{
  "total": 847,
  "topics": [
    {
      "id": 572981,
      "title": "First Competition – Need Feedback on My EDA and Approach",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 324784,
      "title": "What is meant by \"the magic\"?",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87954,
      "title": "How to do EDA?",
      "comment_count": 14,
      "views": 0,
      "votes": 33
    },
    {
      "id": 429188,
      "title": "Model is predicting 0 True Positives if Using PCA",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 89004,
      "title": "The Power of GPU",
      "comment_count": 52,
      "views": 0,
      "votes": 159
    },
    {
      "id": 89003,
      "title": "#1 Solution",
      "comment_count": 123,
      "views": 0,
      "votes": 386
    },
    {
      "id": 88939,
      "title": "2nd place solution",
      "comment_count": 197,
      "views": 0,
      "votes": 343
    },
    {
      "id": 88974,
      "title": "Long journey to 27th place ",
      "comment_count": 27,
      "views": 0,
      "votes": 41
    },
    {
      "id": 312790,
      "title": "Pseudo Labeling with AutoX",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 311105,
      "title": "Santander Customer Transaction Prediction",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84797,
      "title": "Anyone test with Deep Neural Net?",
      "comment_count": 11,
      "views": 0,
      "votes": null
    },
    {
      "id": 290147,
      "title": "automl for Santander, Public Score 0.89327",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 85125,
      "title": "List of Fake Samples and Public/Private LB split",
      "comment_count": 69,
      "views": 0,
      "votes": 185
    },
    {
      "id": 80809,
      "title": "A (general) framework for competitions",
      "comment_count": 36,
      "views": 0,
      "votes": 210
    },
    {
      "id": 90231,
      "title": "5 Things I learned from my first Kaggle Competition",
      "comment_count": 18,
      "views": 0,
      "votes": 32
    },
    {
      "id": 84974,
      "title": "Linear SVM kernel",
      "comment_count": 15,
      "views": 0,
      "votes": 18
    },
    {
      "id": 89320,
      "title": "How do we optimize GBM hyperparameters?",
      "comment_count": 61,
      "views": 0,
      "votes": 67
    },
    {
      "id": 80754,
      "title": "optuna and maximized objective",
      "comment_count": 14,
      "views": 0,
      "votes": 4
    },
    {
      "id": 89012,
      "title": "Tribute to Chris Deotte",
      "comment_count": 41,
      "views": 0,
      "votes": 123
    },
    {
      "id": 87560,
      "title": "Kaggle PSA: These are not probability density functions.",
      "comment_count": 6,
      "views": 0,
      "votes": 9
    },
    {
      "id": 87133,
      "title": "XX Model is capable of XX",
      "comment_count": 37,
      "views": 0,
      "votes": 96
    },
    {
      "id": 88934,
      "title": "Why count is crucial? ",
      "comment_count": 7,
      "views": 0,
      "votes": 7
    },
    {
      "id": 88936,
      "title": "Rapids.ai Solution",
      "comment_count": 22,
      "views": 0,
      "votes": 49
    },
    {
      "id": 163632,
      "title": "Late submission",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 81012,
      "title": "Tired of endless stacks of stacks? Consider going semi-supervised!",
      "comment_count": 25,
      "views": 0,
      "votes": 113
    },
    {
      "id": 160499,
      "title": "How to add team members?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82689,
      "title": "AUC - Explanation + fast Metric for Keras",
      "comment_count": 1,
      "views": 0,
      "votes": 14
    },
    {
      "id": 138037,
      "title": "Can I get in touch with the Competition Host?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 82842,
      "title": "Tanh working better than relu or elu or leakyrelu",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81289,
      "title": "Code Snippet for XGBoost Bayesian Optimization ",
      "comment_count": 2,
      "views": 0,
      "votes": 15
    },
    {
      "id": 128993,
      "title": "Starter",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87846,
      "title": "Let's stop sharing magic",
      "comment_count": 157,
      "views": 0,
      "votes": 134
    },
    {
      "id": 112936,
      "title": "Is this a regression or classification problem",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 80863,
      "title": "Approaches to anonymous feature engineering?",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 80807,
      "title": "Two types of AUC in Keras",
      "comment_count": 13,
      "views": 0,
      "votes": 32
    },
    {
      "id": 117318,
      "title": "data description ",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 87701,
      "title": "Help regarding GPU",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 83791,
      "title": "Error: MemoryError. How use other way for run my model?",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 113957,
      "title": "is there any notebook available for very beginners with models like logistic regression and other basic models",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 88929,
      "title": "5th place partial solution - NN part (Github Updated)",
      "comment_count": 42,
      "views": 0,
      "votes": 82
    },
    {
      "id": 88078,
      "title": "Intuitive description of mathematical objects",
      "comment_count": 13,
      "views": 0,
      "votes": null
    },
    {
      "id": 87618,
      "title": "I am a novice",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 111704,
      "title": "XGBClassifier issue",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 88681,
      "title": "Need help with this LGB error",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81539,
      "title": "I just made my first competition submission!",
      "comment_count": 30,
      "views": 0,
      "votes": 52
    },
    {
      "id": 81189,
      "title": "All you need (more) is bayesian optimization...",
      "comment_count": 7,
      "views": 0,
      "votes": 35
    },
    {
      "id": 88921,
      "title": "What have you learned during this competition?",
      "comment_count": 7,
      "views": 0,
      "votes": 7
    },
    {
      "id": 86736,
      "title": "Everyone is stressed",
      "comment_count": 161,
      "views": 0,
      "votes": 134
    },
    {
      "id": 95688,
      "title": "Why  train the model with 2 features(var_x and frequency_var_x) at a time works?",
      "comment_count": 7,
      "views": 0,
      "votes": 4
    },
    {
      "id": 86716,
      "title": "Bayesian Optimization(Find HyperParams)",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 89034,
      "title": "29th place solution with code",
      "comment_count": 31,
      "views": 0,
      "votes": 99
    },
    {
      "id": 88926,
      "title": "🏅Gold Medal Solutions 🏅",
      "comment_count": 26,
      "views": 0,
      "votes": 99
    },
    {
      "id": 89009,
      "title": "NN as part of 13th place solution (late joiners view)",
      "comment_count": 13,
      "views": 0,
      "votes": 38
    },
    {
      "id": 83786,
      "title": "autumn4577 and autumn45777",
      "comment_count": 90,
      "views": 0,
      "votes": 74
    },
    {
      "id": 85338,
      "title": "Row-wise features engineering",
      "comment_count": 5,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88888,
      "title": "Public LB 0.922 magic notebook",
      "comment_count": 58,
      "views": 0,
      "votes": 179
    },
    {
      "id": 91337,
      "title": "competition disappeared?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 91045,
      "title": "Help! >0.92 in train, dev-train, validate dataset, but 0.68 in Test dataset",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 90088,
      "title": "Question about sm.logit and sklearn LR for Ensemble",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 89515,
      "title": "Final Results?",
      "comment_count": 101,
      "views": 0,
      "votes": 30
    },
    {
      "id": 89692,
      "title": "Finally GM",
      "comment_count": 44,
      "views": 0,
      "votes": 72
    },
    {
      "id": 84632,
      "title": "Top public kernels comparison",
      "comment_count": 53,
      "views": 0,
      "votes": 228
    },
    {
      "id": 89562,
      "title": "Can someone sum up what is magic ?",
      "comment_count": 19,
      "views": 0,
      "votes": 8
    },
    {
      "id": 89678,
      "title": "Congratulations to @30CrMnSiA for Grandmaster status!",
      "comment_count": 18,
      "views": 0,
      "votes": 24
    },
    {
      "id": 88986,
      "title": "My silver medal will disappear ",
      "comment_count": 60,
      "views": 0,
      "votes": 39
    },
    {
      "id": 88897,
      "title": "5TH Place Solution",
      "comment_count": 33,
      "views": 0,
      "votes": 148
    },
    {
      "id": 89023,
      "title": "7th place The Zoo",
      "comment_count": 21,
      "views": 0,
      "votes": 67
    },
    {
      "id": 89677,
      "title": "Result seems finalized, check your profile!",
      "comment_count": 11,
      "views": 0,
      "votes": 10
    },
    {
      "id": 89541,
      "title": "One important Question",
      "comment_count": 13,
      "views": 0,
      "votes": 13
    },
    {
      "id": 89606,
      "title": "Career Path",
      "comment_count": 5,
      "views": 0,
      "votes": 6
    },
    {
      "id": 88968,
      "title": "Congratulations and Thank You to YaG320",
      "comment_count": 24,
      "views": 0,
      "votes": 84
    },
    {
      "id": 89421,
      "title": "HOW did you find magic features?",
      "comment_count": 14,
      "views": 0,
      "votes": 36
    },
    {
      "id": 90017,
      "title": "Did you got re-introduced :)",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 89687,
      "title": "Finally I got my medal! Congratulations!",
      "comment_count": 13,
      "views": 0,
      "votes": 16
    },
    {
      "id": 89371,
      "title": "The Difference that Santander Makes",
      "comment_count": 12,
      "views": 0,
      "votes": 29
    },
    {
      "id": 88944,
      "title": "its not effin magic",
      "comment_count": 144,
      "views": 0,
      "votes": 60
    },
    {
      "id": 89912,
      "title": "The magic is indeed very enlightening",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 88950,
      "title": "A OK solution without peeking test data --  we are hiring talented:)",
      "comment_count": 11,
      "views": 0,
      "votes": 15
    },
    {
      "id": 89059,
      "title": "How you learned Data Science?",
      "comment_count": 8,
      "views": 0,
      "votes": 5
    },
    {
      "id": 89769,
      "title": "Kaggle Competition Points Calculation",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 89007,
      "title": "#12 My exploration on Santander",
      "comment_count": 15,
      "views": 0,
      "votes": 28
    },
    {
      "id": 89070,
      "title": "20th solution",
      "comment_count": 20,
      "views": 0,
      "votes": 19
    },
    {
      "id": 88632,
      "title": "What a finish",
      "comment_count": 78,
      "views": 0,
      "votes": 34
    },
    {
      "id": 89083,
      "title": "#13 Place - Solution and Takeaways",
      "comment_count": 5,
      "views": 0,
      "votes": 23
    },
    {
      "id": 89914,
      "title": "Public and Private LB generators",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 89302,
      "title": "9th place solution (nagiss part)",
      "comment_count": 9,
      "views": 0,
      "votes": 30
    },
    {
      "id": 88913,
      "title": "Really easy jump to LB 0.922",
      "comment_count": 27,
      "views": 0,
      "votes": 44
    },
    {
      "id": 89117,
      "title": "A summary of #18's solution",
      "comment_count": 6,
      "views": 0,
      "votes": 18
    },
    {
      "id": 88953,
      "title": "52nd Solution: \"Is Var_108 a Date?\"",
      "comment_count": 32,
      "views": 0,
      "votes": 69
    },
    {
      "id": 88940,
      "title": "Another ruined Santander competition?",
      "comment_count": 33,
      "views": 0,
      "votes": 35
    },
    {
      "id": 89207,
      "title": "The real winner of this competition is yag320",
      "comment_count": 8,
      "views": 0,
      "votes": 23
    },
    {
      "id": 89255,
      "title": "My Guidelines and tips for teaming up the right way",
      "comment_count": 7,
      "views": 0,
      "votes": 27
    },
    {
      "id": 88946,
      "title": "Why the magic works- is each raw variable actually a mix of two?",
      "comment_count": 25,
      "views": 0,
      "votes": 36
    },
    {
      "id": 88997,
      "title": "10th place solution",
      "comment_count": 5,
      "views": 0,
      "votes": 11
    },
    {
      "id": 88970,
      "title": "part of #4 solution",
      "comment_count": 34,
      "views": 0,
      "votes": 47
    },
    {
      "id": 89168,
      "title": "How did Santander create 200 uncorrelated variables?",
      "comment_count": 13,
      "views": 0,
      "votes": 6
    },
    {
      "id": 89114,
      "title": "Answer to \"Leak vs Magic\" debate",
      "comment_count": 69,
      "views": 0,
      "votes": 10
    },
    {
      "id": 89702,
      "title": "Disqualified. Why?",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 89681,
      "title": "Why was participants finally decreased？",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 89711,
