Discuss, Learn and be Happy דיון בשאלות

help brightness_4 brightness_7 format_textdirection_r_to_l format_textdirection_l_to_r

Which evaluation metric is most directly affected when a model produces a high number of false positives? (2025 מועד א סמסטר ב)

1
done
by
מיין לפי

Which metric is more suitable for unsupervised learning? (מועד א 2025 סמסטר ב)

1
done
by
מיין לפי

Which of the following activation functions is more suitable for the final layer activation of a neural network that aims to solve a binary classification problem? (מועד א 2025 סמסטר ב)

1
done
by
מיין לפי

You trained a binary classifier and obtained the following confusion matrix: Predicted Positive Predicted Negative Actual Positive 80 20 Actual Negative 120 780 Which of the following statements is most accurate based on the confusion matrix? (סידור נכון: TP = 80 , TN=780 , FP=120 תFN=20 ) (מועד א 2025 סמסטר ב)

1
done
by
מיין לפי

Batch-norm might eliminate the need for using dropout (מועד א 2025 סמסטר ב)

1
done
by
מיין לפי

Batch-normalization allows for higher learning rates during training מועד ב 2025 סמסטר ב

1
done
by
מיין לפי

Increasing the depth of a decision tree is likely to decrease error on the training set מועד ב 2025 סמסטר ב

1
done
by
מיין לפי

KNN does not perform a training phase מועד ב 2025 סמסטר ב

1
done
by
מיין לפי

The original BERT model uses only the encoder part of the Transformer architecture מועד ב 2025 סמסטר ב

1
done
by
מיין לפי

We need to employ iterative (numeric) solution to find out the linear regression weights מועד ב סמסטר ב 2025

1
done
by
מיין לפי