Viva prep · Real questions · Student experiences Enroll in Bootcamp

Search viva questions

Level3_20 MLP Level 1 Aug 7, 2026
  1. 1
    Show your ID card.
  2. 2
    Introduction.
  3. 3
    Explain your notebook/project.
  4. 4
    How many models did you train?
  5. 5
    On how many models did you perform Hyperparameter Tuning?
  6. 6
    Why did you perform Hyperparameter Tuning only on those models?
  7. 7
    Which model took the longest to train? Why?
  8. 8
    Explain Gradient Descent.
  9. 9
    Explain Gradient Descent on a whiteboard.
  10. 10
    Why did you use Accuracy instead of F1-Score or ROC-AUC?
  11. 11
    Draw and explain the Confusion Matrix.
  12. 12
    Write and explain the formulas for Precision, Recall, F1-Score, and Accuracy.
  13. 13
    How did you perform feature engineering on date/time features?
  14. 14
    Which hyperparameters did you tune?
  15. 15
    What time series model would you use?
  16. 16
    What more could you have done to improve the project?
  17. 17
    How did you handle class imbalance?
  18. 18
    What does the C parameter in Logistic Regression represent?
  19. 19
    Why did you use Median instead of Mean (or vice versa) for imputation?
  20. 20
    Which imputation technique would you use for a right-skewed feature?
  21. 21
    Which model did you use for the final submission?
  22. 22
    Show your Kaggle leaderboard score and rank.
  23. 23
    Show your Kaggle submissions.
  24. 24
    Questions based on your notebook implementation.
  25. 25
    Suggestions on improving the notebook/project.
  26. 26
    Asked if you have any questions for the proctor.
level3_30 MLP Level 1 Aug 7, 2026
  1. 1
    Introduction.
  2. 2
    Explain your notebook/project.
  3. 3
    How did you handle missing (NaN) values?
  4. 4
    Which imputer should be used when a feature contains outliers?
  5. 5
    How does Logistic Regression work?
  6. 6
    What is the loss function of Logistic Regression?
  7. 7
    How does Linear Regression work?
  8. 8
    What is the loss function of Linear Regression?
  9. 9
    How does Random Forest work?
  10. 10
    How does XGBoost work?
  11. 11
    Difference between Bagging and Boosting.
  12. 12
    How can you handle an imbalanced dataset?
  13. 13
    What is the Pearson Correlation formula?
  14. 14
    Do all trees in a Random Forest receive all features?
  15. 15
    What are the types of feature reduction techniques?
  16. 16
    How does PCA work?
  17. 17
    Is PCA linear or non-linear?
  18. 18
    What are Eigenvectors?
  19. 19
    How do you calculate PCA manually?
  20. 20
    How do you calculate Eigenvectors manually?
  21. 21
    Why does L1 Regularization eliminate features?
  22. 22
    Why does L2 Regularization help prevent overfitting?
  23. 23
    In a poisonous apple detection problem, would you prioritize Precision or Recall? Why?
level1-8 MLP Level 1 Aug 7, 2026
  1. 1
    Explain your EDA.
  2. 2
    Explain your EDA code.
  3. 3
    Suggestions to improve the notebook/code.
  4. 4
    No additional theory or coding questions were asked.
level1_viva2 MLP Level 1 Aug 7, 2026
  1. 1
    Show your ID card.
  2. 2
    Introduction.
  3. 3
    Tell me about yourself.
  4. 4
    Tell me about yourself.
  5. 5
    Explain the problem statement.
  6. 6
    Explain your notebook/project.
  7. 7
    Explain your preprocessing.
  8. 8
    Explain your EDA.
  9. 9
    Explain your feature engineering.
  10. 10
    Explain your hyperparameter tuning.
  11. 11
    Explain your model comparison.
  12. 12
    What insights did you gain from your analysis?
  13. 13
    What difficulties did you face during the project?
  14. 14
    How did you overcome those difficulties?
  15. 15
    What did you learn from the project?
  16. 16
    Show an overview of your Kaggle submissions.
  17. 17
    What models did you use?
  18. 18
    Why did you use those models?
  19. 19
    Why did you use only tree/leaf-based models?
  20. 20
    Why do we perform hyperparameter tuning?
  21. 21
    What are hyperparameters?
  22. 22
    Difference between parameters and hyperparameters.
  23. 23
    How is hyperparameter tuning done?
  24. 24
    Ways to perform hyperparameter tuning.
