return A pair of training and test datasets.
+     * @throws IOException If the datasets failed to load.
+     */
+    public <T extends Output<T>> Pair<Dataset<T>, Dataset<T>> load(OutputFactory<T> outputFactory) throws IOException {
         logger.info(String.format("Loading data from %s", trainingPath));
         Dataset<T> train;
         Dataset<T> test;
@@ -123,13 +203,13 @@ public <T extends Output<T>> Pair<Dataset<T>,Dataset<T>> load(OutputFactory<T> o
                     if (minCount > 0) {
                         logger.info("Found " + train.getFeatureIDMap().size() + " features");
                         logger.info("Removing features that occur fewer than " + minCount + " times.");
-                        train = new MinimumCardinalityDataset<>(train,minCount);
+                        train = new MinimumCardinalityDataset<>(train, minCount);
                     }
                     logger.info(String.format("Loaded %d training examples for %s", train.size(), train.getOutputs().toString()));
                     logger.info("Found " + train.getFeatureIDMap().size() + " features, and " + train.getOutputInfo().size() + " response dimensions");
                     @SuppressWarnings("unchecked")
                     Dataset<T> deserTest = (Dataset<T>) oits.readObject();
-                    test = new ImmutableDataset<>(deserTest,deserTest.getSourceProvenance(),deserTest.getOutputFactory(),train.getFeatureIDMap(),train.getOutputIDInfo(),true);
+                    test = new ImmutableDataset<>(deserTest, deserTest.getSourceProvenance(), deserTest.getOutputFactory(), train.getFeatureIDMap(), train.getOutputIDInfo(), true);
                 } catch (ClassNotFoundException e) {
                     throw new IllegalArgumentException("Unknown class in serialised files", e);
                 }
@@ -137,17 +217,17 @@ public <T extends Output<T>> Pair<Dataset<T>,Dataset<T>> load(OutputFactory<T> o
             case LIBSVM:
                 //
                 // Load the libsvm text-based data format.
-                LibSVMDataSource<T> trainSVMSource = new LibSVMDataSource<>(trainingPath,outputFactory);
+                LibSVMDataSource<T> trainSVMSource = new LibSVMDataSource<>(trainingPath, outputFactory);
                 train = new MutableDataset<>(trainSVMSource);
                 boolean zeroIndexed = trainSVMSource.isZeroIndexed();
                 int maxFeatureID = trainSVMSource.getMaxFeatureID();
                 if (minCount > 0) {
                     logger.info("Removing features that occur fewer than " + minCount + " times.");
-                    train = new MinimumCardinalityDataset<>(train,minCount);
+                    train = new MinimumCardinalityDataset<>(train, minCount);
                 }
                 logger.info(String.format("Loaded %d training examples for %s", train.size(), train.getOutputs().toString()));
                 logger.info("Found " + train.getFeatureIDMap().size() + " features, and " + train.getOutputInfo().size() + " response dimensions");
-                test = new ImmutableDataset<>(new LibSVMDataSource<>(testingPath,outputFactory,zeroIndexed,maxFeatureID), train.getFeatureIDMap(), train.getOutputIDInfo(), false);
+                test = new ImmutableDataset<>(new LibSVMDataSource<>(testingPath, outputFactory, zeroIndexed, maxFeatureID), train.getFeatureIDMap(), train.getOutputIDInfo(), false);
                 break;
             case TEXT:
                 //
@@ -163,7 +243,7 @@ public <T extends Output<T>> Pair<Dataset<T>,Dataset<T>> load(OutputFactory<T> o
                 train = new MutableDataset<>(trainSource);
                 if (minCount > 0) {
                     logger.info("Removing features that occur fewer than " + minCount + " times.");
-                    train = new MinimumCardinalityDataset<>(train,minCount);
+                    train = new MinimumCardinalityDataset<>(train, minCount);
                 }
 
                 logger.info(String.format("Loaded %d training examples for %s", train.size(), train.getOutputs().toString()));
@@ -179,11 +259,11 @@ public <T extends Output<T>> Pair<Dataset<T>,Dataset<T>> load(OutputFactory<T> o
                     throw new IllegalArgumentException("Please supply a response column name");
                 }
                 separator = delimiter.value;
-                CSVLoader<T> loader = new CSVLoader<>(separator,outputFactory);
-                train = new MutableDataset<>(loader.loadDataSource(trainingPath,csvResponseName));
+                CSVLoader<T> loader = new CSVLoader<>(separator, outputFactory);
+                train = new MutableDataset<>(loader.loadDataSource(trainingPath, csvResponseName));
                 logger.info(String.format("Loaded %d training examples for %s", train.size(), train.getOutputs().toString()));
                 logger.info("Found " + train.getFeatureIDMap().size() + " features, and " + train.getOutputInfo().size() + " response dimensions");
-                test = new MutableDataset<>(loader.loadDataSource(testingPath,csvResponseName));
+                test = new MutableDataset<>(loader.loadDataSource(testingPath, csvResponseName));
                 break;
             case COLUMNAR:
                 if (rowProcessor == null) {
@@ -196,24 +276,28 @@ public <T extends Output<T>> Pair<Dataset<T>,Dataset<T>> load(OutputFactory<T> o
                 @SuppressWarnings("unchecked") // checked by the if statement above
                 RowProcessor<T> typedRowProcessor = (RowProcessor<T>) rowProcessor;
                 separator = delimiter.value;
-                train = new MutableDataset<>(new CSVDataSource<>(trainingPath,typedRowProcessor,true,separator,csvQuoteChar));
+                train = new MutableDataset<>(new CSVDataSource<>(trainingPath, typedRowProcessor, true, separator, csvQuoteChar));
                 logger.info(String.format("Loaded %d training examples for %s", train.size(), train.getOutputs().toString()));
                 logger.info("Found " + train.getFeatureIDMap().size() + " features, and " + train.getOutputInfo().size() + " response dimensions");
-                test = new MutableDataset<>(new CSVDataSource<>(testingPath,typedRowProcessor,true,separator,csvQuoteChar));
+                test = new MutableDataset<>(new CSVDataSource<>(testingPath, typedRowProcessor, true, separator, csvQuoteChar));
                 break;
             default:
                 throw new IllegalArgumentException("Unsupported input format " + inputFormat);
         }
         logger.info(String.format("Loaded %d testing examples", test.size()));
-        return new Pair<>(train,test);
+        return new Pair<>(train, test);
     }
 
+    /**
+     * Saves the model out to the path specified in this options.
+     * @param model The model to save.
+     * @param <T> The type of the model output.
+     * @throws IOException If the model could not be saved.
+     */
     public <T extends Output<T>> void saveModel(Model<T> model) throws IOException {
-        FileOutputStream fout = new FileOutputStream(outputPath.toFile());
-        ObjectOutputStream oout = new ObjectOutputStream(fout);
-        oout.writeObject(model);
-        oout.close();
-        fout.close();
-        logger.info("Serialized model to file: " + outputPath);
+        try (ObjectOutputStream objOut = new ObjectOutputStream(new FileOutputStream(outputPath.toFile()))) {
+            objOut.writeObject(model);
+            logger.info("Serialized model to file: " + outputPath);
+        }
     }
 }
diff --git a/Data/src/main/java/org/tribuo/data/DatasetExplorer.java b/Data/src/main/java/org/tribuo/data/DatasetExplorer.java
index 7cc0fd097..1a8be5d5d 100644
--- a/Data/src/main/java/org/tribuo/data/DatasetExplorer.java
+++ b/Data/src/main/java/org/tribuo/data/DatasetExplorer.java
@@ -45,10 +45,13 @@
 public final class DatasetExplorer implements CommandGroup {
     private static final Logger logger = Logger.getLogger(DatasetExplorer.class.getName());
 
-    protected CommandInterpreter shell;
+    private final CommandInterpreter shell;
 
     private Dataset<?> dataset;
 
+    /**
+     * Constructs a dataset explorer.
+     */
     public DatasetExplorer() {
         shell = new CommandInterpreter();
         shell.setPrompt("dataset sh% ");
@@ -64,6 +67,10 @@ public String getDescription() {
         return "Commands for inspecting a Dataset.";
     }
 
+    /**
+     * The filename completer.
+     * @return The completer array.
+     */
     public Completer[] fileCompleter() {
         return new Completer[]{
                 new Completers.FileNameCompleter(),
@@ -79,6 +86,12 @@ public void startShell() {
         shell.start();
     }
 
+    /**
+     * Loads a serialized dataset.
+     * @param ci The command interpreter.
+     * @param path The path to load.
+     * @return A status string.
+     */
     @Command(usage = "<filename> - Load a dataset from disk.", completers="fileCompleter")
     public String loadDataset(CommandInterpreter ci, File path) {
         try (ObjectInputStream ois = new ObjectInputStream(new BufferedInputStream(new FileInputStream(path)))) {
@@ -97,6 +110,12 @@ public String loadDataset(CommandInterpreter ci, File path) {
         return "Loaded dataset from path " + path.toString();
     }
 
+    /**
+     * Shows information on a particular feature.
+     * @param ci The command interpreter.
+     * @param featureName The feature name.
+     * @return The feature information.
+     */
     @Command(usage="Shows the information on a particular feature")
     public String featureInfo(CommandInterpreter ci, String featureName) {
         VariableInfo f = dataset.getFeatureMap().get(featureName);
@@ -107,21 +126,42 @@ public String featureInfo(CommandInterpreter ci, String featureName) {
         }
     }
 
+    /**
+     * Shows the output information.
+     * @param ci The command interpreter.
+     * @return The output information.
+     */
     @Command(usage="Shows the output information.")
     public String outputInfo(CommandInterpreter ci) {
         return dataset.getOutputInfo().toReadableString();
     }
 
+    /**
+     * Shows the number of examples in this dataset.
+     * @param ci The command interpreter.
+     * @return The number of examples.
+     */
     @Command(usage="Shows the number of rows in the dataset")
     public String numExamples(CommandInterpreter ci) {
         return ""+dataset.getData().size();
     }
 
+    /**
+     * Shows the number of features in this dataset.
+     * @param ci The command interpreter.
+     * @return The number of features.
+     */
     @Command(usage="Shows the number of features in the dataset")
     public String numFeatures(CommandInterpreter ci) {
         return ""+dataset.getFeatureMap().size();
     }
 
+    /**
+     * Shows the number of features which occurred more than minCount times in the dataset.
+     * @param ci The command interpreter.
+     * @param minCount The minimum occurrence count.
+     * @return The number of features which occurred more than minCount times.
+     */
     @Command(usage="<min count> - Shows the number of features that occurred more than min count times.")
     public String minCount(CommandInterpreter ci, int minCount) {
         int counter = 0;
@@ -133,11 +173,22 @@ public String minCount(CommandInterpreter ci, int minCount) {
         return counter + " features occurred more than " + minCount + " times.";
     }
 
+    /**
+     * Shows the output statistics.
+     * @param ci The command interpreter.
+     * @return The output statistics string.
+     */
     @Command(usage="Shows the output statistics")
-    public String showLabelStats(CommandInterpreter ci) {
-        return "Label histogram : \n" + dataset.getOutputInfo().toReadableString();
+    public String showOutputStats(CommandInterpreter ci) {
+        return "Output statistics: \n" + dataset.getOutputInfo().toReadableString();
     }
 
+    /**
+     * Saves out the dataset as a CSV file.
+     * @param ci The command interpreter.
+     * @param path The path to save to.
+     * @return A status message.
+     */
     @Command(usage="Saves out the data as a CSV.")
     public String saveCSV(CommandInterpreter ci, String path) {
         CSVSaver saver = new CSVSaver();
@@ -150,6 +201,11 @@ public String saveCSV(CommandInterpreter ci, String path) {
         }
     }
 
+    /**
+     * Shows the dataset provenance.
+     * @param ci The command interpreter.
+     * @return The dataset provenance string.
+     */
     @Command(usage="Shows the dataset provenance")
     public String showProvenance(CommandInterpreter ci) {
         return dataset.getProvenance().toString();
@@ -159,10 +215,17 @@ public String showProvenance(CommandInterpreter ci) {
      * Command line options.
      */
     public static class DatasetExplorerOptions implements Options {
-        @Option(charName='f',longName="filename",usage="Dataset file to load. Optional.")
+        /**
+         * Dataset file to load. Optional.
+         */
+        @Option(charName = 'f', longName = "filename", usage = "Dataset file to load. Optional.")
         public String modelFilename;
     }
 
+    /**
+     * Runs a dataset explorer.
+     * @param args CLI arguments.
+     */
     public static void main(String[] args) {
         DatasetExplorerOptions options = new DatasetExplorerOptions();
         ConfigurationManager cm = new ConfigurationManager(args,options,false);
diff --git a/Data/src/main/java/org/tribuo/data/PreprocessAndSerialize.java b/Data/src/main/java/org/tribuo/data/PreprocessAndSerialize.java
index cdfc066af..1a98e868e 100644
--- a/Data/src/main/java/org/tribuo/data/PreprocessAndSerialize.java
+++ b/Data/src/main/java/org/tribuo/data/PreprocessAndSerialize.java
@@ -45,12 +45,22 @@ private PreprocessAndSerialize() {}
      * Command line options.
      */
     public static class PreprocessAndSerializeOptions implements Options {
-        @Option(charName='d', longName="dataSource", usage="Datasource to load from a config file")
+        /**
+         * Datasource to load from a config file
+         */
+        @Option(charName = 'd', longName = "dataSource", usage = "Datasource to load from a config file")
         public ConfigurableDataSource<? extends Output<?>> dataSource;
-        @Option(charName='o', longName="serialized-dataset", usage="path to serialize the dataset")
+        /**
+         * path to serialize the dataset
+         */
+        @Option(charName = 'o', longName = "serialized-dataset", usage = "path to serialize the dataset")
         public Path output;
     }
 
+    /**
+     * Run the PreprocessAndSerialize CLI.
+     * @param args The CLI args.
+     */
     public static void main(String[] args) {
 
         LabsLogFormatter.setAllLogFormatters();
diff --git a/Data/src/main/java/org/tribuo/data/columnar/ColumnarFeature.java b/Data/src/main/java/org/tribuo/data/columnar/ColumnarFeature.java
index 84266486a..3768d6289 100644
--- a/Data/src/main/java/org/tribuo/data/columnar/ColumnarFeature.java
+++ b/Data/src/main/java/org/tribuo/data/columnar/ColumnarFeature.java
@@ -29,8 +29,14 @@
 public class ColumnarFeature extends Feature {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * The string used as the field name of conjunction features.
+     */
     public static final String CONJUNCTION = "CONJ";
 
+    /**
+     * The joiner between the field name and feature name.
+     */
     public static final String JOINER = "@";
 
     private final String fieldName;
diff --git a/Data/src/main/java/org/tribuo/data/columnar/ColumnarIterator.java b/Data/src/main/java/org/tribuo/data/columnar/ColumnarIterator.java
index 4e80c369c..1a090c598 100644
--- a/Data/src/main/java/org/tribuo/data/columnar/ColumnarIterator.java
+++ b/Data/src/main/java/org/tribuo/data/columnar/ColumnarIterator.java
@@ -36,7 +36,13 @@
 public abstract class ColumnarIterator extends IOSpliterator<ColumnarIterator.Row> implements Iterator<ColumnarIterator.Row> {
     private static final Logger logger = Logger.getLogger(ColumnarIterator.class.getName());
 
+    /**
+     * The column headers for this iterator.
+     */
     protected List<String> fields;
+    /**
+     * The current row.
+     */
     protected Optional<Row> currentRow = Optional.empty();
 
     /**
@@ -50,20 +56,38 @@ public static class Row {
         private final List<String> fields;
         private final Map<String, String> rowData;
 
+        /**
+         * Constructs a row from a columnar source.
+         * @param index The row index.
+         * @param fields The field names, assumed to be immutable.
+         * @param rowData The row data.
+         */
         public Row(long index, List<String> fields, Map<String, String> rowData) {
             this.index = index;
             this.fields = fields;
             this.rowData = Collections.unmodifiableMap(rowData);
         }
 
+        /**
+         * Gets the field headers.
+         * @return The field headers.
+         */
         public List<String> getFields() {
             return fields;
         }
 
+        /**
+         * Gets the row index.
+         * @return The row index.
+         */
         public long getIndex() {
             return index;
         }
 
+        /**
+         * Gets the row data.
+         * @return The row data.
+         */
         public Map<String, String> getRowData() {
             return rowData;
         }
diff --git a/Data/src/main/java/org/tribuo/data/columnar/processors/field/DateFieldProcessor.java b/Data/src/main/java/org/tribuo/data/columnar/processors/field/DateFieldProcessor.java
index e6a9bf3b9..370cf5767 100644
--- a/Data/src/main/java/org/tribuo/data/columnar/processors/field/DateFieldProcessor.java
+++ b/Data/src/main/java/org/tribuo/data/columnar/processors/field/DateFieldProcessor.java
@@ -112,6 +112,11 @@ private DateFeatureType(ToIntFunction<LocalDate> func) {
             this.extractionFunction = func;
         }
 
+        /**
+         * Applies this enum's extraction function to the supplied date.
+         * @param date The date to extract from.
+         * @return The output of the extraction function.
+         */
         public int extract(LocalDate date) {
             return extractionFunction.applyAsInt(date);
         }
diff --git a/Data/src/main/java/org/tribuo/data/columnar/processors/field/RegexFieldProcessor.java b/Data/src/main/java/org/tribuo/data/columnar/processors/field/RegexFieldProcessor.java
index 6c723ce78..92902eccc 100644
--- a/Data/src/main/java/org/tribuo/data/columnar/processors/field/RegexFieldProcessor.java
+++ b/Data/src/main/java/org/tribuo/data/columnar/processors/field/RegexFieldProcessor.java
@@ -51,8 +51,17 @@ public class RegexFieldProcessor implements FieldProcessor {
      * Matching mode.
      */
     public enum Mode {
+        /**
+         * Triggers feature generation if the whole string matches.
+         */
         MATCH_ALL,
+        /**
+         * Triggers feature generation if the string contains a match.
+         */
         MATCH_CONTAINS,
+        /**
+         * Triggers feature generation for each matching group in the string.
+         */
         GROUPS
     }
 
diff --git a/Data/src/main/java/org/tribuo/data/columnar/processors/response/BinaryResponseProcessor.java b/Data/src/main/java/org/tribuo/data/columnar/processors/response/BinaryResponseProcessor.java
index 62cff6e06..8da58bdf3 100644
--- a/Data/src/main/java/org/tribuo/data/columnar/processors/response/BinaryResponseProcessor.java
+++ b/Data/src/main/java/org/tribuo/data/columnar/processors/response/BinaryResponseProcessor.java
@@ -47,7 +47,14 @@ public class BinaryResponseProcessor<T extends Output<T>> implements ResponsePro
     @Config(mandatory = true, description="Output factory to use to create the response.")
     private OutputFactory<T> outputFactory;
 
+    /**
+     * The default positive response.
+     */
     public static final String POSITIVE_NAME = "1";
+
+    /**
+     * The default negative response.
+     */
     public static final String NEGATIVE_NAME = "0";
 
     @Config(description="The positive response to emit.")
diff --git a/Data/src/main/java/org/tribuo/data/columnar/processors/response/EmptyResponseProcessor.java b/Data/src/main/java/org/tribuo/data/columnar/processors/response/EmptyResponseProcessor.java
index 0db9b878e..180812cb5 100644
--- a/Data/src/main/java/org/tribuo/data/columnar/processors/response/EmptyResponseProcessor.java
+++ b/Data/src/main/java/org/tribuo/data/columnar/processors/response/EmptyResponseProcessor.java
@@ -38,6 +38,9 @@
  */
 public final class EmptyResponseProcessor<T extends Output<T>> implements ResponseProcessor<T> {
 
+    /**
+     * The field name this response processor looks for, which is ignored anyway as this processor always returns {@link Optional#empty()}.
+     */
     public static final String FIELD_NAME = "TRIBUO##NULL_RESPONSE_PROCESSOR";
 
     @Config(mandatory = true,description="Output factory to type the columnar loader.")
diff --git a/Data/src/main/java/org/tribuo/data/csv/CSVDataSource.java b/Data/src/main/java/org/tribuo/data/csv/CSVDataSource.java
index 5f885079b..dc3fd4978 100644
--- a/Data/src/main/java/org/tribuo/data/csv/CSVDataSource.java
+++ b/Data/src/main/java/org/tribuo/data/csv/CSVDataSource.java
@@ -305,6 +305,10 @@ <T extends Output<T>> CSVDataSourceProvenance(CSVDataSource<T> host) {
             this.sha256Hash = new HashProvenance(DEFAULT_HASH_TYPE,RESOURCE_HASH,ProvenanceUtil.hashResource(DEFAULT_HASH_TYPE,host.dataPath));
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public CSVDataSourceProvenance(Map<String,Provenance> map) {
             this(extractProvenanceInfo(map));
         }
@@ -316,6 +320,11 @@ private CSVDataSourceProvenance(ExtractedInfo info) {
             this.sha256Hash = (HashProvenance) info.instanceValues.get(RESOURCE_HASH);
         }
 
+        /**
+         * Separates this class's non-configurable fields from the configurable fields.
+         * @param map The provenances.
+         * @return The extracted provenance information.
+         */
         protected static ExtractedInfo extractProvenanceInfo(Map<String,Provenance> map) {
             Map<String,Provenance> configuredParameters = new HashMap<>(map);
             String className = ObjectProvenance.checkAndExtractProvenance(configuredParameters,CLASS_NAME, StringProvenance.class, CSVDataSourceProvenance.class.getSimpleName()).getValue();
diff --git a/Data/src/main/java/org/tribuo/data/csv/CSVLoader.java b/Data/src/main/java/org/tribuo/data/csv/CSVLoader.java
index 6e9dfc5cc..b50b7127b 100644
--- a/Data/src/main/java/org/tribuo/data/csv/CSVLoader.java
+++ b/Data/src/main/java/org/tribuo/data/csv/CSVLoader.java
@@ -363,6 +363,10 @@ public final static class CSVLoaderProvenance implements DataSourceProvenance {
             this.sha256Hash = new HashProvenance(DEFAULT_HASH_TYPE, RESOURCE_HASH, ProvenanceUtil.hashResource(DEFAULT_HASH_TYPE, path));
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public CSVLoaderProvenance(Map<String, Provenance> map) {
             this.className = ObjectProvenance.checkAndExtractProvenance(map, CLASS_NAME, StringProvenance.class, CSVLoaderProvenance.class.getSimpleName());
             this.factoryProvenance = ObjectProvenance.checkAndExtractProvenance(map, OUTPUT_FACTORY, OutputFactoryProvenance.class, CSVLoaderProvenance.class.getSimpleName());
diff --git a/Data/src/main/java/org/tribuo/data/csv/CSVSaver.java b/Data/src/main/java/org/tribuo/data/csv/CSVSaver.java
index f56897b35..8c80705bb 100644
--- a/Data/src/main/java/org/tribuo/data/csv/CSVSaver.java
+++ b/Data/src/main/java/org/tribuo/data/csv/CSVSaver.java
@@ -48,9 +48,13 @@
  */
 public class CSVSaver implements Configurable {
 
-    public final static String DEFAULT_RESPONSE = "Response";
     private static final Logger logger = Logger.getLogger(CSVSaver.class.getName());
 
+    /**
+     * The default response column name.
+     */
+    public final static String DEFAULT_RESPONSE = "Response";
+
     @Config(description="The column separator.")
     private char separator = CSVIterator.SEPARATOR;
     @Config(description="The quote character.")
diff --git a/Data/src/main/java/org/tribuo/data/sql/ResultSetIterator.java b/Data/src/main/java/org/tribuo/data/sql/ResultSetIterator.java
index 1ad1c3030..67c264fc3 100644
--- a/Data/src/main/java/org/tribuo/data/sql/ResultSetIterator.java
+++ b/Data/src/main/java/org/tribuo/data/sql/ResultSetIterator.java
@@ -39,6 +39,11 @@ public class ResultSetIterator extends ColumnarIterator {
 
     private int rowNum = 0;
 
+    /**
+     * Construct a result set iterator over the supplied result set.
+     * @param rs The result set.
+     * @throws SQLException If the result set cannot be inspected.
+     */
     public ResultSetIterator(ResultSet rs) throws SQLException {
         resultSet = rs;
         ResultSetMetaData rsm = resultSet.getMetaData();
@@ -48,6 +53,12 @@ public ResultSetIterator(ResultSet rs) throws SQLException {
         }
     }
 
+    /**
+     * Constructs a result set iterator over the supplied result set using the specified fetch buffer size.
+     * @param rs The result set.
+     * @param fetchSize The fetch size.
+     * @throws SQLException If the result set cannot be inspected.
+     */
     public ResultSetIterator(ResultSet rs, int fetchSize) throws SQLException {
         super(IOSpliterator.DEFAULT_CHARACTERISTICS, fetchSize, Long.MAX_VALUE);
         resultSet = rs;
diff --git a/Data/src/main/java/org/tribuo/data/sql/SQLDBConfig.java b/Data/src/main/java/org/tribuo/data/sql/SQLDBConfig.java
index 93a4b8a1d..a9b5c9907 100644
--- a/Data/src/main/java/org/tribuo/data/sql/SQLDBConfig.java
+++ b/Data/src/main/java/org/tribuo/data/sql/SQLDBConfig.java
@@ -33,16 +33,15 @@
 import java.util.Properties;
 
 /**
- *
  * N.B. This class accepts raw SQL strings and executes them directly via JDBC. It DOES NOT perform
  * any SQL escaping or other injection prevention. It is the user's responsibility to ensure that SQL passed to this
  * class performs as desired.
- *
+ * <p>
  * SQL database configuration. If you specify the {@linkplain SQLDBConfig#host}, {@linkplain SQLDBConfig#port}, and
  * {@linkplain SQLDBConfig#db} strings and use {@code oracle.jdbc.OracleDriver} as your JDBC Driver, then this will
  * automatically generate a connectionString, otherwise it must be specified manually and host, port, and db fields
  * can be omitted.
- *
+ * <p>
  * {@link java.sql.DriverManager}'s default logic will be used to determine which {@link java.sql.Driver} to use for
  * a given connection string.
  */
@@ -70,12 +69,32 @@ public class SQLDBConfig implements Configurable, Provenancable<ConfiguredObject
 
     private SQLDBConfig() {}
 
+    /**
+     * Constructs a SQL database configuration.
+     * <p>
+     * Note it is recommended that wallet based connections are used rather than this constructor using {@link SQLDBConfig(String,Map)}.
+     * @param connectionString The connection string.
+     * @param username The username.
+     * @param password The password.
+     * @param properties The connection properties.
+     */
     public SQLDBConfig(String connectionString, String username, String password, Map<String, String> properties) {
         this(connectionString, properties);
         this.username = username;
         this.password = password;
     }
 
