Viva prep · Real questions · Student experiences Enroll in Bootcamp

Proctor workspace

Proctor level1_viva4

Share your experience Add your viva experience here
85 Questions
2 Sets
0 Topics
0 Reviews

Student reviews

No student reviews for this proctor yet.

Approved viva sets

Download all
  1. 1
    Tell me about yourself.
    Times asked 6
  2. 2
    What is your Kaggle leaderboard score/rank?
    Times asked 1
  3. 3
    Explain your notebook/project.
    Times asked 21
  4. 4
    Explain your EDA.
    Times asked 8
  5. 5
    How many graphs did you create?
    Times asked 1
  6. 6
    Explain each graph you created.
    Times asked 1
  7. 7
    What observations did you make from your graphs?
    Times asked 1
  8. 8
    How do you hide/show values in a correlation heatmap?
    Times asked 1
  9. 9
    Explain your boxplot.
    Times asked 1
  10. 10
    Explain your density regression plot.
    Times asked 1
  11. 11
    What feature engineering did you perform?
    Times asked 2
  12. 12
    Why did you perform that feature engineering?
    Times asked 1
  13. 13
    How did you handle missing values?
    Times asked 4
  14. 14
    How did you handle numerical missing values?
    Times asked 2
  15. 15
    How did you handle categorical missing values?
    Times asked 1
  16. 16
    How did you handle disguised missing values (None, NaN, Unknown, Unspecified)?
    Times asked 1
  17. 17
    Why did you preserve the "Unknown" category?
    Times asked 1
  18. 18
    What numerical transformers did you use?
    Times asked 1
  19. 19
    What categorical transformers did you use?
    Times asked 1
  20. 20
    Why did you use those transformers?
    Times asked 1
  21. 21
    What is Data Leakage?
    Times asked 1
  22. 22
    When does Data Leakage occur?
    Times asked 1
  23. 23
    How did you prevent Data Leakage?
    Times asked 1
  24. 24
    What is an N-gram?
    Times asked 1
  25. 25
    What is Unigram?
    Times asked 1
  26. 26
    What is Bigram?
    Times asked 1
  27. 27
    What is a 6-gram, 8-gram, or 11-gram?
    Times asked 1
  28. 28
    What is TF-IDF?
    Times asked 2
  29. 29
    Difference between word-level and character-level TF-IDF.
    Times asked 1
  30. 30
    What does TF-IDF Vectorizer do?
    Times asked 1
  31. 31
    Explain the parameters of TF-IDF Vectorizer.
    Times asked 1
  32. 32
    Why did you choose 50,000 TF-IDF features?
    Times asked 1
  33. 33
    What models did you use?
    Times asked 3
  34. 34
    Why did you choose those models?
    Times asked 3
  35. 35
    What is your best model?
    Times asked 1
  36. 36
    Why is it your best model?
    Times asked 2
  37. 37
    What is an Ensemble model?
    Times asked 1
  38. 38
    Explain your final modeling approach.
    Times asked 1
  39. 39
    What hyperparameter tuning did you perform?
    Times asked 1
  40. 40
    Which hyperparameters did you tune?
    Times asked 2
  41. 41
    How many model fits will GridSearchCV perform?
    Times asked 1
  42. 42
    How many model fits will RandomizedSearchCV perform?
    Times asked 1
  43. 43
    Difference between GridSearchCV and RandomizedSearchCV.
    Times asked 5
  44. 44
    How do you know how many fits will be performed?
    Times asked 1
  45. 45
    What is n_estimators?
    Times asked 2
  46. 46
    Why did you choose n_estimators = 100?
    Times asked 1
  47. 47
    What happens if n_estimators = 2000?
    Times asked 1
  48. 48
    What is max_depth?
    Times asked 1
  49. 49
    What is learning_rate?
    Times asked 1
  50. 50
    What is n_iter?
    Times asked 1
  51. 51
    What would you do if the model overfits because of max_depth?
    Times asked 1
  52. 52
    How do you prevent overfitting?
    Times asked 1
  53. 53
    How do you improve an underfitting model?
    Times asked 1
  54. 54
    What is IQR?
    Times asked 1
  55. 55
    How does outlier detection help?
    Times asked 1
  56. 56
    How does a boxplot help detect outliers?
    Times asked 1
  57. 57
    How does ColumnTransformer work?
    Times asked 1
  58. 58
    Why didn't you use scaling?
    Times asked 2
  59. 59
    Should SGD be used without scaling?
    Times asked 1
  60. 60
    What is SGD?
    Times asked 1
  61. 61
    Why did you use SGD?
    Times asked 1
  62. 62
    Difference between Accuracy, Precision, Recall, and F1-Score.
    Times asked 1
  63. 63
    Why did you use F1-Score?
    Times asked 1
  64. 64
    What is RMSLE?
    Times asked 1
  65. 65
    What is Blended Score?
    Times asked 1
  66. 66
    Why did your RMSLE score decrease after blending?
    Times asked 1
  67. 67
    What is a Decision Tree?
    Times asked 1
  68. 68
    What algorithm do XGBoost, LightGBM, and CatBoost use?
    Times asked 1
  69. 69
    Difference between XGBoost, LightGBM, and CatBoost.
    Times asked 1
  70. 70
    Are these models linear or non-linear?
    Times asked 1
  71. 71
    Explain your complete preprocessing and modeling pipeline.
    Times asked 1
  72. 72
    How does your complete pipeline work?
    Times asked 1
  73. 73
    What are custom classes?
    Times asked 1
  74. 74
    Why did you create custom classes?
    Times asked 1
  75. 75
    Did you use any LLMs while building the project?
    Times asked 1
  76. 76
    Coding: Impute the most frequent categorical value for each column.
    Times asked 1
  77. 77
    Coding: Perform PCA on numerical features and obtain the top two components.
    Times asked 1
  78. 78
    Coding: Perform K-Means clustering on a dataset.
    Times asked 1
  79. 79
    Coding: Load a dataset, preprocess it, perform train-test split, and build a Linear Regression model.
    Times asked 1
  80. 80
    Coding: Implement RandomizedSearchCV.
    Times asked 1
  81. 81
    Coding: Create a complete Pipeline with ColumnTransformer, Imputers, Transformers, and a model.
    Times asked 1
  1. 1
    Perform the checksum verification.
    Times asked 1
    Official solution