      "title": "My ranking is gone",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 89078,
      "title": "Kaggle addiction",
      "comment_count": 24,
      "views": 0,
      "votes": 17
    },
    {
      "id": 89015,
      "title": "Many thanks to YaG320",
      "comment_count": 11,
      "views": 0,
      "votes": 32
    },
    {
      "id": 88891,
      "title": "Magic feature and other findings",
      "comment_count": 19,
      "views": 0,
      "votes": 51
    },
    {
      "id": 88990,
      "title": "Congratulations to the winners and to the ones that found the magic!",
      "comment_count": 4,
      "views": 0,
      "votes": 7
    },
    {
      "id": 88902,
      "title": "3rd Place Solution Summary and Q&A ",
      "comment_count": 18,
      "views": 0,
      "votes": 40
    },
    {
      "id": 89506,
      "title": "Why creating duplicated new features can improve",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 88907,
      "title": "Congratulations to all Santander Winning Magicians.",
      "comment_count": 8,
      "views": 0,
      "votes": 21
    },
    {
      "id": 89767,
      "title": "Pixie Cull",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 89112,
      "title": "#19 solution",
      "comment_count": 10,
      "views": 0,
      "votes": 14
    },
    {
      "id": 88927,
      "title": "14th solution and magic feature(double peak)",
      "comment_count": 3,
      "views": 0,
      "votes": 21
    },
    {
      "id": 81190,
      "title": "90% accuracy",
      "comment_count": 36,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88904,
      "title": "Simple 5 line code transformation gives 0.922",
      "comment_count": 14,
      "views": 0,
      "votes": 32
    },
    {
      "id": 88623,
      "title": "0.910",
      "comment_count": 16,
      "views": 0,
      "votes": 21
    },
    {
      "id": 86768,
      "title": "where should we novices go?",
      "comment_count": 36,
      "views": 0,
      "votes": 17
    },
    {
      "id": 88794,
      "title": "Any ideas 924s?",
      "comment_count": 34,
      "views": 0,
      "votes": 26
    },
    {
      "id": 89739,
      "title": "Congratulations to all !!",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88886,
      "title": "#8 team solution",
      "comment_count": 2,
      "views": 0,
      "votes": 43
    },
    {
      "id": 89016,
      "title": "Thanks to Branden Murray",
      "comment_count": 7,
      "views": 0,
      "votes": 26
    },
    {
      "id": 89145,
      "title": "If it's not a magic, could you explain why it works? ",
      "comment_count": 5,
      "views": 0,
      "votes": 5
    },
    {
      "id": 89142,
      "title": "When can the cheater removal be finished?",
      "comment_count": 11,
      "views": 0,
      "votes": 19
    },
    {
      "id": 88382,
      "title": "Any ideas 0.901's?",
      "comment_count": 96,
      "views": 0,
      "votes": 9
    },
    {
      "id": 89111,
      "title": "Is Data Scientists job really overhyped?",
      "comment_count": 11,
      "views": 0,
      "votes": 5
    },
    {
      "id": 88981,
      "title": "Applause for YaG320",
      "comment_count": 3,
      "views": 0,
      "votes": 23
    },
    {
      "id": 88884,
      "title": "Congratulations and kernel with LB 0.922!",
      "comment_count": 3,
      "views": 0,
      "votes": 23
    },
    {
      "id": 88943,
      "title": "Best graduation gift(#24 solution to consider magic)",
      "comment_count": 10,
      "views": 0,
      "votes": 28
    },
    {
      "id": 89187,
      "title": "NN question input shape ",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 89275,
      "title": "Never give up ! (157th)",
      "comment_count": 4,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88956,
      "title": "30th solution",
      "comment_count": 12,
      "views": 0,
      "votes": 21
    },
    {
      "id": 88906,
      "title": "12th solution and magic ",
      "comment_count": 3,
      "views": 0,
      "votes": 29
    },
    {
      "id": 88857,
      "title": "You couldn't have escaped Santander tonight",
      "comment_count": 2,
      "views": 0,
      "votes": 31
    },
    {
      "id": 88826,
      "title": "Champions league vs Kaggle",
      "comment_count": 18,
      "views": 0,
      "votes": 20
    },
    {
      "id": 89537,
      "title": "How did you do EDA in this game?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88861,
      "title": "Fingers Crossed But Shout out to CPMP and Chris",
      "comment_count": 6,
      "views": 0,
      "votes": 23
    },
    {
      "id": 88585,
      "title": "To those who share magic/tricks without evidence",
      "comment_count": 21,
      "views": 0,
      "votes": -1
    },
    {
      "id": 88230,
      "title": "Found the magic :D LB 0.909",
      "comment_count": 51,
      "views": 0,
      "votes": 65
    },
    {
      "id": 88937,
      "title": "How did I find the first FE (unique value counts) in the last day",
      "comment_count": 12,
      "views": 0,
      "votes": 14
    },
    {
      "id": 88479,
      "title": "My advice in the last day",
      "comment_count": 34,
      "views": 0,
      "votes": 30
    },
    {
      "id": 88895,
      "title": "Simple manipulation to achieve 0.922",
      "comment_count": 9,
      "views": 0,
      "votes": 21
    },
    {
      "id": 89504,
      "title": "Data Imbalance.",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 89011,
      "title": "Some useful hints",
      "comment_count": 1,
      "views": 0,
      "votes": 14
    },
    {
      "id": 88923,
      "title": "Could we know a little bit background story from Santander",
      "comment_count": 2,
      "views": 0,
      "votes": 20
    },
    {
      "id": 87815,
      "title": "Tips for Breaking .901",
      "comment_count": 59,
      "views": 0,
      "votes": 162
    },
    {
      "id": 89233,
      "title": "First silver medal? Still waiting !",
      "comment_count": 7,
      "views": 0,
      "votes": 4
    },
    {
      "id": 87486,
      "title": "Magic Hint? : What M.L \"can't see\"",
      "comment_count": 50,
      "views": 0,
      "votes": 120
    },
    {
      "id": 88523,
      "title": "Make plan for system outage before deadline",
      "comment_count": 7,
      "views": 0,
      "votes": 23
    },
    {
      "id": 88318,
      "title": "Insanity",
      "comment_count": 26,
      "views": 0,
      "votes": 48
    },
    {
      "id": 88909,
      "title": "Collating Magic Solutions",
      "comment_count": 3,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88967,
      "title": "#38 solution",
      "comment_count": 12,
      "views": 0,
      "votes": 18
    },
    {
      "id": 89002,
      "title": "About traps of finding the magic",
      "comment_count": 10,
      "views": 0,
      "votes": 16
    },
    {
      "id": 89361,
      "title": "A Review Of Synthetic Data",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 89050,
      "title": "Congratulation",
      "comment_count": 4,
      "views": 0,
      "votes": 16
    },
    {
      "id": 88675,
      "title": "Shake up is coming?",
      "comment_count": 17,
      "views": 0,
      "votes": 18
    },
    {
      "id": 89094,
      "title": "How was your teaming up experience",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 89344,
      "title": "Why my solution did not work?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88339,
      "title": "Don't give up, the MAGIC is on your side :)",
      "comment_count": 29,
      "views": 0,
      "votes": 20
    },
    {
      "id": 88800,
      "title": "This competition is so challenging",
      "comment_count": 7,
      "views": 0,
      "votes": 11
    },
    {
      "id": 88073,
      "title": "Santander's (likely) motivation",
      "comment_count": 8,
      "views": 0,
      "votes": 20
    },
    {
      "id": 88922,
      "title": "Simple count feature + aug can reach 0.920",
      "comment_count": 4,
      "views": 0,
      "votes": 16
    },
    {
      "id": 88737,
      "title": "Model tune can utilize the magic",
      "comment_count": 26,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88602,
      "title": "The top magicians with 0.927 is so quiet...",
      "comment_count": 25,
      "views": 0,
      "votes": 11
    },
    {
      "id": 88394,
      "title": "Can I really get a job easily If I get a gold on Kaggle?",
      "comment_count": 17,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88938,
      "title": "My Story-140th place",
      "comment_count": 5,
      "views": 0,
      "votes": 9
    },
    {
      "id": 89090,
      "title": "Kernel work smart > work hard [LB 0.920]",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 88343,
      "title": "What about second magic?",
      "comment_count": 11,
      "views": 0,
      "votes": 19
    },
    {
      "id": 88863,
      "title": "Most shared \"magic\" in this competition",
      "comment_count": 4,
      "views": 0,
      "votes": 12
    },
    {
      "id": 88919,
      "title": "Amazing magic, But why it works",
      "comment_count": 14,
      "views": 0,
      "votes": 5
    },
    {
      "id": 88758,
      "title": "How many submission you have left?",
      "comment_count": 21,
      "views": 0,
      "votes": 3
    },
    {
      "id": 89013,
      "title": "Is it really \"correct\" to include information from the test data for training?",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 89025,
      "title": "Why there were fake rows in test data?",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88978,
      "title": "Learned \"magic\" after reading all posts from top teams",
      "comment_count": 4,
      "views": 0,
      "votes": 14
    },
    {
      "id": 88849,
      "title": "9000 teams, new history record of Kaggle",
      "comment_count": 5,
      "views": 0,
      "votes": 6
    },
    {
      "id": 88755,
      "title": "At last I find some so-called 'magic'",
      "comment_count": 11,
      "views": 0,
      "votes": 12
    },
    {
      "id": 88893,
      "title": "105th place submission :)",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 88640,
      "title": "Is Kaggle server broken? I can't access my kernels.",
      "comment_count": 22,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88980,
      "title": "Magic FE in a nutshell",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88847,
      "title": "2 more people to break 9k barrier",
      "comment_count": 8,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88475,
      "title": "This competition's winners should write a textbook on ML",
      "comment_count": 10,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88858,
      "title": "Very talanted and gifted people",
      "comment_count": 5,
      "views": 0,
      "votes": 9
    },
    {
      "id": 87395,
      "title": "Maybe memes are the solution?",
      "comment_count": 15,
      "views": 0,
      "votes": 25
    },
    {
      "id": 88246,
      "title": "How to win, even if you lose...",
      "comment_count": 1,
      "views": 0,
      "votes": 49
    },
    {
      "id": 88819,
      "title": "What do we have to do after the game is over?",
      "comment_count": 14,
      "views": 0,
      "votes": null
    },
    {
      "id": 88842,
      "title": "EDA: The Most Influencial Feature?",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 89234,
      "title": "Grouping client",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87912,
      "title": "Final discussion on failed FE",
      "comment_count": 56,
      "views": 0,
      "votes": 22
    },
    {
      "id": 89224,
      "title": "Competition fun",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88969,
      "title": "How will Santander benefit from this competition",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88942,
      "title": "How EDA helped you in FE",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88975,
      "title": "Withdrawal Symptoms",
      "comment_count": 5,
      "views": 0,
      "votes": 6
    },
    {
      "id": 88327,
      "title": "Did you give up or will you search a magic till the end?",
      "comment_count": 28,
      "views": 0,
      "votes": 9
    },
    {
      "id": 88811,
      "title": "Quit with honor",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 89045,
      "title": "Have we found all the data properties so far?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88885,
      "title": "It's over",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 87843,
      "title": "mean encoding tutorials of coursera",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88682,
      "title": "Those who can't kill me will eventually make me lose hair.",
      "comment_count": 10,
      "views": 0,
      "votes": 7
    },
    {
      "id": 88889,
      "title": "Lightning Guide to .908+: Magical Counting",
      "comment_count": 1,
      "views": 0,
      "votes": 9
    },
    {
      "id": 89019,
      "title": "Hyperparameter optimization",
      "comment_count": 7,
      "views": 0,
      "votes": 5
    },
    {
      "id": 88901,
      "title": "# 21 Magics",
      "comment_count": 1,
      "views": 0,
      "votes": 17
    },
    {
      "id": 89222,
      "title": "Why neural network work in this problem?",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 88333,
      "title": "This competition is a perfect illustration of  \"The old man and the sea\"",
