{
  "total": 74,
  "topics": [
    {
      "id": 369672,
      "title": "Wrapping up Tabular Playground Series 2022",
      "comment_count": 11,
      "views": 0,
      "votes": 30
    },
    {
      "id": 361941,
      "title": "Calling for Feedback on Tabular Playground Series",
      "comment_count": 29,
      "views": 0,
      "votes": 28
    },
    {
      "id": 361940,
      "title": "New to Machine Learning or to Kaggle? Check this out.",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 363778,
      "title": "Probability calibration by isotonic regression",
      "comment_count": 4,
      "views": 0,
      "votes": 25
    },
    {
      "id": 364480,
      "title": "How to evaluate model calibration without graphs",
      "comment_count": 6,
      "views": 0,
      "votes": 9
    },
    {
      "id": 369674,
      "title": "1st Place Solution",
      "comment_count": 21,
      "views": 0,
      "votes": 52
    },
    {
      "id": 363649,
      "title": "Stacking and blending for beginners",
      "comment_count": 11,
      "views": 0,
      "votes": 15
    },
    {
      "id": 370126,
      "title": "3rd place solution",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 369741,
      "title": "MORE activation",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 369731,
      "title": "7th place solution",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 369242,
      "title": "❗️Be careful with final submissions",
      "comment_count": 8,
      "views": 0,
      "votes": 14
    },
    {
      "id": 370006,
      "title": "TPS December 2022 ???",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 369671,
      "title": "The shakeup has happened again!",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 368905,
      "title": "Feature selection: Jensen-Shannon distance",
      "comment_count": 5,
      "views": 0,
      "votes": 7
    },
    {
      "id": 368894,
      "title": "Analysis of blending strategies ",
      "comment_count": 2,
      "views": 0,
      "votes": 9
    },
    {
      "id": 369248,
      "title": "📚 TPS-Nov-22 winning solutions",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 368941,
      "title": "A small idea about the final way to boost your score(with sample code)",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 368090,
      "title": "Viewing the name of the submission file?",
      "comment_count": 2,
      "views": 0,
      "votes": 9
    },
    {
      "id": 367591,
      "title": "Understanding the evaluation metric[LogLoss]",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 364013,
      "title": "The overlooked piece of information",
      "comment_count": 23,
      "views": 0,
      "votes": 54
    },
    {
      "id": 363968,
      "title": "Understanding Ensemble Learning, Blending, and Stacking Generalization",
      "comment_count": 20,
      "views": 0,
      "votes": 56
    },
    {
      "id": 366139,
      "title": "Improving score by increasing and decreasing prediction probabilities by small values",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 367596,
      "title": " TPS November 2022: Beginner friendly compilation",
      "comment_count": 5,
      "views": 0,
      "votes": 11
    },
    {
      "id": 368243,
      "title": "stacking model",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 363683,
      "title": "Collection of helpful ideas - TPS Nov 2022",
      "comment_count": 8,
      "views": 0,
      "votes": 13
    },
    {
      "id": 367512,
      "title": "Dilemma (Chasing two rabbits)",
      "comment_count": 4,
      "views": 0,
      "votes": 8
    },
    {
      "id": 367413,
      "title": "Idea: cluster the correlation matrix, take simple average over each cluster, only then optimize weights",
      "comment_count": 0,
      "views": 0,
      "votes": 14
    },
    {
      "id": 367070,
      "title": "Model embedding",
      "comment_count": 0,
      "views": 0,
      "votes": 14
    },
    {
      "id": 366518,
      "title": "New activation: Growing Cosine Unit",
      "comment_count": 4,
      "views": 0,
      "votes": 10
    },
    {
      "id": 366100,
      "title": "#Folds vs CV score - important note",
      "comment_count": 2,
      "views": 0,
      "votes": 24
    },
    {
      "id": 364074,
      "title": "Tabular NNs in LightAutoML - 0.51522 on LB",
      "comment_count": 11,
      "views": 0,
      "votes": 17
    },
    {
      "id": 365103,
      "title": "Host Solution, private LB on the previous TPS. Just in case you missed it.",
      "comment_count": 4,
      "views": 0,
      "votes": 17
    },
    {
      "id": 366494,
      "title": "Difference between cross-val-score and loop model.",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 366685,
      "title": "Adversarial Validation",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 365849,
      "title": "📌 Feature Engineering Without Overfitting - PCA 🔥🔥🔥",
      "comment_count": 4,
      "views": 0,
      "votes": 11
    },
    {
      "id": 366152,
      "title": "How did I improve my score on the leaderboard, from 0.515 to 0.514 to 0.513",
      "comment_count": 2,
      "views": 0,
      "votes": 11
    },
    {
      "id": 366987,
