{
  "total": 442,
  "topics": [
    {
      "id": 289993,
      "title": "New to Kaggle or Machine Learning? Come Say Hi!",
      "comment_count": 64,
      "views": 0,
      "votes": 45
    },
    {
      "id": 308248,
      "title": "Competition is Finalized - Congratulations to our Winners; Recap",
      "comment_count": 19,
      "views": 0,
      "votes": 36
    },
    {
      "id": 307760,
      "title": "2nd Solution - YOLOv5",
      "comment_count": 35,
      "views": 0,
      "votes": 116
    },
    {
      "id": 297282,
      "title": "CUDA out of memory when training YOLOX-l and YOLOX-m",
      "comment_count": 12,
      "views": 0,
      "votes": 4
    },
    {
      "id": 291373,
      "title": "Resources for Learning TensorFlow 🤓",
      "comment_count": 9,
      "views": 0,
      "votes": 19
    },
    {
      "id": 300638,
      "title": "[LB 0.579] Yolov5(higher resolution) is all you need",
      "comment_count": 260,
      "views": 0,
      "votes": 180
    },
    {
      "id": 289999,
      "title": "🔥🔥Winning solution of Previous Object Detection Kaggle Challenges",
      "comment_count": 64,
      "views": 0,
      "votes": 256
    },
    {
      "id": 297863,
      "title": "My Kaggle Lessons Learned in 2021",
      "comment_count": 101,
      "views": 0,
      "votes": 237
    },
    {
      "id": 307619,
      "title": "6th place solution ,  yolov5m6/s6,  trust cv, post classification.",
      "comment_count": 19,
      "views": 0,
      "votes": 51
    },
    {
      "id": 290016,
      "title": "🔥🔥Kaggle Starter Notebooks for Object Detection",
      "comment_count": 31,
      "views": 0,
      "votes": 152
    },
    {
      "id": 307878,
      "title": "Trust CV -- 1st Place Solution",
      "comment_count": 77,
      "views": 0,
      "votes": 192
    },
    {
      "id": 322456,
      "title": "Anyone can explain why training on smaller size then inference on bigger works?",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 290956,
      "title": "TensorFlow Object Detection API",
      "comment_count": 1,
      "views": 0,
      "votes": 15
    },
    {
      "id": 307831,
      "title": "🎖 Summary of all solutions shared + Tricks🎖",
      "comment_count": 2,
      "views": 0,
      "votes": 30
    },
    {
      "id": 307626,
      "title": "4th Place Solution - CenterNet",
      "comment_count": 58,
      "views": 0,
      "votes": 108
    },
    {
      "id": 307735,
      "title": "8th place solution",
      "comment_count": 10,
      "views": 0,
      "votes": 36
    },
    {
      "id": 295862,
      "title": "[Guide] - How to ensemble object detection models?",
      "comment_count": 3,
      "views": 0,
      "votes": 39
    },
    {
      "id": 309065,
      "title": "And in the end... who really cares about the Great Barrier Reef?",
      "comment_count": 16,
      "views": 0,
      "votes": 12
    },
    {
      "id": 318810,
      "title": "Access to test data",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 296976,
      "title": "Augmentations for Object Detection",
      "comment_count": 12,
      "views": 0,
      "votes": 12
    },
    {
      "id": 305663,
      "title": "Has anyone else tried a multi-object tracker like Bytetrack or deepsort?",
      "comment_count": 9,
      "views": 0,
      "votes": 6
    },
    {
      "id": 308007,
      "title": "5th place solution, poisson blending,detection and tracking",
      "comment_count": 26,
      "views": 0,
      "votes": 141
    },
    {
      "id": 301747,
      "title": "EfficientDet Pytorch Starter(Train & Infer)",
      "comment_count": 3,
      "views": 0,
      "votes": 16
    },
    {
      "id": 307800,
      "title": "63rd place solution - Single Cascade R-CNN + Norfair",
      "comment_count": 1,
      "views": 0,
      "votes": 9
    },
    {
      "id": 307609,
      "title": "69th Place - WBF with Threshold",
      "comment_count": 52,
      "views": 0,
      "votes": 93
    },
    {
      "id": 311848,
      "title": "EDA to Yolo v5🚀  Great Barrier Reef - Blog⭐🐬⛵",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 307718,
      "title": "11th Place solution - Team COTS ",
      "comment_count": 9,
      "views": 0,
      "votes": 34
    },
    {
      "id": 307786,
      "title": "7th Place Solution - Cascade RCNN+Tracking ( Public LB is not all you need.) ",
      "comment_count": 18,
      "views": 0,
      "votes": 45
    },
    {
      "id": 307756,
      "title": "10th place solution — 2xYOLOv5 + tracking",