+    /**
+     * Constructs a SQL database configuration.
+     * <p>
+     * Note it is recommended that wallet based connections are used rather than this constructor using {@link SQLDBConfig(String,Map)}.
+     * @param host The host to connect to.
+     * @param port The port to connect on.
+     * @param db The db name.
+     * @param username The username.
+     * @param password The password.
+     * @param properties The connection properties.
+     */
     public SQLDBConfig(String host, String port, String db, String username, String password, Map<String, String> properties) {
         this(makeConnectionString(host, port, db), properties);
         this.host = host;
@@ -85,6 +104,11 @@ public SQLDBConfig(String host, String port, String db, String username, String
         this.password = password;
     }
 
+    /**
+     * Constructs a SQL database configuration.
+     * @param connectionString The connection string.
+     * @param properties The connection properties.
+     */
     public SQLDBConfig(String connectionString, Map<String, String> properties) {
         this.connectionString = connectionString;
         this.propMap = properties;
diff --git a/Data/src/main/java/org/tribuo/data/sql/SQLDataSource.java b/Data/src/main/java/org/tribuo/data/sql/SQLDataSource.java
index 58eda93ab..0e2919f09 100644
--- a/Data/src/main/java/org/tribuo/data/sql/SQLDataSource.java
+++ b/Data/src/main/java/org/tribuo/data/sql/SQLDataSource.java
@@ -68,8 +68,20 @@ public class SQLDataSource<T extends Output<T>> extends ColumnarDataSource<T> im
 
     private final Set<Statement> statements = new HashSet<>();
 
+    /**
+     * For OLCUT.
+     */
     private SQLDataSource() {}
 
+    /**
+     * Constructs a SQLDataSource.
+     * @param sqlString The SQL query to run.
+     * @param sqlConfig The SQL connection configuration.
+     * @param outputFactory The output factory to use.
+     * @param rowProcessor The row processor to convert the returned rows into examples.
+     * @param outputRequired Is an output required from this source.
+     * @throws SQLException If the database could not be read.
+     */
     public SQLDataSource(String sqlString, SQLDBConfig sqlConfig, OutputFactory<T> outputFactory, RowProcessor<T> rowProcessor, boolean outputRequired) throws SQLException {
         super(outputFactory, rowProcessor, outputRequired);
         this.sqlConfig = sqlConfig;
@@ -122,6 +134,10 @@ <T extends Output<T>> SQLDataSourceProvenance(SQLDataSource<T> host) {
             this.dataSourceCreationTime = new DateTimeProvenance(DATASOURCE_CREATION_TIME,OffsetDateTime.now());
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public SQLDataSourceProvenance(Map<String,Provenance> map) {
             this(extractProvenanceInfo(map));
         }
diff --git a/Data/src/main/java/org/tribuo/data/sql/SQLToCSV.java b/Data/src/main/java/org/tribuo/data/sql/SQLToCSV.java
index fe9eb417e..98ce66cb0 100644
--- a/Data/src/main/java/org/tribuo/data/sql/SQLToCSV.java
+++ b/Data/src/main/java/org/tribuo/data/sql/SQLToCSV.java
@@ -55,17 +55,35 @@ public class SQLToCSV {
      * Command line options.
      */
     public static class SQLToCSVOptions implements Options {
-        @Option(charName='n', longName="connection", usage="Connection string to the SQL database")
+        /**
+         * Connection string to the SQL database
+         */
+        @Option(charName = 'n', longName = "connection", usage = "Connection string to the SQL database")
         public String connString;
-        @Option(charName='p', longName="password", usage="Password for the SQL database")
+        /**
+         * Password for the SQL database
+         */
+        @Option(charName = 'p', longName = "password", usage = "Password for the SQL database")
         public String password;
-        @Option(charName='u', longName="username", usage="Username for the SQL database")
+        /**
+         * Username for the SQL database
+         */
+        @Option(charName = 'u', longName = "username", usage = "Username for the SQL database")
         public String username;
-        @Option(charName='i', longName="input-sql", usage="SQL File to run as a query, defaults to stdin")
+        /**
+         * SQL File to run as a query, defaults to stdin
+         */
+        @Option(charName = 'i', longName = "input-sql", usage = "SQL File to run as a query, defaults to stdin")
         public Path inputPath;
-        @Option(charName='o', longName="output-csv", usage="File to write query results as CSV, defaults to stdout")
+        /**
+         * File to write query results as CSV, defaults to stdout
+         */
+        @Option(charName = 'o', longName = "output-csv", usage = "File to write query results as CSV, defaults to stdout")
         public Path outputPath;
-        @Option(longName="db-config", usage="Name of the DBConfig to use")
+        /**
+         * Name of the DBConfig to use
+         */
+        @Option(longName = "db-config", usage = "Name of the DBConfig to use")
         public SQLDBConfig dbConfig;
 
     }
diff --git a/Data/src/main/java/org/tribuo/data/text/DirectoryFileSource.java b/Data/src/main/java/org/tribuo/data/text/DirectoryFileSource.java
index 931e85f2a..82fa6e5bd 100644
--- a/Data/src/main/java/org/tribuo/data/text/DirectoryFileSource.java
+++ b/Data/src/main/java/org/tribuo/data/text/DirectoryFileSource.java
@@ -110,20 +110,15 @@ protected DirectoryFileSource() {}
      * document preprocessors on the data read from the files in the directories
      * representing classes.
      *
+     * @param dataDir The directory to inspect.
      * @param outputFactory The output factory used to generate the outputs.
      * @param extractor The text feature extractor that will run on the
      * documents.
      * @param preprocessors Pre-processors that we will run on the documents
      * before extracting their features.
      */
-    public DirectoryFileSource(OutputFactory<T> outputFactory, TextFeatureExtractor<T> extractor, DocumentPreprocessor... preprocessors) {
-        this.outputFactory = outputFactory;
-        this.extractor = extractor;
-        this.preprocessors.addAll(Arrays.asList(preprocessors));
-    }
-
-    public DirectoryFileSource(Path newsDir, OutputFactory<T> outputFactory, TextFeatureExtractor<T> extractor, DocumentPreprocessor... preprocessors) {
-        this.dataDir = newsDir;
+    public DirectoryFileSource(Path dataDir, OutputFactory<T> outputFactory, TextFeatureExtractor<T> extractor, DocumentPreprocessor... preprocessors) {
+        this.dataDir = dataDir;
         this.outputFactory = outputFactory;
         this.extractor = extractor;
         this.preprocessors.addAll(Arrays.asList(preprocessors));
@@ -269,6 +264,10 @@ <T extends Output<T>> DirectoryFileSourceProvenance(DirectoryFileSource<T> host)
             this.dataSourceCreationTime = new DateTimeProvenance(DATASOURCE_CREATION_TIME,OffsetDateTime.now());
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public DirectoryFileSourceProvenance(Map<String,Provenance> map) {
             this(extractProvenanceInfo(map));
         }
@@ -279,6 +278,11 @@ private DirectoryFileSourceProvenance(ExtractedInfo info) {
             this.fileModifiedTime = (DateTimeProvenance) info.instanceValues.get(FILE_MODIFIED_TIME);
         }
 
+        /**
+         * Splits the provenance into configured and non-configured values.
+         * @param map the provenances.
+         * @return The extracted information.
+         */
         protected static ExtractedInfo extractProvenanceInfo(Map<String,Provenance> map) {
             Map<String,Provenance> configuredParameters = new HashMap<>(map);
             String className = ObjectProvenance.checkAndExtractProvenance(configuredParameters,CLASS_NAME, StringProvenance.class, DirectoryFileSourceProvenance.class.getSimpleName()).getValue();
diff --git a/Data/src/main/java/org/tribuo/data/text/SplitTextData.java b/Data/src/main/java/org/tribuo/data/text/SplitTextData.java
index 925b53822..5a79fc087 100644
--- a/Data/src/main/java/org/tribuo/data/text/SplitTextData.java
+++ b/Data/src/main/java/org/tribuo/data/text/SplitTextData.java
@@ -55,18 +55,39 @@ public static class TrainTestSplitOptions implements Options {
         public String getOptionsDescription() {
             return "Splits a standard text format dataset in two.";
         }
-        @Option(charName='s',longName="split-fraction",usage="Split fraction.")
+
+        /**
+         * Split fraction.
+         */
+        @Option(charName = 's', longName = "split-fraction", usage = "Split fraction.")
         public float splitFraction;
-        @Option(charName='i',longName="input-file",usage="Input data file in standard text format.")
+        /**
+         * Input data file in standard text format.
+         */
+        @Option(charName = 'i', longName = "input-file", usage = "Input data file in standard text format.")
         public Path inputPath;
-        @Option(charName='t',longName="training-output-file",usage="Output training data file.")
+        /**
+         * Output training data file.
+         */
+        @Option(charName = 't', longName = "training-output-file", usage = "Output training data file.")
         public Path trainPath;
-        @Option(charName='v',longName="validation-output-file",usage="Output validation data file.")
+        /**
+         * Output validation data file.
+         */
+        @Option(charName = 'v', longName = "validation-output-file", usage = "Output validation data file.")
         public Path validationPath;
-        @Option(charName='r',longName="rng-seed",usage="Seed for the RNG.")
+        /**
+         * Seed for the RNG.
+         */
+        @Option(charName = 'r', longName = "rng-seed", usage = "Seed for the RNG.")
         public long seed = 1;
     }
 
+    /**
+     * Runs the SplitTextData CLI.
+     * @param args The CLI arguments.
+     * @throws IOException If the files could not be read or written to.
+     */
     public static void main(String[] args) throws IOException {
         
         //
diff --git a/Data/src/main/java/org/tribuo/data/text/TextDataSource.java b/Data/src/main/java/org/tribuo/data/text/TextDataSource.java
index a264e76a3..d0cb1e442 100644
--- a/Data/src/main/java/org/tribuo/data/text/TextDataSource.java
+++ b/Data/src/main/java/org/tribuo/data/text/TextDataSource.java
@@ -30,34 +30,34 @@
 import java.util.List;
 
 /**
- * A base class for textual data sets. We assume that all textual data is 
+ * A base class for textual data sets. We assume that all textual data is
  * written and read using UTF-8.
  */
 public abstract class TextDataSource<T extends Output<T>> implements ConfigurableDataSource<T> {
-    
+
     /**
      * Document preprocessors that should be run on the documents that make up
      * this data set.
      */
-    @Config(description="The document preprocessors to run on each document in the data source.")
+    @Config(description = "The document preprocessors to run on each document in the data source.")
     protected List<DocumentPreprocessor> preprocessors = new ArrayList<>();
-    
+
     /**
      * The path that data was read from.
      */
-    @Config(mandatory=true,description="The path to read the data from.")
+    @Config(mandatory = true, description = "The path to read the data from.")
     protected Path path;
 
     /**
      * The factory that converts a String into an {@link Output}.
      */
-    @Config(mandatory=true,description="The factory that converts a String into an Output instance.")
+    @Config(mandatory = true, description = "The factory that converts a String into an Output instance.")
     protected OutputFactory<T> outputFactory;
 
     /**
      * The extractor that we'll use to turn text into examples.
      */
-    @Config(mandatory=true,description="The feature extractor that generates Features from text.")
+    @Config(mandatory = true, description = "The feature extractor that generates Features from text.")
     protected TextFeatureExtractor<T> extractor;
 
     /**
@@ -68,15 +68,17 @@ public abstract class TextDataSource<T extends Output<T>> implements Configurabl
     /**
      * for olcut
      */
-    protected TextDataSource() {}
+    protected TextDataSource() {
+    }
 
     /**
      * Creates a text data set by reading it from a path.
-     * @param path the path to read data from
-     * @param outputFactory the output factory used to generate the outputs.
-     * @param extractor The feature extractor to run on the text.
-     * @param preprocessors processors that will be run on the data before it
-     * is added as examples.
+     *
+     * @param path          The path to read data from
+     * @param outputFactory The output factory used to generate the outputs.
+     * @param extractor     The feature extractor to run on the text.
+     * @param preprocessors Processors that will be run on the data before it
+     *                      is added as examples.
      */
     public TextDataSource(Path path, OutputFactory<T> outputFactory, TextFeatureExtractor<T> extractor, DocumentPreprocessor... preprocessors) {
         this.path = path;
@@ -84,11 +86,20 @@ public TextDataSource(Path path, OutputFactory<T> outputFactory, TextFeatureExtr
         this.extractor = extractor;
         this.preprocessors.addAll(Arrays.asList(preprocessors));
     }
-    
+
+    /**
+     * Creates a text data set by reading it from a file.
+     *
+     * @param file          The file to read data from
+     * @param outputFactory The output factory used to generate the outputs.
+     * @param extractor     The feature extractor to run on the text.
+     * @param preprocessors Processors that will be run on the data before it
+     *                      is added as examples.
+     */
     public TextDataSource(File file, OutputFactory<T> outputFactory, TextFeatureExtractor<T> extractor, DocumentPreprocessor... preprocessors) {
         this(file.toPath(), outputFactory, extractor, preprocessors);
     }
-    
+
     @Override
     public Iterator<Example<T>> iterator() {
         if (!data.isEmpty()) {
@@ -131,6 +142,7 @@ protected String handleDoc(String doc) {
 
     /**
      * Reads the data from the Path.
+     *
      * @throws java.io.IOException if there is any error reading the data.
      */
     protected abstract void read() throws java.io.IOException;
diff --git a/Data/src/main/java/org/tribuo/data/text/TextProcessingException.java b/Data/src/main/java/org/tribuo/data/text/TextProcessingException.java
index 2dd8241dd..9a10b4126 100644
--- a/Data/src/main/java/org/tribuo/data/text/TextProcessingException.java
+++ b/Data/src/main/java/org/tribuo/data/text/TextProcessingException.java
@@ -22,14 +22,27 @@
 public class TextProcessingException extends Exception {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * Creates a TextProcessingException with the specified message.
+     * @param message The exception message.
+     */
     public TextProcessingException(String message) {
         super(message);
     }
 
+    /**
+     * Creates a TextProcessingException wrapping the supplied throwable with the specified message.
+     * @param message The exception message.
+     * @param throwable The throwable to wrap.
+     */
     public TextProcessingException(String message, Throwable throwable) {
         super(message,throwable);
     }
 
+    /**
+     * Creates a TextProcessingException wrapping the supplied throwable.
+     * @param throwable The throwable to wrap.
+     */
     public TextProcessingException(Throwable throwable) {
         super(throwable);
     }
diff --git a/Data/src/main/java/org/tribuo/data/text/impl/BasicPipeline.java b/Data/src/main/java/org/tribuo/data/text/impl/BasicPipeline.java
index 40ad59d0c..338ca951b 100644
--- a/Data/src/main/java/org/tribuo/data/text/impl/BasicPipeline.java
+++ b/Data/src/main/java/org/tribuo/data/text/impl/BasicPipeline.java
@@ -17,6 +17,7 @@
 package org.tribuo.data.text.impl;
 
 import com.oracle.labs.mlrg.olcut.config.Config;
+import com.oracle.labs.mlrg.olcut.config.PropertyException;
 import com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance;
 import com.oracle.labs.mlrg.olcut.provenance.impl.ConfiguredObjectProvenanceImpl;
 import org.tribuo.Feature;
@@ -47,6 +48,12 @@ public class BasicPipeline implements TextPipeline {
     @Config(description="n in the n-gram to emit.")
     private int ngram = 2;
 
+    /**
+     * Constructs a basic text pipeline which tokenizes the input and generates word
+     * n-gram features in the range 1 to {@code ngram}.
+     * @param tokenizer The tokenizer.
+     * @param ngram The size of the n-grams to generate.
+     */
     public BasicPipeline(Tokenizer tokenizer, int ngram) {
         this.tokenizer = tokenizer;
         this.ngram = ngram;
@@ -54,7 +61,7 @@ public BasicPipeline(Tokenizer tokenizer, int ngram) {
     }
 
     /**
-     * For olcut
+     * For OLCUT.
      */
     private BasicPipeline() {}
 
@@ -63,6 +70,9 @@ private BasicPipeline() {}
      */
     @Override
     public void postConfig() {
+        if (ngram < 1) {
+            throw new PropertyException("","ngram","ngram must be positive, found " + ngram);
+        }
         for (int i = 1; i <= ngram; ++i) {
             processors.add(new NgramProcessor(tokenizer,i,1.0));
         }
@@ -84,7 +94,6 @@ public List<Feature> process(String tag, String data) {
                 logger.log(Level.INFO, String.format("TextProcessingException thrown by processor %s with text %s",p,data), e);
             }
         }
-        //logger.log(Level.INFO,features.toString());
 
         return aggregator.aggregate(features);
     }
diff --git a/Data/src/main/java/org/tribuo/data/text/impl/CasingPreprocessor.java b/Data/src/main/java/org/tribuo/data/text/impl/CasingPreprocessor.java
index 4a0e84939..d9cc1078a 100644
--- a/Data/src/main/java/org/tribuo/data/text/impl/CasingPreprocessor.java
+++ b/Data/src/main/java/org/tribuo/data/text/impl/CasingPreprocessor.java
@@ -30,7 +30,14 @@ public class CasingPreprocessor implements DocumentPreprocessor {
      * The possible casing operations.
      */
     public enum CasingOperation {
-        LOWERCASE, UPPERCASE;
+        /**
+         * Lowercase the input text.
+         */
+        LOWERCASE,
+        /**
+         * Uppercase the input text.
+         */
+        UPPERCASE;
 
         /**
          * Apply the appropriate casing operation.
diff --git a/Data/src/main/java/org/tribuo/data/text/impl/FeatureHasher.java b/Data/src/main/java/org/tribuo/data/text/impl/FeatureHasher.java
index e8832ca5a..2450e83a8 100644
--- a/Data/src/main/java/org/tribuo/data/text/impl/FeatureHasher.java
+++ b/Data/src/main/java/org/tribuo/data/text/impl/FeatureHasher.java
@@ -29,6 +29,8 @@
 
 /**
  * Hashes the feature names to reduce the dimensionality.
+ * <p>
+ * Uses murmurhash3_x86_32 as the hashing function for the feature names.
  */
 public class FeatureHasher implements FeatureTransformer {
 
@@ -36,7 +38,11 @@ public class FeatureHasher implements FeatureTransformer {
 
     @Config(mandatory = true,description="Dimension to map the hash into.")
     private int dimension;
-    
+
+    /**
+     * Constructs a feature hasher using the supplied hash dimension.
+     * @param dimension The dimension to reduce the hashed features into.
+     */
     public FeatureHasher(int dimension) {
         this.dimension = dimension;
     }
diff --git a/Data/src/main/java/org/tribuo/data/text/impl/SimpleStringDataSource.java b/Data/src/main/java/org/tribuo/data/text/impl/SimpleStringDataSource.java
index 6432c67d6..6deae4fe5 100644
--- a/Data/src/main/java/org/tribuo/data/text/impl/SimpleStringDataSource.java
+++ b/Data/src/main/java/org/tribuo/data/text/impl/SimpleStringDataSource.java
@@ -60,6 +60,12 @@ public class SimpleStringDataSource<T extends Output<T>> extends SimpleTextDataS
      */
     private SimpleStringDataSource() {}
 
+    /**
+     * Constructs a simple string data source from the supplied lines.
+     * @param rawLines The lines to parse.
+     * @param outputFactory The output factory.
+     * @param extractor The feature extractor.
+     */
     public SimpleStringDataSource(List<String> rawLines, OutputFactory<T> outputFactory, TextFeatureExtractor<T> extractor) {
         super(outputFactory, extractor);
         this.rawLines = rawLines;
@@ -119,6 +125,10 @@ <T extends Output<T>> SimpleStringDataSourceProvenance(SimpleStringDataSource<T>
             this.sha256Hash = new HashProvenance(DEFAULT_HASH_TYPE,RESOURCE_HASH, ProvenanceUtil.hashList(DEFAULT_HASH_TYPE,host.rawLines));
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public SimpleStringDataSourceProvenance(Map<String, Provenance> map) {
             this(extractProvenanceInfo(map));
         }
diff --git a/Data/src/main/java/org/tribuo/data/text/impl/SimpleTextDataSource.java b/Data/src/main/java/org/tribuo/data/text/impl/SimpleTextDataSource.java
index 537cf999d..7a6a3b350 100644
--- a/Data/src/main/java/org/tribuo/data/text/impl/SimpleTextDataSource.java
+++ b/Data/src/main/java/org/tribuo/data/text/impl/SimpleTextDataSource.java
@@ -76,11 +76,25 @@ public class SimpleTextDataSource<T extends Output<T>> extends TextDataSource<T>
      */
     protected SimpleTextDataSource() {}
 
+    /**
+     * Constructs a simple text data source by reading lines from the supplied path.
+     * @param path The path to load.
+     * @param outputFactory The output factory to use.
+     * @param extractor The feature extractor.
+     * @throws IOException If the path could not be read.
+     */
     public SimpleTextDataSource(Path path, OutputFactory<T> outputFactory, TextFeatureExtractor<T> extractor) throws IOException {
         super(path, outputFactory, extractor);
         postConfig();
     }
 
+    /**
+     * Constructs a simple text data source by reading lines from the supplied file.
+     * @param file The file to load.
+     * @param outputFactory The output factory to use.
+     * @param extractor The feature extractor.
+     * @throws IOException If the file could not be read.
+     */
     public SimpleTextDataSource(File file, OutputFactory<T> outputFactory, TextFeatureExtractor<T> extractor) throws IOException {
         super(file, outputFactory, extractor);
         postConfig();
@@ -165,6 +179,10 @@ <T extends Output<T>> SimpleTextDataSourceProvenance(SimpleTextDataSource<T> hos
             this.sha256Hash = new HashProvenance(DEFAULT_HASH_TYPE,RESOURCE_HASH,ProvenanceUtil.hashResource(DEFAULT_HASH_TYPE,host.path));
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public SimpleTextDataSourceProvenance(Map<String,Provenance> map) {
             this(extractProvenanceInfo(map));
         }
diff --git a/Data/src/main/java/org/tribuo/data/text/impl/TextFeatureExtractorImpl.java b/Data/src/main/java/org/tribuo/data/text/impl/TextFeatureExtractorImpl.java
index 83e68f16a..897648608 100644
--- a/Data/src/main/java/org/tribuo/data/text/impl/TextFeatureExtractorImpl.java
+++ b/Data/src/main/java/org/tribuo/data/text/impl/TextFeatureExtractorImpl.java
@@ -25,13 +25,12 @@
 import org.tribuo.data.text.TextFeatureExtractor;
 import org.tribuo.data.text.TextPipeline;
 import org.tribuo.impl.ArrayExample;
-import org.tribuo.impl.ListExample;
 
 import java.util.List;
 
 /**
  * An implementation of {@link TextFeatureExtractor} that takes a
- * {@link TextPipeline} and generates {@link ListExample}.
+ * {@link TextPipeline} and generates {@link ArrayExample}.
  */
 public class TextFeatureExtractorImpl<T extends Output<T>> implements TextFeatureExtractor<T> {
 
@@ -42,7 +41,11 @@ public class TextFeatureExtractorImpl<T extends Output<T>> implements TextFeatur
      * for olcut
      */
     private TextFeatureExtractorImpl() {}
-    
+
+    /**
+     * Constructs a text feature extractor wrapping the supplied text pipeline.
+     * @param pipeline The text processing pipeline.
+     */
     public TextFeatureExtractorImpl(TextPipeline pipeline) {
         this.pipeline = pipeline;
     }
diff --git a/Data/src/main/java/org/tribuo/data/text/impl/UniqueAggregator.java b/Data/src/main/java/org/tribuo/data/text/impl/UniqueAggregator.java
index 103036d2f..874f8d10b 100644
--- a/Data/src/main/java/org/tribuo/data/text/impl/UniqueAggregator.java
+++ b/Data/src/main/java/org/tribuo/data/text/impl/UniqueAggregator.java
@@ -37,10 +37,19 @@ public class UniqueAggregator implements FeatureAggregator {
     @Config(description="Value to emit, if unset emits the last value observed for that token.")
     private double value = Double.NaN;
 
+    /**
+     * Constructs an aggregator that replaces all features with the same
+     * name with a single feature with the specified value.
+     * @param value The inserted feature value.
+     */
     public UniqueAggregator(double value) {
         this.value = value;
     }
 
+    /**
+     * Constructs an aggregator that replaces all features with the same
+     * name with a single feature with the last observed value of that feature.
+     */
     public UniqueAggregator() { }
 
     @Override
diff --git a/Interop/Core/src/main/java/org/tribuo/interop/ExternalDatasetProvenance.java b/Interop/Core/src/main/java/org/tribuo/interop/ExternalDatasetProvenance.java
index 97f9aa471..855488f22 100644
--- a/Interop/Core/src/main/java/org/tribuo/interop/ExternalDatasetProvenance.java
+++ b/Interop/Core/src/main/java/org/tribuo/interop/ExternalDatasetProvenance.java
@@ -35,10 +35,23 @@
 public class ExternalDatasetProvenance extends DatasetProvenance {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * An empty provenance used as a placeholder for externally trained models.
+     * @param description The model description.
+     * @param factory The output factory.
+     * @param isSequence Is it a sequence model?
+     * @param numFeatures The number of features.
+     * @param numOutputs The output dimensionality.
+     * @param <T> The type of the output.
+     */
     public <T extends Output<T>> ExternalDatasetProvenance(String description, OutputFactory<T> factory, boolean isSequence, int numFeatures, int numOutputs) {
         super(new SimpleDataSourceProvenance(description, OffsetDateTime.now(),factory), new ListProvenance<>(), Dataset.class.getName(), false, isSequence, -1, numFeatures, numOutputs);
     }
 
+    /**
+     * Deserialization constructor.
+     * @param map The provenances.
+     */
     public ExternalDatasetProvenance(Map<String, Provenance> map) {
         super(map);
     }
diff --git a/Interop/Core/src/main/java/org/tribuo/interop/ExternalModel.java b/Interop/Core/src/main/java/org/tribuo/interop/ExternalModel.java
index 7ec1a624b..2e141d420 100644
--- a/Interop/Core/src/main/java/org/tribuo/interop/ExternalModel.java
+++ b/Interop/Core/src/main/java/org/tribuo/interop/ExternalModel.java
@@ -58,6 +58,15 @@ public abstract class ExternalModel<T extends Output<T>,U,V> extends Model<T> {
 
     private int batchSize = DEFAULT_BATCH_SIZE;
 
+    /**
+     * Constructs an external model from a model trained outside of Tribuo.
+     * @param name The model name.
+     * @param provenance The model provenance.
+     * @param featureIDMap The feature domain.
+     * @param outputIDInfo The output domain.
+     * @param generatesProbabilities Does this model generate probabilistic predictions.
+     * @param featureMapping The mapping from Tribuo's feature names to the model's feature indices.
+     */
     protected ExternalModel(String name, ModelProvenance provenance,
                             ImmutableFeatureMap featureIDMap, ImmutableOutputInfo<T> outputIDInfo,
                             boolean generatesProbabilities, Map<String,Integer> featureMapping) {
@@ -88,6 +97,16 @@ protected ExternalModel(String name, ModelProvenance provenance,
         }
     }
 
