number of folds (default = 10) for cross validation optimisation
seed
for random split to train / test (default 666)
logging_level
print extra data while training 0 - no data, 1 - gridSearch data (default), 2 - all data
Value
vector with average accuracy for chosen parameters, and a list of the best parameter combination: (accuracy, (num_iter, beta, lambda, loss_type))
Details
Finds optimised parameters for deepboost training.
using grid search techniques over:
- predefined, battle tested parameter possible values
- cross validation over k folds