    SimpleChecksum file ka hash (jaise SHA256) — file change hui ya nahi verify.

    Portal pe submitted ZIP ka checksum diya hota hai. Local file ka hash nikaalo, match hona chahiye.
    Linux: sha256sum project.zip
    Mismatch = galat file / corrupt download. Examiner integrity check karta hai ki tumhari submitted copy hi run ho rahi hai.

  2. 2
    Demonstrate the core functionalities of the application.
    Times asked 2
    Official solution

    SimpleDemo script pehle se practice karo, 5-7 minute, saari roles.

    Suggested flow

    1) Register/login user 2) Admin login — CRUD 3) Approval/blacklist 4) Booking/apply 5) Search 6) Validation edge (slots 0, duplicate) 7) Logout
    Bolte-bolte dikhao: 'Yahan overbooking rokta hoon'. Code tab kholna jab poochhein.
    Crash ho to debug calmly: terminal error, typo, DB path.
    Mandatory features skip mat karna — examiner list tick karta hai.

  3. 3
    Show and explain your models.
    Times asked 1
    Official solution

    Simplemodels.py har table ki class hai — columns, PK/FK, relationships. Yeh Model layer hai.

    Viva mein file khol ke bolo:
    1) Kaunsi classes/tables hain (User, Role-specific, Booking...)
    2) Har table ka kaam 1 line
    3) Relationships: 1-1 / 1-M / M-M, backref, cascade, lazy
    4) Constraints: unique email, nullable, default status

    Example skeleton

    class User(db.Model):
        id = db.Column(db.Integer, primary_key=True)
        email = db.Column(db.String(120), unique=True, nullable=False)
        role = db.Column(db.String(20), default='user')
        bookings = db.relationship('Booking', backref='user', cascade='all, delete-orphan')

    ER diagram: rectangles entities, lines relationships, PK/FK mark. Schema = yahi structure.
    Tables create: db.create_all() ya migrations. SQLite file SQL viewer se dikha sakte ho.

  4. 4
    Which database did you use in your project?
    Times asked 1
    Official solution

    SimpleSQLite file-based chhota DB, MySQL/Postgres server-based production DB.

    SQLite fayde: zero setup, ek .db file, MAD1/dev ke liye perfect, portable.

    Nuksaanconcurrent writes kam, user/password nahi, network access nahi, production scale nahi.

    MySQL: users, permissions, concurrent, replica. App factory mein URI change karke switch: sqlite:///app.db → mysql://...

Maintaine By Lazy IITians Team + IITM BS students Regualr for More data you have