+    /**
+     * Constructs an external model from a model trained outside of Tribuo.
+     * @param name The model name.
+     * @param provenance The model provenance.
+     * @param featureIDMap The feature domain.
+     * @param outputIDInfo The output domain.
+     * @param generatesProbabilities Does this model generate probabilistic predictions.
+     * @param featureForwardMapping The mapping from Tribuo's indices to the model's indices.
+     * @param featureBackwardMapping The mapping from the model's indices to Tribuo's indices.
+     */
     protected ExternalModel(String name, ModelProvenance provenance, ImmutableFeatureMap featureIDMap, ImmutableOutputInfo<T> outputIDInfo, int[] featureForwardMapping, int[] featureBackwardMapping, boolean generatesProbabilities) {
         super(name,provenance,featureIDMap,outputIDInfo,generatesProbabilities);
         this.featureBackwardMapping = Arrays.copyOf(featureBackwardMapping,featureBackwardMapping.length);
diff --git a/Interop/ONNX/src/main/java/org/tribuo/interop/onnx/ImageTransformer.java b/Interop/ONNX/src/main/java/org/tribuo/interop/onnx/ImageTransformer.java
index 4dcceb1b6..5630172ac 100644
--- a/Interop/ONNX/src/main/java/org/tribuo/interop/onnx/ImageTransformer.java
+++ b/Interop/ONNX/src/main/java/org/tribuo/interop/onnx/ImageTransformer.java
@@ -55,6 +55,12 @@ public class ImageTransformer implements ExampleTransformer {
      */
     private ImageTransformer() {}
 
+    /**
+     * Constructs an image transformer with the specified parameters.
+     * @param channels The number of colour channels.
+     * @param height The height.
+     * @param width The width.
+     */
     public ImageTransformer(int channels, int height, int width) {
         if (width < 1 || height < 1 || channels < 1) {
             throw new PropertyException("","Inputs must be positive integers, found [c="+channels+",h="+height+",w="+width+"]");
diff --git a/Interop/ONNX/src/main/java/org/tribuo/interop/onnx/extractors/BERTFeatureExtractor.java b/Interop/ONNX/src/main/java/org/tribuo/interop/onnx/extractors/BERTFeatureExtractor.java
index 208a484bd..756a5be2c 100644
--- a/Interop/ONNX/src/main/java/org/tribuo/interop/onnx/extractors/BERTFeatureExtractor.java
+++ b/Interop/ONNX/src/main/java/org/tribuo/interop/onnx/extractors/BERTFeatureExtractor.java
@@ -93,24 +93,57 @@ public enum OutputPooling {
     }
 
     // BERT input names
+    /**
+     * Input name for the token ids.
+     */
     public static final String INPUT_IDS = "input_ids";
+    /**
+     * Input name for the attention mask.
+     */
     public static final String ATTENTION_MASK = "attention_mask";
+    /**
+     * Input name for the token type ids.
+     */
     public static final String TOKEN_TYPE_IDS = "token_type_ids";
 
     // BERT output names
+    /**
+     * Output name for the token level outputs.
+     */
     public static final String TOKEN_OUTPUT = "output_0";
+    /**
+     * Output name for the classification token output.
+     */
     public static final String CLS_OUTPUT = "output_1";
 
     // Token names
+    /**
+     * Default classification token name.
+     */
     public static final String CLASSIFICATION_TOKEN = "[CLS]";
+    /**
+     * Default separator token name.
+     */
     public static final String SEPARATOR_TOKEN = "[SEP]";
+    /**
+     * Default unknown token name.
+     */
     public static final String UNKNOWN_TOKEN = "[UNK]";
 
     // Metadata name for the token
+    /**
+     * Metadata key for the token value stored in a Tribuo {@link org.tribuo.Example}.
+     */
     public static final String TOKEN_METADATA = "Token";
 
     // Values expected by BERT inputs
+    /**
+     * Mask value.
+     */
     public static final long MASK_VALUE = 1;
+    /**
+     * Token type value for the first sentence.
+     */
     public static final long TOKEN_TYPE_VALUE = 0;
 
     @Config(mandatory = true,description="Output factory to use.")
@@ -978,11 +1011,20 @@ static final class BERTResult {
      * CLI options for running BERT.
      */
     public static class BERTFeatureExtractorOptions implements Options {
-        @Option(charName='b',longName="bert",usage="BERTFeatureExtractor instance")
+        /**
+         * BERTFeatureExtractor instance
+         */
+        @Option(charName = 'b', longName = "bert", usage = "BERTFeatureExtractor instance")
         public BERTFeatureExtractor<?> bert;
-        @Option(charName='i',longName="input-file",usage="Input file to read, one doc per line")
+        /**
+         * Input file to read, one doc per line
+         */
+        @Option(charName = 'i', longName = "input-file", usage = "Input file to read, one doc per line")
         public Path inputFile;
-        @Option(charName='o',longName="output-file",usage="Output json file.")
+        /**
+         * Output json file.
+         */
+        @Option(charName = 'o', longName = "output-file", usage = "Output json file.")
         public Path outputFile;
     }
 
diff --git a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/FeatureConverter.java b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/FeatureConverter.java
index 1a01ca81b..13f1d08fa 100644
--- a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/FeatureConverter.java
+++ b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/FeatureConverter.java
@@ -67,7 +67,6 @@ public interface FeatureConverter extends Configurable, Provenancable<Configured
 
     /**
      * Converts a list of {@link SGDVector}s representing a batch of features into a {@link TensorMap}.
-     * <p>
      * @param vectors The batch of features to convert.
      * @return A TensorMap (similar to a TF Python feed_dict) representing this minibatch.
      */
diff --git a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/MultiLabelConverter.java b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/MultiLabelConverter.java
index 2c4afce66..1ce8fd50e 100644
--- a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/MultiLabelConverter.java
+++ b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/MultiLabelConverter.java
@@ -59,6 +59,9 @@ public class MultiLabelConverter implements OutputConverter<MultiLabel> {
     private static final long serialVersionUID = 1L;
     private static final Logger logger = Logger.getLogger(MultiLabelConverter.class.getName());
 
+    /**
+     * The threshold to determine if a label has been predicted.
+     */
     public static final double THRESHOLD = 0.5;
 
     /**
diff --git a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TensorFlowTrainer.java b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TensorFlowTrainer.java
index d3c4cab04..64542f1d3 100644
--- a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TensorFlowTrainer.java
+++ b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TensorFlowTrainer.java
@@ -597,10 +597,19 @@ public TrainerProvenance getProvenance() {
         return new TensorFlowTrainerProvenance(this);
     }
 
+    /**
+     * Provenance for {@link TensorFlowTrainer}.
+     */
     public static final class TensorFlowTrainerProvenance extends SkeletalTrainerProvenance {
         private static final long serialVersionUID = 1L;
 
+        /**
+         * The name of the provenance field containing the graph hash.
+         */
         public static final String GRAPH_HASH = "graph-hash";
+        /**
+         * The name of the provenance field containing the graph modified timestamp.
+         */
         public static final String GRAPH_LAST_MOD = "graph-last-modified";
 
         private final HashProvenance graphHash;
@@ -618,6 +627,10 @@ <T extends Output<T>> TensorFlowTrainerProvenance(TensorFlowTrainer<T> host) {
             }
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public TensorFlowTrainerProvenance(Map<String,Provenance> map) {
             this(extractTFProvenanceInfo(map));
         }
diff --git a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TensorFlowUtil.java b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TensorFlowUtil.java
index ac4a24efc..b392cb31f 100644
--- a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TensorFlowUtil.java
+++ b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TensorFlowUtil.java
@@ -42,10 +42,25 @@
 public abstract class TensorFlowUtil {
     private static final Logger logger = Logger.getLogger(TensorFlowUtil.class.getName());
 
+    /**
+     * The name of the variable op.
+     */
     public static final String VARIABLE_V2 = "VariableV2";
+    /**
+     * The name of the assignment op.
+     */
     public static final String ASSIGN_OP = "Assign";
+    /**
+     * The name given to the assignment operation from the placeholders.
+     */
     public static final String ASSIGN_PLACEHOLDER = "Assign_from_Placeholder";
+    /**
+     * The name of the placeholder op.
+     */
     public static final String PLACEHOLDER = "Placeholder";
+    /**
+     * The name of the data type.
+     */
     public static final String DTYPE = "dtype";
 
     private TensorFlowUtil() {}
@@ -190,8 +205,17 @@ public static void restoreMarshalledVariables(Session session, Map<String, Tenso
     public static final class TensorTuple implements Serializable {
         private static final long serialVersionUID = 1L;
 
+        /**
+         * The tensor class name.
+         */
         public final String className;
+        /**
+         * The shape of the tensor.
+         */
         public final long[] shape;
+        /**
+         * The tensor data.
+         */
         public final byte[] data;
 
         /**
diff --git a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TrainTest.java b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TrainTest.java
index 6d1e224bb..f6af12566 100644
--- a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TrainTest.java
+++ b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/TrainTest.java
@@ -90,6 +90,7 @@ private static Pair<Dataset<Label>,Dataset<Label>> load(Path trainingPath, Path
     public static class TensorflowOptions implements Options {
         private static List<String> DEFAULT_PARAM_NAMES = new ArrayList<>();
         private static List<Float> DEFAULT_PARAM_VALUES = new ArrayList<>();
+
         static {
             DEFAULT_PARAM_NAMES.add("learningRate");
             DEFAULT_PARAM_NAMES.add("initialAccumulatorValue");
@@ -101,54 +102,107 @@ public static class TensorflowOptions implements Options {
         public String getOptionsDescription() {
             return "Trains and tests a Tensorflow classification model.";
         }
-        @Option(charName='f',longName="model-output-path",usage="Path to serialize model to.")
+
+        /**
+         * Path to serialize model to.
+         */
+        @Option(charName = 'f', longName = "model-output-path", usage = "Path to serialize model to.")
         public Path outputPath;
-        @Option(charName='u',longName="training-file",usage="Path to the libsvm format training file.")
+        /**
+         * Path to the libsvm format training file.
+         */
+        @Option(charName = 'u', longName = "training-file", usage = "Path to the libsvm format training file.")
         public Path trainingPath;
-        @Option(charName='v',longName="testing-file",usage="Path to the libsvm format testing file.")
+        /**
+         * Path to the libsvm format testing file.
+         */
+        @Option(charName = 'v', longName = "testing-file", usage = "Path to the libsvm format testing file.")
         public Path testingPath;
 
-        @Option(charName='i',longName="init-name",usage="Name of the initialisation operation.")
+        /**
+         * Name of the initialisation operation.
+         */
+        @Option(charName = 'i', longName = "init-name", usage = "Name of the initialisation operation.")
         public String initName = Init.DEFAULT_NAME;
-        @Option(charName='l',longName="output-name",usage="Name of the output operation.")
+        /**
+         * Name of the output operation.
+         */
+        @Option(charName = 'l', longName = "output-name", usage = "Name of the output operation.")
         public String outputName;
-        @Option(longName="optimizer-param-names",usage="Gradient optimizer param names, see org.tribuo.interop.tensorflow.GradientOptimiser.")
+        /**
+         * Gradient optimizer param names, see org.tribuo.interop.tensorflow.GradientOptimiser.
+         */
+        @Option(longName = "optimizer-param-names", usage = "Gradient optimizer param names, see org.tribuo.interop.tensorflow.GradientOptimiser.")
         public List<String> gradientParamNames = DEFAULT_PARAM_NAMES;
-        @Option(longName="optimizer-param-values",usage="Gradient optimizer param values, see org.tribuo.interop.tensorflow.GradientOptimiser.")
+        /**
+         * Gradient optimizer param values, see org.tribuo.interop.tensorflow.GradientOptimiser.
+         */
+        @Option(longName = "optimizer-param-values", usage = "Gradient optimizer param values, see org.tribuo.interop.tensorflow.GradientOptimiser.")
         public List<Float> gradientParamValues = DEFAULT_PARAM_VALUES;
-        @Option(charName='g',longName="gradient-optimizer",usage="The gradient optimizer to use.")
+        /**
+         * The gradient optimizer to use.
+         */
+        @Option(charName = 'g', longName = "gradient-optimizer", usage = "The gradient optimizer to use.")
         public GradientOptimiser optimiser = GradientOptimiser.ADAGRAD;
-        @Option(longName="test-batch-size",usage="Test time minibatch size.")
+        /**
+         * Test time minibatch size.
+         */
+        @Option(longName = "test-batch-size", usage = "Test time minibatch size.")
         public int testBatchSize = 16;
-        @Option(charName='b',longName="batch-size",usage="Minibatch size.")
+        /**
+         * Minibatch size.
+         */
+        @Option(charName = 'b', longName = "batch-size", usage = "Minibatch size.")
         public int batchSize = 128;
-        @Option(charName='e',longName="num-epochs",usage="Number of gradient descent epochs.")
+        /**
+         * Number of gradient descent epochs.
+         */
+        @Option(charName = 'e', longName = "num-epochs", usage = "Number of gradient descent epochs.")
         public int epochs = 5;
-        @Option(longName="logging-interval",usage="Interval between logging the loss.")
+        /**
+         * Interval between logging the loss.
+         */
+        @Option(longName = "logging-interval", usage = "Interval between logging the loss.")
         public int loggingInterval = 1000;
-        @Option(charName='n',longName="input-name",usage="Name of the input placeholder.")
+        /**
+         * Name of the input placeholder.
+         */
+        @Option(charName = 'n', longName = "input-name", usage = "Name of the input placeholder.")
         public String inputName;
-        @Option(longName="image-format",usage="Image format, in [W,H,C]. Defaults to MNIST.")
+        /**
+         * Image format, in [W,H,C]. Defaults to MNIST.
+         */
+        @Option(longName = "image-format", usage = "Image format, in [W,H,C]. Defaults to MNIST.")
         public String imageFormat = "28,28,1";
-        @Option(charName='t',longName="input-type",usage="Input type.")
+        /**
+         * Input type.
+         */
+        @Option(charName = 't', longName = "input-type", usage = "Input type.")
         public InputType inputType = InputType.IMAGE;
-        @Option(charName='m',longName="model-protobuf",usage="Path to the protobuf containing the network description.")
+        /**
+         * Path to the protobuf containing the network description.
+         */
+        @Option(charName = 'm', longName = "model-protobuf", usage = "Path to the protobuf containing the network description.")
         public Path protobufPath;
-        @Option(charName='p',longName="checkpoint-dir",usage="Path to the checkpoint base directory.")
+        /**
+         * Path to the checkpoint base directory.
+         */
+        @Option(charName = 'p', longName = "checkpoint-dir", usage = "Path to the checkpoint base directory.")
         public Path checkpointPath;
 
         /**
          * Zips the gradient parameter names and values.
+         *
          * @return The gradient parameter map.
          */
-        public Map<String,Float> getGradientParams() {
+        public Map<String, Float> getGradientParams() {
             if (gradientParamNames.size() != gradientParamValues.size()) {
                 throw new IllegalArgumentException("Must supply both name and value for the gradient parameters, " +
                         "found " + gradientParamNames.size() + " names, and " + gradientParamValues.size() + "values.");
             }
-            Map<String,Float> output = new HashMap<>();
-            for (int i = 0 ; i < gradientParamNames.size(); i++) {
-                output.put(gradientParamNames.get(i),gradientParamValues.get(i));
+            Map<String, Float> output = new HashMap<>();
+            for (int i = 0; i < gradientParamNames.size(); i++) {
+                output.put(gradientParamNames.get(i), gradientParamValues.get(i));
             }
             return output;
         }
diff --git a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/example/GraphDefTuple.java b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/example/GraphDefTuple.java
index 7d85679ad..b58871092 100644
--- a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/example/GraphDefTuple.java
+++ b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/example/GraphDefTuple.java
@@ -24,9 +24,21 @@
  * Will be a record one day.
  */
 public final class GraphDefTuple {
+    /**
+     * The graph definition protobuf.
+     */
     public final GraphDef graphDef;
+    /**
+     * Name of the input operation.
+     */
     public final String inputName;
+    /**
+     * Name of the output operation.
+     */
     public final String outputName;
+    /**
+     * Name of the initialization operation.
+     */
     public final String initName;
 
     /**
diff --git a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/sequence/TensorFlowSequenceTrainer.java b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/sequence/TensorFlowSequenceTrainer.java
index 1dae26259..20b452339 100644
--- a/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/sequence/TensorFlowSequenceTrainer.java
+++ b/Interop/Tensorflow/src/main/java/org/tribuo/interop/tensorflow/sequence/TensorFlowSequenceTrainer.java
@@ -99,6 +99,21 @@ public class TensorFlowSequenceTrainer<T extends Output<T>> implements SequenceT
 
     protected int trainInvocationCounter;
 
+    /**
+     * Constructs a TensorFlowSequenceTrainer using the specified parameters.
+     * @param graphPath The path to the TF graph.
+     * @param featureConverter The feature conversion object.
+     * @param outputConverter The output conversion object.
+     * @param minibatchSize The training minibatch size.
+     * @param epochs The number of training epochs.
+     * @param loggingInterval The logging interval.
+     * @param seed The RNG seed.
+     * @param initOp The name of the init operation.
+     * @param trainOp The name of the training operation.
+     * @param getLossOp The name of the loss operation.
+     * @param predictOp The name of the prediction operation.
+     * @throws IOException If the graph can not be read from disk.
+     */
     public TensorFlowSequenceTrainer(Path graphPath,
                                      SequenceFeatureConverter featureConverter,
                                      SequenceOutputConverter<T> outputConverter,
@@ -279,10 +294,19 @@ public TrainerProvenance getProvenance() {
         return new TensorFlowSequenceTrainerProvenance(this);
     }
 
+    /**
+     * Provenance for {@link TensorFlowSequenceTrainer}.
+     */
     public static class TensorFlowSequenceTrainerProvenance extends SkeletalTrainerProvenance {
         private static final long serialVersionUID = 1L;
 
+        /**
+         * The name of the provenance field containing the graph hash.
+         */
         public static final String GRAPH_HASH = "graph-hash";
+        /**
+         * The name of the provenance field containing the graph modified timestamp.
+         */
         public static final String GRAPH_LAST_MOD = "graph-last-modified";
 
         private final StringProvenance graphHash;
@@ -295,6 +319,10 @@ <T extends Output<T>> TensorFlowSequenceTrainerProvenance(TensorFlowSequenceTrai
             this.graphLastModified = new DateTimeProvenance(GRAPH_LAST_MOD,OffsetDateTime.ofInstant(Instant.ofEpochMilli(host.graphPath.toFile().lastModified()), ZoneId.systemDefault()));
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public TensorFlowSequenceTrainerProvenance(Map<String,Provenance> map) {
             this(extractTFProvenanceInfo(map));
         }
diff --git a/Json/src/main/java/org/tribuo/json/JsonDataSource.java b/Json/src/main/java/org/tribuo/json/JsonDataSource.java
index 5610d51e6..853b2923c 100644
--- a/Json/src/main/java/org/tribuo/json/JsonDataSource.java
+++ b/Json/src/main/java/org/tribuo/json/JsonDataSource.java
@@ -145,6 +145,10 @@ <T extends Output<T>> JsonDataSourceProvenance(JsonDataSource<T> host) {
             this.sha256Hash = new HashProvenance(DEFAULT_HASH_TYPE,RESOURCE_HASH,ProvenanceUtil.hashResource(DEFAULT_HASH_TYPE,host.dataPath));
         }
 
+        /**
+         * Deserialization constructor.
+         * @param map The provenances.
+         */
         public JsonDataSourceProvenance(Map<String,Provenance> map) {
             this(extractProvenanceInfo(map));
         }
@@ -156,6 +160,11 @@ private JsonDataSourceProvenance(ExtractedInfo info) {
             this.sha256Hash = (HashProvenance) info.instanceValues.get(RESOURCE_HASH);
         }
 
+        /**
+         * Splits the provenance into configurable and non-configurable provenances.
+         * @param map The provenances to split.
+         * @return The extracted information.
+         */
         protected static ExtractedInfo extractProvenanceInfo(Map<String,Provenance> map) {
             Map<String,Provenance> configuredParameters = new HashMap<>(map);
             String className = ObjectProvenance.checkAndExtractProvenance(configuredParameters,CLASS_NAME, StringProvenance.class, JsonDataSourceProvenance.class.getSimpleName()).getValue();
diff --git a/Json/src/main/java/org/tribuo/json/StripProvenance.java b/Json/src/main/java/org/tribuo/json/StripProvenance.java
index 34bc47a7c..dbdbdab7a 100644
--- a/Json/src/main/java/org/tribuo/json/StripProvenance.java
+++ b/Json/src/main/java/org/tribuo/json/StripProvenance.java
@@ -249,16 +249,34 @@ public String getOptionsDescription() {
             return "A program for removing Provenance information from a Tribuo Model or SequenceModel.";
         }
 
+        /**
+         * Stores a hash of the model provenance in the stripped model.
+         */
         @Option(charName = 'h', longName = "store-provenance-hash", usage = "Stores a hash of the model provenance in the stripped model.")
         public boolean storeHash;
+        /**
+         * The model to load.
+         */
         @Option(charName = 'i', longName = "input-model-path", usage = "The model to load.")
         public File inputModel;
+        /**
+         * The location to write out the stripped model.
+         */
         @Option(charName = 'o', longName = "output-model-path", usage = "The location to write out the stripped model.")
         public File outputModel;
+        /**
+         * Write out the stripped provenance as json.
+         */
         @Option(charName = 'p', longName = "provenance-path", usage = "Write out the stripped provenance as json.")
         public File provenanceFile;
+        /**
+         * The provenances to remove
+         */
         @Option(charName = 'r', longName = "remove-provenances", usage = "The provenances to remove")
         public EnumSet<ProvenanceTypes> removeProvenances = EnumSet.noneOf(ProvenanceTypes.class);
+        /**
+         * The hash type to use.
+         */
         @Option(charName = 't', longName = "hash-type", usage = "The hash type to use.")
         public ProvenanceUtil.HashType hashType = ObjectProvenance.DEFAULT_HASH_TYPE;
     }
diff --git a/Math/src/main/java/org/tribuo/math/la/DenseMatrix.java b/Math/src/main/java/org/tribuo/math/la/DenseMatrix.java
index 0dbf1258a..95c9b4bf6 100644
--- a/Math/src/main/java/org/tribuo/math/la/DenseMatrix.java
+++ b/Math/src/main/java/org/tribuo/math/la/DenseMatrix.java
@@ -171,6 +171,11 @@ public double get(int i, int j) {
         return values[i][j];
     }
 
+    /**
+     * Constructs a dense vector by gathering values across dimension 1.
+     * @param elements The indices to gather.
+     * @return A dense vector.
+     */
     public DenseVector gatherAcrossDim1(int[] elements) {
         if (elements.length != dim2) {
             throw new IllegalArgumentException("Invalid number of elements to gather, must select one per value of dim2");
@@ -184,6 +189,11 @@ public DenseVector gatherAcrossDim1(int[] elements) {
         return new DenseVector(outputValues);
     }
 
+    /**
+     * Constructs a dense vector by gathering values across dimension 2.
+     * @param elements The indices to gather.
+     * @return A dense vector.
+     */
     public DenseVector gatherAcrossDim2(int[] elements) {
         if (elements.length != dim1) {
             throw new IllegalArgumentException("Invalid number of elements to gather, must select one per value of dim1");
@@ -512,6 +522,11 @@ public void add(int i, int j, double value) {
         values[i][j] += value;
     }
 
+    /**
+     * Adds the specified value to the specified elements across dimension 1.
+     * @param indices The indices to update.
+     * @param value The value to add.
+     */
     public void addAcrossDim1(int[] indices, double value) {
         if (indices.length != dim2) {
             throw new IllegalArgumentException("Invalid number of elements to add, must select one per value of dim2");
@@ -521,6 +536,11 @@ public void addAcrossDim1(int[] indices, double value) {
         }
     }
 
+    /**
+     * Adds the specified value to the specified elements across dimension 2.
+     * @param indices The indices to update.
+     * @param value The value to add.
+     */
     public void addAcrossDim2(int[] indices, double value) {
         if (indices.length != dim1) {
             throw new IllegalArgumentException("Invalid number of elements to indices, must select one per value of dim1");
diff --git a/Math/src/main/java/org/tribuo/math/la/DenseSparseMatrix.java b/Math/src/main/java/org/tribuo/math/la/DenseSparseMatrix.java
index 0e8c10149..84819bf7a 100644
--- a/Math/src/main/java/org/tribuo/math/la/DenseSparseMatrix.java
+++ b/Math/src/main/java/org/tribuo/math/la/DenseSparseMatrix.java
@@ -42,16 +42,29 @@ public class DenseSparseMatrix implements Matrix {
         this.shape = new int[]{dim1,dim2};
     }
 
+    /***
+     * Constructs a DenseSparseMatrix out of the supplied sparse vector list.
+     * <p>
+     * Throws IllegalArgumentException if the list forms a ragged matrix.
+     * @param values The list of sparse vectors.
+     */
     public DenseSparseMatrix(List<SparseVector> values) {
         this.values = new SparseVector[values.size()];
         this.dim1 = values.size();
         this.dim2 = values.get(0).size();
         this.shape = new int[]{dim1,dim2};
         for (int i = 0; i < values.size(); i++) {
+            if (values.get(i).size() != dim2) {
+                throw new IllegalArgumentException("Unexpected size, found " + values.get(i).size() + ", expected " + dim2);
+            }
             this.values[i] = values.get(i);
         }
     }
 
+    /**
+     * Creates a new DenseSparseMatrix by deep copying the supplied DenseSparseMatrix.
+     * @param other The matrix to copy.
+     */
     public DenseSparseMatrix(DenseSparseMatrix other) {
         this.dim1 = other.dim1;
         this.dim2 = other.dim2;
diff --git a/Math/src/main/java/org/tribuo/math/la/DenseVector.java b/Math/src/main/java/org/tribuo/math/la/DenseVector.java
index 9242574c4..22b4c3372 100644
--- a/Math/src/main/java/org/tribuo/math/la/DenseVector.java
+++ b/Math/src/main/java/org/tribuo/math/la/DenseVector.java
@@ -39,10 +39,19 @@ public class DenseVector implements SGDVector {
     private final int[] shape;
     protected final double[] elements;
 
+    /**
+     * Creates an empty dense vector of the specified size.
+     * @param size The vector size.
+     */
     public DenseVector(int size) {
         this(size,0.0);
     }
 