      "title": "Dataset reduced to the limit",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 365583,
      "title": "PLS-DA and LogisticRegression model to reduce number of variables",
      "comment_count": 6,
      "views": 0,
      "votes": 12
    },
    {
      "id": 364834,
      "title": "OOF Ensembling: Forward Selection",
      "comment_count": 5,
      "views": 0,
      "votes": 21
    },
    {
      "id": 365928,
      "title": "Question about selecting variables based on correlation coeffient",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 364082,
      "title": "There are 7515 ones and 7485 zeros in private LB test set",
      "comment_count": 10,
      "views": 0,
      "votes": 30
    },
    {
      "id": 366388,
      "title": "How to read Kaggle data into local repository",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 366356,
      "title": "clipping values is wrong",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 364931,
      "title": "Additional Features in Submission Files",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 364083,
      "title": "Selecting Best base leaners  ",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 363309,
      "title": "Initial thoughts and references",
      "comment_count": 10,
      "views": 0,
      "votes": 24
    },
    {
      "id": 364357,
      "title": "UMAP Dimensionality Reduction",
      "comment_count": 3,
      "views": 0,
      "votes": 12
    },
    {
      "id": 363488,
      "title": "# Features vs LB score competition",
      "comment_count": 14,
      "views": 0,
      "votes": 13
    },
    {
      "id": 363990,
      "title": "LASSO as Feature Selector",
      "comment_count": 2,
      "views": 0,
      "votes": 9
    },
    {
      "id": 364048,
      "title": "📌 Advanced Dataset - Pseudo Labels 🎭🎭🎭",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 364425,
      "title": "\"Calibrated\" best prediction. logloss 0.51869 ",
      "comment_count": 0,
      "views": 0,
      "votes": 8
    },
    {
      "id": 363937,
      "title": "Ideas for weighting ",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 364154,
      "title": "How to augment the data we have?",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 363381,
      "title": "Great advices/tips from Veterans. Embrace their GMasters Class!",
      "comment_count": 7,
      "views": 0,
      "votes": 28
    },
    {
      "id": 363483,
      "title": "5k features? Let's use feature selection",
      "comment_count": 5,
      "views": 0,
      "votes": 20
    },
    {
      "id": 363763,
      "title": "Don't forget to drop (almost) perfectly correlated predictions",
      "comment_count": 0,
      "views": 0,
      "votes": 13
    },
    {
      "id": 363743,
      "title": "Submission files may have systematic bias",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 363721,
      "title": "Don't trust in log loss to do feature selection!",
      "comment_count": 0,
      "views": 0,
      "votes": 13
    },
    {
      "id": 363308,
      "title": "💉🩺🧬 Nice Resources to refer to for November 2022 TPS🌡📈💢 ",
      "comment_count": 9,
      "views": 0,
      "votes": 16
    },
    {
      "id": 364058,
      "title": "Handling the incorrect values",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 363372,
      "title": "Feather Dataset (Read all at once)",
      "comment_count": 5,
      "views": 0,
      "votes": 13
    },
    {
      "id": 363447,
      "title": "Illustration of concept: The score of the labeled rows matches the file name",
      "comment_count": 7,
      "views": 0,
      "votes": 8
    },
    {
      "id": 363511,
      "title": "Looking for Techniques to Eliminate Features",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 363311,
      "title": "Competition metric- binary log-loss- references and explanations",
      "comment_count": 5,
      "views": 0,
      "votes": 14
    },
    {
      "id": 363703,
      "title": "Is there a better way of building a DF?",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 363300,
      "title": "File 0.7016449929.csv has an additional column",
      "comment_count": 8,
      "views": 0,
      "votes": 14
    },
    {
      "id": 363371,
      "title": "📌 Mean Methods Comparisons 🔥🔥🔥",
      "comment_count": 3,
      "views": 0,
      "votes": 12
    },
    {
      "id": 363334,
      "title": "Outliers can ruin your model!",
      "comment_count": 2,
      "views": 0,
      "votes": 14
    },
    {
      "id": 363564,
      "title": "ROC-AUC for Features Selection",
      "comment_count": 0,
      "views": 0,
      "votes": 8
    },
    {
      "id": 363675,
      "title": "All About Blending and Stacking",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 363672,
      "title": "Correlation Matrix",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 363880,
      "title": "Target permutation & hierarchical clustering to choose features",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 363469,
      "title": "Initial idea on how to avoid overfitting",
      "comment_count": 8,
      "views": 0,
      "votes": 5
    },
    {
      "id": 363988,
      "title": "A simple question!",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    }
  ],
  "errors": []
}