      "comment_count": 8,
      "views": 0,
      "votes": 15
    },
    {
      "id": 89057,
      "title": "[Shake up report] - Drop of 8364 places...is this the biggest drop in LB placing ever",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88877,
      "title": "Power of team work",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88434,
      "title": "I've realized that I am not a magician but a warrior.",
      "comment_count": 10,
      "views": 0,
      "votes": 18
    },
    {
      "id": 88815,
      "title": "Late to the party...",
      "comment_count": 5,
      "views": 0,
      "votes": 6
    },
    {
      "id": 88830,
      "title": "Howcome the teams are increasing after merger deadline.",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88914,
      "title": "Private LB 0.92156 -- 2.5 magics",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 88962,
      "title": "One tiny framework for Kaggle competitions",
      "comment_count": 1,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88807,
      "title": "This competition was a long journey. ",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 88915,
      "title": "Magic kernel from 26th ",
      "comment_count": 1,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88631,
      "title": "The Cat Returns",
      "comment_count": 8,
      "views": 0,
      "votes": 8
    },
    {
      "id": 88792,
      "title": "Thanks for the competition",
      "comment_count": 1,
      "views": 0,
      "votes": 15
    },
    {
      "id": 88837,
      "title": "How much does ensemble help?",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88743,
      "title": "Last day of madness",
      "comment_count": 7,
      "views": 0,
      "votes": 9
    },
    {
      "id": 88928,
      "title": "just fix var_x value can give you 0.916lb.[only 200 dim]",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 88866,
      "title": "Collective write up",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 88581,
      "title": "How long does it take to jump from 0.910 to 0.920?",
      "comment_count": 12,
      "views": 0,
      "votes": 6
    },
    {
      "id": 87115,
      "title": "Hints for late joiners",
      "comment_count": 36,
      "views": 0,
      "votes": 23
    },
    {
      "id": 89191,
      "title": "Doubt",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88879,
      "title": "My team's leader is disconnected",
      "comment_count": 14,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88874,
      "title": "What a competition",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88854,
      "title": "shake-up",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88841,
      "title": "More hype about magic than GOT season 8",
      "comment_count": 3,
      "views": 0,
      "votes": 9
    },
    {
      "id": 88806,
      "title": "Is the 'magic' a transformation of value frequency?",
      "comment_count": 11,
      "views": 0,
      "votes": -3
    },
    {
      "id": 88881,
      "title": "To admin, just to confirm, the competition is ending in 30 min?",
      "comment_count": 9,
      "views": 0,
      "votes": null
    },
    {
      "id": 88637,
      "title": "Is it Okay ?",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88329,
      "title": "A simple (silly) question about overfit/ underfit",
      "comment_count": 18,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88838,
      "title": "May break the .901, but ran out of submission",
      "comment_count": 8,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88896,
      "title": "final sub  in 3 seconds, get 92club tho!!!!!!!!!!!",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 87952,
      "title": "4 things I will remember from this competition ",
      "comment_count": 23,
      "views": 0,
      "votes": 33
    },
    {
      "id": 88620,
      "title": "Hopefully there will be more than 9000 teams",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 88875,
      "title": "Finally give up, thank you all",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 89062,
      "title": "Token of Gratitude (109th Place Solution)",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 88563,
      "title": "What is your CV?",
      "comment_count": 21,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88168,
      "title": "Tipps for selecting the final submission",
      "comment_count": 16,
      "views": 0,
      "votes": 12
    },
    {
      "id": 88916,
      "title": "List of FE methods",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 88984,
      "title": "The Real Magic",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 89065,
      "title": "Thanks to all of you!!!",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 89060,
      "title": "Thank U, Next!",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88873,
      "title": "When will the final results be in?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88560,
      "title": "The deadline \"2 days to go\" is not precise enough",
      "comment_count": 5,
      "views": 0,
      "votes": 8
    },
    {
      "id": 88583,
      "title": "Any ideas, 0.910s ?",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 89088,
      "title": "Will the results of this private Leaderboard change?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88613,
      "title": "You act like I'm a wizard or something",
      "comment_count": 1,
      "views": 0,
      "votes": 14
    },
    {
      "id": 89017,
      "title": "What an Amazing Competition",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88297,
      "title": "A magic-hunting music playlist...",
      "comment_count": 18,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88670,
      "title": "DIFFERENT COUNT PLOTS TRAN VS TEST",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88871,
      "title": "Capturing top-50 rankings before the LB shakeup",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 88305,
      "title": "Question to all Grandmasters and Uncles out there",
      "comment_count": 10,
      "views": 0,
      "votes": 12
    },
    {
      "id": 88344,
      "title": "Is it possible that Private LB will be completely different ranking than Public LB?",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 88278,
      "title": "50% of features are not informative",
      "comment_count": 9,
      "views": 0,
      "votes": 10
    },
    {
      "id": 89109,
      "title": "Jumped way down",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 88947,
      "title": "Congratulations to the winners !!",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 89038,
      "title": "Awsomely close  ...",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88925,
      "title": "What is the meaning of fake data? How did you find this data?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88930,
      "title": "Share you journey - how did you find the magic?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88932,
      "title": "Congratulations",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88734,
      "title": "I still have hope",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88903,
      "title": "What does it mean for Santander",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88892,
      "title": "Let's start sharing magic",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 89073,
      "title": "fastai tabular model stuck at 0.86 for private lb",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88880,
      "title": "Over 9000",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87792,
      "title": "Team up? 0.917LB looking for 0.915+.",
      "comment_count": 14,
      "views": 0,
      "votes": 6
    },
    {
      "id": 88870,
      "title": "Final selections and hindsight bias",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88869,
      "title": "I wish all kagglers the fair final without shakeup)",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88459,
      "title": "Thursday will be Christmas",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 89046,
      "title": "1st competition -> 1st silver . . . almost :/",
      "comment_count": 0,
      "views": 0,
      "votes": -1
    },
    {
      "id": 88851,
      "title": "FINALLY - .911",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88852,
      "title": "Giving up",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88733,
      "title": "Can we use PCA dimension reduced variables for LGBM?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88480,
      "title": "A reminder for all magicians",
      "comment_count": 14,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88825,
      "title": "See you in the next one!",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 89005,
      "title": "How far have you been from finding leak/magic?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 87920,
      "title": "Not Magic, But Motivation Is All We Need",
      "comment_count": 12,
      "views": 0,
      "votes": 39
    },
    {
      "id": 88622,
      "title": "Do you want magic?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88985,
      "title": "How do you go about finding fake data?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 88973,
      "title": "Congratulations to everyone",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88972,
      "title": "Is there any advices for choosing the submission for the private LB",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88797,
      "title": "Deadline of Santander competition coincides with Kaggledays @ San Francisco ",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88635,
      "title": "Looks like magic, but...",
      "comment_count": 11,
      "views": 0,
      "votes": null
    },
    {
      "id": 88966,
      "title": "Plots for unique counting # VS samples from Normal distribution",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88514,
      "title": "Sacrifices for the Santander: Whatever it takes",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88961,
      "title": "Feature engineering that did not worked",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88960,
      "title": "Model that overfitted the data",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88423,
      "title": "cv:0.907 LB0.882 what happened????",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 88084,
      "title": "Awesome thing on Interpretability",
      "comment_count": 8,
      "views": 0,
      "votes": 15
    },
    {
      "id": 88578,
      "title": "Two ways to utilize the magic?",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88931,
      "title": "\"boosting algorithms like LightGBM are not very good in detecting the frequencies of the features\"",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88933,
      "title": "I feel lost... I think I need to refresh...",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88920,
      "title": "Not be clear about how the PB results is generated ",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88289,
      "title": "this compete really makes ME despairing ",
      "comment_count": 8,
      "views": 0,
      "votes": 9
    },
    {
      "id": 88908,
      "title": "Can someone help me on my code to calculate the unique count of each values ",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 88910,
      "title": "Variety in top kernels",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88898,
      "title": "Muggle vs Wizard (Over 9000 teams ! )",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88506,
      "title": "Just never give up.  Not sure is the magic. ",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 88660,
      "title": "It's Deadline Day and Kaggle is offline",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88634,
      "title": "Choosing best submission with kfolds variance",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87546,
      "title": "Enough of this Magic! Sorry top LB holders, I am revealing it right away.",
      "comment_count": 20,
      "views": 0,
      "votes": 106
    },
    {
      "id": 88477,
      "title": "Less is more, more is less",
      "comment_count": 10,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88071,
      "title": "What is your \"build time\"?",
      "comment_count": 12,
      "views": 0,
      "votes": 10
    },
    {
      "id": 88829,
      "title": "How does final submission work",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 85360,
      "title": "Fake Training Data?",
      "comment_count": 47,
      "views": 0,
      "votes": 87
    },
    {
      "id": 88580,
      "title": "Question about santander's data transformation",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88817,
      "title": "Choice of error",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88049,
      "title": "Did I find the magic?",
      "comment_count": 49,
      "views": 0,
      "votes": 8
    },
    {
      "id": 88663,
      "title": "How many times we can submit until the deadline?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 88764,
      "title": "When will the magic be revealed?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88210,
      "title": "Variable Interactions as a result of 400 seperate models",
      "comment_count": 7,
      "views": 0,
      "votes": 11
    },
    {
      "id": 88535,
      "title": "can't select submissions",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87057,