+    /**
+     * Creates a dense vector of the specified size where each element is initialised to the specified value.
+     * @param size The vector size.
+     * @param value The initial value.
+     */
     public DenseVector(int size, double value) {
         this.elements = new double[size];
         Arrays.fill(this.elements,value);
@@ -376,6 +385,11 @@ public double sum() {
         return sum;
     }
 
+    /**
+     * Sums this vector, applying the supplied function to each element first.
+     * @param f The function to apply to the elements.
+     * @return The sum of f(x).
+     */
     public double sum(DoubleUnaryOperator f) {
         double sum = 0.0;
         for (int i = 0; i < elements.length; i++) {
diff --git a/Math/src/main/java/org/tribuo/math/la/MatrixTuple.java b/Math/src/main/java/org/tribuo/math/la/MatrixTuple.java
index 907d90cbc..6668130b1 100644
--- a/Math/src/main/java/org/tribuo/math/la/MatrixTuple.java
+++ b/Math/src/main/java/org/tribuo/math/la/MatrixTuple.java
@@ -25,17 +25,35 @@
  */
 public class MatrixTuple {
 
+    /**
+     * The i index.
+     */
     public int i;
+    /**
+     * The j index.
+     */
     public int j;
 
+    /**
+     * The value.
+     */
     public double value;
 
+    /**
+     * Constructs an empty matrix tuple.
+     */
     public MatrixTuple() {
         this.i = -1;
         this.j = -1;
         this.value = Double.NaN;
     }
 
+    /**
+     * Constructs a matrix tuple with the specified values.
+     * @param i The i index.
+     * @param j The j index.
+     * @param value The value.
+     */
     public MatrixTuple(int i, int j, int value) {
         this.i = i;
         this.j = j;
diff --git a/Math/src/main/java/org/tribuo/math/la/SparseVector.java b/Math/src/main/java/org/tribuo/math/la/SparseVector.java
index 8b6ef79a0..480ac403e 100644
--- a/Math/src/main/java/org/tribuo/math/la/SparseVector.java
+++ b/Math/src/main/java/org/tribuo/math/la/SparseVector.java
@@ -90,6 +90,12 @@ private SparseVector(SparseVector other) {
         this.shape = new int[]{size};
     }
 
+    /**
+     * Creates a sparse vector of the specified size, with the supplied value at each of the indices.
+     * @param size The vector size.
+     * @param indices The indices of the sparse vector.
+     * @param value The initial value.
+     */
     public SparseVector(int size, int[] indices, double value) {
         this.indices = Arrays.copyOf(indices,indices.length);
         this.values = new double[indices.length];
@@ -810,6 +816,13 @@ public double l1Distance(SGDVector other) {
         return distance(other,Math::abs,DoubleUnaryOperator.identity());
     }
 
+    /**
+     * Computes the distance between this vector and the other vector.
+     * @param other The other vector.
+     * @param transformFunc The transformation function to apply to each paired dimension difference.
+     * @param normalizeFunc The normalization to apply after summing the transformed differences.
+     * @return The distance between the two vectors.
+     */
     public double distance(SGDVector other, DoubleUnaryOperator transformFunc, DoubleUnaryOperator normalizeFunc) {
         if (other.size() != size) {
             throw new IllegalArgumentException("Can't measure the distance between two vectors of different lengths, this = " + size + ", other = " + other.size());
diff --git a/Math/src/main/java/org/tribuo/math/la/Tensor.java b/Math/src/main/java/org/tribuo/math/la/Tensor.java
index 33b7cf284..b0cd22efa 100644
--- a/Math/src/main/java/org/tribuo/math/la/Tensor.java
+++ b/Math/src/main/java/org/tribuo/math/la/Tensor.java
@@ -25,6 +25,11 @@
  */
 public interface Tensor extends Serializable {
 
+    /**
+     * The number of elements in this shape, i.e., the product of the shape array.
+     * @param shape The tensor shape.
+     * @return The number of elements in the shape.
+     */
     public static int shapeSum(int[] shape) {
         int sum = 1;
         for (int i = 0; i < shape.length; i++) {
@@ -33,6 +38,15 @@ public static int shapeSum(int[] shape) {
         return sum;
     }
 
+    /**
+     * Checks that the two tensors have compatible shapes.
+     * <p>
+     * Compatible shapes are those which are exactly equal, as Tribuo does not
+     * support broadcasting.
+     * @param first The first tensor.
+     * @param second The second tensor.
+     * @return True if the shapes are the same.
+     */
     public static boolean shapeCheck(Tensor first, Tensor second) {
         if ((first != null) && (second != null)) {
             return Arrays.equals(first.getShape(),second.getShape());
diff --git a/Math/src/main/java/org/tribuo/math/la/VectorTuple.java b/Math/src/main/java/org/tribuo/math/la/VectorTuple.java
index 6aff3e7f2..f95a574e0 100644
--- a/Math/src/main/java/org/tribuo/math/la/VectorTuple.java
+++ b/Math/src/main/java/org/tribuo/math/la/VectorTuple.java
@@ -19,21 +19,40 @@
 /**
  * A mutable tuple used to avoid allocation when iterating a vector.
  * <p>
+ * It's basically a cursor into a vector.
+ * <p>
  * While it does implement hashcode, don't use it as a key as the hashcode is data dependent.
  */
 public class VectorTuple {
 
+    /**
+     * The tolerance for equality in value comparisons.
+     */
     public static final double DELTA = 1e-12;
 
+    /**
+     * The current index.
+     */
     public int index;
 
+    /**
+     * The current value.
+     */
     public double value;
 
+    /**
+     * Creates an empty vector tuple.
+     */
     public VectorTuple() {
         this.index = -1;
         this.value = Double.NaN;
     }
 
+    /**
+     * Creates a vector tuple with the specified values.
+     * @param index The current index.
+     * @param value The current value.
+     */
     public VectorTuple(int index, int value) {
         this.index = index;
         this.value = value;
diff --git a/Math/src/main/java/org/tribuo/math/optimisers/AdaGrad.java b/Math/src/main/java/org/tribuo/math/optimisers/AdaGrad.java
index b724ff6d7..91eb2932d 100644
--- a/Math/src/main/java/org/tribuo/math/optimisers/AdaGrad.java
+++ b/Math/src/main/java/org/tribuo/math/optimisers/AdaGrad.java
@@ -52,21 +52,41 @@ public class AdaGrad implements StochasticGradientOptimiser {
 
     private Tensor[] gradsSquared;
 
-    public AdaGrad(double initialLearningRate, double epsilon) {
+    /**
+     * Creates an AdaGrad optimiser using the specified learning rate, epsilon and initial accumulator value.
+     * @param initialLearningRate The learning rate.
+     * @param epsilon The epsilon value for stabilising the gradient inversion.
+     * @param initialValue The initial value for the gradient accumulator.
+     */
+    public AdaGrad(double initialLearningRate, double epsilon, double initialValue) {
         this.initialLearningRate = initialLearningRate;
         this.epsilon = epsilon;
+        this.initialValue = initialValue;
+    }
+
+    /**
+     * Creates an AdaGrad optimiser using the specified learning rate and epsilon.
+     * <p>
+     * Sets the initial value for the accumulator to zero.
+     * @param initialLearningRate The learning rate.
+     * @param epsilon The epsilon value for stabilising the gradient inversion.
+     */
+    public AdaGrad(double initialLearningRate, double epsilon) {
+        this(initialLearningRate,epsilon,0.0);
     }
 
     /**
-     * Sets epsilon to 1e-6.
+     * Creates an AdaGrad optimiser using the specified initial learning rate.
+     * <p>
+     * Sets epsilon to 1e-6, and the initial accumulator value to zero.
      * @param initialLearningRate The learning rate.
      */
     public AdaGrad(double initialLearningRate) {
-        this(initialLearningRate,1e-6);
+        this(initialLearningRate,1e-6,0.0);
     }
 
     /**
-     * For olcut.
+     * For OLCUT.
      */
     private AdaGrad() { }
 
diff --git a/Math/src/main/java/org/tribuo/math/optimisers/AdaGradRDA.java b/Math/src/main/java/org/tribuo/math/optimisers/AdaGradRDA.java
index 9351d1b16..6fa7f13f8 100644
--- a/Math/src/main/java/org/tribuo/math/optimisers/AdaGradRDA.java
+++ b/Math/src/main/java/org/tribuo/math/optimisers/AdaGradRDA.java
@@ -71,6 +71,14 @@ public class AdaGradRDA implements StochasticGradientOptimiser {
 
     private Parameters parameters = null;
 
+    /**
+     * Creates an AdaGradRDA optimiser with the specified parameter values.
+     * @param initialLearningRate The learning rate.
+     * @param epsilon The epsilon value for stabilising the gradient inversion.
+     * @param l1 The l1 penalty.
+     * @param l2 The l2 penalty.
+     * @param numExamples The number of examples to scale the l1 and l2 penalties.
+     */
     public AdaGradRDA(double initialLearningRate, double epsilon, double l1, double l2, int numExamples) {
         this.initialLearningRate = initialLearningRate;
         this.epsilon = epsilon;
@@ -79,12 +87,19 @@ public AdaGradRDA(double initialLearningRate, double epsilon, double l1, double
         this.numExamples = numExamples;
     }
 
+    /**
+     * Creates an AdaGradRDA optimiser with the specified parameter values.
+     * <p>
+     * Sets the regularisation parameters to zero.
+     * @param initialLearningRate The learning rate.
+     * @param epsilon The epsilon value for stabilising the gradient inversion.
+     */
     public AdaGradRDA(double initialLearningRate, double epsilon) {
         this(initialLearningRate,epsilon,0,0,1);
     }
 
     /**
-     * For olcut.
+     * For OLCUT.
      */
     private AdaGradRDA() { }
 
@@ -152,8 +167,18 @@ public ConfiguredObjectProvenance getProvenance() {
      * An interface which tags a {@link Tensor} with a convertToDense method.
      */
     private static interface AdaGradRDATensor {
+        /**
+         * Returns a dense copy of this tensor.
+         * @return A dense copy.
+         */
         public Tensor convertToDense();
 
+        /**
+         * Zeros the value around the threshold.
+         * @param input The value to truncate.
+         * @param threshold The threshold to zero.
+         * @return The input value with the region around the threshold removed.
+         */
         public static double truncate(double input, double threshold) {
             if (input > threshold) {
                 return input - threshold;
diff --git a/Math/src/main/java/org/tribuo/math/optimisers/GradientOptimiserOptions.java b/Math/src/main/java/org/tribuo/math/optimisers/GradientOptimiserOptions.java
index 30855caf4..be652c410 100644
--- a/Math/src/main/java/org/tribuo/math/optimisers/GradientOptimiserOptions.java
+++ b/Math/src/main/java/org/tribuo/math/optimisers/GradientOptimiserOptions.java
@@ -32,47 +32,93 @@ public class GradientOptimiserOptions implements Options {
      * Type of the gradient optimisers available in CLIs.
      */
     public enum StochasticGradientOptimiserType {
+        /**
+         * The AdaDelta optimiser.
+         */
         ADADELTA,
+        /**
+         * The AdaGrad optimiser.
+         */
         ADAGRAD,
+        /**
+         * The AdaGrad Regularised Dual Averaging optimiser.
+         */
         ADAGRADRDA,
+        /**
+         * The Adam optimiser.
+         */
         ADAM,
+        /**
+         * The Pegasos optimiser.
+         */
         PEGASOS,
+        /**
+         * The RMSProp optimiser.
+         */
         RMSPROP,
+        /**
+         * SGD with a constant learning rate.
+         */
         CONSTANTSGD,
+        /**
+         * SGD with a linear-decay learning rate.
+         */
         LINEARSGD,
+        /**
+         * SGD with a square root decay learning rate.
+         */
         SQRTSGD
     }
 
     @Option(longName = "sgo-type", usage = "Selects the gradient optimiser. Defaults to ADAGRAD.")
     private StochasticGradientOptimiserType optimiserType = StochasticGradientOptimiserType.ADAGRAD;
 
+    /**
+     * Learning rate for AdaGrad, AdaGradRDA, Adam, Pegasos.
+     */
     @Option(longName = "sgo-learning-rate", usage = "Learning rate for AdaGrad, AdaGradRDA, Adam, Pegasos.")
     public double learningRate = 0.18;
 
+    /**
+     * Epsilon for AdaDelta, AdaGrad, AdaGradRDA, Adam.
+     */
     @Option(longName = "sgo-epsilon", usage = "Epsilon for AdaDelta, AdaGrad, AdaGradRDA, Adam.")
     public double epsilon = 0.066;
 
+    /**
+     * Rho for RMSProp, AdaDelta, SGD with Momentum.
+     */
     @Option(longName = "sgo-rho", usage = "Rho for RMSProp, AdaDelta, SGD with Momentum.")
     public double rho = 0.95;
 
+    /**
+     * Lambda for Pegasos.
+     */
     @Option(longName = "sgo-lambda", usage = "Lambda for Pegasos.")
     public double lambda = 1e-2;
 
-    @Option(longName="sgo-parameter-averaging",usage="Use parameter averaging.")
+    /**
+     * Use parameter averaging.
+     */
+    @Option(longName = "sgo-parameter-averaging", usage = "Use parameter averaging.")
     public boolean paramAve = false;
 
-    @Option(longName="sgo-momentum",usage="Use momentum in SGD.")
+    /**
+     * Use momentum in SGD.
+     */
+    @Option(longName = "sgo-momentum", usage = "Use momentum in SGD.")
     public SGD.Momentum momentum = SGD.Momentum.NONE;
 
     /**
      * Gets the configured gradient optimiser.
+     *
      * @return The gradient optimiser.
      */
     public StochasticGradientOptimiser getOptimiser() {
         StochasticGradientOptimiser sgo;
-        switch(optimiserType) {
+        switch (optimiserType) {
             case ADADELTA:
-                sgo = new AdaDelta(rho,epsilon);
+                sgo = new AdaDelta(rho, epsilon);
                 break;
             case ADAGRAD:
                 sgo = new AdaGrad(learningRate, epsilon);
@@ -81,25 +127,25 @@ public StochasticGradientOptimiser getOptimiser() {
                 sgo = new AdaGradRDA(learningRate, epsilon);
                 break;
             case ADAM:
-                sgo = new Adam(learningRate,epsilon);
+                sgo = new Adam(learningRate, epsilon);
                 break;
             case PEGASOS:
-                sgo = new Pegasos(learningRate,lambda);
+                sgo = new Pegasos(learningRate, lambda);
                 break;
             case RMSPROP:
-                sgo = new RMSProp(learningRate,rho);
+                sgo = new RMSProp(learningRate, rho);
                 break;
             case CONSTANTSGD:
-                sgo = SGD.getSimpleSGD(learningRate,rho,momentum);
+                sgo = SGD.getSimpleSGD(learningRate, rho, momentum);
                 break;
             case LINEARSGD:
-                sgo = SGD.getLinearDecaySGD(learningRate,rho,momentum);
+                sgo = SGD.getLinearDecaySGD(learningRate, rho, momentum);
                 break;
             case SQRTSGD:
-                sgo = SGD.getSqrtDecaySGD(learningRate,rho,momentum);
+                sgo = SGD.getSqrtDecaySGD(learningRate, rho, momentum);
                 break;
             default:
-                throw new IllegalArgumentException("Unhandled StochasticGradientOptimiser type: "+optimiserType);
+                throw new IllegalArgumentException("Unhandled StochasticGradientOptimiser type: " + optimiserType);
         }
         if (paramAve) {
             logger.info("Using parameter averaging");
diff --git a/Math/src/main/java/org/tribuo/math/optimisers/Pegasos.java b/Math/src/main/java/org/tribuo/math/optimisers/Pegasos.java
index 5082b2cfd..7818977ef 100644
--- a/Math/src/main/java/org/tribuo/math/optimisers/Pegasos.java
+++ b/Math/src/main/java/org/tribuo/math/optimisers/Pegasos.java
@@ -63,6 +63,11 @@ public class Pegasos implements StochasticGradientOptimiser {
      */
     private Pegasos() { }
 
+    /**
+     * Constructs a Pegasos optimiser with the specified parameters.
+     * @param baseRate The base learning rate.
+     * @param lambda The regularisation parameter.
+     */
     public Pegasos(double baseRate, double lambda) {
         this.baseRate = baseRate;
         this.lambda = lambda;
diff --git a/Math/src/main/java/org/tribuo/math/optimisers/RMSProp.java b/Math/src/main/java/org/tribuo/math/optimisers/RMSProp.java
index 1e70a91b0..691c9c060 100644
--- a/Math/src/main/java/org/tribuo/math/optimisers/RMSProp.java
+++ b/Math/src/main/java/org/tribuo/math/optimisers/RMSProp.java
@@ -66,6 +66,13 @@ public class RMSProp implements StochasticGradientOptimiser {
 
     private DoubleUnaryOperator square;
 
+    /**
+     * Constructs an RMSProp gradient optimiser using the specified parameters.
+     * @param initialLearningRate The initial learning rate.
+     * @param rho The momentum parameter.
+     * @param epsilon The epsilon to ensure division stability.
+     * @param decay The decay parameter.
+     */
     public RMSProp(double initialLearningRate, double rho, double epsilon, double decay) {
         this.initialLearningRate = initialLearningRate;
         this.rho = rho;
@@ -75,6 +82,11 @@ public RMSProp(double initialLearningRate, double rho, double epsilon, double de
         postConfig();
     }
 
+    /**
+     * Constructs an RMSProp gradient optimiser using the specified parameters with epsilon set to 1e-8 and decay to 0.0.
+     * @param initialLearningRate The initial learning rate.
+     * @param rho The momentum parameter.
+     */
     public RMSProp(double initialLearningRate, double rho) {
         this(initialLearningRate,rho,1e-8,0.0);
     }
diff --git a/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingMatrix.java b/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingMatrix.java
index 1b7cc4c1d..3b61fb9e4 100644
--- a/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingMatrix.java
+++ b/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingMatrix.java
@@ -41,6 +41,14 @@ public class ShrinkingMatrix extends DenseMatrix implements ShrinkingTensor {
     private int iteration;
     private double multiplier;
 
+    /**
+     * Constructs a shrinking matrix copy of the supplied dense matrix.
+     * <p>
+     * This matrix shrinks during each call to {@link #intersectAndAddInPlace(Tensor, DoubleUnaryOperator)}.
+     * @param v The matrix to copy.
+     * @param baseRate The base amount of shrinking to apply after each update.
+     * @param scaleShrinking If true reduce the shrinking value over time proportionally to the number of updates.
+     */
     public ShrinkingMatrix(DenseMatrix v, double baseRate, boolean scaleShrinking) {
         super(v);
         this.baseRate = baseRate;
@@ -52,6 +60,15 @@ public ShrinkingMatrix(DenseMatrix v, double baseRate, boolean scaleShrinking) {
         this.multiplier = 1.0;
     }
 
+    /**
+     * Constructs a shrinking matrix copy of the supplied dense matrix.
+     * <p>
+     * This matrix shrinks during each call to {@link #intersectAndAddInPlace(Tensor, DoubleUnaryOperator)},
+     * and then reprojects the matrix so it has the same twoNorm.
+     * @param v The matrix to copy.
+     * @param baseRate The base rate of shrinkage.
+     * @param lambda The lambda value (see {@link org.tribuo.math.optimisers.Pegasos}).
+     */
     public ShrinkingMatrix(DenseMatrix v, double baseRate, double lambda) {
         super(v);
         this.baseRate = baseRate;
diff --git a/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingTensor.java b/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingTensor.java
index 8beef61a6..678ecc460 100644
--- a/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingTensor.java
+++ b/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingTensor.java
@@ -22,8 +22,15 @@
  * An interface which tags a {@link Tensor} with a convertToDense method.
  */
 public interface ShrinkingTensor {
+    /**
+     * The tolerance below which the scale factor is applied to the stored values and reset to 1.0.
+     */
     public static final double tolerance = 1e-6;
 
+    /**
+     * Converts the tensor into a dense tensor.
+     * @return A dense tensor copy of this shrinking tensor.
+     */
     public Tensor convertToDense();
 }
 
diff --git a/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingVector.java b/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingVector.java
index 4ff9819b9..38ab7ae2d 100644
--- a/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingVector.java
+++ b/Math/src/main/java/org/tribuo/math/optimisers/util/ShrinkingVector.java
@@ -39,6 +39,14 @@ public class ShrinkingVector extends DenseVector implements ShrinkingTensor {
     private int iteration;
     private double multiplier;
 
+    /**
+     * Constructs a shrinking vector copy of the supplied dense matrix.
+     * <p>
+     * This vector shrinks during each call to {@link #intersectAndAddInPlace(Tensor, DoubleUnaryOperator)}.
+     * @param v The vector to copy.
+     * @param baseRate The base amount of shrinking to apply after each update.
+     * @param scaleShrinking If true reduce the shrinking value over time proportionally to the number of updates.
+     */
     public ShrinkingVector(DenseVector v, double baseRate, boolean scaleShrinking) {
         super(v);
         this.baseRate = baseRate;
@@ -49,6 +57,15 @@ public ShrinkingVector(DenseVector v, double baseRate, boolean scaleShrinking) {
         this.multiplier = 1.0;
     }
 
+    /**
+     * Constructs a shrinking vector copy of the supplied dense vector.
+     * <p>
+     * This vector shrinks during each call to {@link #intersectAndAddInPlace(Tensor, DoubleUnaryOperator)},
+     * and then reprojects the vector so it has the same twoNorm.
+     * @param v The vector to copy.
+     * @param baseRate The base rate of shrinkage.
+     * @param lambda The lambda value (see {@link org.tribuo.math.optimisers.Pegasos}).
+     */
     public ShrinkingVector(DenseVector v, double baseRate, double lambda) {
         super(v);
         this.baseRate = baseRate;
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/MultiLabel.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/MultiLabel.java
index 7884ef653..7c974ff17 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/MultiLabel.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/MultiLabel.java
@@ -50,7 +50,11 @@
 public class MultiLabel implements Classifiable<MultiLabel> {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * The string for the binary negative label.
+     */
     public static final String NEGATIVE_LABEL_STRING = "ML##NEGATIVE";
+
     /**
      * A Label representing the binary negative label. Used in binary
      * approaches to multi-label classification to represent the absence
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/MultiLabelFactory.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/MultiLabelFactory.java
index 6cad405dc..1ef213b1d 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/MultiLabelFactory.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/MultiLabelFactory.java
@@ -40,6 +40,9 @@
 public final class MultiLabelFactory implements OutputFactory<MultiLabel> {
     private static final long serialVersionUID = 1L;
 
+    /**
+     * The sentinel unknown multi-label output used to signal there is no ground truth value.
+     */
     public static final MultiLabel UNKNOWN_MULTILABEL = new MultiLabel(LabelFactory.UNKNOWN_LABEL);
 
     private static final MultiLabelFactoryProvenance provenance = new MultiLabelFactoryProvenance();
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/ClassifierChainModel.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/ClassifierChainModel.java
index 9de07754d..fd13bc2b5 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/ClassifierChainModel.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/ClassifierChainModel.java
@@ -58,7 +58,7 @@
  * <p>
  * See:
  * <pre>
- * Read, J., Pfahringer, B., Holmes, G., & Frank, E.
+ * Read, J., Pfahringer, B., Holmes, G., &amp; Frank, E.
  * "Classifier Chains for Multi-Label Classification"
  * Machine Learning, pages 333-359, 2011.
  * </pre>
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/ClassifierChainTrainer.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/ClassifierChainTrainer.java
index 1cd4465b8..7cc8f417d 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/ClassifierChainTrainer.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/ClassifierChainTrainer.java
@@ -65,7 +65,7 @@
  * <p>
  * See:
  * <pre>
- * Read, J., Pfahringer, B., Holmes, G., & Frank, E.
+ * Read, J., Pfahringer, B., Holmes, G., &amp; Frank, E.
  * "Classifier Chains for Multi-Label Classification"
  * Machine Learning, pages 333-359, 2011.
  * </pre>
@@ -73,9 +73,21 @@
 public final class ClassifierChainTrainer implements Trainer<MultiLabel> {
     private static final Logger logger = Logger.getLogger(ClassifierChainTrainer.class.getName());
 
+    /**
+     * The prefix for classifier chain added features.
+     */
     public static final String CC_PREFIX = "CC_FEATURES";
+    /**
+     * The string used in the feature name for positive labels.
+     */
     public static final String CC_POSITIVE = "POSITIVE";
+    /**
+     * The string used in the feature name for negative labels.
+     */
     public static final String CC_NEGATIVE = "NEGATIVE";
+    /**
+     * The joiner character for classifier chain added features.
+     */
     public static final String CC_SEPARATOR = "_";
 
     @Config(mandatory = true, description = "The trainer to use.")
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/IndependentMultiLabelModel.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/IndependentMultiLabelModel.java
index 3a7bd8a53..f4aeb8462 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/IndependentMultiLabelModel.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/IndependentMultiLabelModel.java
@@ -54,14 +54,15 @@ public class IndependentMultiLabelModel extends Model<MultiLabel> {
 
     /**
      * The list of Label and list of Model must be in the same order, and have a bijection.
-     * @param labels The list of labels this model was trained on.
-     * @param models The list of individual binary models.
+     *
+     * @param labels      The list of labels this model was trained on.
+     * @param models      The list of individual binary models.
      * @param description A description of the trainer.
-     * @param featureMap The feature domain used in training.
-     * @param labelInfo The label domain used in training.
+     * @param featureMap  The feature domain used in training.
+     * @param labelInfo   The label domain used in training.
      */
     IndependentMultiLabelModel(List<Label> labels, List<Model<Label>> models, ModelProvenance description, ImmutableFeatureMap featureMap, ImmutableOutputInfo<MultiLabel> labelInfo) {
-        super("binary-relevance",description,featureMap,labelInfo,models.get(0).generatesProbabilities());
+        super("binary-relevance", description, featureMap, labelInfo, models.get(0).generatesProbabilities());
         this.labels = labels;
         this.models = models;
     }
@@ -69,7 +70,7 @@ public class IndependentMultiLabelModel extends Model<MultiLabel> {
     @Override
     public Prediction<MultiLabel> predict(Example<MultiLabel> example) {
         Set<Label> predictedLabels = new HashSet<>();
-        BinaryExample e = new BinaryExample(example,null);
+        BinaryExample e = new BinaryExample(example, null);
         int numUsed = 0;
         for (Model<Label> m : models) {
             Prediction<Label> p = m.predict(e);
@@ -80,7 +81,7 @@ public Prediction<MultiLabel> predict(Example<MultiLabel> example) {
                 predictedLabels.add(p.getOutput());
             }
         }
-        return new Prediction<>(new MultiLabel(predictedLabels),numUsed,example);
+        return new Prediction<>(new MultiLabel(predictedLabels), numUsed, example);
     }
 
