Grid search Sample Clauses

Grid search. In order to evaluate the model, the data is split into two parts: training set (80%) and test set (20%). This ensures that, while training the model with the training set, the evaluation can be performed on an out of sample data set. For the hyperparameter tuning, or grid search, 10-fold cross validation is used on the training set. The parameters tuned in this process, with the respective ranges and the optimal value are shown in Table 3a. Parameter Range Optimum Number of neurons [16,32,64,128] 32 Batch size [32,64,128,256] 128 Number of epochs [5:5:30] 25 Embedding size [20,40,60,80] 60 Model Tuned parameters k-NN k = 34 GLM C = 1 SVM kernel = ‘RBF’, C = 10, γ = 1e− 5 XGBoost η = 0.2, max_depth = 5 (a) Bi-LSTM model