      "title": "Collating Magic",
      "comment_count": 46,
      "views": 0,
      "votes": 134
    },
    {
      "id": 88509,
      "title": "Found it; however at this stage we should be carful",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 88407,
      "title": "Is magic has been harnessed optimally ?",
      "comment_count": 19,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88686,
      "title": "CV-0.912，but...",
      "comment_count": 8,
      "views": 0,
      "votes": null
    },
    {
      "id": 88707,
      "title": "How do i make my final submission?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 87128,
      "title": "In Search of Weirdness",
      "comment_count": 28,
      "views": 0,
      "votes": 88
    },
    {
      "id": 88697,
      "title": "Model ensembling starter code with mlens superlearner",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88183,
      "title": "What Competition you will do next?",
      "comment_count": 16,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87869,
      "title": "One thing I learned is..Never start by forking the best Kernel.",
      "comment_count": 15,
      "views": 0,
      "votes": 12
    },
    {
      "id": 88657,
      "title": "Really mismatched metrics, CV 0.902 , LB 0.897 and Train AUC at 0.97ish",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88227,
      "title": "Some thoughts",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 85065,
      "title": "Dexter's Lab is back!",
      "comment_count": 124,
      "views": 0,
      "votes": 48
    },
    {
      "id": 88406,
      "title": "What should be the gaps between train AUC and valid AUC",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88780,
      "title": "magic",
      "comment_count": 9,
      "views": 0,
      "votes": -20
    },
    {
      "id": 87476,
      "title": "This is a game about finding ONE PIECE",
      "comment_count": 17,
      "views": 0,
      "votes": 32
    },
    {
      "id": 88796,
      "title": "Why team number still is increasing?",
      "comment_count": 2,
      "views": 0,
      "votes": -3
    },
    {
      "id": 88057,
      "title": "Team The ZOO - Truly Inspiring.",
      "comment_count": 7,
      "views": 0,
      "votes": 14
    },
    {
      "id": 88574,
      "title": "What LGBM does not see? ",
      "comment_count": 5,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88321,
      "title": "CV auc wayyy different from LB auc",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88484,
      "title": "Only two days .....",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87304,
      "title": "Visual Features Correlation",
      "comment_count": 18,
      "views": 0,
      "votes": 23
    },
    {
      "id": 88562,
      "title": "Cross validation and a target leak",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 88266,
      "title": "Can anyone tell me where scores come from?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88517,
      "title": "the gap between train and valid",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88135,
      "title": "Here is why your model is overfitting/not making progress",
      "comment_count": 3,
      "views": 0,
      "votes": 11
    },
    {
      "id": 88493,
      "title": "There are two days left before the end of the game",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 88251,
      "title": "Maybe I need help",
      "comment_count": 13,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88277,
      "title": "This comp is sooooooo hard!",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88428,
      "title": "Current validation scheme can trick you ?",
      "comment_count": 2,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87320,
      "title": "My latest idea on weirdness - unique counts vs simulated",
      "comment_count": 25,
      "views": 0,
      "votes": 32
    },
    {
      "id": 88381,
      "title": "Why overfitting in LB?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 88255,
      "title": "Using public score to predict final outcome",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 88369,
      "title": "Have no idea of blending...I don't know why",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87819,
      "title": "Whats your training score?",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 86776,
      "title": "Is this magic ?",
      "comment_count": 8,
      "views": 0,
      "votes": 12
    },
    {
      "id": 87767,
      "title": "Maximum Kernel running time?",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 88053,
      "title": "Is .901 score a lottery / where will we finish in private leaderboard?",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88342,
      "title": "Can I run bayesian optimization with part of my CPU?",
      "comment_count": 3,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87930,
      "title": "Explaining feature engineering ",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 85426,
      "title": "Data clusters",
      "comment_count": 48,
      "views": 0,
      "votes": 94
    },
    {
      "id": 88072,
      "title": "Target Encoding",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 88232,
      "title": "Is training AUC 0.993 normal for this competition?",
      "comment_count": 12,
      "views": 0,
      "votes": null
    },
    {
      "id": 88122,
      "title": "My magicbook took the magic away...",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88211,
      "title": "Kagglers still moving up the leaderboard when i had already given up",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 88270,
      "title": "A really good approach for the competition , explained",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87916,
      "title": "Reason ranking top in Permutation Importance, but no CV increased",
      "comment_count": 15,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87557,
      "title": "Searching for kernel",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 88167,
      "title": "Best single Model",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87699,
      "title": "Is this a part of the 'magic'?",
      "comment_count": 7,
      "views": 0,
      "votes": 9
    },
    {
      "id": 87446,
      "title": "Team up?",
      "comment_count": 27,
      "views": 0,
      "votes": 8
    },
    {
      "id": 88006,
      "title": "What only Santander can teach me",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 88170,
      "title": "How to sync your CV and LB",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 86912,
      "title": "A part of the magic?",
      "comment_count": 8,
      "views": 0,
      "votes": 28
    },
    {
      "id": 88253,
      "title": "Big Magic on Train and empty on Test ",
      "comment_count": 6,
      "views": 0,
      "votes": -9
    },
    {
      "id": 83848,
      "title": "var_12 and var_81",
      "comment_count": 50,
      "views": 0,
      "votes": 28
    },
    {
      "id": 87319,
      "title": "Why none of the participants are trying CatBoost or XGBoost anymore?",
      "comment_count": 15,
      "views": 0,
      "votes": null
    },
    {
      "id": 88144,
      "title": "Discussion upvotes: how long till the medal?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87780,
      "title": "One Quarter Magic Club",
      "comment_count": 10,
      "views": 0,
      "votes": 5
    },
    {
      "id": 87785,
      "title": "how to tune lightgbm after adding features?",
      "comment_count": 8,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87518,
      "title": "KDE Classifier",
      "comment_count": 5,
      "views": 0,
      "votes": 12
    },
    {
      "id": 86958,
      "title": "How many hours have you invested in this competition?",
      "comment_count": 31,
      "views": 0,
      "votes": 10
    },
    {
      "id": 87597,
      "title": "Need more teams to compensate the climbing LB",
      "comment_count": 12,
      "views": 0,
      "votes": 5
    },
    {
      "id": 87430,
      "title": "Do you upvote kernels/discussion posts that help you to improve your score?",
      "comment_count": 21,
      "views": 0,
      "votes": 12
    },
    {
      "id": 87413,
      "title": "Competition frustration (and hopefully how to deal with it)",
      "comment_count": 2,
      "views": 0,
      "votes": 20
    },
    {
      "id": 87821,
      "title": "How many times have you reread discussion?",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87016,
      "title": "Chris's hints may be helpful",
      "comment_count": 41,
      "views": 0,
      "votes": 29
    },
    {
      "id": 88056,
      "title": "After merge, it's like all you best friends get married...",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87839,
      "title": "Question about LGBM overfitting",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 87847,
      "title": "Want to team up - My only differential is handling Keras very well.",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87998,
      "title": "Association analysis reveals some patterns",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 87679,
      "title": "Kernel: how long till the gold medal?",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 87989,
      "title": "Why do I do a lot of nn results are still not very good?",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 88017,
      "title": "Neural net approach with fast.ai fails at testing",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87988,
      "title": "Std feature importance very high",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 87941,
      "title": "Difference between training mean AUC and local CV",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87720,
      "title": "something seems wrong",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87807,
      "title": "Keep teaming up",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87917,
      "title": "CV - 0.923 but LB - 0.884 ..Why!??",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87596,
      "title": "Is this Normal !!",
      "comment_count": 11,
      "views": 0,
      "votes": 4
    },
    {
      "id": 87330,
      "title": "Stack War",
      "comment_count": 21,
      "views": 0,
      "votes": 16
    },
    {
      "id": 87781,
      "title": "I'd like to team up BUT...",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 86950,
      "title": "Data anonymization 101",
      "comment_count": 9,
      "views": 0,
      "votes": 42
    },
    {
      "id": 87810,
      "title": "Looking for team mates",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 87924,
      "title": "Why stop sharing the magic?",
      "comment_count": 35,
      "views": 0,
      "votes": -41
    },
    {
      "id": 87608,
      "title": "Let's merge to find the magic!",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87861,
      "title": "Submit numbers",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87768,
      "title": "Looking for Team Up",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 87370,
      "title": "The submission should be binary or probability",
      "comment_count": 5,
      "views": 0,
      "votes": 14
    },
    {
      "id": 87196,
      "title": "StackNet",
      "comment_count": 11,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87625,
      "title": "Magic is there indeed, all the hints are here in discussion but not the process.",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87676,
      "title": "high CV(0.94690)， but low LB(0.770)",
      "comment_count": 14,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87585,
      "title": "Looks like Santander will take over Titanic soon",
      "comment_count": 5,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87687,
      "title": "Looking to team up",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87799,
      "title": "Did anyone tried coloring transform ?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87638,
      "title": "var6 and var81 interaction always correlates with target better than others",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 87416,
      "title": "Z Scores to replace original features",
      "comment_count": 8,
      "views": 0,
      "votes": 4
    },
    {
      "id": 87619,
      "title": "How does this number sound.... 9.999999999976694e-05? Any help would be appreciated :)",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87786,
      "title": "Some folds are very different from others",
      "comment_count": 11,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87118,
      "title": "Torus Shape Data after denoising dataset",
      "comment_count": 8,
      "views": 0,
      "votes": 11
    },
    {
      "id": 86951,
      "title": "When you have bimodal distribution",
      "comment_count": 12,
      "views": 0,
      "votes": 9
    },
    {
      "id": 87641,
      "title": "submission problem",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87764,
      "title": "Looking For Teammate - Last 15 Hours",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 86792,
      "title": "t-SNE embedding by a neural network",
      "comment_count": 11,
      "views": 0,
      "votes": 43
    },
    {
      "id": 87088,
      "title": "Times of ATM transactions",
      "comment_count": 4,
      "views": 0,
      "votes": 7
    },
    {
      "id": 87694,
      "title": "How are 0.901 submissions ranked?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87739,