     /**
@@ -88,23 +89,26 @@ public Prediction<MultiLabel> predict(Example<MultiLabel> example) {
      * <p>
      * If the individual models support per label features, then only the features
      * for the positive label are aggregated.
-     * <p>
+     *
      * @param n the number of features to return. If this value is less than 0,
-     * all features should be returned for each class.
+     *          all features should be returned for each class.
      * @return The top n features.
      */
     @Override
     public Map<String, List<Pair<String, Double>>> getTopFeatures(int n) {
-        Map<String,List<Pair<String,Double>>> map = new HashMap<>();
+        Map<String, List<Pair<String, Double>>> map = new HashMap<>();
         for (int i = 0; i < models.size(); i++) {
             Model<Label> m = models.get(i);
             String label = labels.get(i).getLabel();
-            Map<String,List<Pair<String,Double>>> modelMap = m.getTopFeatures(n);
+            Map<String, List<Pair<String, Double>>> modelMap = m.getTopFeatures(n);
             if (modelMap != null) {
                 if (modelMap.size() == 1) {
-                    map.put(label,modelMap.get(Model.ALL_OUTPUTS));
+                    map.put(label, modelMap.get(Model.ALL_OUTPUTS));
                 } else {
-                    map.merge(label,modelMap.get(label),(List<Pair<String,Double>> l, List<Pair<String,Double>> r) -> {l.addAll(r); return l;});
+                    map.merge(label, modelMap.get(label), (List<Pair<String, Double>> l, List<Pair<String, Double>> r) -> {
+                        l.addAll(r);
+                        return l;
+                    });
                 }
             }
         }
@@ -123,6 +127,6 @@ protected IndependentMultiLabelModel copy(String newName, ModelProvenance newPro
         for (Model<Label> e : models) {
             newModels.add(e.copy());
         }
-        return new IndependentMultiLabelModel(labels,newModels,newProvenance,featureIDMap,outputIDInfo);
+        return new IndependentMultiLabelModel(labels, newModels, newProvenance, featureIDMap, outputIDInfo);
     }
 }
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/IndependentMultiLabelTrainer.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/IndependentMultiLabelTrainer.java
index d28c9501f..a814cca63 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/IndependentMultiLabelTrainer.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/baseline/IndependentMultiLabelTrainer.java
@@ -61,6 +61,10 @@ public class IndependentMultiLabelTrainer implements Trainer<MultiLabel> {
      */
     private IndependentMultiLabelTrainer() {}
 
+    /**
+     * Constructs an independent multi-label trainer wrapped around the supplied classification trainer.
+     * @param innerTrainer The trainer to use for each individual label.
+     */
     public IndependentMultiLabelTrainer(Trainer<Label> innerTrainer) {
         this.innerTrainer = innerTrainer;
     }
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/ensemble/CCEnsembleTrainer.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/ensemble/CCEnsembleTrainer.java
index 9f1d821eb..00feeafa4 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/ensemble/CCEnsembleTrainer.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/ensemble/CCEnsembleTrainer.java
@@ -53,7 +53,7 @@
  * <p>
  * See:
  * <pre>
- * Read, J., Pfahringer, B., Holmes, G., & Frank, E.
+ * Read, J., Pfahringer, B., Holmes, G., &amp; Frank, E.
  * "Classifier Chains for Multi-Label Classification"
  * Machine Learning, pages 333-359, 2011.
  * </pre>
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/evaluation/MultiLabelConfusionMatrix.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/evaluation/MultiLabelConfusionMatrix.java
index 4627c7995..cdd5f5154 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/evaluation/MultiLabelConfusionMatrix.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/evaluation/MultiLabelConfusionMatrix.java
@@ -56,6 +56,11 @@ public final class MultiLabelConfusionMatrix implements ConfusionMatrix<MultiLab
     private final DenseMatrix[] mcm;
     private final DenseMatrix confusion;
 
+    /**
+     * Constructs a multi-label confusion matrix for the specified model and predictions.
+     * @param model The model.
+     * @param predictions The predictions.
+     */
     public MultiLabelConfusionMatrix(Model<MultiLabel> model, List<Prediction<MultiLabel>> predictions) {
         this(model.getOutputIDInfo(), predictions);
     }
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/evaluation/MultiLabelMetric.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/evaluation/MultiLabelMetric.java
index 05b12f53b..43b03b3c1 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/evaluation/MultiLabelMetric.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/evaluation/MultiLabelMetric.java
@@ -38,6 +38,12 @@ public class MultiLabelMetric implements EvaluationMetric<MultiLabel, MultiLabel
     private final String name;
     private final BiFunction<MetricTarget<MultiLabel>, Context, Double> impl;
 
+    /**
+     * Constructs a multi-label metric.
+     * @param target The metric target.
+     * @param name The name of the metric.
+     * @param impl The implementing function for this metric.
+     */
     public MultiLabelMetric(MetricTarget<MultiLabel> target, String name, BiFunction<MetricTarget<MultiLabel>, Context, Double> impl) {
         this.target = target;
         this.name = name;
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/example/MultiLabelDataGenerator.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/example/MultiLabelDataGenerator.java
index f5ee11c68..cef67dd51 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/example/MultiLabelDataGenerator.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/example/MultiLabelDataGenerator.java
@@ -38,9 +38,12 @@ public class MultiLabelDataGenerator {
 
     private static final MultiLabelFactory factory = new MultiLabelFactory();
 
-    private MultiLabelDataGenerator() {
-    }
+    private MultiLabelDataGenerator() {}
 
+    /**
+     * Simple training data for checking multi-label trainers.
+     * @return Simple 3 class training data.
+     */
     public static Dataset<MultiLabel> generateTrainData() {
         DataSourceProvenance provenance = new SimpleDataSourceProvenance("TrainingData", OffsetDateTime.now(), factory);
         MutableDataset<MultiLabel> dataset = new MutableDataset<>(provenance, factory);
@@ -90,6 +93,10 @@ public static Dataset<MultiLabel> generateTrainData() {
         return dataset;
     }
 
+    /**
+     * Simple test data for checking multi-label trainers.
+     * @return Simple 3 class test data.
+     */
     public static Dataset<MultiLabel> generateTestData() {
         DataSourceProvenance provenance = new SimpleDataSourceProvenance("TestingData", OffsetDateTime.now(), factory);
         MutableDataset<MultiLabel> dataset = new MutableDataset<>(provenance,factory);
@@ -121,6 +128,10 @@ public static Dataset<MultiLabel> generateTestData() {
         return dataset;
     }
 
+    /**
+     * Generate training and testing datasets.
+     * @return A pair of datasets.
+     */
     public static Pair<Dataset<MultiLabel>,Dataset<MultiLabel>> generateDataset() {
         return new Pair<>(generateTrainData(),generateTestData());
     }
diff --git a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/example/MultiLabelGaussianDataSource.java b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/example/MultiLabelGaussianDataSource.java
index ff7bacd26..a6c124f7d 100644
--- a/MultiLabel/Core/src/main/java/org/tribuo/multilabel/example/MultiLabelGaussianDataSource.java
+++ b/MultiLabel/Core/src/main/java/org/tribuo/multilabel/example/MultiLabelGaussianDataSource.java
@@ -52,9 +52,9 @@
  * <p>
  * The functions are:
  * <ul>
- *     <li>y_0 is positive if N(w_00*x_0 + w_01*x_1 + w_02*x_1*x_0 + w_03*x_1*x_1*x_1,variance) > threshold_0.</li>
- *     <li>y_1 is positive if N(w_10*x_0 + w_11*x_1 + w_12*x_1*x_0 + w_13*x_1*x_1*x_1,variance) < threshold_1.</li>
- *     <li>y_2 is positive if N(w_20*x_0 + w_21*x_2 + w_22*x_1*x_0 + w_23*x_1*x_2*x_2,variance) > threshold_2.</li>
+ *     <li>y_0 is positive if N(w_00*x_0 + w_01*x_1 + w_02*x_1*x_0 + w_03*x_1*x_1*x_1,variance) &gt; threshold_0.</li>
+ *     <li>y_1 is positive if N(w_10*x_0 + w_11*x_1 + w_12*x_1*x_0 + w_13*x_1*x_1*x_1,variance) &lt; threshold_1.</li>
+ *     <li>y_2 is positive if N(w_20*x_0 + w_21*x_2 + w_22*x_1*x_0 + w_23*x_1*x_2*x_2,variance) &gt; threshold_2.</li>
  * </ul>
  * There are 4 features x_0, x_1, x_2, x_3. x_0 and x_1 are used by all three labels,
  * x_2 is only used by y_2, and x_3 is irrelevant.
@@ -279,9 +279,9 @@ public static MutableDataset<MultiLabel> generateDataset(int numSamples, float[]
      * <p>
      * The functions are:
      * <ul>
-     *     <li>y_0 is positive if N(w_00*x_0 + w_01*x_1 + w_02*x_1*x_0 + w_03*x_1*x_1*x_1,variance) > threshold_0.</li>
-     *     <li>y_1 is positive if N(w_10*x_0 + w_11*x_1 + w_12*x_1*x_0 + w_13*x_1*x_1*x_1,variance) < threshold_1.</li>
-     *     <li>y_2 is positive if N(w_20*x_0 + w_21*x_2 + w_22*x_1*x_0 + w_23*x_1*x_2*x_2,variance) > threshold_2.</li>
+     *     <li>y_0 is positive if N(w_00*x_0 + w_01*x_1 + w_02*x_1*x_0 + w_03*x_1*x_1*x_1,variance) &gt; threshold_0.</li>
+     *     <li>y_1 is positive if N(w_10*x_0 + w_11*x_1 + w_12*x_1*x_0 + w_13*x_1*x_1*x_1,variance) &lt; threshold_1.</li>
+     *     <li>y_2 is positive if N(w_20*x_0 + w_21*x_2 + w_22*x_1*x_0 + w_23*x_1*x_2*x_2,variance) &gt; threshold_2.</li>
      * </ul>
      * There are 4 features x_0, x_1, x_2, x_3. x_0 and x_1 are used by all three labels,
      * x_2 is only used by y_2, and x_3 is irrelevant.
diff --git a/MultiLabel/SGD/src/main/java/org/tribuo/multilabel/sgd/linear/LinearSGDOptions.java b/MultiLabel/SGD/src/main/java/org/tribuo/multilabel/sgd/linear/LinearSGDOptions.java
index e564755f1..114c45b98 100644
--- a/MultiLabel/SGD/src/main/java/org/tribuo/multilabel/sgd/linear/LinearSGDOptions.java
+++ b/MultiLabel/SGD/src/main/java/org/tribuo/multilabel/sgd/linear/LinearSGDOptions.java
@@ -36,23 +36,51 @@ public class LinearSGDOptions implements Options {
     /**
      * Available loss types.
      */
-    public enum LossEnum {HINGE, SIGMOID}
+    public enum LossEnum {
+        /**
+         * Hinge loss.
+         */
+        HINGE,
+        /**
+         * Log loss, i.e., binary cross-entropy.
+         */
+        SIGMOID
+    }
 
+    /**
+     * The gradient descent optimiser options.
+     */
     public GradientOptimiserOptions sgoOptions;
 
+    /**
+     * Number of SGD epochs.
+     */
     @Option(longName = "sgd-epochs", usage = "Number of SGD epochs.")
     public int sgdEpochs = 5;
+    /**
+     * Loss function.
+     */
     @Option(longName = "sgd-objective", usage = "Loss function.")
     public LossEnum sgdObjective = LossEnum.SIGMOID;
+    /**
+     * Log the objective after n examples.
+     */
     @Option(longName = "sgd-logging-interval", usage = "Log the objective after <int> examples.")
     public int sgdLoggingInterval = 100;
+    /**
+     * Minibatch size.
+     */
     @Option(longName = "sgd-minibatch-size", usage = "Minibatch size.")
     public int sgdMinibatchSize = 1;
+    /**
+     * Sets the random seed for the LinearSGDTrainer.
+     */
     @Option(longName = "sgd-seed", usage = "Sets the random seed for the LinearSGDTrainer.")
-    private long sgdSeed = Trainer.DEFAULT_SEED;
+    public long sgdSeed = Trainer.DEFAULT_SEED;
 
     /**
      * Returns the loss function specified in the arguments.
+     *
      * @return The loss function.
      */
     public MultiLabelObjective getLoss() {
@@ -66,6 +94,10 @@ public MultiLabelObjective getLoss() {
         }
     }
 
+    /**
+     * Gets the LinearSGDTrainer specified by the options in this object.
+     * @return The configured trainer.
+     */
     public LinearSGDTrainer getTrainer() {
         logger.info(String.format("Set logging interval to %d", sgdLoggingInterval));
         return new LinearSGDTrainer(getLoss(), sgoOptions.getOptimiser(), sgdEpochs, sgdLoggingInterval, sgdMinibatchSize, sgdSeed);
diff --git a/Regression/Core/src/main/java/org/tribuo/regression/MutableRegressionInfo.java b/Regression/Core/src/main/java/org/tribuo/regression/MutableRegressionInfo.java
index 8d2a5e1c1..f6f09b875 100644
--- a/Regression/Core/src/main/java/org/tribuo/regression/MutableRegressionInfo.java
+++ b/Regression/Core/src/main/java/org/tribuo/regression/MutableRegressionInfo.java
@@ -33,6 +33,10 @@ public class MutableRegressionInfo extends RegressionInfo implements MutableOutp
         super();
     }
 
+    /**
+     * Constructs a mutable copy of the supplied regression info.
+     * @param info The info to copy.
+     */
     public MutableRegressionInfo(RegressionInfo info) {
         super(info);
     }
diff --git a/Regression/Core/src/main/java/org/tribuo/regression/Regressor.java b/Regression/Core/src/main/java/org/tribuo/regression/Regressor.java
index cb70053ff..708fab840 100644
--- a/Regression/Core/src/main/java/org/tribuo/regression/Regressor.java
+++ b/Regression/Core/src/main/java/org/tribuo/regression/Regressor.java
@@ -237,7 +237,6 @@ public Optional<DimensionTuple> getDimension(String name) {
 
     /**
      * Returns a dimension tuple for the requested dimension index.
-     * <p>
      * @throws IndexOutOfBoundsException if the index is outside the range.
      * @param idx The dimension index.
      * @return A tuple representing that dimension.
diff --git a/Regression/Core/src/main/java/org/tribuo/regression/evaluation/RegressionSufficientStatistics.java b/Regression/Core/src/main/java/org/tribuo/regression/evaluation/RegressionSufficientStatistics.java
index 74b98ccd9..4a395038d 100644
--- a/Regression/Core/src/main/java/org/tribuo/regression/evaluation/RegressionSufficientStatistics.java
+++ b/Regression/Core/src/main/java/org/tribuo/regression/evaluation/RegressionSufficientStatistics.java
@@ -47,6 +47,12 @@ public final class RegressionSufficientStatistics {
     // if useExampleWeights is false, weightSum == n
     final float weightSum;
 
+    /**
+     * Constructs the sufficient statistics for regression metrics.
+     * @param domain The output domain.
+     * @param predictions The predictions.
+     * @param useExampleWeights Should example weights be used.
+     */
     public RegressionSufficientStatistics(ImmutableOutputInfo<Regressor> domain, List<Prediction<Regressor>> predictions, boolean useExampleWeights) {
         this.domain = domain;
         this.n = predictions.size();
diff --git a/Regression/Core/src/main/java/org/tribuo/regression/example/RegressionDataGenerator.java b/Regression/Core/src/main/java/org/tribuo/regression/example/RegressionDataGenerator.java
index 14b2d35d3..fbcbec86f 100644
--- a/Regression/Core/src/main/java/org/tribuo/regression/example/RegressionDataGenerator.java
+++ b/Regression/Core/src/main/java/org/tribuo/regression/example/RegressionDataGenerator.java
@@ -37,9 +37,18 @@
  */
 public abstract class RegressionDataGenerator {
 
+    /**
+     * Name of the first output dimension.
+     */
     public static final String firstDimensionName = "dim1";
+    /**
+     * Name of the second output dimension.
+     */
     public static final String secondDimensionName= "dim2";
 
+    /**
+     * Name of the single dimension.
+     */
     public static final String SINGLE_DIM_NAME = "REGRESSOR";
 
     private static final RegressionFactory REGRESSION_FACTORY = new RegressionFactory();
@@ -50,6 +59,11 @@ public abstract class RegressionDataGenerator {
      */
     private RegressionDataGenerator() {}
 
+    /**
+     * Generates a train/test dataset pair which is dense in the features,
+     * each example has 4 features,{A,B,C,D}.
+     * @return A pair of datasets.
+     */
     public static Pair<Dataset<Regressor>,Dataset<Regressor>> multiDimDenseTrainTest() {
         return multiDimDenseTrainTest(-1.0);
     }
@@ -106,6 +120,11 @@ public static Pair<Dataset<Regressor>,Dataset<Regressor>> multiDimDenseTrainTest
         return new Pair<>(train,test);
     }
 
+    /**
+     * Generates a pair of datasets, where the features are sparse,
+     * and unknown features appear in the test data.
+     * @return A pair of datasets.
+     */
     public static Pair<Dataset<Regressor>,Dataset<Regressor>> multiDimSparseTrainTest() {
         return multiDimSparseTrainTest(-1.0);
     }
@@ -194,6 +213,11 @@ public static Example<Regressor> emptyMultiDimExample() {
         return new ArrayExample<>(new Regressor(dimensionNames,new double[]{1,-1}),new String[]{},new double[]{});
     }
 
+    /**
+     * Generates a train/test dataset pair which is dense in the features,
+     * each example has 4 features,{A,B,C,D}.
+     * @return A pair of datasets.
+     */
     public static Pair<Dataset<Regressor>,Dataset<Regressor>> denseTrainTest() {
         return denseTrainTest(-1.0);
     }
@@ -252,6 +276,11 @@ public static Pair<Dataset<Regressor>,Dataset<Regressor>> denseTrainTest(double
         return new Pair<>(train,test);
     }
 
+    /**
+     * Generates a pair of datasets, where the features are sparse,
+     * and unknown features appear in the test data.
+     * @return A pair of datasets.
+     */
     public static Pair<Dataset<Regressor>,Dataset<Regressor>> sparseTrainTest() {
         return sparseTrainTest(-1.0);
     }
diff --git a/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/LibLinearRegressionTrainer.java b/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/LibLinearRegressionTrainer.java
index d2e7f5a1c..cda9482d6 100644
--- a/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/LibLinearRegressionTrainer.java
+++ b/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/LibLinearRegressionTrainer.java
@@ -70,6 +70,10 @@ public LibLinearRegressionTrainer() {
         this(new LinearRegressionType(LinearType.L2R_L2LOSS_SVR));
     }
 
+    /**
+     * Creates a trainer using the default values (1, 1000, 0.1, 0.1) and specified algorithm.
+     * @param trainerType The linear regression algorithm.
+     */
     public LibLinearRegressionTrainer(LinearRegressionType trainerType) {
         this(trainerType,1.0,1000,0.1,0.1);
     }
diff --git a/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/LinearRegressionType.java b/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/LinearRegressionType.java
index 918df70c7..0bd28031f 100644
--- a/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/LinearRegressionType.java
+++ b/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/LinearRegressionType.java
@@ -73,6 +73,10 @@ public SolverType getSolverType() {
      */
     private LinearRegressionType() {}
 
+    /**
+     * Constructs a LinearRegressionType with the specified LibLinear algorithm.
+     * @param type The liblinear algorithm.
+     */
     public LinearRegressionType(LinearType type) {
         this.type = type;
     }
diff --git a/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/TrainTest.java b/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/TrainTest.java
index 330524cac..9e40a39ff 100644
--- a/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/TrainTest.java
+++ b/Regression/LibLinear/src/main/java/org/tribuo/regression/liblinear/TrainTest.java
@@ -50,21 +50,41 @@ public static class LibLinearOptions implements Options {
         public String getOptionsDescription() {
             return "Trains and tests a LibLinear regression model on the specified datasets.";
         }
+
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
 
-        @Option(charName='p',longName="cost-penalty",usage="Cost penalty for SVM.")
+        /**
+         * Cost penalty for SVM.
+         */
+        @Option(charName = 'p', longName = "cost-penalty", usage = "Cost penalty for SVM.")
         public double cost = 1.0;
+        /**
+         * Max iterations over the data.
+         */
         @Option(longName = "max-iterations", usage = "Max iterations over the data.")
         public int maxIterations = 1000;
-        @Option(longName="epsilon-insensitivity",usage="Regression value insensitivity for margin.")
+        /**
+         * Regression value insensitivity for margin.
+         */
+        @Option(longName = "epsilon-insensitivity", usage = "Regression value insensitivity for margin.")
         public double epsilon = 0.1;
-        @Option(charName='e',longName="termination-criterion",usage="Tolerance of the optimization termination criterion.")
+        /**
+         * Tolerance of the optimization termination criterion.
+         */
+        @Option(charName = 'e', longName = "termination-criterion", usage = "Tolerance of the optimization termination criterion.")
         public double terminationCriterion = 0.01;
-        @Option(charName='t',longName="algorithm",usage="Type of SVR.")
+        /**
+         * Type of SVR.
+         */
+        @Option(charName = 't', longName = "algorithm", usage = "Type of SVR.")
         public LinearType algorithm = LinearType.L2R_L2LOSS_SVR;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
      * @throws IOException if there is any error reading the examples.
      */
diff --git a/Regression/LibSVM/src/main/java/org/tribuo/regression/libsvm/SVMRegressionType.java b/Regression/LibSVM/src/main/java/org/tribuo/regression/libsvm/SVMRegressionType.java
index 8a505728a..9feb804d5 100644
--- a/Regression/LibSVM/src/main/java/org/tribuo/regression/libsvm/SVMRegressionType.java
+++ b/Regression/LibSVM/src/main/java/org/tribuo/regression/libsvm/SVMRegressionType.java
@@ -58,6 +58,10 @@ public enum SVMMode {
      */
     private SVMRegressionType() {}
 
+    /**
+     * Constructs an SVMRegressionType using the specified SVM algorithm.
+     * @param type The SVM algorithm.
+     */
     public SVMRegressionType(SVMMode type) {
         this.type = type;
     }
diff --git a/Regression/LibSVM/src/main/java/org/tribuo/regression/libsvm/TrainTest.java b/Regression/LibSVM/src/main/java/org/tribuo/regression/libsvm/TrainTest.java
index 149961d87..959ab7a99 100644
--- a/Regression/LibSVM/src/main/java/org/tribuo/regression/libsvm/TrainTest.java
+++ b/Regression/LibSVM/src/main/java/org/tribuo/regression/libsvm/TrainTest.java
@@ -52,23 +52,46 @@ public static class LibSVMOptions implements Options {
         public String getOptionsDescription() {
             return "Trains and tests a LibSVM regression model on the specified datasets.";
         }
+
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
 
-        @Option(longName="coefficient",usage="Intercept in kernel function.")
+        /**
+         * Intercept in kernel function.
+         */
+        @Option(longName = "coefficient", usage = "Intercept in kernel function.")
         public double coeff = 1.0;
-        @Option(charName='d',longName="degree",usage="Degree in polynomial kernel.")
+        /**
+         * Degree in polynomial kernel.
+         */
+        @Option(charName = 'd', longName = "degree", usage = "Degree in polynomial kernel.")
         public int degree = 3;
-        @Option(charName='g',longName="gamma",usage="Gamma value in kernel function.")
+        /**
+         * Gamma value in kernel function.
+         */
+        @Option(charName = 'g', longName = "gamma", usage = "Gamma value in kernel function.")
         public double gamma = 0.0;
-        @Option(charName='k',longName="kernel",usage="Type of SVM kernel.")
+        /**
+         * Type of SVM kernel.
+         */
+        @Option(charName = 'k', longName = "kernel", usage = "Type of SVM kernel.")
         public KernelType kernelType = KernelType.LINEAR;
-        @Option(charName='t',longName="type",usage="Type of SVM.")
+        /**
+         * Type of SVM.
+         */
+        @Option(charName = 't', longName = "type", usage = "Type of SVM.")
         public SVMRegressionType.SVMMode svmType = SVMMode.EPSILON_SVR;
-        @Option(longName="standardize",usage="Standardize the regression outputs internally to the SVM")
+        /**
+         * Standardize the regression outputs internally to the SVM
+         */
+        @Option(longName = "standardize", usage = "Standardize the regression outputs internally to the SVM")
         public boolean standardize = false;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
      * @throws IOException if there is any error reading the examples.
      */
diff --git a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/TrainTest.java b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/TrainTest.java
index 58b20fd50..382476c5b 100644
--- a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/TrainTest.java
+++ b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/TrainTest.java
@@ -82,32 +82,65 @@ public static class RegressionTreeOptions implements Options {
         public String getOptionsDescription() {
             return "Trains and tests a CART regression model on the specified datasets.";
         }
+
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
-        @Option(longName="csv-response-split-char",usage="Character to split the CSV response on to generate multiple regression dimensions. Defaults to ':'.")
+        /**
+         * Character to split the CSV response on to generate multiple regression dimensions. Defaults to ':'.
+         */
+        @Option(longName = "csv-response-split-char", usage = "Character to split the CSV response on to generate multiple regression dimensions. Defaults to ':'.")
         public char splitChar = ':';
-        @Option(charName='d',longName="max-depth",usage="Maximum depth in the decision tree.")
+        /**
+         * Maximum depth in the decision tree.
+         */
+        @Option(charName = 'd', longName = "max-depth", usage = "Maximum depth in the decision tree.")
         public int depth = 6;
-        @Option(charName='e',longName="split-fraction",usage="Fraction of features in split.")
+        /**
+         * Fraction of features in split.
+         */
+        @Option(charName = 'e', longName = "split-fraction", usage = "Fraction of features in split.")
         public float fraction = 1.0f;
-        @Option(charName='m',longName="min-child-weight",usage="Minimum child weight.")
+        /**
+         * Minimum child weight.
+         */
+        @Option(charName = 'm', longName = "min-child-weight", usage = "Minimum child weight.")
         public float minChildWeight = 5.0f;
-        @Option(charName='p',longName="min-impurity-decrease",usage="Minimumum decrease in impurity required in order" +
-                " for the node to be split.")
+        /**
+         * Minimumum decrease in impurity required in order for the node to be split.
+         */
+        @Option(charName = 'p', longName = "min-impurity-decrease", usage = "Minimumum decrease in impurity required in order for the node to be split.")
         public float minImpurityDecrease = 0.0f;
-        @Option(charName='r',longName="use-random-split-points",usage="Whether to choose split points for features at" +
-                " random.")
+        /**
+         * Whether to choose split points for features at random.
+         */
+        @Option(charName = 'r', longName = "use-random-split-points", usage = "Whether to choose split points for features at random.")
         public boolean useRandomSplitPoints = false;
-        @Option(charName='n',longName="normalize",usage="Normalize the leaf outputs so each leaf sums to 1.0.")
+        /**
+         * Normalize the leaf outputs so each leaf sums to 1.0.
+         */
+        @Option(charName = 'n', longName = "normalize", usage = "Normalize the leaf outputs so each leaf sums to 1.0.")
         public boolean normalize = false;
-        @Option(charName='i',longName="impurity",usage="Impurity measure to use. Defaults to MSE.")
+        /**
+         * Impurity measure to use. Defaults to MSE.
+         */
+        @Option(charName = 'i', longName = "impurity", usage = "Impurity measure to use. Defaults to MSE.")
         public ImpurityType impurityType = ImpurityType.MSE;
-        @Option(charName='t',longName="tree-type",usage="Tree type.")
+        /**
+         * Tree type.
+         */
+        @Option(charName = 't', longName = "tree-type", usage = "Tree type.")
         public TreeType treeType = TreeType.CART_INDEPENDENT;
-        @Option(longName="print-tree",usage="Prints the decision tree.")
+        /**
+         * Prints the decision tree.
+         */
+        @Option(longName = "print-tree", usage = "Prints the decision tree.")
         public boolean printTree;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
      * @throws IOException if there is any error reading the examples.
      */
diff --git a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/InvertedFeature.java b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/InvertedFeature.java
index 3d3a2d453..c8877c70b 100644
--- a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/InvertedFeature.java
+++ b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/InvertedFeature.java
@@ -36,6 +36,9 @@ public class InvertedFeature implements Comparable<InvertedFeature> {
 
     private static final int DEFAULT_SIZE = 8;
 
