דסבת

Discuss, Learn and be Happy דיון בשאלות

• «
• 1
• 2
• ...
• 6
• 7
• 8
• 9
• 10
• 11
• 12
• ...
• 36
• 37
• »

What does LOOCV (Leave-One-Out Cross-Validation) involve?

-

1

+

שאלה #131966

		It involves splitting the data into K folds and training the model K times, each time using a different fold as the test set.
		It randomly selects a portion of the data to be used for training and the remaining data for testing.
done		It iteratively trains the model on all data points except one, and then evaluates the model's performance on the omitted data point.
		It involves dividing the data into a train set and a validation set, with the validation set used for hyperparameter tuning.

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

What distinguishes the train dataset from the validation dataset in machine learning model development?

-

1

+

שאלה #131967

done		The train dataset serves as the primary dataset for training the model, whereas the validation dataset is employed for assessing the model's performance and generalization to unseen data during hyperparameter tuning.
		The train dataset is used exclusively for training the model parameters, while the validation dataset is utilized for fine-tuning the model architecture.
		The train dataset contains labeled data used for both model training and evaluation, while the validation dataset consists of unlabeled data used only for testing.
		The train dataset is typically larger in size and used for cross-validation, whereas the validation dataset is smaller and employed for model testing to estimate the model's accuracy.

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

How does nested cross-validation differ from traditional cross-validation techniques?

-

1

+

שאלה #131968

		Nested cross-validation does not involve hyperparameter tuning.
		Nested cross-validation uses a single train-test split for model evaluation.
done		Nested cross-validation incorporates an inner loop for hyperparameter tuning within each iteration of the outer loop.
		Nested cross-validation requires a larger dataset compared to traditional cross-validation.

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

Why is nested cross-validation considered more reliable for model evaluation?

-

1

+

שאלה #131969

		Because it provides a biased estimate of the model's performance.
		Because it requires less computational resources compared to traditional cross-validation.
		Because it focuses solely on hyperparameter tuning without evaluating the model's generalization ability.
done		Because it accounts for both hyperparameter tuning and model evaluation on unseen data, resulting in a more accurate performance estimate.

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

What is data leakage with respect to feature engineering/transformation in machine learning?

-

1

+

שאלה #131970

done		Unintentional inclusion of information from the validation or test set during feature engineering or transformation steps.
		Including irrelevant features in the model.
		Using a small subset of features for model training.
		Transforming features without considering their importance.

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

Why is it important to preserve class ratios during data splitting?

-

1

+

שאלה #131971

		To increase computational efficiency.
done		To ensure that each fold of the dataset is representative of the overall distribution of classes.
		To decrease the complexity of the model.
		To reduce overfitting in the model.

נערך לפני 6 חודשים Shachar Adam

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

What can happen if time-dependent records are not preserved during random splitting of data?

-

1

+

שאלה #131972

		The model may generalize well to unseen data.
		The model may become too complex.
done		The temporal relationships between data points may be disrupted, leading to inaccurate model evaluations.
		The model may suffer from underfitting.

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

How can data leakage be prevented during feature engineering/transformation?

-

1

+

שאלה #131973

		By using a different feature engineering/transformation technique for each fold of the dataset.
		By transforming features without considering their relevance to the target variable.
		By including all features from the validation set in the training set.
done		By ensuring that feature engineering/transformation is performed only on the training set, without using information from the validation or test set.

נערך לפני 6 חודשים Shachar Adam

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

What is the primary purpose of preserving class ratios during data splitting?

-

1

+

שאלה #131974

		To increase the model's complexity.
done		To prevent biases in model evaluation by ensuring that each fold represents the overall distribution of classes.
		To decrease the computational resources required for model training.
		To maximize the number of data points in the validation set.

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

What is the primary purpose of the Receiver Operating Characteristic (ROC) curve?

-

1

+

שאלה #131975

		Classifying instances into multiple categories.
done		Visualizing, organizing, and selecting classifiers based on their performance.
		Identifying outliers in the dataset.
		Calculating feature importance in machine learning models.

• gavel
• mood_bad
• share

הגב

מיין לפי דירוג חדש ישן

• «
• 1
• 2
• ...
• 6
• 7
• 8
• 9
• 10
• 11
• 12
• ...
• 36
• 37
• »