zero-one.geni.ml.tuning — zero.one/geni 0.4.0

avg-metrics clj

(avg-metrics model)

A cross-validator model's metric for each of its estimator's param maps, averaged over the folds, in the order of :estimator-param-maps.

cross-validator clj

(cross-validator {:keys [estimator evaluator estimator-param-maps num-folds seed
                         parallelism collect-sub-models]})

K-fold cross validation performs model selection by splitting the dataset into a set of non-overlapping randomly partitioned folds which are used as separate training and test datasets e.g., with k=3 folds, K-fold cross validation will generate 3 (training, test) dataset pairs, each of which uses 2/3 of the data for training and 1/3 for testing. Each fold is used as the test set exactly once.

Source: https://spark.apache.org/docs/3.0.1/api/scala/org/apache/spark/ml/tuning/CrossValidator.html

Timestamp: 2020-10-19T01:55:48.855Z

param-grid clj

(param-grid grids)

Builder for a param grid used in grid search-based model selection.

Source: https://spark.apache.org/docs/3.0.1/api/scala/org/apache/spark/ml/tuning/ParamGridBuilder.html

Timestamp: 2020-10-19T01:55:49.184Z

param-grid-builder clj

(param-grid-builder grids)

Builder for a param grid used in grid search-based model selection.

Source: https://spark.apache.org/docs/3.0.1/api/scala/org/apache/spark/ml/tuning/ParamGridBuilder.html

Timestamp: 2020-10-19T01:55:49.184Z

sub-models clj

(sub-models model)

The models that a cross-validator or a train-validation split fitted for each param map, with :collect-sub-models true: a vector per fold of a model per param map for a cross-validator, and a model per param map for a split. Nil when it didn't keep them.

train-validation-split clj

(train-validation-split {:keys [estimator evaluator estimator-param-maps seed
                                parallelism collect-sub-models train-ratio]})

Validation for hyper-parameter tuning. Randomly splits the input dataset into train and validation sets, and uses evaluation metric on the validation set to select the best model. Similar to CrossValidator, but only splits the set once.

Source: https://spark.apache.org/docs/3.0.1/api/scala/org/apache/spark/ml/tuning/TrainValidationSplit.html

Timestamp: 2020-10-19T01:55:49.563Z

validation-metrics clj

(validation-metrics model)

A train-validation split model's metric for each of its estimator's param maps, in the order of :estimator-param-maps.