  25. 25
    Difference between Pipeline and ColumnTransformer.
  26. 26
    Difference between Pipeline and Transformer.
  27. 27
    Why did you use StandardScaler?
  28. 28
    Which scaling method is better for handling outliers?
  29. 29
    Why do machine learning models use numerical features?
  30. 30
    What is PCA?
  31. 31
    Difference between XGBoost and LightGBM.
  32. 32
    What is overfitting?
  33. 33
    How do you reduce overfitting?
  34. 34
    What is the Bias-Variance Tradeoff?
  35. 35
    Difference between Bagging and Boosting.
  36. 36
    Which sklearn API contains OneHotEncoder?
  37. 37
    Which sklearn API contains StandardScaler?
  38. 38
    Coding: Implement the preprocessing Pipeline used in your notebook.
  39. 39
    Coding: Write a Decision Tree classifier snippet.
  40. 40
    Coding: Load the California Housing dataset.
  41. 41
    Coding: Perform a train-test split on the California Housing dataset.
  42. 42
    Coding: Load the Iris dataset.
  43. 43
    Coding: Perform a train-test split on the Iris dataset.
  44. 44
    Coding: Train a Decision Tree classifier on the Iris dataset.
  45. 45
    Coding: Print the Accuracy Score.
  46. 46
    Coding: Use the describe() function on the California Housing dataset.
mlp_level1_viva9 MLP Level 1 Aug 7, 2026
  1. 1
    Show your ID card.
  2. 2
    Explain your notebook/project.
  3. 3
    Explain your EDA.
  4. 4
    Explain your univariate analysis.
  5. 5
    Explain your bivariate analysis.
  6. 6
    Explain the graphs you created and the insights from them.
  7. 7
    How did you handle missing values?
  8. 8
    What is SimpleImputer?
  9. 9
    How does SimpleImputer work?
  10. 10
    Explain your preprocessing pipeline.
  11. 11
    Have you used Pipeline?
  12. 12
    Why did you use (or not use) Pipeline?
  13. 13
    Explain your feature engineering.
  14. 14
    Did you create any new features?
  15. 15
    What encoding technique did you use?
  16. 16
    Why did you choose that encoder?
  17. 17
    Replace OrdinalEncoder with OneHotEncoder.
  18. 18
    Difference between Label Encoding and OneHot Encoding.
  19. 19
    What scaler did you use?
  20. 20
    Why did you choose MinMaxScaler over StandardScaler?
  21. 21
    Explain StandardScaler.
  22. 22
    Difference between StandardScaler and MinMaxScaler.
  23. 23
    What models did you try?
  24. 24
    Which model performed the best?
  25. 25
    Why didn't you explore linear models?
  26. 26
    Explain your model comparison.
  27. 27
    Explain the hyperparameters you tuned.
  28. 28
    What is the learning rate?
  29. 29
    Is a higher or lower learning rate better? Why?
  30. 30
    What is the Bias-Variance Tradeoff?
  31. 31
    What is Boosting?
  32. 32
    What is Gradient Boosting?
  33. 33
    What algorithm does Boosting use?
  34. 34
    What is Bagging?
  35. 35
    Difference between Bagging and Boosting.
  36. 36
    Difference between XGBoost and LightGBM.
  37. 37
    What is R² Score?
  38. 38
    Write the formula for R² Score.
  39. 39
    From which library do you import r2_score?
  40. 40
    From which library do you import OneHotEncoder?
  41. 41
    From which library do you import XGBoost?
  42. 42
    Why is Logistic Regression called "Regression" in a classification task?
  43. 43
    What is F1-Score?
  44. 44
    Write the formulas for Accuracy, Precision, and F1-Score.
  45. 45
    What are Ensemble methods?
  46. 46
    What resources did you use while building the project?
  47. 47
    Coding: Create a dataframe containing only categorical (object) columns.
  48. 48
    Coding: Apply OneHotEncoder to categorical columns with ≤10 unique values and find the new number of columns.
  49. 49
    Coding: Split the training data into numerical and categorical dataframes.
  50. 50
    Coding: Build separate preprocessing pipelines for numerical and categorical features using ColumnTransformer.
  51. 51
    Coding: Perform a train-test split.
  52. 52
    Coding: Print all numerical and non-numerical columns.
  53. 53
    Coding: Print Recall Score instead of Accuracy Score.
  54. 54
    Coding: Implement RandomForestRegressor (import, initialize, fit, predict).
