Jovian
Sign In
Learn practical skills, build real-world projects, and advance your career

Table of Contents

  1. Models implementation and evaluation
    1. Implementation
    2. Experiments
    3. Conclusion
  2. Interpret GradientBoostingRegressor model
    1. Global Interpretability
      1. Variable Importance Plot
      2. SHAP Dependence Plot
    2. Local Interpretability
      1. Heatmap plot
      2. Test Case 1
        1. Individual SHAP Value Plot
        2. Waterfall plot
        3. Individual bar plot
        4. Decision plot
      3. Test Case 2
        1. Individual SHAP Value Plot
        2. Waterfall plot
        3. Individual bar plot
        4. Decision plot
      4. Test Case 3
        1. Individual SHAP Value Plot
        2. Waterfall plot
        3. Individual bar plot
        4. Decision plot
  3. Interpret StackingClassifier model
    1. Global Interpretability
    2. Local Interpretability
      1. Test Case 1
      2. Test Case 2
    3. Diverse Counterfactual Explanations
      1. Test Case 1
      2. Test Case 2
    4. Association rule mining
      1. Top rules
      2. Rules related to correct and incorrect predition
      3. Rules related to the cooperation
      4. Conclusion
  4. Conclusion
from google.colab import drive
drive.mount('/content/drive')
Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).

Models implementation and evaluation (to top)

Implementation (to top)

Regarding the suggestion of model implementation, we dropped the features ('popularity', 'vote_average', 'vote_count') that are still not available before a movie is released. Then, we redid the features engineering, data preprocessing and retrained all the models for regression and classification tasks.