What is a hyperparameter?, What is a hyperparameter?, B) Set before training, C) Output prediction, D) Final score, Which of these is a hyperparameter in Decision Trees?, A) Weights, B) Bias term, C) max_depth, D) Prediction, In SVM, a small C usually means: Wide margin / high bias, True, False, Why should you NOT tune using the test set?, A) It makes training faster, B) It causes information leakage, C) It reduces accuracy, D) It changes labels, If GridSearchCV has 3 values for depth and 2 values for leaf size, how many combinations?, 3, 5, 7, 6, What is a good practice before experiments?, A) Change labels, B) Set random_state, C) Remove target column, D) Use test set repeatedly, Validation curves show performance versus:, A) Number of classes, B) One hyperparameter value, D) Epoch number only, C) File size, Cross-validation splits data into:, A) Colors, C) Labels, B) k fold, D) Classes only, If training score is high but validation score is low, the model is:, A) Overfitting, B) Underfitting, C) Perfect, D) Balanced, Learning curves plot performance against:, A) CPU speed, B) Number of training samples, C) Tree depth only, D) Labels count.

WEEK4 A1

podle
Tisk
Vložit

Výsledková tabule/Žebříček

Vizuální styl

Možnosti

Přepnout šablonu

Obnovit automatické uložení: ?