  55. 55
    Coding: Build a simple preprocessing pipeline.
  56. 56
    Coding: Write a custom preprocessing pipeline.
  57. 57
    Coding: Apply Label Encoding to ["green", "blue", "white", "blue", "green"].
  58. 58
    Coding: Filter rows where RegionCode = "Florida".
  59. 59
    Coding: Filter rows where RegionCode = "Florida" and TargetValue > 50000.
  60. 60
    Coding: Create a subset of the data for RegionCode = "Florida".
  61. 61
    Coding: Drop two specified columns from the training dataset.
level1_viva8_MLP_Viva MLP Level 1 Aug 7, 2026
  1. 1
    Introduction
  2. 2
    Start explaining your notebook from the EDA section.
  3. 3
    Explain every line of your code in detail.
  4. 4
    No additional theory questions were asked.
leve2_74 MLP Level 1 Aug 7, 2026
  1. 1
    Introduction.
  2. 2
    Explain the problem statement.
  3. 3
    Explain your notebook from start to finish.
  4. 4
    What is the learning rate?
  5. 5
    How does the learning rate affect overfitting and underfitting?
  6. 6
    Coding: Create a dummy Pipeline and add it to a ColumnTransformer.
  7. 7
    Coding: Use a different scaling method and a different imputation strategy.
  8. 8
    What is Data Preprocessing?
  9. 9
    What is EDA?
  10. 10
    What is Feature Engineering?
  11. 11
    What is Hyperparameter Tuning?
  12. 12
    What does random_state = 42 mean?
  13. 13
    Questions based on your notebook implementation.
Level3_46 MLP Level 1 Aug 7, 2026
  1. 1
    Explain your notebook/project.
  2. 2
    How did you handle outliers?
  3. 3
    Is removing outliers always a good practice?
  4. 4
    What can you do instead of removing outliers?
  5. 5
    What is collinearity?
  6. 6
    If two features are highly correlated, what should you do?
  7. 7
    Explain Bagging.
  8. 8
    Explain Boosting.
  9. 9
    How does Bagging work?
  10. 10
    How does Boosting work?
  11. 11
    Explain Decision Tree.
  12. 12
    How does a Decision Tree work?
  13. 13
    What is Entropy?
  14. 14
    Why are Bagging and Boosting better than a single Decision Tree?
  15. 15
    What are different evaluation metrics?
  16. 16
    Write the formulas for evaluation metrics.
  17. 17
    Why is Accuracy not always a good evaluation metric?
  18. 18
    For a diabetes dataset, which evaluation metric would you choose and why?
  19. 19
    How does Logistic Regression work?
  20. 20
    How does KNN work?
level3_36 MLP Level 1 Aug 7, 2026
  1. 1
    What is the problem statement?
  2. 2
    Explain your notebook/project.
  3. 3
    Explain Precision.
  4. 4
    Explain Recall.
  5. 5
    Explain F1-Score.
  6. 6
    Why did you use Label Encoding instead of other encoding techniques?
  7. 7
    Explain the hyperparameters you used.
  8. 8
    Explain Logistic Regression.
  9. 9
    Explain SVM.
  10. 10
    What is a kernel in SVM?
  11. 11
    Explain KNN.
  12. 12
    Why is KNN called a lazy learner?
  13. 13
    What are the different types of SVM kernels?
  14. 14
    Explain Bagging.
  15. 15
    Explain Boosting.
  16. 16
    Why did you use Random Forest?
  17. 17
    What are feature selection techniques?
  18. 18
    What is ROC-AUC?
  19. 19
    Why didn't you use ROC-AUC as the scoring metric?
level2_107_MLP_Viva MLP Level 1 Aug 7, 2026
  1. 1
    What are the techniques for handling null values?
  2. 2
    How do you handle imbalanced datasets?
  3. 3
    Why did you try multiple models?
  4. 4
    What is the relationship between Information Gain and Entropy?
  5. 5
    Explain the Decision Tree building process.
  6. 6
    Why did you choose that learning rate for LightGBM?
  7. 7
    Coding: On the MNIST dataset, create a Pipeline with StandardScaler and SVM, print the Classification Report, then apply PCA with SVM and compare the scores.
  8. 8
    Name different activation functions.
  9. 9
    Explain the Tanh activation function (input and output).
Most Asked Topics

No tags yet.

View all →
Top Proctors
Created for educational purposes only. Questions are based on students' personal experiences and may not reflect actual exam content.