+    /**
+     * The feature value of this object.
+     */
     public final double value;
 
     /**
@@ -49,12 +52,22 @@ public class InvertedFeature implements Comparable<InvertedFeature> {
      */
     private int index;
 
+    /**
+     * Constructs an inverted feature for the specified value which occurs at the specified indices.
+     * @param value The value.
+     * @param indices The indices where the value occurs.
+     */
     public InvertedFeature(double value, int[] indices) {
         this.value = value;
         this.indices = indices;
         this.curSize = indices.length;
     }
 
+    /**
+     * Constructs an inverted feature for the specifed value which occurs at a single index.
+     * @param value The value.
+     * @param index The index where the value occurs.
+     */
     public InvertedFeature(double value, int index) {
         this.value = value;
         this.index = index;
@@ -71,6 +84,10 @@ private InvertedFeature(InvertedFeature other) {
         }
     }
 
+    /**
+     * Adds an index where the feature value occurs.
+     * @param index The index.
+     */
     public void add(int index) {
         if (indices == null) {
             initArrays();
@@ -87,6 +104,10 @@ private void append(int index) {
         curSize++;
     }
 
+    /**
+     * Gets the indices where this feature value occurs.
+     * @return The indices.
+     */
     public int[] indices() {
         if (indices != null) {
             return indices;
@@ -97,6 +118,11 @@ public int[] indices() {
         }
     }
 
+    /**
+     * Fixes the size of the backing array.
+     * <p>
+     * Used when all the feature values have been observed.
+     */
     public void fixSize() {
         if (indices != null) {
             indices = Arrays.copyOf(indices, curSize);
@@ -219,6 +245,10 @@ public String toString() {
         }
     }
 
+    /**
+     * Copies this inverted feature.
+     * @return A copy of this feature.
+     */
     public InvertedFeature deepCopy() {
         return new InvertedFeature(this);
     }
diff --git a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/RegressorTrainingNode.java b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/RegressorTrainingNode.java
index 20862975d..0e2355846 100644
--- a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/RegressorTrainingNode.java
+++ b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/RegressorTrainingNode.java
@@ -76,6 +76,17 @@ public class RegressorTrainingNode extends AbstractTrainingNode<Regressor> {
 
     private final float weightSum;
 
+    /**
+     * Constructs a tree training node for regression problems.
+     * @param impurity The impurity function.
+     * @param tuple The data tuple.
+     * @param dimIndex The output dimension index of this node.
+     * @param dimName The output dimension name.
+     * @param numExamples The number of examples.
+     * @param featureIDMap The feature domain.
+     * @param outputInfo The output domain.
+     * @param leafDeterminer The leaf determination parameters.
+     */
     public RegressorTrainingNode(RegressorImpurity impurity, InvertedData tuple, int dimIndex, String dimName,
                                  int numExamples, ImmutableFeatureMap featureIDMap,
                                  ImmutableOutputInfo<Regressor> outputInfo, LeafDeterminer leafDeterminer) {
diff --git a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/TreeFeature.java b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/TreeFeature.java
index 5946405b0..34c0e83b7 100644
--- a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/TreeFeature.java
+++ b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impl/TreeFeature.java
@@ -41,6 +41,10 @@ public class TreeFeature implements Iterable<InvertedFeature> {
 
     private boolean sorted = true;
 
+    /**
+     * Constructs an inverted feature with the specified feature id.
+     * @param id The feature id.
+     */
     public TreeFeature(int id) {
         this.id = id;
         this.feature = new ArrayList<>();
@@ -78,6 +82,10 @@ public Iterator<InvertedFeature> iterator() {
         return feature.iterator();
     }
 
+    /**
+     * Gets the inverted feature values for this feature.
+     * @return The list of feature values.
+     */
     public List<InvertedFeature> getFeature() {
         return feature;
     }
diff --git a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impurity/RegressorImpurity.java b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impurity/RegressorImpurity.java
index 20f19651d..6c3bc1ddd 100644
--- a/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impurity/RegressorImpurity.java
+++ b/Regression/RegressionTree/src/main/java/org/tribuo/regression/rtree/impurity/RegressorImpurity.java
@@ -101,12 +101,23 @@ default public double impurity(IntArrayContainer indices, float[] targets, float
     }
 
     /**
-     * Tuple class for the impurity and summed weight.
+     * Tuple class for the impurity and summed weight. Will be a record one day.
      */
     public static class ImpurityTuple {
+        /**
+         * The impurity value.
+         */
         public final float impurity;
+        /**
+         * The sum of the weights.
+         */
         public final float weight;
 
+        /**
+         * Construct an impurity tuple.
+         * @param impurity The impurity value.
+         * @param weight The sum of the weights.
+         */
         public ImpurityTuple(float impurity, float weight) {
             this.impurity = impurity;
             this.weight = weight;
diff --git a/Regression/SGD/src/main/java/org/tribuo/regression/sgd/TrainTest.java b/Regression/SGD/src/main/java/org/tribuo/regression/sgd/TrainTest.java
index abe32e815..f8ea6ca11 100644
--- a/Regression/SGD/src/main/java/org/tribuo/regression/sgd/TrainTest.java
+++ b/Regression/SGD/src/main/java/org/tribuo/regression/sgd/TrainTest.java
@@ -73,20 +73,40 @@ public static class SGDOptions implements Options {
         public String getOptionsDescription() {
             return "Trains and tests a linear SGD regression model on the specified datasets.";
         }
+
+        /**
+         * The dataset loading options.
+         */
         public DataOptions general;
+        /**
+         * Gradient descent options.
+         */
         public GradientOptimiserOptions gradientOptions;
 
-        @Option(charName='i',longName="epochs",usage="Number of SGD epochs. Defaults to 5.")
+        /**
+         * Number of SGD epochs. Defaults to 5.
+         */
+        @Option(charName = 'i', longName = "epochs", usage = "Number of SGD epochs. Defaults to 5.")
         public int epochs = 5;
-        @Option(charName='o',longName="objective",usage="Loss function. Defaults to SQUARED.")
+        /**
+         * Loss function. Defaults to SQUARED.
+         */
+        @Option(charName = 'o', longName = "objective", usage = "Loss function. Defaults to SQUARED.")
         public LossEnum loss = LossEnum.SQUARED;
-        @Option(charName='p',longName="logging-interval",usage="Log the objective after <int> examples. Defaults to 100.")
+        /**
+         * Log the objective after n examples. Defaults to 100.
+         */
+        @Option(charName = 'p', longName = "logging-interval", usage = "Log the objective after <int> examples. Defaults to 100.")
         public int loggingInterval = 100;
-        @Option(charName='z',longName="minibatch-size",usage="Minibatch size. Defaults to 1.")
+        /**
+         * Minibatch size. Defaults to 1.
+         */
+        @Option(charName = 'z', longName = "minibatch-size", usage = "Minibatch size. Defaults to 1.")
         public int minibatchSize = 1;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
      * @throws IOException if there is any error reading the examples.
      */
diff --git a/Regression/SGD/src/main/java/org/tribuo/regression/sgd/objectives/Huber.java b/Regression/SGD/src/main/java/org/tribuo/regression/sgd/objectives/Huber.java
index e6148ad6b..3d0ad2e88 100644
--- a/Regression/SGD/src/main/java/org/tribuo/regression/sgd/objectives/Huber.java
+++ b/Regression/SGD/src/main/java/org/tribuo/regression/sgd/objectives/Huber.java
@@ -32,6 +32,9 @@
  */
 public class Huber implements RegressionObjective {
 
+    /**
+     * The default cost beyond which the function is linear.
+     */
     public static final double DEFAULT_COST = 5;
 
     @Config(description="Cost beyond which the loss function is linear.")
diff --git a/Regression/SLM/src/main/java/org/tribuo/regression/slm/ElasticNetCDTrainer.java b/Regression/SLM/src/main/java/org/tribuo/regression/slm/ElasticNetCDTrainer.java
index d7c3f2366..236872458 100644
--- a/Regression/SLM/src/main/java/org/tribuo/regression/slm/ElasticNetCDTrainer.java
+++ b/Regression/SLM/src/main/java/org/tribuo/regression/slm/ElasticNetCDTrainer.java
@@ -88,14 +88,34 @@ public class ElasticNetCDTrainer implements SparseTrainer<Regressor> {
      */
     private ElasticNetCDTrainer() { }
 
+    /**
+     * Constructs an elastic net trainer using the supplied parameters, with a tolerance of 1e-4, max iterations of 500, randomising the feature choice and using {@link Trainer#DEFAULT_SEED} as the RNG seed.
+     * @param alpha The regularisation strength.
+     * @param l1Ratio The ratio between the l1 and l2 penalties.
+     */
     public ElasticNetCDTrainer(double alpha, double l1Ratio) {
         this(alpha,l1Ratio,1e-4,500,false,Trainer.DEFAULT_SEED);
     }
 
+    /**
+     * Constructs an elastic net trainer using the supplied parameters, with a tolerance of 1e-4, max iterations of 500, and randomising the feature choice.
+     * @param alpha The regularisation strength.
+     * @param l1Ratio The ratio between the l1 and l2 penalties.
+     * @param seed The RNG seed.
+     */
     public ElasticNetCDTrainer(double alpha, double l1Ratio, long seed) {
         this(alpha,l1Ratio,1e-4,500,true,seed);
     }
 
+    /**
+     * Constructs an elastic net trainer using the supplied parameters.
+     * @param alpha The regularisation strength.
+     * @param l1Ratio The ratio between the l1 and l2 penalties.
+     * @param tolerance The convergence tolerance.
+     * @param maxIterations The maximum number of iterations.
+     * @param randomise Randomise the feature choice order.
+     * @param seed The RNG seed.
+     */
     public ElasticNetCDTrainer(double alpha, double l1Ratio, double tolerance, int maxIterations, boolean randomise, long seed) {
         this.alpha = alpha;
         this.l1Ratio = l1Ratio;
diff --git a/Regression/SLM/src/main/java/org/tribuo/regression/slm/TrainTest.java b/Regression/SLM/src/main/java/org/tribuo/regression/slm/TrainTest.java
index 2a0795d58..9c0bdae4b 100644
--- a/Regression/SLM/src/main/java/org/tribuo/regression/slm/TrainTest.java
+++ b/Regression/SLM/src/main/java/org/tribuo/regression/slm/TrainTest.java
@@ -77,21 +77,41 @@ public static class SLMOptions implements Options {
         public String getOptionsDescription() {
             return "Trains and tests a sparse linear regression model on the specified datasets.";
         }
+
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
 
-        @Option(charName='m',longName="max-features-num", usage="Set the maximum number of features.")
+        /**
+         * Set the maximum number of features.
+         */
+        @Option(charName = 'm', longName = "max-features-num", usage = "Set the maximum number of features.")
         public int maxNumFeatures = -1;
-        @Option(charName='a',longName="algorithm", usage="Choose the training algorithm (stepwise forward selection or least angle regression).")
+        /**
+         * Choose the training algorithm (stepwise forward selection or least angle regression).
+         */
+        @Option(charName = 'a', longName = "algorithm", usage = "Choose the training algorithm (stepwise forward selection or least angle regression).")
         public SLMType algorithm = SLMType.LARS;
-        @Option(charName='b',longName="alpha", usage="Regularisation strength in the Elastic Net.")
+        /**
+         * Regularisation strength in the Elastic Net.
+         */
+        @Option(charName = 'b', longName = "alpha", usage = "Regularisation strength in the Elastic Net.")
         public double alpha = 1.0;
-        @Option(charName='l',longName="l1Ratio", usage="Ratio between the l1 and l2 penalties in the Elastic Net. Must be between 0 and 1.")
+        /**
+         * Ratio between the l1 and l2 penalties in the Elastic Net. Must be between 0 and 1.
+         */
+        @Option(charName = 'l', longName = "l1Ratio", usage = "Ratio between the l1 and l2 penalties in the Elastic Net. Must be between 0 and 1.")
         public double l1Ratio = 1.0;
-        @Option(longName="iterations",usage="Iterations of Elastic Net.")
+        /**
+         * Iterations of Elastic Net.
+         */
+        @Option(longName = "iterations", usage = "Iterations of Elastic Net.")
         public int iterations = 500;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
      * @throws IOException if there is any error reading the examples.
      */
diff --git a/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/TrainTest.java b/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/TrainTest.java
index b612eb49c..3213aef9f 100644
--- a/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/TrainTest.java
+++ b/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/TrainTest.java
@@ -49,37 +49,78 @@ public static class XGBoostOptions implements Options {
         public String getOptionsDescription() {
             return "Trains and tests an XGBoost regression model on the specified datasets.";
         }
+
+        /**
+         * The data loading options.
+         */
         public DataOptions general;
 
-        @Option(longName="regression-metric", usage="Regression type to use. Defaults to LINEAR.")
+        /**
+         * Regression type to use. Defaults to LINEAR.
+         */
+        @Option(longName = "regression-metric", usage = "Regression type to use. Defaults to LINEAR.")
         public RegressionType rType = RegressionType.LINEAR;
-        @Option(charName='m',longName="ensemble-size",usage="Number of trees in the ensemble.")
+        /**
+         * Number of trees in the ensemble.
+         */
+        @Option(charName = 'm', longName = "ensemble-size", usage = "Number of trees in the ensemble.")
         public int ensembleSize = -1;
-        @Option(charName='a',longName="alpha",usage="L1 regularization term for weights (default 0).")
+        /**
+         * L1 regularization term for weights (default 0).
+         */
+        @Option(charName = 'a', longName = "alpha", usage = "L1 regularization term for weights (default 0).")
         public float alpha = 0.0f;
-        @Option(longName="min-weight",usage="Minimum sum of instance weights needed in a leaf (default 1, range [0,inf]).")
+        /**
+         * Minimum sum of instance weights needed in a leaf (default 1, range [0,inf]).
+         */
+        @Option(longName = "min-weight", usage = "Minimum sum of instance weights needed in a leaf (default 1, range [0,inf]).")
         public float minWeight = 1;
-        @Option(charName='d',longName="max-depth",usage="Max tree depth (default 6, range (0,inf]).")
+        /**
+         * Max tree depth (default 6, range (0,inf]).
+         */
+        @Option(charName = 'd', longName = "max-depth", usage = "Max tree depth (default 6, range (0,inf]).")
         public int depth = 6;
-        @Option(charName='e',longName="eta",usage="Step size shrinkage parameter (default 0.3, range [0,1]).")
+        /**
+         * Step size shrinkage parameter (default 0.3, range [0,1]).
+         */
+        @Option(charName = 'e', longName = "eta", usage = "Step size shrinkage parameter (default 0.3, range [0,1]).")
         public float eta = 0.3f;
-        @Option(longName="subsample-features",usage="Subsample features for each tree (default 1, range (0,1]).")
+        /**
+         * Subsample features for each tree (default 1, range (0,1]).
+         */
+        @Option(longName = "subsample-features", usage = "Subsample features for each tree (default 1, range (0,1]).")
         public float subsampleFeatures = 1.0f;
-        @Option(charName='g',longName="gamma",usage="Minimum loss reduction to make a split (default 0, range [0,inf]).")
+        /**
+         * Minimum loss reduction to make a split (default 0, range [0,inf]).
+         */
+        @Option(charName = 'g', longName = "gamma", usage = "Minimum loss reduction to make a split (default 0, range [0,inf]).")
         public float gamma = 0.0f;
-        @Option(charName='l',longName="lambda",usage="L2 regularization term for weights (default 1).")
+        /**
+         * L2 regularization term for weights (default 1).
+         */
+        @Option(charName = 'l', longName = "lambda", usage = "L2 regularization term for weights (default 1).")
         public float lambda = 1.0f;
-        @Option(charName='q',longName="quiet",usage="Make the XGBoost training procedure quiet.")
+        /**
+         * Make the XGBoost training procedure quiet.
+         */
+        @Option(charName = 'q', longName = "quiet", usage = "Make the XGBoost training procedure quiet.")
         public boolean quiet;
-        @Option(longName="subsample",usage="Subsample size for each tree (default 1, range (0,1]).")
+        /**
+         * Subsample size for each tree (default 1, range (0,1]).
+         */
+        @Option(longName = "subsample", usage = "Subsample size for each tree (default 1, range (0,1]).")
         public float subsample = 1.0f;
-        @Option(charName='t',longName="num-threads",usage="Number of threads to use (default 4, range (1, num hw threads)).")
+        /**
+         * Number of threads to use (default 4, range (1, num hw threads)).
+         */
+        @Option(charName = 't', longName = "num-threads", usage = "Number of threads to use (default 4, range (1, num hw threads)).")
         public int numThreads = 4;
     }
 
     /**
+     * Runs a TrainTest CLI.
      * @param args the command line arguments
-     * @throws java.io.IOException if there is any error reading the examples.
+     * @throws IOException if there is any error reading the examples.
      */
     public static void main(String[] args) throws IOException {
         //
diff --git a/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/XGBoostOptions.java b/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/XGBoostOptions.java
index f1dccac9a..b25fadbf3 100644
--- a/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/XGBoostOptions.java
+++ b/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/XGBoostOptions.java
@@ -25,38 +25,75 @@
  * CLI options for configuring an XGBoost regression trainer.
  */
 public class XGBoostOptions implements Options {
-    @Option(longName="xgb-regression-metric", usage="Regression type to use. Defaults to LINEAR.")
+    /**
+     * Regression type to use. Defaults to LINEAR.
+     */
+    @Option(longName = "xgb-regression-metric", usage = "Regression type to use. Defaults to LINEAR.")
     public RegressionType rType = RegressionType.LINEAR;
-    @Option(longName="xgb-ensemble-size",usage="Number of trees in the ensemble.")
+    /**
+     * Number of trees in the ensemble.
+     */
+    @Option(longName = "xgb-ensemble-size", usage = "Number of trees in the ensemble.")
     public int ensembleSize = -1;
-    @Option(longName="xgb-alpha",usage="L1 regularization term for weights (default 0).")
+    /**
+     * L1 regularization term for weights (default 0).
+     */
+    @Option(longName = "xgb-alpha", usage = "L1 regularization term for weights (default 0).")
     public float alpha = 0.0f;
-    @Option(longName="xgb-min-weight",usage="Minimum sum of instance weights needed in a leaf (default 1, range [0,inf]).")
+    /**
+     * Minimum sum of instance weights needed in a leaf (default 1, range [0,inf]).
+     */
+    @Option(longName = "xgb-min-weight", usage = "Minimum sum of instance weights needed in a leaf (default 1, range [0,inf]).")
     public float minWeight = 1;
-    @Option(longName="xgb-max-depth",usage="Max tree depth (default 6, range (0,inf]).")
+    /**
+     * Max tree depth (default 6, range (0,inf]).
+     */
+    @Option(longName = "xgb-max-depth", usage = "Max tree depth (default 6, range (0,inf]).")
     public int depth = 6;
-    @Option(longName="xgb-eta",usage="Step size shrinkage parameter (default 0.3, range [0,1]).")
+    /**
+     * Step size shrinkage parameter (default 0.3, range [0,1]).
+     */
+    @Option(longName = "xgb-eta", usage = "Step size shrinkage parameter (default 0.3, range [0,1]).")
     public float eta = 0.3f;
-    @Option(longName="xgb-subsample-features",usage="Subsample features for each tree (default 1, range (0,1]).")
+    /**
+     * Subsample features for each tree (default 1, range (0,1]).
+     */
+    @Option(longName = "xgb-subsample-features", usage = "Subsample features for each tree (default 1, range (0,1]).")
     public float subsampleFeatures;
-    @Option(longName="xgb-gamma",usage="Minimum loss reduction to make a split (default 0, range [0,inf]).")
+    /**
+     * Minimum loss reduction to make a split (default 0, range [0,inf]).
+     */
+    @Option(longName = "xgb-gamma", usage = "Minimum loss reduction to make a split (default 0, range [0,inf]).")
     public float gamma = 0.0f;
-    @Option(longName="xgb-lambda",usage="L2 regularization term for weights (default 1).")
+    /**
+     * L2 regularization term for weights (default 1).
+     */
+    @Option(longName = "xgb-lambda", usage = "L2 regularization term for weights (default 1).")
     public float lambda = 1.0f;
-    @Option(longName="xgb-quiet",usage="Make the XGBoost training procedure quiet.")
+    /**
+     * Make the XGBoost training procedure quiet.
+     */
+    @Option(longName = "xgb-quiet", usage = "Make the XGBoost training procedure quiet.")
     public boolean quiet;
-    @Option(longName="xgb-subsample",usage="Subsample size for each tree (default 1, range (0,1]).")
+    /**
+     * Subsample size for each tree (default 1, range (0,1]).
+     */
+    @Option(longName = "xgb-subsample", usage = "Subsample size for each tree (default 1, range (0,1]).")
     public float subsample = 1.0f;
-    @Option(longName="xgb-num-threads",usage="Number of threads to use (default 4, range (1, num hw threads)).")
+    /**
+     * Number of threads to use (default 4, range (1, num hw threads)).
+     */
+    @Option(longName = "xgb-num-threads", usage = "Number of threads to use (default 4, range (1, num hw threads)).")
     public int numThreads;
     @Option(longName = "xgb-seed", usage = "Sets the random seed for XGBoost.")
     private long seed = Trainer.DEFAULT_SEED;
 
     /**
      * Gets the configured XGBoostRegressionTrainer.
+     *
      * @return The configured trainer.
      */
     public XGBoostRegressionTrainer getTrainer() {
-        return new XGBoostRegressionTrainer(rType,ensembleSize,eta,gamma,depth,minWeight,subsample,subsampleFeatures,lambda,alpha,numThreads,quiet,seed);
+        return new XGBoostRegressionTrainer(rType, ensembleSize, eta, gamma, depth, minWeight, subsample, subsampleFeatures, lambda, alpha, numThreads, quiet, seed);
     }
 }
diff --git a/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/XGBoostRegressionTrainer.java b/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/XGBoostRegressionTrainer.java
index fc4b18676..53b14c820 100644
--- a/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/XGBoostRegressionTrainer.java
+++ b/Regression/XGBoost/src/main/java/org/tribuo/regression/xgboost/XGBoostRegressionTrainer.java
@@ -93,6 +93,9 @@ public enum RegressionType {
          */
         PSEUDOHUBER("reg:pseudohubererror");
 
+        /**
+         * The parameter name used by the XGBoost native library.
+         */
         public final String paramName;
 
         RegressionType(String paramName) {
@@ -103,14 +106,33 @@ public enum RegressionType {
     @Config(description="The type of regression.")
     private RegressionType rType = RegressionType.LINEAR;
 
+    /**
+     * Create an XGBoost trainer using the squared error loss function.
+     *
+     * @param numTrees Number of trees to boost.
+     */
     public XGBoostRegressionTrainer(int numTrees) {
         this(RegressionType.LINEAR, numTrees);
     }
 
+    /**
+     * Create an XGBoost trainer.
+     *
+     * @param rType The type of regression to build.
+     * @param numTrees Number of trees to boost.
+     */
     public XGBoostRegressionTrainer(RegressionType rType, int numTrees) {
         this(rType, numTrees, 0.3, 0, 6, 1, 1, 1, 1, 0, 4, true, Trainer.DEFAULT_SEED);
     }
 
+    /**
+     * Create an XGBoost trainer.
+     *
+     * @param rType The type of regression to build.
+     * @param numTrees Number of trees to boost.
+     * @param numThreads Number of threads to use (default 4).
+     * @param silent Silence the training output text.
+     */
     public XGBoostRegressionTrainer(RegressionType rType, int numTrees, int numThreads, boolean silent) {
         this(rType, numTrees, 0.3, 0, 6, 1, 1, 1, 1, 0, numThreads, silent, Trainer.DEFAULT_SEED);
     }
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/InformationTheory.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/InformationTheory.java
index 50b89bdf0..787e40622 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/InformationTheory.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/InformationTheory.java
@@ -48,9 +48,21 @@
 public final class InformationTheory {
     private static final Logger logger = Logger.getLogger(InformationTheory.class.getName());
 
+    /**
+     * The ratio of samples to symbols before emitting a warning.
+     */
     public static final double SAMPLES_RATIO = 5.0;
+    /**
+     * The initial size of the various maps.
+     */
     public static final int DEFAULT_MAP_SIZE = 20;
+    /**
+     * Log base 2.
+     */
     public static final double LOG_2 = Math.log(2);
+    /**
+     * Log base e.
+     */
     public static final double LOG_E = Math.log(Math.E);
 
     /**
@@ -528,13 +540,18 @@ public static double calculateEntropy(DoubleStream vector) {
 
     /**
      * A tuple of the information theoretic value, along with the number of
-     * states in the random variable.
+     * states in the random variable. Will be a record one day.
      */
     private static class ScoreStateCountTuple {
         public final double score;
         public final int stateCount;
 
-        public ScoreStateCountTuple(double score, int stateCount) {
+        /**
+         * Construct a score state tuple
+         * @param score The score.
+         * @param stateCount The number of states.
+         */
+        ScoreStateCountTuple(double score, int stateCount) {
             this.score = score;
             this.stateCount = stateCount;
         }
@@ -547,12 +564,30 @@ public String toString() {
 
     /**
      * An immutable named tuple containing the statistics from a G test.
+     * <p>
+     * Will be a record one day.
      */
     public static final class GTestStatistics {
+        /**
+         * The G test statistic.
+         */
         public final double gStatistic;
+        /**
+         * The number of states.
+         */
         public final int numStates;
+        /**
+         * The probability of that statistic.
+         */
         public final double probability;
 