      "title": "Multiple model optimization",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 86709,
      "title": "My thought on why 'shuffle'  and 'augment' can work",
      "comment_count": 22,
      "views": 0,
      "votes": 29
    },
    {
      "id": 88014,
      "title": "Joined late and....",
      "comment_count": 1,
      "views": 0,
      "votes": -21
    },
    {
      "id": 87491,
      "title": "Is there any hyperparameter tuning way or framework which is like GridSearchCV and can save its predicted values",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87581,
      "title": "Just Thinking Loud",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87017,
      "title": "Still looking for magic : Timestamp ",
      "comment_count": 8,
      "views": 0,
      "votes": 18
    },
    {
      "id": 87086,
      "title": "#1 leaderboard changed and the 0.924 mark",
      "comment_count": 14,
      "views": 0,
      "votes": 7
    },
    {
      "id": 87451,
      "title": "Feature engineering - Model vars as many normal distributions",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 87377,
      "title": "Leader Board Shakeup",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87488,
      "title": "How many features did you add ",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84825,
      "title": "添加一些新的特征",
      "comment_count": 13,
      "views": 0,
      "votes": -1
    },
    {
      "id": 87193,
      "title": "Team Merger Dilemma ",
      "comment_count": 8,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87450,
      "title": "A Kaggler's Guide to Model Stacking in Practice",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87455,
      "title": "Is this for real !!",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 86965,
      "title": "Peaks in the distributions of variables",
      "comment_count": 10,
      "views": 0,
      "votes": 12
    },
    {
      "id": 87478,
      "title": "Does Discussion already find out the \"weird\"?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87475,
      "title": "How should deal with peaks and bimodal distribution? ",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87343,
      "title": "Waiting for the 0.925 comments ;)",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87494,
      "title": "Why ？I added several features.",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87498,
      "title": "feature",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87211,
      "title": "AUC score",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87250,
      "title": "Bimodal distribution of (CV-LB) values?",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 87482,
      "title": "AUC error",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87341,
      "title": "Training takes longer than the session limit time",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 87409,
      "title": "1 and 0?",
      "comment_count": 2,
      "views": 0,
      "votes": -1
    },
    {
      "id": 86844,
      "title": "Luck is matter....",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87406,
      "title": "Attempt to understand the pre processing better",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87347,
      "title": "Nice website to learn statistics",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 85419,
      "title": "Var PDF's",
      "comment_count": 24,
      "views": 0,
      "votes": 26
    },
    {
      "id": 84912,
      "title": "Submission file output in binary or probability?",
      "comment_count": 10,
      "views": 0,
      "votes": 9
    },
    {
      "id": 87333,
      "title": "Why plot is a different?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 86487,
      "title": "A key to improve your score",
      "comment_count": 30,
      "views": 0,
      "votes": 19
    },
    {
      "id": 87422,
      "title": "Trying to understand something",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 87270,
      "title": "shared weights one more time",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 87202,
      "title": "Collating  weird findings.",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 84450,
      "title": "var_68 is date?",
      "comment_count": 28,
      "views": 0,
      "votes": 40
    },
    {
      "id": 87026,
      "title": "Different solutions with same PL score",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87100,
      "title": "Is this the real deal?",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 86604,
      "title": "Looking for teammate...",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 86789,
      "title": "Is there only a single magic?",
      "comment_count": 23,
      "views": 0,
      "votes": 4
    },
    {
      "id": 87301,
      "title": "Do you think the data is the result of a PCA?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 87181,
      "title": "Peaks are replaced missing values?",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 87288,
      "title": "Grouping features by similar statistics",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87236,
      "title": "Augmentation and Stacking ",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87282,
      "title": "8K+ First Time Ever in the History of Kaggle (Featured)",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 82621,
      "title": "lgb: Which hyperparameters to tune and what tools do you use for tunning?",
      "comment_count": 8,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87059,
      "title": "Looking for a team",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87109,
      "title": "The cumsum of Data",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87149,
      "title": "looking for teamates...let's dig magic...",
      "comment_count": 6,
      "views": 0,
      "votes": -11
    },
    {
      "id": 86498,
      "title": "Weekend effect",
      "comment_count": 5,
      "views": 0,
      "votes": 21
    },
    {
      "id": 87099,
      "title": "One more thing about magic",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 86880,
      "title": "How to deal with bumps of var_12, var_81, etc",
      "comment_count": 8,
      "views": 0,
      "votes": 5
    },
    {
      "id": 87251,
      "title": "Concept Drift",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87130,
      "title": "Synthetic Data Vault (SDV) - should we synthesize a larger and cleaner dataset?",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 87003,
      "title": "Negative And Positive Features",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 82237,
      "title": "Some feature engineerings I have done",
      "comment_count": 40,
      "views": 0,
      "votes": 90
    },
    {
      "id": 87034,
      "title": "Homer Simpson Magic",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 86940,
      "title": "cannot submit new results",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 86994,
      "title": "Training score vs valid score?",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 86653,
      "title": "Discover The Magic Of Santander",
      "comment_count": 22,
      "views": 0,
      "votes": 4
    },
    {
      "id": 86983,
      "title": "best way to bin numerical columns",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 87125,
      "title": "How about this?",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 86670,
      "title": "winning competitions be like...",
      "comment_count": 7,
      "views": 0,
      "votes": 14
    },
    {
      "id": 87144,
      "title": "Weighted Rank for submission",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 87240,
      "title": "Counts of Unique Values",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 87004,
      "title": "Hypothesis: it IS time series data",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 83882,
      "title": "Shuffling the features",
      "comment_count": 75,
      "views": 0,
      "votes": 168
    },
    {
      "id": 87101,
      "title": "Why k-folds auc may be so different?",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 85251,
      "title": "Got LB 0.869 without LGB or NN, even no need to train :)",
      "comment_count": 19,
      "views": 0,
      "votes": 21
    },
    {
      "id": 80996,
      "title": "Best single model",
      "comment_count": 249,
      "views": 0,
      "votes": 70
    },
    {
      "id": 86747,
      "title": "This is one of the better contests in Kaggle",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 85033,
      "title": "A proof of synthetic data",
      "comment_count": 14,
      "views": 0,
      "votes": 15
    },
    {
      "id": 86989,
      "title": "Fake Data?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 85411,
      "title": "Weak feature interactions are there",
      "comment_count": 6,
      "views": 0,
      "votes": 15
    },
    {
      "id": 87195,
      "title": "Markov Chains?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 86889,
      "title": "Because of the anonymizing of the data, competitions became kind of (guess what this variable stands for),",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87348,
      "title": "I need to join a group...",
      "comment_count": 0,
      "views": 0,
      "votes": -6
    },
    {
      "id": 86797,
      "title": "Bad Seeds?",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 84961,
      "title": "Out of ideas - sharing a couple of cool things",
      "comment_count": 32,
      "views": 0,
      "votes": 45
    },
    {
      "id": 87131,
      "title": "Looking for merge with NN score >= 0.900",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 86979,
      "title": "How to deal with anonymous data?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 86849,
      "title": "Retrosynthesis competition?",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 87092,
      "title": "Performance tuning tips - General guidance & For competition Santander Customer Transaction Prediction",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 87078,
      "title": "I'm looking for a kernel or a method",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84780,
      "title": "My method to double check the submit result",
      "comment_count": 11,
      "views": 0,
      "votes": 35
    },
    {
      "id": 86660,
      "title": "Kalman filter for denoising features?",
      "comment_count": 4,
      "views": 0,
      "votes": 14
    },
    {
      "id": 86720,
      "title": "Some features could be predictions from internal classification models !",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 86633,
      "title": "LightGBM Naive Bayes - 0.900",
      "comment_count": 1,
      "views": 0,
      "votes": 14
    },
    {
      "id": 87007,
      "title": "Questions about submission",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 84514,
      "title": "Information regarding Submission & Scoring",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 86654,
      "title": "I am on holidays... Teamup?",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 84638,
      "title": "Kaggler's Terminology",
      "comment_count": 6,
      "views": 0,
      "votes": 26
    },
    {
      "id": 86780,
      "title": "Cluster Feature",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 85039,
      "title": "When you're on the bronze area for the first time in your life xD",
      "comment_count": 15,
      "views": 0,
      "votes": 15
    },
    {
      "id": 82515,
      "title": "[update] LGB - Best Params",
      "comment_count": 29,
      "views": 0,
      "votes": 63
    },
    {
      "id": 86790,
      "title": "Feature Engineering! Tips or suggestions to newbies...",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 86772,
      "title": "How do you measure correlation?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 86793,
      "title": "Is there anyone using the GPU version of LightGBM?",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83288,
      "title": "Silver and bronze medals have no value at this point",
      "comment_count": 67,
      "views": 0,
      "votes": 38
    },
    {
      "id": 86552,
      "title": "How many features used with your best model?",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 86694,
      "title": "Data augmentation using imbalanced-learn",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84998,
      "title": "GTC 2019 Keynote",
      "comment_count": 10,
      "views": 0,
      "votes": 29
    },
    {
      "id": 86500,
      "title": "This is now the competition with the highest number of teams in any Kaggle competition",
      "comment_count": 13,
      "views": 0,
      "votes": 3
    },
    {
      "id": 86522,
      "title": "FE that Requires Interaction between Columns Will not Help",
      "comment_count": 4,
      "views": 0,
      "votes": 7
    },
    {
      "id": 84517,
      "title": "Help",
      "comment_count": 1,
      "views": 0,
      "votes": -3
    },
    {
      "id": 85434,
      "title": "Train and test distributions, again",
      "comment_count": 7,
      "views": 0,
      "votes": 5
    },
    {
      "id": 86861,
      "title": "No output print display in the kernal",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 86787,
      "title": "Does random seed also affect accuracy?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 86728,
      "title": "The last piece of the jigsaw",
