Definisci misure aggregate

Ora andrai a definire delle misure di performance. Il dataset knowledge_train_data è già stato caricato per te, così come i pacchetti mlr e tidyverse. Inoltre, è già stato eseguito il seguente codice:

task <- makeClassifTask(data = knowledge_train_data, 
                        target = "UNS")

lrn <- makeLearner(cl = "classif.nnet", fix.factors.prediction = TRUE)

param_set <- makeParamSet(
  makeIntegerParam("size", lower = 1, upper = 5),
  makeIntegerParam("maxit", lower = 1, upper = 300),
  makeNumericParam("decay", lower = 0.0001, upper = 1)
)

ctrl_random <- makeTuneControlRandom(maxit = 10)

Questo esercizio fa parte del corso

Ottimizzazione degli iperparametri in R

Visualizza il corso

Istruzioni dell'esercizio

Usa la funzione setAggregation, che aggregherà la deviazione standard delle metriche di performance.
Applica setAggregation all’errore medio di misclassificazione e all’accuratezza dopo il resampling.
Ottimizza il modello in base all’errore medio di misclassificazione. Ricorda che il primo argomento viene usato per l’ottimizzazione.

Esercizio pratico interattivo

Prova a risolvere questo esercizio completando il codice di esempio.

# Create holdout sampling
holdout <- makeResampleDesc("Holdout", predict = "both")

# Perform tuning
lrn_tune <- tuneParams(learner = lrn, 
                       task = task, 
                       resampling = holdout, 
                       control = ctrl_random, 
                       par.set = param_set,
                       measures = list(___, ___(___, train.mean), ___, ___(___, train.mean)))

Modifica ed esegui il codice

Questo esercizio fa parte del corso

Ottimizzazione degli iperparametri in R

AvançadoNível de habilidade

4.8+

Inizia il corso gratis

Why do we use the strange word "hyperparameter"? What makes it hyper? Here, you will understand what model parameters are, and why they are different from hyperparameters in machine learning. You will then see why we would want to tune them and how the default setting of caret automatically includes hyperparameter tuning.

Exercise 1: Parameters vs hyperparameters Exercise 2: Model parameters vs. hyperparameters Exercise 3: Hyperparameters in linear models Exercise 4: What are the coefficients?Exercise 5: Recap of machine learning basics Exercise 6: Machine learning with caret Exercise 7: Resampling schemes Exercise 8: Hyperparameter tuning in caret Exercise 9: Hyperparameters in Stochastic Gradient Boosting Exercise 10: Changing the number of hyperparameters to tune Exercise 11: Tune hyperparameters manually

In this chapter, you will learn how to tune hyperparameters with a Cartesian grid. Then, you will implement faster and more efficient approaches. You will use Random Search and adaptive resampling to tune the parameter grid, in a way that concentrates on values in the neighborhood of the optimal settings.

Exercise 1: Hyperparameter tuning in caret Exercise 2: Finding hyperparameters Exercise 3: Cartesian grid search in caret Exercise 4: Plot hyperparameter model output Exercise 5: Grid vs. Random Search Exercise 6: Grid search with range of hyperparameters Exercise 7: Find train() option for random search Exercise 8: Random search with caret Exercise 9: Adaptive resampling Exercise 10: Advantages of Adaptive Resampling Exercise 11: Adaptive Resampling with caret

Here, you will use another package for machine learning that has very convenient hyperparameter tuning functions. You will define a Cartesian grid or perform Random Search, as well as advanced techniques. You will also learn different ways to plot and evaluate models with different hyperparameters.

Exercise 1: Machine Learning con mlr Exercise 2: Machine Learning con mlr Exercise 3: Modeling con mlr Exercise 4: Grid e random search con mlr Exercise 5: Random search con mlr Exercise 6: Esegui l'hyperparameter tuning con mlr Exercise 7: Valutare gli iperparametri con mlr Exercise 8: Perché valutare il tuning?Exercise 9: Valutare i risultati del tuning degli iperparametri Exercise 10: Ottimizzazione avanzata con mlr Exercise 11: Definisci controlli di tuning avanzati Exercise 12: Definisci misure aggregate

Esercizio in corso

Exercise 13: Impostare gli iperparametri

In this final chapter, you will use h2o, another package for machine learning with very convenient hyperparameter tuning functions. You will use it to train different models and define a Cartesian grid. Then, You will implement a Random Search use stopping criteria. Finally, you will learn AutoML, an h2o interface which allows for very fast and convenient model and hyperparameter tuning with just one function.

Exercise 1: Machine learning with h2o Exercise 2: Prepare data for modelling with h2o Exercise 3: Modeling with h2o Exercise 4: Grid and random search with h2o Exercise 5: Grid search with h2o Exercise 6: Random search with h2o Exercise 7: Stopping criteria Exercise 8: Automatic machine learning with H2O Exercise 9: AutoML in h2o Exercise 10: Scoring the leaderboard Exercise 11: Extract h2o models and evaluate performance Exercise 12: Wrap-up