+        /**
+         * Constructs a GTestStatistics tuple with the supplied values.
+         * @param gStatistic The g test statistic.
+         * @param numStates The number of states.
+         * @param probability The probability of that statistic.
+         */
+        // TODO should be package private.
         public GTestStatistics(double gStatistic, int numStates, double probability) {
             this.gStatistic = gStatistic;
             this.numStates = numStates;
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/WeightedInformationTheory.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/WeightedInformationTheory.java
index abaac7a7e..80ae99a29 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/WeightedInformationTheory.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/WeightedInformationTheory.java
@@ -46,9 +46,21 @@
 public final class WeightedInformationTheory {
     private static final Logger logger = Logger.getLogger(WeightedInformationTheory.class.getName());
 
+    /**
+     * The ratio of samples to symbols before emitting a warning.
+     */
     public static final double SAMPLES_RATIO = 5.0;
+    /**
+     * The initial size of the various maps.
+     */
     public static final int DEFAULT_MAP_SIZE = 20;
+    /**
+     * Log base 2.
+     */
     public static final double LOG_2 = Math.log(2);
+    /**
+     * Log base e.
+     */
     public static final double LOG_E = Math.log(Math.E);
 
     /**
@@ -61,7 +73,18 @@ public final class WeightedInformationTheory {
      * Chooses which variable is the one with associated weights.
      */
     public enum VariableSelector {
-        FIRST, SECOND, THIRD
+        /**
+         * The first variable is weighted.
+         */
+        FIRST,
+        /**
+         * The second variable is weighted.
+         */
+        SECOND,
+        /**
+         * The third variable is weighted.
+         */
+        THIRD
     }
 
     /**
@@ -79,7 +102,7 @@ private WeightedInformationTheory() {}
      * @param second Another array of values.
      * @param target Target array of values.
      * @param weights Array of weight values.
-     * @return The mutual information I(first,second;joint)
+     * @return The weighted mutual information I_w(first,second;joint)
      */
     public static <T1,T2,T3> double jointMI(List<T1> first, List<T2> second, List<T3> target, List<Double> weights) {
         WeightedTripleDistribution<T1, T2, T3> tripleRV = WeightedTripleDistribution.constructFromLists(first, second, target, weights);
@@ -87,6 +110,15 @@ public static <T1,T2,T3> double jointMI(List<T1> first, List<T2> second, List<T3
         return jointMI(tripleRV);
     }
 
+    /**
+     * Calculates the discrete weighted joint mutual information, using
+     * histogram probability estimators.
+     * @param tripleRV The weighted triple distribution.
+     * @param <T1> The first element type.
+     * @param <T2> The second element type.
+     * @param <T3> The third element type.
+     * @return The weighted mutual information I_w(first,second;joint)
+     */
     public static <T1,T2,T3> double jointMI(WeightedTripleDistribution<T1,T2,T3> tripleRV) {
         Map<CachedTriple<T1,T2,T3>, WeightCountTuple> jointCount = tripleRV.getJointCount();
         Map<CachedPair<T1,T2>,WeightCountTuple> abCount = tripleRV.getABCount();
@@ -114,6 +146,17 @@ public static <T1,T2,T3> double jointMI(WeightedTripleDistribution<T1,T2,T3> tri
         return jmi;
     }
 
+    /**
+     * Calculates the discrete weighted joint mutual information, using
+     * histogram probability estimators.
+     * @param rv The triple distribution.
+     * @param weights The weights for one of the variables.
+     * @param vs The weighted variable id.
+     * @param <T1> The first element type.
+     * @param <T2> The second element type.
+     * @param <T3> The third element type.
+     * @return The weighted mutual information I_w(first,second;joint)
+     */
     public static <T1,T2,T3> double jointMI(TripleDistribution<T1,T2,T3> rv, Map<?,Double> weights, VariableSelector vs){
         Double boxedWeight;
         double vecLength = rv.count;
@@ -167,7 +210,7 @@ public static <T1,T2,T3> double jointMI(TripleDistribution<T1,T2,T3> rv, Map<?,D
      * @param second Another array of values.
      * @param condition Array to condition upon.
      * @param weights Array of weight values.
-     * @return The conditional mutual information I(first;second|condition)
+     * @return The weighted conditional mutual information I_w(first;second|condition)
      */
     public static <T1,T2,T3> double conditionalMI(List<T1> first, List<T2> second, List<T3> condition, List<Double> weights) {
         if ((first.size() == second.size()) && (first.size() == condition.size()) && (first.size() == weights.size())) {
@@ -179,6 +222,15 @@ public static <T1,T2,T3> double conditionalMI(List<T1> first, List<T2> second, L
         }
     }
 
+    /**
+     * Calculates the discrete weighted conditional mutual information, using
+     * histogram probability estimators.
+     * @param tripleRV The weighted triple distribution.
+     * @param <T1> The first element type.
+     * @param <T2> The second element type.
+     * @param <T3> The condition element type.
+     * @return The weighted conditional mutual information I_w(first;second|condition)
+     */
     public static <T1,T2,T3> double conditionalMI(WeightedTripleDistribution<T1,T2,T3> tripleRV) {
         Map<CachedTriple<T1,T2,T3>,WeightCountTuple> jointCount = tripleRV.getJointCount();
         Map<CachedPair<T1,T3>,WeightCountTuple> acCount = tripleRV.getACCount();
@@ -209,6 +261,17 @@ public static <T1,T2,T3> double conditionalMI(WeightedTripleDistribution<T1,T2,T
         return cmi;
     }
 
+    /**
+     * Calculates the discrete weighted conditional mutual information, using
+     * histogram probability estimators.
+     * @param rv The triple distribution.
+     * @param weights The element weights.
+     * @param vs The variable to apply the weights to.
+     * @param <T1> The first element type.
+     * @param <T2> The second element type.
+     * @param <T3> The condition element type.
+     * @return The weighted conditional mutual information I_w(first;second|condition)
+     */
     public static <T1,T2,T3> double conditionalMI(TripleDistribution<T1,T2,T3> rv, Map<?,Double> weights, VariableSelector vs) {
         Double boxedWeight;
         Map<CachedTriple<T1,T2,T3>,MutableLong> jointCount = rv.getJointCount();
@@ -264,7 +327,7 @@ public static <T1,T2,T3> double conditionalMI(TripleDistribution<T1,T2,T3> rv, M
      * @param first An array of values
      * @param second Another array of values
      * @param weights Array of weight values.
-     * @return The mutual information I(first;Second)
+     * @return The weighted mutual information I_w(first;Second)
      */
     public static <T1,T2> double mi(ArrayList<T1> first, ArrayList<T2> second, ArrayList<Double> weights) {
         if ((first.size() == second.size()) && (first.size() == weights.size())) {
@@ -275,6 +338,14 @@ public static <T1,T2> double mi(ArrayList<T1> first, ArrayList<T2> second, Array
         }
     }
 
+    /**
+     * Calculates the discrete weighted mutual information, using histogram
+     * probability estimators.
+     * @param jointDist The weighted joint distribution.
+     * @param <T1> Type of the first element.
+     * @param <T2> Type of the second element.
+     * @return The weighted mutual information I_w(first;Second)
+     */
     public static <T1,T2> double mi(WeightedPairDistribution<T1,T2> jointDist) {
         double vectorLength = jointDist.count;
         double mi = 0.0;
@@ -301,6 +372,16 @@ public static <T1,T2> double mi(WeightedPairDistribution<T1,T2> jointDist) {
         return mi;
     }
 
+    /**
+     * Calculates the discrete weighted mutual information, using histogram
+     * probability estimators.
+     * @param pairDist The joint distribution.
+     * @param weights The element weights.
+     * @param vs The variable to apply the weights to.
+     * @param <T1> Type of the first element.
+     * @param <T2> Type of the second element.
+     * @return The weighted mutual information I_w(first;Second)
+     */
     public static <T1,T2> double mi(PairDistribution<T1,T2> pairDist, Map<?,Double> weights, VariableSelector vs) {
         if (vs == VariableSelector.THIRD) {
             throw new IllegalArgumentException("MI only has two variables");
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/example/InformationTheoryDemo.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/example/InformationTheoryDemo.java
index ceeb23a5e..487bbcf76 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/example/InformationTheoryDemo.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/example/InformationTheoryDemo.java
@@ -141,10 +141,18 @@ public static class DemoOptions implements Options {
         public String getOptionsDescription() {
             return "A demo class showing how to calculate various mutual informations from different inputs.";
         }
-        @Option(charName='t',longName="type",usage="The type of the input distribution.")
+
+        /**
+         * The type of the input distribution.
+         */
+        @Option(charName = 't', longName = "type", usage = "The type of the input distribution.")
         public DistributionType type = DistributionType.RANDOM;
     }
 
+    /**
+     * Runs a simple demo of the information theory functions.
+     * @param args The CLI arguments.
+     */
     public static void main(String[] args) {
 
         DemoOptions options = new DemoOptions();
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/CachedPair.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/CachedPair.java
index 27ca391d8..0fc7f181a 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/CachedPair.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/CachedPair.java
@@ -32,7 +32,12 @@ public class CachedPair<T1, T2> extends Pair<T1,T2> {
     private static final long serialVersionUID = 1L;
 
     private final int cachedHash;
-    
+
+    /**
+     * Constructs a CachedPair.
+     * @param a The first element.
+     * @param b The second element.
+     */
     public CachedPair(T1 a, T2 b) {
         super(a,b);
         this.cachedHash = super.hashCode();
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/CachedTriple.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/CachedTriple.java
index 88cc8170a..c7884f587 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/CachedTriple.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/CachedTriple.java
@@ -44,6 +44,12 @@ public class CachedTriple<T1, T2, T3> implements Serializable {
 
     private final int cachedHash;
 
+    /**
+     * Constructs a CachedTriple.
+     * @param a The first element.
+     * @param b The second element.
+     * @param c The third element.
+     */
     public CachedTriple(T1 a, T2 b, T3 c) {
         this.a = a;
         this.b = b;
@@ -54,26 +60,50 @@ public CachedTriple(T1 a, T2 b, T3 c) {
         this.cachedHash = calculateHashCode();
     }
 
+    /**
+     * Gets the first element.
+     * @return The first element.
+     */
     public T1 getA() {
         return a;
     }
 
+    /**
+     * Gets the second element.
+     * @return The second element.
+     */
     public T2 getB() {
         return b;
     }
-    
+
+    /**
+     * Gets the third element.
+     * @return The third element.
+     */
     public T3 getC() {
         return c;
     }
 
+    /**
+     * Gets the pair of the first and second elements.
+     * @return A pair of the first and second elements.
+     */
     public CachedPair<T1,T2> getAB() {
         return ab;
     }
 
+    /**
+     * Gets the pair of the first and third elements.
+     * @return A pair of the first and third elements.
+     */
     public CachedPair<T1,T3> getAC() {
         return ac;
     }
 
+    /**
+     * Gets the pair of the second and third elements.
+     * @return A pair of the second and third elements.
+     */
     public CachedPair<T2,T3> getBC() {
         return bc;
     }
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/PairDistribution.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/PairDistribution.java
index c311c302e..643b2e101 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/PairDistribution.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/PairDistribution.java
@@ -32,12 +32,31 @@
  */
 public class PairDistribution<T1,T2> {
 
+    /**
+     * The number of samples this distribution has seen.
+     */
     public final long count;
 
+    /**
+     * The joint distribution.
+     */
     public final Map<CachedPair<T1,T2>,MutableLong> jointCounts;
+    /**
+     * The first marginal distribution.
+     */
     public final Map<T1,MutableLong> firstCount;
+    /**
+     * The second marginal distribution.
+     */
     public final Map<T2,MutableLong> secondCount;
 
+    /**
+     * Constructs a pair distribution.
+     * @param count The total sample count.
+     * @param jointCounts The joint counts.
+     * @param firstCount The first variable count.
+     * @param secondCount The second variable count.
+     */
     public PairDistribution(long count, Map<CachedPair<T1,T2>,MutableLong> jointCounts, Map<T1,MutableLong> firstCount, Map<T2,MutableLong> secondCount) {
         this.count = count;
         this.jointCounts = new LinkedHashMap<>(jointCounts);
@@ -45,6 +64,13 @@ public PairDistribution(long count, Map<CachedPair<T1,T2>,MutableLong> jointCoun
         this.secondCount = new LinkedHashMap<>(secondCount);
     }
 
+    /**
+     * Constructs a pair distribution.
+     * @param count The total sample count.
+     * @param jointCounts The joint counts.
+     * @param firstCount The first variable count.
+     * @param secondCount The second variable count.
+     */
     public PairDistribution(long count, LinkedHashMap<CachedPair<T1,T2>,MutableLong> jointCounts, LinkedHashMap<T1,MutableLong> firstCount, LinkedHashMap<T2,MutableLong> secondCount) {
         this.count = count;
         this.jointCounts = jointCounts;
@@ -91,6 +117,13 @@ public static <T1,T2> PairDistribution<T1,T2> constructFromLists(List<T1> first,
         }
     }
 
+    /**
+     * Constructs a distribution from a joint count.
+     * @param jointCount The joint count.
+     * @param <T1> The type of the first variable.
+     * @param <T2> The type of the second variable.
+     * @return A pair distribution.
+     */
     public static <T1,T2> PairDistribution<T1,T2> constructFromMap(Map<CachedPair<T1,T2>,MutableLong> jointCount) {
         Map<T1,MutableLong> aCount = new HashMap<>(InformationTheory.DEFAULT_MAP_SIZE);
         Map<T2,MutableLong> bCount = new HashMap<>(InformationTheory.DEFAULT_MAP_SIZE);
@@ -98,6 +131,15 @@ public static <T1,T2> PairDistribution<T1,T2> constructFromMap(Map<CachedPair<T1
         return constructFromMap(jointCount,aCount,bCount);
     }
 
+    /**
+     * Constructs a distribution from a joint count.
+     * @param jointCount The joint count.
+     * @param aSize The initial size of the first marginal hash map.
+     * @param bSize The initial size of the second marginal hash map.
+     * @param <T1> The type of the first variable.
+     * @param <T2> The type of the second variable.
+     * @return A pair distribution.
+     */
     public static <T1,T2> PairDistribution<T1,T2> constructFromMap(Map<CachedPair<T1,T2>,MutableLong> jointCount, int aSize, int bSize) {
         Map<T1,MutableLong> aCount = new HashMap<>(aSize);
         Map<T2,MutableLong> bCount = new HashMap<>(bSize);
@@ -105,6 +147,15 @@ public static <T1,T2> PairDistribution<T1,T2> constructFromMap(Map<CachedPair<T1
         return constructFromMap(jointCount,aCount,bCount);
     }
 
+    /**
+     * Constructs a joint distribution from the counts.
+     * @param jointCount The joint count.
+     * @param aCount The first marginal count.
+     * @param bCount The second marginal count.
+     * @param <T1> The type of the first variable.
+     * @param <T2> The type of the second variable.
+     * @return A pair distribution.
+     */
     public static <T1,T2> PairDistribution<T1,T2> constructFromMap(Map<CachedPair<T1,T2>,MutableLong> jointCount,
                                                                            Map<T1,MutableLong> aCount,
                                                                            Map<T2,MutableLong> bCount) {
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/RowList.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/RowList.java
index cf6a47191..f5563f3bc 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/RowList.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/RowList.java
@@ -37,8 +37,12 @@
  */
 public final class RowList<T> implements List<Row<T>> {
     private final Set<List<T>> set;
-    private final int size;    
+    private final int size;
 
+    /**
+     * Constructs a RowList from a set of lists.
+     * @param set The feature lists.
+     */
     public RowList(Set<List<T>> set) {
         this.set = Collections.unmodifiableSet(new LinkedHashSet<>(set));
         size = set.iterator().next().size();
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/TripleDistribution.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/TripleDistribution.java
index fa8895fea..cc0c29075 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/TripleDistribution.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/TripleDistribution.java
@@ -27,13 +27,21 @@
 /**
  * Generates the counts for a triplet of vectors. Contains the joint
  * count, the three pairwise counts, and the three marginal counts.
+ * <p>
+ * Relies upon hashCode and equals to determine element equality for counting.
  * @param <T1> Type of the first list.
  * @param <T2> Type of the second list.
  * @param <T3> Type of the third list.
  */
 public class TripleDistribution<T1,T2,T3> {
+    /**
+     * The default map size to initialise the marginalised count maps with.
+     */
     public static final int DEFAULT_MAP_SIZE = 20;
 
+    /**
+     * The number of samples in this distribution.
+     */
     public final long count;
 
     private final Map<CachedTriple<T1,T2,T3>,MutableLong> jointCount;
@@ -44,6 +52,17 @@ public class TripleDistribution<T1,T2,T3> {
     private final Map<T2,MutableLong> bCount;
     private final Map<T3,MutableLong> cCount;
 
+    /**
+     * Constructs a triple distribution from the supplied distributions.
+     * @param count The sample count.
+     * @param jointCount The joint distribution over the three variables.
+     * @param abCount The joint distribution over A and B.
+     * @param acCount The joint distribution over A and C.
+     * @param bcCount The joint distribution over B and C.
+     * @param aCount The marginal distribution over A.
+     * @param bCount The marginal distribution over B.
+     * @param cCount The marginal distribution over C.
+     */
     public TripleDistribution(long count, Map<CachedTriple<T1,T2,T3>,MutableLong> jointCount, Map<CachedPair<T1,T2>,MutableLong> abCount, Map<CachedPair<T1,T3>,MutableLong> acCount, Map<CachedPair<T2,T3>,MutableLong> bcCount, Map<T1,MutableLong> aCount, Map<T2,MutableLong> bCount, Map<T3,MutableLong> cCount) {
         this.count = count;
         this.jointCount = jointCount;
@@ -55,34 +74,74 @@ public TripleDistribution(long count, Map<CachedTriple<T1,T2,T3>,MutableLong> jo
         this.cCount = cCount;
     }
 
+    /**
+     * The joint distribution over the three variables.
+     * @return The joint distribution.
+     */
     public Map<CachedTriple<T1,T2,T3>,MutableLong> getJointCount() {
         return jointCount;
     }
-    
+
+    /**
+     * The joint distribution over the first and second variables.
+     * @return The joint distribution over A and B.
+     */
     public Map<CachedPair<T1,T2>,MutableLong> getABCount() {
         return abCount;
     }
-    
+
+    /**
+     * The joint distribution over the first and third variables.
+     * @return The joint distribution over A and C.
+     */
     public Map<CachedPair<T1,T3>,MutableLong> getACCount() {
         return acCount;
     }
-    
+
+    /**
+     * The joint distribution over the second and third variables.
+     * @return The joint distribution over B and C.
+     */
     public Map<CachedPair<T2,T3>,MutableLong> getBCCount() {
         return bcCount;
     }
-    
+
+    /**
+     * The marginal distribution over the first variable.
+     * @return The marginal distribution for A.
+     */
     public Map<T1,MutableLong> getACount() {
         return aCount;
     }
-    
+
+    /**
+     * The marginal distribution over the second variable.
+     * @return The marginal distribution for B.
+     */
     public Map<T2,MutableLong> getBCount() {
         return bCount;
     }
-    
+
+    /**
+     * The marginal distribution over the third variable.
+     * @return The marginal distribution for C.
+     */
     public Map<T3,MutableLong> getCCount() {
         return cCount;
     }
-    
+
+    /**
+     * Constructs a TripleDistribution from three lists of the same length.
+     * <p>
+     * If they are not the same length it throws IllegalArgumentException.
+     * @param first The first list.
+     * @param second The second list.
+     * @param third The third list.
+     * @param <T1> The first type.
+     * @param <T2> The second type.
+     * @param <T3> The third type.
+     * @return The TripleDistribution.
+     */
     public static <T1,T2,T3> TripleDistribution<T1,T2,T3> constructFromLists(List<T1> first, List<T2> second, List<T3> third) {
         Map<CachedTriple<T1,T2,T3>,MutableLong> jointCount = new LinkedHashMap<>(DEFAULT_MAP_SIZE);
         Map<CachedPair<T1,T2>,MutableLong> abCount = new HashMap<>(DEFAULT_MAP_SIZE);
@@ -128,6 +187,14 @@ public static <T1,T2,T3> TripleDistribution<T1,T2,T3> constructFromLists(List<T1
         }
     }
 
+    /**
+     * Constructs a TripleDistribution by marginalising the supplied joint distribution.
+     * @param jointCount The joint distribution.
+     * @param <T1> The type of A.
+     * @param <T2> The type of B.
+     * @param <T3> The type of C.
+     * @return A TripleDistribution.
+     */
     public static <T1,T2,T3> TripleDistribution<T1,T2,T3> constructFromMap(Map<CachedTriple<T1,T2,T3>,MutableLong> jointCount) {
         Map<CachedPair<T1,T2>,MutableLong> abCount = new HashMap<>(DEFAULT_MAP_SIZE);
         Map<CachedPair<T1,T3>,MutableLong> acCount = new HashMap<>(DEFAULT_MAP_SIZE);
@@ -139,6 +206,22 @@ public static <T1,T2,T3> TripleDistribution<T1,T2,T3> constructFromMap(Map<Cache
         return constructFromMap(jointCount,abCount,acCount,bcCount,aCount,bCount,cCount);
     }
 
+    /**
+     * Constructs a TripleDistribution by marginalising the supplied joint distribution.
+     * <p>
+     * Sizes are used to preallocate the HashMaps.
+     * @param jointCount The joint distribution.
+     * @param abSize The number of unique AB states.
+     * @param acSize The number of unique AC states.
+     * @param bcSize The number of unique BC states.
+     * @param aSize The number of unique A states.
+     * @param bSize The number of unique B states.
+     * @param cSize The number of unique C states.
+     * @param <T1> The type of A.
+     * @param <T2> The type of B.
+     * @param <T3> The type of C.
+     * @return A TripleDistribution.
+     */
     public static <T1,T2,T3> TripleDistribution<T1,T2,T3> constructFromMap(Map<CachedTriple<T1,T2,T3>,MutableLong> jointCount,
                                                                            int abSize, int acSize, int bcSize,
                                                                            int aSize, int bSize, int cSize) {
@@ -152,6 +235,22 @@ public static <T1,T2,T3> TripleDistribution<T1,T2,T3> constructFromMap(Map<Cache
         return constructFromMap(jointCount,abCount,acCount,bcCount,aCount,bCount,cCount);
     }
 
+    /**
+     * Constructs a TripleDistribution by marginalising the supplied joint distribution.
+     * <p>
+     * Sizes are used to preallocate the HashMaps.
+     * @param jointCount The joint distribution.
+     * @param abCount An empty hashmap for AB.
+     * @param acCount An empty hashmap for AC.
+     * @param bcCount An empty hashmap for BC.
+     * @param aCount An empty hashmap for A.
+     * @param bCount An empty hashmap for B.
+     * @param cCount An empty hashmap for C.
+     * @param <T1> The type of A.
+     * @param <T2> The type of B.
+     * @param <T3> The type of C.
+     * @return A TripleDistribution.
+     */
     public static <T1,T2,T3> TripleDistribution<T1,T2,T3> constructFromMap(Map<CachedTriple<T1,T2,T3>,MutableLong> jointCount,
                                                                            Map<CachedPair<T1,T2>,MutableLong> abCount,
                                                                            Map<CachedPair<T1,T3>,MutableLong> acCount,
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightCountTuple.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightCountTuple.java
index ae92a87ca..98bd46e3a 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightCountTuple.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightCountTuple.java
@@ -17,18 +17,32 @@
 package org.tribuo.util.infotheory.impl;
 
 /**
- * An mutable tuple of a double and a long.
+ * A mutable tuple of a double and a long.
  */
 public final class WeightCountTuple {
 
+    /**
+     * The current weight.
+     */
     public double weight;
+    /**
+     * The current count.
+     */
     public long count;
 
+    /**
+     * Creates a zeroed WeightCountTuple.
+     */
     public WeightCountTuple() {
         this.weight = 0.0;
         this.count = 0;
     }
 
+    /**
+     * Creates a WeightCountTuple with the specifed values.
+     * @param weight The current weight.
+     * @param count The current count.
+     */
     public WeightCountTuple(double weight, long count) {
         this.weight = weight;
         this.count = count;
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightedPairDistribution.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightedPairDistribution.java
index fcaac1b1c..7316c70af 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightedPairDistribution.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightedPairDistribution.java
@@ -32,12 +32,24 @@
  */
 public class WeightedPairDistribution<T1,T2> {
 
+    /**
+     * The sample count.
+     */
     public final long count;
 
     private final Map<CachedPair<T1,T2>,WeightCountTuple> jointCounts;
     private final Map<T1,WeightCountTuple> firstCount;
     private final Map<T2,WeightCountTuple> secondCount;
 
+    /**
+     * Constructs a weighted pair distribution from the supplied values.
+     * <p>
+     * Copies the maps out into LinkedHashMaps for iteration speed.
+     * @param count The sample count.
+     * @param jointCounts The joint distribution.
+     * @param firstCount The first marginal distribution.
+     * @param secondCount The second marginal distribution.
+     */
     public WeightedPairDistribution(long count, Map<CachedPair<T1,T2>,WeightCountTuple> jointCounts, Map<T1,WeightCountTuple> firstCount, Map<T2,WeightCountTuple> secondCount) {
         this.count = count;
         this.jointCounts = new LinkedHashMap<>(jointCounts);
@@ -45,21 +57,40 @@ public WeightedPairDistribution(long count, Map<CachedPair<T1,T2>,WeightCountTup
         this.secondCount = new LinkedHashMap<>(secondCount);
     }
 
+    /**
+     * Constructs a weighted pair distribution from the supplied values.
+     * @param count The sample count.
+     * @param jointCounts The joint distribution.
+     * @param firstCount The first marginal distribution.
+     * @param secondCount The second marginal distribution.
+     */
     public WeightedPairDistribution(long count, LinkedHashMap<CachedPair<T1,T2>,WeightCountTuple> jointCounts, LinkedHashMap<T1,WeightCountTuple> firstCount, LinkedHashMap<T2,WeightCountTuple> secondCount) {
         this.count = count;
         this.jointCounts = jointCounts;
         this.firstCount = firstCount;
         this.secondCount = secondCount;
     }
-    
+
+    /**
+     * Gets the joint distribution.
+     * @return The joint distribution.
+     */
     public Map<CachedPair<T1,T2>,WeightCountTuple> getJointCounts() {
         return jointCounts;
     }
-    
+
+    /**
+     * Gets the first marginal distribution.
+     * @return The first marginal distribution.
+     */
     public Map<T1,WeightCountTuple> getFirstCount() {
         return firstCount;
     }
-    
+
+    /**
+     * Gets the second marginal distribution.
+     * @return The second marginal distribution.
+     */
     public Map<T2,WeightCountTuple> getSecondCount() {
         return secondCount;
     }
diff --git a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightedTripleDistribution.java b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightedTripleDistribution.java
index 40eefb8bb..2ee48a042 100644
--- a/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightedTripleDistribution.java
+++ b/Util/InformationTheory/src/main/java/org/tribuo/util/infotheory/impl/WeightedTripleDistribution.java
@@ -31,8 +31,14 @@
  * @param <T3> Type of the third list.
  */
 public class WeightedTripleDistribution<T1,T2,T3> {
+    /**
+     * The default map size.
+     */
     public static final int DEFAULT_MAP_SIZE = 20;
 