      "comment_count": 9,
      "views": 0,
      "votes": -1
    },
    {
      "id": 86673,
      "title": "Kaggle New Kernel Editor Style",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 85218,
      "title": "Various way to extract magic feature",
      "comment_count": 3,
      "views": 0,
      "votes": 12
    },
    {
      "id": 86836,
      "title": "Stacking ",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 86496,
      "title": "What can we do with a date-time feature?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83755,
      "title": "How come they are stuck at 0.924",
      "comment_count": 61,
      "views": 0,
      "votes": 33
    },
    {
      "id": 86532,
      "title": "How to implement oof predictions model in real life?",
      "comment_count": 11,
      "views": 0,
      "votes": 2
    },
    {
      "id": 85323,
      "title": "5 teams did break the 0.920 score",
      "comment_count": 10,
      "views": 0,
      "votes": 6
    },
    {
      "id": 86674,
      "title": "Submitting Type",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 86795,
      "title": "Deep expertise in one technology vs broad set of tools",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 86785,
      "title": "FE to multiple features drops CV score",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84382,
      "title": "Submission File",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 86666,
      "title": "Are we sure that there are no interactions between the variables?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 86769,
      "title": "The appropriate number of cross-validation?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 85400,
      "title": "For Tilli GP Features",
      "comment_count": 5,
      "views": 0,
      "votes": 6
    },
    {
      "id": 85394,
      "title": "Some meaningless FE I did",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 85152,
      "title": "To eliminate or not?",
      "comment_count": 3,
      "views": 0,
      "votes": 11
    },
    {
      "id": 86507,
      "title": "Do you want to have an extra month to run it? I would ",
      "comment_count": 23,
      "views": 0,
      "votes": -3
    },
    {
      "id": 84614,
      "title": "What does \"leak\" mean?",
      "comment_count": 18,
      "views": 0,
      "votes": 12
    },
    {
      "id": 83570,
      "title": "List of experiments that haven't broken 0.9",
      "comment_count": 64,
      "views": 0,
      "votes": 90
    },
    {
      "id": 86623,
      "title": "how to utilize stackings?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83489,
      "title": "we should team up earlier",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83398,
      "title": "Looking for team mates!",
      "comment_count": 15,
      "views": 0,
      "votes": 1
    },
    {
      "id": 85309,
      "title": "Are variables 12, 53 and 126 diff pics of the same thing?",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 86608,
      "title": "Is there any easy(less time) method to tune XGBoost/CatBoost  Hyperparamteres ??",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 86541,
      "title": "Can you give me a little hint? ",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 85194,
      "title": "How can I get magic feature ???",
      "comment_count": 10,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82899,
      "title": "Looking for a team",
      "comment_count": 10,
      "views": 0,
      "votes": 0
    },
    {
      "id": 86523,
      "title": "Var_68 and the other ones",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 84977,
      "title": "Sorting",
      "comment_count": 13,
      "views": 0,
      "votes": 4
    },
    {
      "id": 85429,
      "title": "Rankdata kills for CV but it's a huge fail on LB",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 82011,
      "title": "New kaggler readme/guide",
      "comment_count": 5,
      "views": 0,
      "votes": 27
    },
    {
      "id": 86621,
      "title": "Predictive Analysis -Santander ",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 85114,
      "title": "Is shuffle augmentation + bagging the way to go?",
      "comment_count": 6,
      "views": 0,
      "votes": 8
    },
    {
      "id": 86479,
      "title": "A little question of lightgbm model",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 85414,
      "title": "3D scatter plots? ",
      "comment_count": 4,
      "views": 0,
      "votes": 0
    },
    {
      "id": 86556,
      "title": "Mean target ",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 85339,
      "title": "Boost LB score by adjusting probability",
      "comment_count": 14,
      "views": 0,
      "votes": 1
    },
    {
      "id": 86501,
      "title": "Var0 *  Var1 gives a  more symmetric normal dist but lowers score!",
      "comment_count": 3,
      "views": 0,
      "votes": 0
    },
    {
      "id": 85353,
      "title": "Data has correlation at the extremes",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 86488,
      "title": "Here it goes 7200+",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84847,
      "title": "Any explanation why shuffling augmentation works?",
      "comment_count": 13,
      "views": 0,
      "votes": 10
    },
    {
      "id": 86506,
      "title": "Teaming up for better score ?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 85131,
      "title": "Why Naive Bayes is worse than model one feature at a time?",
      "comment_count": 11,
      "views": 0,
      "votes": 1
    },
    {
      "id": 86463,
      "title": "How does the system compute auc score?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 83048,
      "title": "What's so different about var_68?",
      "comment_count": 38,
      "views": 0,
      "votes": 27
    },
    {
      "id": 84612,
      "title": "What is stacking?? How it helps?",
      "comment_count": 17,
      "views": 0,
      "votes": 7
    },
    {
      "id": 84996,
      "title": "problem when editing kaggle kernels",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 85096,
      "title": "Code to unlock trainset",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 84886,
      "title": "overfitting in stacking",
      "comment_count": 21,
      "views": 0,
      "votes": 2
    },
    {
      "id": 85441,
      "title": "kernel can not use",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 85121,
      "title": " Boosting creativity towards feature engineering",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 85327,
      "title": "How to approach feature engineering here ?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 85229,
      "title": "Does picking up some folds instead of all folds  help?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 85314,
      "title": "the same accuracy rate0.900, but the ranking is different?",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 85182,
      "title": "How long does it take to train LGBM?",
      "comment_count": 4,
      "views": 0,
      "votes": -2
    },
    {
      "id": 80539,
      "title": "Don't you have a life?",
      "comment_count": 35,
      "views": 0,
      "votes": 61
    },
    {
      "id": 84802,
      "title": "K Fold gave a better result than Stratified K folds.",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 85188,
      "title": "data leak & EDA?",
      "comment_count": 12,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84761,
      "title": "\"Shuffling the features\" seems the key magic",
      "comment_count": 14,
      "views": 0,
      "votes": 7
    },
    {
      "id": 84804,
      "title": "Genetically Engineered Features",
      "comment_count": 4,
      "views": 0,
      "votes": 14
    },
    {
      "id": 85244,
      "title": "Low correlation between data, how to improve?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 85028,
      "title": "SMOTE oversampling overfitting",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82904,
      "title": "Added features and LB score increased",
      "comment_count": 30,
      "views": 0,
      "votes": 45
    },
    {
      "id": 83774,
      "title": "Some thoughts",
      "comment_count": 18,
      "views": 0,
      "votes": 10
    },
    {
      "id": 84794,
      "title": "Why do not imbalanced-learn?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 85223,
      "title": "fastai doesnt work on kernels",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 85179,
      "title": "How helpful is kaggle for our machine learning skills?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84830,
      "title": "Why is everything so \"Normal\"?",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 85178,
      "title": "Stacking alternatives",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 85003,
      "title": "Best Single Model",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84970,
      "title": "Are we maybe dealing with heatmaps data?",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 84633,
      "title": "Evaluation will be Done on Binary Values.. How is that possible ?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84743,
      "title": "Decision Tree Visualization Of the Data set",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 84610,
      "title": "Most popular competition on Kaggle by number of participants",
      "comment_count": 20,
      "views": 0,
      "votes": 2
    },
    {
      "id": 84658,
      "title": "Will Santander Transactions beat Home Default Credit Risk",
      "comment_count": 8,
      "views": 0,
      "votes": 2
    },
    {
      "id": 84178,
      "title": "Augmentation by [shuffling the features]",
      "comment_count": 19,
      "views": 0,
      "votes": 15
    },
    {
      "id": 82674,
      "title": "Importance of new features.",
      "comment_count": 9,
      "views": 0,
      "votes": 4
    },
    {
      "id": 84639,
      "title": "Hyperparameter Tuning for LightGBM Models",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84542,
      "title": "Categorical features?",
      "comment_count": 13,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84678,
      "title": "Gradient Boosting predict the same in rounded data",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84701,
      "title": "Should I trust my CV or LB?",
      "comment_count": 9,
      "views": 0,
      "votes": null
    },
    {
      "id": 84747,
      "title": "how to add the focal loss to the custom loss function of lightgbm",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 84828,
      "title": "How to break 0.899?",
      "comment_count": 5,
      "views": 0,
      "votes": 0
    },
    {
      "id": 83363,
      "title": "focal_loss for neural network when class is imbalance",
      "comment_count": 9,
      "views": 0,
      "votes": 10
    },
    {
      "id": 84619,
      "title": "Looking for Team",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84806,
      "title": "Access LightGBM training score",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 83864,
      "title": "Leakhunters...?",
      "comment_count": 10,
      "views": 0,
      "votes": 12
    },
    {
      "id": 84686,
      "title": "feature engineering 0.42corr, but no better score, ideas?",
      "comment_count": 5,
      "views": 0,
      "votes": 0
    },
    {
      "id": 81217,
      "title": "Why is this a competition?",
      "comment_count": 55,
      "views": 0,
      "votes": 27
    },
    {
      "id": 84004,
      "title": "CV versus LB",
      "comment_count": 20,
      "views": 0,
      "votes": 6
    },
    {
      "id": 84348,
      "title": "Something interesting about exp transform",
      "comment_count": 12,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81870,
      "title": "Looking for team",
      "comment_count": 11,
      "views": 0,
      "votes": 0
    },
    {
      "id": 84495,
      "title": "Can I stack different n_splits",
      "comment_count": 13,
      "views": 0,
      "votes": 0
    },
    {
      "id": 84657,
      "title": "Santander submission error",
      "comment_count": 10,
      "views": 0,
      "votes": null
    },
    {
      "id": 84309,
      "title": "Take home tips",
      "comment_count": 1,
      "views": 0,
      "votes": 18
    },
    {
      "id": 84710,
      "title": "Are there some clues in ensembling, with CatBoost, Xgboot, LightGBM",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 84703,
      "title": "How comes my LB so poor ?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 84668,
      "title": "So, what is happening with the Data?",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84548,
      "title": "New features better than var_81?",
      "comment_count": 8,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83606,
      "title": "Got CV 0.912+ but 0.892LB",
      "comment_count": 31,
      "views": 0,
      "votes": 13
    },
    {
      "id": 80770,
      "title": "Is it segmentation problem?",
      "comment_count": 19,
      "views": 0,
      "votes": 39
    },
    {
      "id": 84575,
      "title": "Are Hyper Parameters machine dependent?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84546,
      "title": "High dimensional properties of data",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 81681,
      "title": "Help! I want to join in a existing team",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 83779,
      "title": "5200+ Teams",
      "comment_count": 13,
      "views": 0,
      "votes": 16
    },
    {
      "id": 84406,
      "title": "Bigger n_splits value",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 81059,
      "title": "Joining a team",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 84447,
      "title": "Sammon Mapping",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82499,
      "title": "Best neural network model",
      "comment_count": 9,
      "views": 0,
      "votes": 25
    },
    {
      "id": 84281,
      "title": "Features' transformations to suggest?",
      "comment_count": 7,
      "views": 0,
      "votes": 4
    },
    {
      "id": 84428,
      "title": "This competition is getting funny on lines of code.",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83121,
      "title": "0.924!!!!! The top score keeps going up",
      "comment_count": 86,
      "views": 0,
      "votes": 37
    },
    {
      "id": 84263,
      "title": "Is it encoded ?",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 84537,
      "title": "A/B testing data? ",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84464,
      "title": "Some features having lot of repeated values",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84433,
      "title": "Resources for tuning LightGBM",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84512,
      "title": "Is lightgbm  sensitive for outliers in variables?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 83033,
      "title": "For those who have missed...",
      "comment_count": 45,
      "views": 0,
      "votes": 65
    },
    {
      "id": 84226,
      "title": "IS THER ANY distribution measurement statics of train test ",
      "comment_count": 8,
      "views": 0,
      "votes": 0
    },
    {
      "id": 84496,
      "title": "Just 24 days are left, and almost 2,000 people are stuck at 0.900!!",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84443,
      "title": "DAE?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 84488,
      "title": "about num_boost_round in LightGBM",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 84415,
      "title": "The competition rules say the submission should be binary",
      "comment_count": 2,
      "views": 0,
      "votes": -1
    },
    {
      "id": 84445,
      "title": "Similar competitions",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 84019,
      "title": "[update] Why am I falling at LB? ",
      "comment_count": 14,
      "views": 0,
      "votes": 3
    },
    {
      "id": 84448,
      "title": "Quick question on Leaderboard arrangement",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 84261,
      "title": "Think about dropping features",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84418,
      "title": "Effect of Joint Distribution on Targets",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 84489,
      "title": "Genuine Thought with ?",
      "comment_count": 4,
      "views": 0,
      "votes": -2
    },
    {
      "id": 84028,
      "title": "the saddest thing is ....",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 83465,
      "title": "Is Data Nonstationary?",
      "comment_count": 10,
      "views": 0,
      "votes": 7
    },
    {
      "id": 84010,
      "title": "OverSampling/UnderSampling",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 84333,
      "title": "Some CV folds processing?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84344,
      "title": "What's your AUC, F1 Score and Best Threshold",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 84295,
      "title": "Stuck, not reaching 90% accuracy even in CV",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84310,
      "title": "How to do target encoding correctly?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 84146,
      "title": "Drop unimportant features, My score also dropped :(",
      "comment_count": 6,
      "views": 0,
      "votes": 0
    },
    {
      "id": 83497,
      "title": "More CV folds, better score?",
      "comment_count": 15,
      "views": 0,
      "votes": 8
    },
    {
      "id": 84213,
      "title": "There is something tricky about test data for public LB ",
      "comment_count": 4,
      "views": 0,
      "votes": -1
    },
    {
      "id": 84211,
      "title": "Overfitting - Random Seed and N Splits",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84093,
      "title": "something is wrong",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84140,
      "title": "Strange features distributions",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 83821,
      "title": "How does CV improve performance?",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83911,
      "title": "Finding lagged columns in anonymized data?",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 84241,
      "title": "Need help on post image editing",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 84034,
      "title": "Will this be another MS Malware?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83700,
      "title": "Found Something That I Can't Understand",
      "comment_count": 10,
      "views": 0,
      "votes": 4
    },
    {
      "id": 83978,
      "title": "Feature Removing brings Improvement??",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 83684,
      "title": "Semi-Supervised Learning?",
      "comment_count": 4,
      "views": 0,
      "votes": 0
    },
    {
      "id": 84147,
      "title": "Pseudo Labeling",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84099,
      "title": "Did any one used PCA",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83866,
      "title": "Did any one try Boruta or Recursive Feature Selection",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 84033,
      "title": "I think, I chose the wrong competition to be my first one",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83070,
      "title": "The Way to Reach 0.900 AUC",
      "comment_count": 17,
      "views": 0,
      "votes": 35
    },
    {
      "id": 80555,
      "title": "Something is odd",
      "comment_count": 51,
      "views": 0,
      "votes": 39
    },
    {
      "id": 83877,
      "title": "How do I quickly check engineered features' value?",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83525,
      "title": "5 Fold vs. 10 Fold CV is the Difference Between 29th and 265th Place",
      "comment_count": 10,
      "views": 0,
      "votes": 8
    },
    {
      "id": 83397,
      "title": "Darius Barušauskas - Kaggle Days Paris - \"My journey to becoming a Kaggle Grandmaster\"",
      "comment_count": 8,
      "views": 0,
      "votes": 15
    },
    {
      "id": 83996,
      "title": "Summarize approaches, please add more",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 82307,
      "title": "A view from the business side might help !",
      "comment_count": 20,
      "views": 0,
      "votes": 74
    },
    {
      "id": 83730,
      "title": "Homogeneity detector ",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 83512,
      "title": "Should the original variables be used when stacking?",
      "comment_count": 7,
      "views": 0,
      "votes": 4
    },
    {
      "id": 83949,
      "title": "Hunt for useful feature!",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83850,
      "title": "CV progress and LB",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 84174,
      "title": "No one USES XGB to improve accuracy?",
      "comment_count": 0,
      "views": 0,
      "votes": -2
    },
    {
      "id": 83989,
      "title": "Any idea how the Public/Private test dataset is split?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 84045,
      "title": "Hypothesis - Columns are dates/times and/or date/time differences",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 84154,
      "title": "Here is an interesting idea",
      "comment_count": 1,
      "views": 0,
      "votes": -2
    },
    {
      "id": 84012,
      "title": "Maybe we should think about it.",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83815,
      "title": "Rank() of probability prediction 0.900??",
      "comment_count": 7,
      "views": 0,
      "votes": -2
    },
    {
      "id": 83388,
      "title": "they just reach 0.922",
      "comment_count": 12,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83810,
      "title": "How to data science academic knowledge",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81377,
      "title": "Submission file - confused",
      "comment_count": 12,
      "views": 0,
      "votes": 10
    },
    {
      "id": 83505,
      "title": "Selecting the best trained model per fold",
      "comment_count": 8,
      "views": 0,
      "votes": 3
    },
    {
      "id": 83621,
      "title": "Submit (1 - submit['target']) and same LB score",
      "comment_count": 14,
      "views": 0,
      "votes": -1
    },
    {
      "id": 83841,
      "title": "Unable to submit my results",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 83843,
      "title": "Late Submission Timeline",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 80663,
      "title": "This deserved to be a Kernel Competition",
      "comment_count": 10,
      "views": 0,
      "votes": 36
    },
    {
      "id": 83849,
      "title": "I found some magic features by some way",
      "comment_count": 16,
      "views": 0,
      "votes": -27
    },
    {
      "id": 83783,
      "title": "How to go far?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 83699,
      "title": "How is the ranking done in the Leaderboard for the same score ?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83355,
      "title": "0.919 Score Hypothesis",
      "comment_count": 7,
      "views": 0,
      "votes": 7
    },
    {
      "id": 82995,
      "title": "Break the CV Threshold?",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 83763,
      "title": "Would you add new features to one model or stack multiple models?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 83453,
      "title": "What's your best nn cv/lb score？",
      "comment_count": 10,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83697,
      "title": "Binary Prediction or Probability",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 83677,
      "title": "New feature creation code block (ratios)",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 83687,
      "title": "Sort by Best Score vs LB rank?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82277,
      "title": "Tricks that work when training NN.",
      "comment_count": 27,
      "views": 0,
      "votes": 44
    },
    {
      "id": 83669,
      "title": "What to submit?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 83588,
      "title": "Stranger things at submissions",
      "comment_count": 6,
      "views": 0,
      "votes": -1
    },
    {
      "id": 80630,
      "title": "Adversarial Validation tells that it's hard to distinguish train/test",
      "comment_count": 5,
      "views": 0,
      "votes": 10
    },
    {
      "id": 83663,
      "title": "AUC in Keras for R",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 83270,
      "title": "Simple Stack Advanced Me 50+ Positions.",
      "comment_count": 9,
      "views": 0,
      "votes": 8
    },
    {
      "id": 81159,
      "title": "Feature Selection?",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83394,
      "title": "SMOTE vs StratifiedKFold for balancing dataset",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 82959,
      "title": "How to know the meaning of each variable ?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83626,
      "title": "Any inside about how they generate or convert the Data? ",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 83471,
      "title": "Stacking models with class probabilities as features",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 81938,
      "title": "Learnings Today",
      "comment_count": 5,
      "views": 0,
      "votes": 4
    },
    {
      "id": 83559,
      "title": "how do you tune stacking model parameters?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 83516,
      "title": "What tools and methods to use for data exploration?",
      "comment_count": 1,
      "views": 0,
      "votes": -1
    },
    {
      "id": 83308,
      "title": "I am getting over 90% accuracy but submission is telling different story",
      "comment_count": 13,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82698,
      "title": "GB or NN? This is a problem!",
      "comment_count": 25,
      "views": 0,
      "votes": 12
    },
    {
      "id": 83004,
      "title": "I would be so nice to see more decimals on the LB :)",
      "comment_count": 7,
      "views": 0,
      "votes": 11
    },
    {
      "id": 83213,
      "title": "Don't you have a life (?) revisited",
      "comment_count": 15,
      "views": 0,
      "votes": -11
    },
    {
      "id": 82841,
      "title": "No Clear Segments",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83216,
      "title": "AUC vs Target ???",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83423,
      "title": "High scoring public kernels forked and submitted",
      "comment_count": 3,
      "views": 0,
      "votes": -1
    },
    {
      "id": 83137,
      "title": "Important features mean nothing alone?",
      "comment_count": 7,
      "views": 0,
      "votes": 5
    },
    {
      "id": 82983,
      "title": "How confident are we?",
      "comment_count": 10,
      "views": 0,
      "votes": 7
    },
    {
      "id": 83047,