+    /**
+     * The sample count.
+     */
     public final long count;
 
     private final Map<CachedTriple<T1,T2,T3>,WeightCountTuple> jointCount;
@@ -43,6 +49,17 @@ public class WeightedTripleDistribution<T1,T2,T3> {
     private final Map<T2,WeightCountTuple> bCount;
     private final Map<T3,WeightCountTuple> cCount;
 
+    /**
+     * Constructs a weighted triple distribution from the supplied values.
+     * @param count The sample count.
+     * @param jointCount The ABC joint distribution.
+     * @param abCount The AB joint distribution.
+     * @param acCount The AC joint distribution.
+     * @param bcCount The BC joint distribution.
+     * @param aCount The A marginal distribution.
+     * @param bCount The B marginal distribution.
+     * @param cCount The C marginal distribution.
+     */
     public WeightedTripleDistribution(long count, Map<CachedTriple<T1,T2,T3>,WeightCountTuple> jointCount, Map<CachedPair<T1,T2>,WeightCountTuple> abCount, Map<CachedPair<T1,T3>,WeightCountTuple> acCount, Map<CachedPair<T2,T3>,WeightCountTuple> bcCount, Map<T1,WeightCountTuple> aCount, Map<T2,WeightCountTuple> bCount, Map<T3,WeightCountTuple> cCount) {
         this.count = count;
         this.jointCount = jointCount;
@@ -54,34 +71,75 @@ public WeightedTripleDistribution(long count, Map<CachedTriple<T1,T2,T3>,WeightC
         this.cCount = cCount;
     }
 
+    /**
+     * The joint distribution over the three variables.
+     * @return The joint distribution.
+     */
     public Map<CachedTriple<T1,T2,T3>,WeightCountTuple> getJointCount() {
         return jointCount;
     }
-    
+
+    /**
+     * The joint distribution over the first and second variables.
+     * @return The joint distribution over A and B.
+     */
     public Map<CachedPair<T1,T2>,WeightCountTuple> getABCount() {
         return abCount;
     }
-    
+
+    /**
+     * The joint distribution over the first and third variables.
+     * @return The joint distribution over A and C.
+     */
     public Map<CachedPair<T1,T3>,WeightCountTuple> getACCount() {
         return acCount;
     }
-    
+
+    /**
+     * The joint distribution over the second and third variables.
+     * @return The joint distribution over B and C.
+     */
     public Map<CachedPair<T2,T3>,WeightCountTuple> getBCCount() {
         return bcCount;
     }
-    
+
+    /**
+     * The marginal distribution over the first variable.
+     * @return The marginal distribution for A.
+     */
     public Map<T1,WeightCountTuple> getACount() {
         return aCount;
     }
-    
+
+    /**
+     * The marginal distribution over the second variable.
+     * @return The marginal distribution for B.
+     */
     public Map<T2,WeightCountTuple> getBCount() {
         return bCount;
     }
-    
+
+    /**
+     * The marginal distribution over the third variable.
+     * @return The marginal distribution for C.
+     */
     public Map<T3,WeightCountTuple> getCCount() {
         return cCount;
     }
-    
+
+    /**
+     * Constructs a WeightedTripleDistribution from three lists of the same length and a list of weights of the same length.
+     * <p>
+     * If they are not the same length it throws IllegalArgumentException.
+     * @param first The first list.
+     * @param second The second list.
+     * @param third The third list.
+     * @param weights The weight list.
+     * @param <T1> The first type.
+     * @param <T2> The second type.
+     * @param <T3> The third type.
+     * @return The WeightedTripleDistribution.
+     */
     public static <T1,T2,T3> WeightedTripleDistribution<T1,T2,T3> constructFromLists(List<T1> first, List<T2> second, List<T3> third, List<Double> weights) {
         Map<CachedTriple<T1,T2,T3>,WeightCountTuple> jointCount = new HashMap<>(DEFAULT_MAP_SIZE);
         Map<CachedPair<T1,T2>,WeightCountTuple> abCount = new HashMap<>(DEFAULT_MAP_SIZE);
@@ -147,6 +205,14 @@ public static <T1,T2,T3> WeightedTripleDistribution<T1,T2,T3> constructFromLists
         }
     }
 
+    /**
+     * Constructs a WeightedTripleDistribution by marginalising the supplied joint distribution.
+     * @param jointCount The joint distribution.
+     * @param <T1> The type of A.
+     * @param <T2> The type of B.
+     * @param <T3> The type of C.
+     * @return A WeightedTripleDistribution.
+     */
     public static <T1,T2,T3> WeightedTripleDistribution<T1,T2,T3> constructFromMap(Map<CachedTriple<T1,T2,T3>,WeightCountTuple> jointCount) {
         Map<CachedPair<T1,T2>,WeightCountTuple> abCount = new HashMap<>(DEFAULT_MAP_SIZE);
         Map<CachedPair<T1,T3>,WeightCountTuple> acCount = new HashMap<>(DEFAULT_MAP_SIZE);
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/Token.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/Token.java
index 2a23f33eb..53703e776 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/Token.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/Token.java
@@ -19,13 +19,25 @@
 /**
  * A single token extracted from a String.
  * <p>
- * Tokens are immutable.
+ * Tokens are immutable, and may be records one day.
  */
 public class Token {
 
+    /**
+     * The token text.
+     */
     public final String text;
+    /**
+     * The start index.
+     */
     public final int start;
+    /**
+     * The end index.
+     */
     public final int end;
+    /**
+     * The token type.
+     */
     public final TokenType type;
 
     /**
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/TokenizationException.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/TokenizationException.java
index 48c21e3f9..27d1123e4 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/TokenizationException.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/TokenizationException.java
@@ -20,17 +20,29 @@
  * Wraps exceptions thrown by tokenizers.
  */
 public class TokenizationException extends RuntimeException {
-
     private static final long serialVersionUID = 1L;
 
+    /**
+     * Creates a TokenizationException with the specified message.
+     * @param message The exception message.
+     */
     public TokenizationException(String message) {
         super(message);
     }
 
+    /**
+     * Creates a TokenizationException wrapping the supplied throwable with the specified message.
+     * @param message The exception message.
+     * @param throwable The throwable to wrap.
+     */
     public TokenizationException(String message, Throwable throwable) {
-        super(message, throwable);
+        super(message,throwable);
     }
 
+    /**
+     * Creates a TokenizationException wrapping the supplied throwable.
+     * @param throwable The throwable to wrap.
+     */
     public TokenizationException(Throwable throwable) {
         super(throwable);
     }
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/Tokenizer.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/Tokenizer.java
index c74491ec0..3b95cc36a 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/Tokenizer.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/Tokenizer.java
@@ -39,6 +39,11 @@
  */
 public interface Tokenizer extends Configurable, Cloneable, Provenancable<ConfiguredObjectProvenance> {
 
+    /**
+     * Creates a supplier from the specified tokenizer by cloning it.
+     * @param tokenizer The tokenizer to copy.
+     * @return A supplier of tokenizers.
+     */
     static Supplier<Tokenizer> createSupplier(Tokenizer tokenizer) {
         Supplier<Tokenizer> supplier = () -> {
             try {
@@ -50,6 +55,11 @@ static Supplier<Tokenizer> createSupplier(Tokenizer tokenizer) {
         return supplier;
     }
 
+    /**
+     * Creates a thread local source of tokenizers by making a Tokenizer supplier using {@link #createSupplier(Tokenizer)}.
+     * @param tokenizer The tokenizer to copy.
+     * @return A thread local for tokenizers.
+     */
     static ThreadLocal<Tokenizer> createThreadLocal(Tokenizer tokenizer) {
         return ThreadLocal.withInitial(createSupplier(tokenizer));
     }
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/BreakIteratorTokenizer.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/BreakIteratorTokenizer.java
index d996cfeeb..fd1d344ea 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/BreakIteratorTokenizer.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/BreakIteratorTokenizer.java
@@ -30,7 +30,7 @@
  */
 public class BreakIteratorTokenizer implements Tokenizer {
 
-    @Config(mandatory = true)
+    @Config(mandatory = true, description="The locale language tag string.")
     private String localeStr;
 
     private Locale locale;
@@ -52,9 +52,12 @@ public class BreakIteratorTokenizer implements Tokenizer {
      * Default constructor for configuration system.
      */
     @SuppressWarnings("unused")
-    private BreakIteratorTokenizer() {
-    }
+    private BreakIteratorTokenizer() {}
 
+    /**
+     * Constructs a BreakIteratorTokenizer using the specified locale.
+     * @param locale The locale to use.
+     */
     public BreakIteratorTokenizer(Locale locale) {
         this.locale = locale;
         this.localeStr = locale.toLanguageTag();
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/NonTokenizer.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/NonTokenizer.java
index 4f7d0b803..c123678dd 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/NonTokenizer.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/NonTokenizer.java
@@ -32,6 +32,9 @@ public class NonTokenizer implements Tokenizer {
 
     private boolean done = false;
 
+    /**
+     * Constructs a NonTokenizer.
+     */
     public NonTokenizer() { }
 
     @Override
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/ShapeTokenizer.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/ShapeTokenizer.java
index 9cbd68f8e..8f7b14b30 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/ShapeTokenizer.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/ShapeTokenizer.java
@@ -57,6 +57,9 @@ public class ShapeTokenizer implements Tokenizer {
 
     private boolean ready;
 
+    /**
+     * Constructs a ShapeTokenizer.
+     */
     public ShapeTokenizer() { }
 
     @Override
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/SplitCharactersTokenizer.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/SplitCharactersTokenizer.java
index c567b9cc0..206348908 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/SplitCharactersTokenizer.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/SplitCharactersTokenizer.java
@@ -39,8 +39,14 @@
  */
 public class SplitCharactersTokenizer extends SplitFunctionTokenizer {
 
+    /**
+     * The default split characters.
+     */
     public static final char[] DEFAULT_SPLIT_CHARACTERS = new char[] { '*', '(', ')', '&', '[', ']', '{', '}', '`',
             '\'', '|', ';', ':', '\\', '!', '-', '?' };
+    /**
+     * The default characters which don't cause splits inside digits.
+     */
     public static final char[] DEFAULT_SPLIT_EXCEPTING_IN_DIGITS_CHARACTERS = new char[] { '.', ',', '/', };
 
     /**
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/SplitFunctionTokenizer.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/SplitFunctionTokenizer.java
index 041d58529..b87d2f5b5 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/SplitFunctionTokenizer.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/SplitFunctionTokenizer.java
@@ -79,34 +79,118 @@ public enum SplitType {
      * already been determined.
      */
     public enum SplitResult {
-        NO_SPLIT_WORD(SplitType.NO_SPLIT, TokenType.WORD), 
+        /**
+         * Not a split, is a word.
+         */
+        NO_SPLIT_WORD(SplitType.NO_SPLIT, TokenType.WORD),
+        /**
+         * Not a split, is a ngram.
+         */
         NO_SPLIT_NGRAM(SplitType.NO_SPLIT, TokenType.NGRAM),
+        /**
+         * Not a split, is punctuation.
+         */
         NO_SPLIT_PUNCTUATION(SplitType.NO_SPLIT, TokenType.PUNCTUATION),
+        /**
+         * Not a split, is whitespace.
+         */
         NO_SPLIT_WHITESPACE(SplitType.NO_SPLIT, TokenType.WHITESPACE),
+        /**
+         * Not a split, is a prefix.
+         */
         NO_SPLIT_PREFIX(SplitType.NO_SPLIT, TokenType.PREFIX),
+        /**
+         * Not a split, is a suffix.
+         */
         NO_SPLIT_SUFFIX(SplitType.NO_SPLIT, TokenType.SUFFIX),
-        NO_SPLIT_INFIX(SplitType.NO_SPLIT, TokenType.INFIX), 
+        /**
+         * Not a split, is infix.
+         */
+        NO_SPLIT_INFIX(SplitType.NO_SPLIT, TokenType.INFIX),
+        /**
+         * Not a split, is unknown.
+         */
         NO_SPLIT_UNKNOWN(SplitType.NO_SPLIT, TokenType.UNKNOWN),
+        /**
+         * Split at.
+         */
         SPLIT_AT(SplitType.SPLIT_AT, TokenType.WORD), //the token type is ignored
+        /**
+         * Split before.
+         */
         SPLIT_BEFORE(SplitType.SPLIT_BEFORE, TokenType.WORD), //the token type is ignored
+        /**
+         * Split after a word.
+         */
         SPLIT_AFTER_WORD(SplitType.SPLIT_AFTER, TokenType.WORD),
+        /**
+         * Split after a ngram.
+         */
         SPLIT_AFTER_NGRAM(SplitType.SPLIT_AFTER, TokenType.NGRAM),
+        /**
+         * Split after punctuation.
+         */
         SPLIT_AFTER_PUNCTUATION(SplitType.SPLIT_AFTER, TokenType.PUNCTUATION),
+        /**
+         * Split after whitespace.
+         */
         SPLIT_AFTER_WHITESPACE(SplitType.SPLIT_AFTER, TokenType.WHITESPACE),
+        /**
+         * Split after a prefix.
+         */
         SPLIT_AFTER_PREFIX(SplitType.SPLIT_AFTER, TokenType.PREFIX),
+        /**
+         * Split after a suffix.
+         */
         SPLIT_AFTER_SUFFIX(SplitType.SPLIT_AFTER, TokenType.SUFFIX),
+        /**
+         * Split after infix.
+         */
         SPLIT_AFTER_INFIX(SplitType.SPLIT_AFTER, TokenType.INFIX),
+        /**
+         * Split after an unknown value.
+         */
         SPLIT_AFTER_UNKNOWN(SplitType.SPLIT_AFTER, TokenType.UNKNOWN),
+        /**
+         * Split before and after a word.
+         */
         SPLIT_BEFORE_AND_AFTER_WORD(SplitType.SPLIT_BEFORE_AND_AFTER, TokenType.WORD),
+        /**
+         * Split before and after a ngram.
+         */
         SPLIT_BEFORE_AND_AFTER_NGRAM(SplitType.SPLIT_BEFORE_AND_AFTER, TokenType.NGRAM),
+        /**
+         * Split before and after punctuation.
+         */
         SPLIT_BEFORE_AND_AFTER_PUNCTUATION(SplitType.SPLIT_BEFORE_AND_AFTER, TokenType.PUNCTUATION),
+        /**
+         * Split before and after whitespace.
+         */
         SPLIT_BEFORE_AND_AFTER_WHITESPACE(SplitType.SPLIT_BEFORE_AND_AFTER, TokenType.WHITESPACE),
+        /**
+         * Split before and after prefix.
+         */
         SPLIT_BEFORE_AND_AFTER_PREFIX(SplitType.SPLIT_BEFORE_AND_AFTER, TokenType.PREFIX),
+        /**
+         * Split before and after suffix.
+         */
         SPLIT_BEFORE_AND_AFTER_SUFFIX(SplitType.SPLIT_BEFORE_AND_AFTER, TokenType.SUFFIX),
+        /**
+         * Split before and after infix.
+         */
         SPLIT_BEFORE_AND_AFTER_INFIX(SplitType.SPLIT_BEFORE_AND_AFTER, TokenType.INFIX),
+        /**
+         * Split before and after unknown.
+         */
         SPLIT_BEFORE_AND_AFTER_UNKNOWN(SplitType.SPLIT_BEFORE_AND_AFTER, TokenType.UNKNOWN);
 
+        /**
+         * The split type.
+         */
         public final SplitType splitType;
+        /**
+         * The token type.
+         */
         public final TokenType tokenType;
 
         SplitResult(SplitType splitType, TokenType tokenType) {
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/WhitespaceTokenizer.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/WhitespaceTokenizer.java
index a0d223323..e74862c13 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/WhitespaceTokenizer.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/WhitespaceTokenizer.java
@@ -21,15 +21,20 @@
 
 /**
  * A simple tokenizer that splits on whitespace.  This tokenizer does not create
- * tokens that corresond to whitespace - only those spans of text delimited by 
+ * tokens that correspond to whitespace - only those spans of text delimited by
  * whitespace.  For example, the text "a b" will result in two tokens "a" and "b". 
- *
  */
 public class WhitespaceTokenizer extends SplitFunctionTokenizer {
 
-    public static SplitFunction whitespaceSplitCharacterFunction = (codepoint, index,
+    /**
+     * The splitting function for whitespace, using {@link Character#isWhitespace(char)}.
+     */
+    public static final SplitFunction whitespaceSplitCharacterFunction = (codepoint, index,
             cs) -> Character.isWhitespace(codepoint) ? SplitResult.SPLIT_AT : SplitResult.NO_SPLIT_WORD;
 
+    /**
+     * Constructs a tokenizer that splits on whitespace.
+     */
     public WhitespaceTokenizer() {
         super(whitespaceSplitCharacterFunction);
     }
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/wordpiece/Wordpiece.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/wordpiece/Wordpiece.java
index e91f15563..02c10c9e8 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/wordpiece/Wordpiece.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/impl/wordpiece/Wordpiece.java
@@ -45,6 +45,9 @@
  */
 public class Wordpiece implements Configurable {
 
+    /**
+     * The default unknown token string.
+     */
     public static final String DEFAULT_UNKNOWN_TOKEN = "[UNK]";
 
     @Config(mandatory=true, description="path to a vocabulary data file.")
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/BreakIteratorTokenizerOptions.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/BreakIteratorTokenizerOptions.java
index ca68ee156..c62130a5e 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/BreakIteratorTokenizerOptions.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/BreakIteratorTokenizerOptions.java
@@ -27,6 +27,9 @@
  */
 public class BreakIteratorTokenizerOptions implements TokenizerOptions {
 
+    /**
+     * BreakIteratorTokenizer - The language tag of the locale to be used.
+     */
     @Option(longName = "bi-tokenizer-language-tag", usage = "BreakIteratorTokenizer - The language tag of the locale to be used.")
     public String languageTag;
 
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/CoreTokenizerOptions.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/CoreTokenizerOptions.java
index 9319a3212..e321a298a 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/CoreTokenizerOptions.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/CoreTokenizerOptions.java
@@ -30,9 +30,21 @@
 public class CoreTokenizerOptions implements TokenizerOptions {
 
     private static final Logger logger = Logger.getLogger(CoreTokenizerOptions.class.getName());
+    /**
+     * Options for the break iterator tokenizer.
+     */
     public BreakIteratorTokenizerOptions breakIteratorOptions;
+    /**
+     * Options for the split characters tokenizer.
+     */
     public SplitCharactersTokenizerOptions splitCharactersTokenizerOptions;
+    /**
+     * Options for the split pattern tokenizer.
+     */
     public SplitPatternTokenizerOptions splitPatternTokenizerOptions;
+    /**
+     * Type of tokenizer
+     */
     @Option(longName = "core-tokenizer-type", usage = "Type of tokenizer")
     public CoreTokenizerType coreTokenizerType = CoreTokenizerType.SPLIT_CHARACTERS;
 
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/SplitCharactersTokenizerOptions.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/SplitCharactersTokenizerOptions.java
index b73023f28..a165d7f61 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/SplitCharactersTokenizerOptions.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/options/SplitCharactersTokenizerOptions.java
@@ -25,10 +25,16 @@
  */
 public class SplitCharactersTokenizerOptions implements TokenizerOptions {
 
+    /**
+     * The characters to split on.
+     */
     @Option(longName = "sc-tokenizer-split-characters", usage = "The characters to split on.")
     public char[] splitChars = SplitCharactersTokenizer.DEFAULT_SPLIT_CHARACTERS;
     ;
 
+    /**
+     * Characters to split on unless they appear between digits
+     */
     @Option(longName = "sc-tokenizer-split-x-digits", usage = "Characters to split on unless they appear between digits")
     public char[] splitXDigitsChars = SplitCharactersTokenizer.DEFAULT_SPLIT_EXCEPTING_IN_DIGITS_CHARACTERS;
 
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/universal/Range.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/universal/Range.java
index 329a0026f..62e605970 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/universal/Range.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/universal/Range.java
@@ -22,15 +22,39 @@
  * A range currently being segmented.
  */
 public final class Range implements CharSequence {
+    /**
+     * The character buffer.
+     */
     public char[] buff = new char[16];
+    /**
+     * The token length.
+     */
     public int len;
+    /**
+     * The start index.
+     */
     public int start;
+    /**
+     * The end index.
+     */
     public int end;
+    /**
+     * The value to increment by.
+     */
     public int incr;
+    /**
+     * The current token type.
+     */
     public Token.TokenType type;
 
     Range() {}
 
+    /**
+     * Sets the first two characters in the range, and the type to NGRAM.
+     * @param c1 The first character.
+     * @param c2 The second character.
+     * @param start The start value.
+     */
     public void set(char c1, char c2, int start) {
         buff[0] = c1;
         buff[1] = c2;
@@ -41,6 +65,11 @@ public void set(char c1, char c2, int start) {
         this.type = Token.TokenType.NGRAM;
     }
 
+    /**
+     * Sets the first character in the range.
+     * @param c The first character.
+     * @param start The start value.
+     */
     public void set(char c, int start) {
         buff[0] = c;
         this.start = start;
@@ -50,6 +79,12 @@ public void set(char c, int start) {
         this.type = Token.TokenType.WORD;
     }
 
+    /**
+     * Sets the character range.
+     * @param buff The characters.
+     * @param len The length of the character buffer.
+     * @param start The start index.
+     */
     public void set(char[] buff, int len, int start) {
         if (this.buff.length < buff.length) {
             this.buff = new char[buff.length + 1];
@@ -62,6 +97,11 @@ public void set(char[] buff, int len, int start) {
         this.type = Token.TokenType.WORD;
     }
 
+    /**
+     * Sets this range to represent a punctuation character.
+     * @param p The punctuation character.
+     * @param start The start index.
+     */
     public void punct(char p, int start) {
         buff[0] = p;
         this.len = 1;
@@ -71,6 +111,10 @@ public void punct(char p, int start) {
         this.type = Token.TokenType.PUNCTUATION;
     }
 
+    /**
+     * Sets the token type.
+     * @param type The token type.
+     */
     public void setType(Token.TokenType type) {
         this.type = type;
     }
diff --git a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/universal/UniversalTokenizer.java b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/universal/UniversalTokenizer.java
index 5821647f7..29727383d 100644
--- a/Util/Tokenization/src/main/java/org/tribuo/util/tokens/universal/UniversalTokenizer.java
+++ b/Util/Tokenization/src/main/java/org/tribuo/util/tokens/universal/UniversalTokenizer.java
@@ -137,7 +137,7 @@ public class UniversalTokenizer implements Tokenizer {
      * Is the tokenizer ready?
      */
     private boolean ready;
-    @Config
+    @Config(description="Send punctuation through as tokens.")
     private boolean sendPunct = false;
     /**
      * A set of tokens that were generated and need to be returned.
@@ -150,6 +150,7 @@ public class UniversalTokenizer implements Tokenizer {
     private char c;
 
     /**
+     * Constructs a universal tokenizer.
      * @param sendPunct if sendPunct is true, then the tokenizer will generate punctuation tokens.
      */
     public UniversalTokenizer(boolean sendPunct) {
@@ -279,26 +280,50 @@ public static boolean isNgram(char c) {
         }
     }
 
+    /**
+     * Does this tokenizer generate unigrams?
+     * @return True if the tokenizer generates unigram tokens.
+     */
     public boolean isGenerateUnigrams() {
         return generateUnigrams;
     }
 
+    /**
+     * Controls if the tokenizer generates unigrams.
+     * @param generateUnigrams If true generates unigram tokens.
+     */
     public void setGenerateUnigrams(boolean generateUnigrams) {
         this.generateUnigrams = generateUnigrams;
     }
 
+    /**
+     * Does this tokenizer generate ngrams?
+     * @return True if the tokenizer generates ngram tokens.
+     */
     public boolean isGenerateNgrams() {
         return generateNgrams;
     }
 
+    /**
+     * Controls if the tokenizer generates ngrams.
+     * @param generateNgrams If true generates ngram tokens.
+     */
     public void setGenerateNgrams(boolean generateNgrams) {
         this.generateNgrams = generateNgrams;
     }
 
+    /**
+     * Returns the maximum token length this tokenizer will generate.
+     * @return The maximum token length.
+     */
     public int getMaxTokenLength() {
         return maxTokenLength;
     }
 
+    /**
+     * Sets the maximum token length this tokenizer will generate.
+     * @param maxTokenLength The maximum token length.
+     */
     public void setMaxTokenLength(int maxTokenLength) {
         this.maxTokenLength = maxTokenLength;
     }
@@ -554,6 +579,10 @@ public Token.TokenType getType() {
         }
     }
 
+    /**
+     * Gets the current position in the input.
+     * @return The current position.
+     */
     public int getPos() {
         return currPos;
     }
diff --git a/Util/Tokenization/src/test/java/org/tribuo/util/tokens/options/CoreTokenizerOptionsTest.java b/Util/Tokenization/src/test/java/org/tribuo/util/tokens/options/CoreTokenizerOptionsTest.java
index a2d3ce7b3..11acb95a7 100644
--- a/Util/Tokenization/src/test/java/org/tribuo/util/tokens/options/CoreTokenizerOptionsTest.java
+++ b/Util/Tokenization/src/test/java/org/tribuo/util/tokens/options/CoreTokenizerOptionsTest.java
@@ -114,6 +114,9 @@ public void testComma() {
 
     public static class CommaOptions implements Options {
 
+        /**
+         * The characters.
+         */
         @Option(longName = "my-chars", usage = "The characters.")
         public char[] myChars;
 
diff --git a/tests/src/main/java/org/tribuo/tests/Resources.java b/tests/src/main/java/org/tribuo/tests/Resources.java
index f16715672..a69e3e554 100644
--- a/tests/src/main/java/org/tribuo/tests/Resources.java
+++ b/tests/src/main/java/org/tribuo/tests/Resources.java
@@ -31,9 +31,14 @@
  * Utils for working with classpath resources at test time.
  */
 public final class Resources {
-    private Resources() {
-    }
+    private Resources() {}
 
+    /**
+     * Copies a classpath resource to a temporary file.
+     * @param resource The resource to copy.
+     * @return The path of the temporary file.
+     * @throws IOException If the resource could not be copied.
+     */
     public static Path copyResourceToTmp(String resource) throws IOException {
         Path path = Files.createTempFile("test", ".csv");
         path.toFile().deleteOnExit();