      "title": "What if the data is timestamp?",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83286,
      "title": "Do experts have their secret tools",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83344,
      "title": "Newbie - Hyperparameter search",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 83279,
      "title": "how can I tune my parameters in this case?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 82570,
      "title": "Do we need to focus on feature engineering?",
      "comment_count": 20,
      "views": 0,
      "votes": 4
    },
    {
      "id": 83160,
      "title": "Work environment and code organization",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 81807,
      "title": "Learnings So Far",
      "comment_count": 15,
      "views": 0,
      "votes": 8
    },
    {
      "id": 83062,
      "title": "what is your local CV and LB",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 83134,
      "title": "How is the competition metric decided ?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83236,
      "title": "Thoughts on what to explore next?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82181,
      "title": "Auc stuck at 75",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 83122,
      "title": "Found some new FE features, but no improvement. Why?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82965,
      "title": "Predictions as Probabilities or Classes",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 83141,
      "title": "We are trying to over-fit LB until better single model comes out ",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 83132,
      "title": "Model Stacking where to go next?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 82885,
      "title": "A Deep Understanding of LGBM",
      "comment_count": 0,
      "views": 0,
      "votes": 8
    },
    {
      "id": 82186,
      "title": "Find something wired with CNN",
      "comment_count": 6,
      "views": 0,
      "votes": 8
    },
    {
      "id": 82839,
      "title": "Why do standard sklearn classifiers all underestimate the result?",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82890,
      "title": "Variation in oof auc only by shuffling the features",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83038,
      "title": "Why model has high AUC score but low f1 score",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83018,
      "title": "How to break through 0.899",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83005,
      "title": "Something I can't understand",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 83003,
      "title": "ValueError: feature_names mismatch after oversampling with SMOTE",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 81158,
      "title": "Categorial (pseudo) features",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 82915,
      "title": "can my position on LB go up itself?",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 82975,
      "title": "CatBoost vs LightGBM",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 82394,
      "title": "Stacked Generalization",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 82940,
      "title": "magic 0.899",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 82678,
      "title": "LB score is slightly higher than local CV",
      "comment_count": 9,
      "views": 0,
      "votes": null
    },
    {
      "id": 82945,
      "title": "LightGBM max leaf size",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 82818,
      "title": "What is feature engineering and where can I learn the basics?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81301,
      "title": "Yet Another Data \"Property\"",
      "comment_count": 21,
      "views": 0,
      "votes": 22
    },
    {
      "id": 82758,
      "title": "why so much 0.899？",
      "comment_count": 2,
      "views": 0,
      "votes": 0
    },
    {
      "id": 82867,
      "title": "Low Boosting Complexity",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82781,
      "title": "How to get the 0.9 AUC in the santander customer transaction prediction",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81921,
      "title": "How do you guys know what to do?",
      "comment_count": 12,
      "views": 0,
      "votes": 6
    },
    {
      "id": 82457,
      "title": "PCA for SCTP",
      "comment_count": 9,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82742,
      "title": "How to Find Iterations Value in CatBoost?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 82721,
      "title": "ROC curve",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 82617,
      "title": "THE SUBMISSION FILE",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 82481,
      "title": "Beginner frustration ",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 80668,
      "title": "maybe show  5 after the decimal point is better",
      "comment_count": 15,
      "views": 0,
      "votes": 11
    },
    {
      "id": 81937,
      "title": "Weird overfitting when using RandomForestClassifier",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 82387,
      "title": "When Adding a Feature Makes Your Model Worse",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 82420,
      "title": "Tuning hyperparameters and avoiding overfitting",
      "comment_count": 8,
      "views": 0,
      "votes": null
    },
    {
      "id": 82324,
      "title": "Prediction format",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 80627,
      "title": "Nature of the features",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81942,
      "title": "My Neural net",
      "comment_count": 8,
      "views": 0,
      "votes": 6
    },
    {
      "id": 82339,
      "title": "First competition: Feature Engineering vs. Hyperparameter tuning",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82395,
      "title": "Over-fitting or something else?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 82193,
      "title": "\"timeout or by a memory leak.\", UserWarning",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 82110,
      "title": "Get Started: Beginners",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 82370,
      "title": "Manual vs Automatic Hyperparameter tuning",
      "comment_count": 3,
      "views": 0,
      "votes": -1
    },
    {
      "id": 82456,
      "title": "Why is it so difficult to upgrade?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82134,
      "title": "Max, min for same cols in train and test",
      "comment_count": 1,
      "views": 0,
      "votes": 11
    },
    {
      "id": 82282,
      "title": "Handling Class Imbalance",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 82389,
      "title": "Submission (Probability or Binary prediction?)",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 82296,
      "title": "Stacking",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 82373,
      "title": "Overfitting the leaderboard",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81916,
      "title": "how can I evaluate stacking ?",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 82058,
      "title": "It is better to show the 4th digit of AUC.",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 82240,
      "title": "Pytorch--NNbaseline",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 82222,
      "title": "Why my submission always fail?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 82192,
      "title": "Some problem ",
      "comment_count": 9,
      "views": 0,
      "votes": null
    },
    {
      "id": 82044,
      "title": "Threshold",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 82024,
      "title": "Submissions per Day",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 82116,
      "title": "Misunderstanding Results using NNs",
      "comment_count": 6,
      "views": 0,
      "votes": 0
    },
    {
      "id": 82005,
      "title": "Similar Competitions?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81554,
      "title": "Pseudo-Labeling",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 80959,
      "title": "3 submissions per day?",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 80706,
      "title": "\"Explained variance ratio\" ?",
      "comment_count": 5,
      "views": 0,
      "votes": 5
    },
    {
      "id": 81880,
      "title": "Problem in Understanding GridSearchCV",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81444,
      "title": "Data interpretation",
      "comment_count": 9,
      "views": 0,
      "votes": 4
    },
    {
      "id": 81769,
      "title": "Given dataset",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81836,
      "title": "Similar competitions or datasets (Classification + all numeric or maybe PCA components) ",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 81641,
      "title": "Any thoughts on these few specific cols?",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81905,
      "title": "Just signed up!",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 81571,
      "title": "Why is it so hard to improve just 0.001?",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 81874,
      "title": "Unique Co variance matrices. Something is fishy!",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 81872,
      "title": "Bayesian optimization",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 81048,
      "title": "IMPORTANT NOTICE: Kaggle Downtime",
      "comment_count": 11,
      "views": 0,
      "votes": 13
    },
    {
      "id": 81842,
      "title": "The only possible way......",
      "comment_count": 0,
      "views": 0,
      "votes": -1
    },
    {
      "id": 81834,
      "title": "OSError reading with pandas",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 81823,
      "title": "Considering on mixing LGB and NN",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 81332,
      "title": "Why is the Data so Uncorrelated?",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 81385,
      "title": "Run time Issues",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 81269,
      "title": "NN's ?",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 81205,
      "title": "Upsampling",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81552,
      "title": "Hidden Categorical Features & Reverse Engineering ?",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 80679,
      "title": "Team Up",
      "comment_count": 11,
      "views": 0,
      "votes": 0
    },
    {
      "id": 81121,
      "title": "Odd entry on the leaderboard",
      "comment_count": 8,
      "views": 0,
      "votes": 7
    },
    {
      "id": 80980,
      "title": "How to get best parameters for lightgbm",
      "comment_count": 18,
      "views": 0,
      "votes": 10
    },
    {
      "id": 81564,
      "title": "Different Population for test and train data",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81062,
      "title": "Racing for the 4th digit",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81340,
      "title": "Variances for different target Values",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 81355,
      "title": "Submission",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81370,
      "title": "Not able to submit",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 81376,
      "title": "I cant upload my submission.",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 81350,
      "title": "Who can explain the jerk of three participants to score: 0.903-0.904?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 80976,
      "title": "Treating columns as a Sequence ? ",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81206,
      "title": "The 0.900 wall has been breached",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 81276,
      "title": "How is the data is masked?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 80864,
      "title": "Ensembling",
      "comment_count": 5,
      "views": 0,
      "votes": 5
    },
    {
      "id": 81060,
      "title": "Should I use the entire training dataset?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 81210,
      "title": "Are LB scores rounded or floored ?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 81103,
      "title": "Feature Engineering & Preprocessing?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 81076,
      "title": "LightGBM + Bayesian optimization for finding parameters +Rank average",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 80867,
      "title": "no target variable in test set?",
      "comment_count": 3,
      "views": 0,
      "votes": -10
    },
    {
      "id": 80763,
      "title": "Why does submit probabilities work?",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 80797,
      "title": "Are you planning to do up-sampling of minority class?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 80615,
      "title": "Well balanced train and test data",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 80834,
      "title": "Selecting Models for Stacking",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 80489,
      "title": "Interesting that ALL the values are masked here.",
      "comment_count": 8,
      "views": 0,
      "votes": null
    },
    {
      "id": 80655,
      "title": "Kaggle kernel Catboost gpu",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 80528,
      "title": "LightGBM base line (0.898)",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    }
  ],
  "errors": []
}