      "comment_count": 16,
      "views": 0,
      "votes": 55
    },
    {
      "id": 308336,
      "title": "16th Simple Solution - Only Yolov5",
      "comment_count": 9,
      "views": 0,
      "votes": 26
    },
    {
      "id": 307871,
      "title": "9th place solution: Tiled Training + Seq-NMS",
      "comment_count": 10,
      "views": 0,
      "votes": 24
    },
    {
      "id": 308329,
      "title": "A Feedback: About Accuracy of This Competition’s Labels",
      "comment_count": 44,
      "views": 0,
      "votes": 33
    },
    {
      "id": 308656,
      "title": "12th place solution (mysteriously disqualified) - Single ConvNeXt CascadeRCNN",
      "comment_count": 15,
      "views": 0,
      "votes": 35
    },
    {
      "id": 303529,
      "title": "🔥Congratulations to Tensor Girl for becoming 3x GrandMaster🔥 ",
      "comment_count": 283,
      "views": 0,
      "votes": 136
    },
    {
      "id": 307707,
      "title": "3rd place solution - Team Hydrogen",
      "comment_count": 31,
      "views": 0,
      "votes": 130
    },
    {
      "id": 308531,
      "title": "Object Detection Models - Top notebooks for training / inference in this competition",
      "comment_count": 6,
      "views": 0,
      "votes": 14
    },
    {
      "id": 308027,
      "title": "Question about CV",
      "comment_count": 17,
      "views": 0,
      "votes": 12
    },
    {
      "id": 307504,
      "title": "⭐️⭐️⭐️ Experiments list - competition summary from our team perspective ⭐️⭐️⭐️⭐️",
      "comment_count": 34,
      "views": 0,
      "votes": 78
    },
    {
      "id": 308455,
      "title": "Why we got disqualified from the competition: Clarification",
      "comment_count": 8,
      "views": 0,
      "votes": 15
    },
    {
      "id": 307605,
      "title": "simplest way to \"overfit\" LB (dangerous!!)",
      "comment_count": 27,
      "views": 0,
      "votes": 31
    },
    {
      "id": 307392,
      "title": "Can't wait to learn how people get such high score in LB",
      "comment_count": 68,
      "views": 0,
      "votes": 70
    },
    {
      "id": 307984,
      "title": "Stagnation Concern For Tensorflow (Attn. TF Team)",
      "comment_count": 18,
      "views": 0,
      "votes": 37
    },
    {
      "id": 307613,
      "title": "Finally got the elusive gold! ",
      "comment_count": 22,
      "views": 0,
      "votes": 40
    },
    {
      "id": 308943,
      "title": "Just for fun: ScaledYOLOv4 for COTS",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 307628,
      "title": "Lost my silver medal and my first lesson on trusting public LB 😥",
      "comment_count": 11,
      "views": 0,
      "votes": 17
    },
    {
      "id": 307689,
      "title": "🤝🤝🤝 TEAM WORK 🤝🤝🤝 during Kaggle competition - summary ",
      "comment_count": 8,
      "views": 0,
      "votes": 31
    },
    {
      "id": 307691,
      "title": "15th place solution: YOLO-X only, Seq-NMS",
      "comment_count": 6,
      "views": 0,
      "votes": 33
    },
    {
      "id": 307669,
      "title": "37th place solution - T0m part [Segment Copy-Paste & Progressive Learning]",
      "comment_count": 10,
      "views": 0,
      "votes": 30
    },
    {
      "id": 307045,
      "title": "Some ideas that worked/not worked for me",
      "comment_count": 33,
      "views": 0,
      "votes": 60
    },
    {
      "id": 308332,
      "title": "One more shakeup after LB finalization?",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 307379,
      "title": "Shake up is coming",
      "comment_count": 29,
      "views": 0,
      "votes": 28
    },
    {
      "id": 307825,
      "title": "17th position on public LB to 45th in private [Beginners approach and mistakes we did]",
      "comment_count": 3,
      "views": 0,
      "votes": 13
    },
    {
      "id": 307807,
      "title": "41st place Solution (7th Public LB) YOLO-Z Inspired Model with Tracking",
      "comment_count": 3,
      "views": 0,
      "votes": 11
    },
    {
      "id": 307869,
      "title": "YOLOX on steroids solution (6th on Public LB / 16th on Private LB) [Placeholder]",
      "comment_count": 5,
      "views": 0,
      "votes": 9
    },
    {
      "id": 307622,
      "title": "22th place solution - Train4Ever",
      "comment_count": 10,
      "views": 0,
      "votes": 30
    },
    {
      "id": 307607,
      "title": "My BBox annotation theory that might explain the CV-LB discrepancy and the shakeup",
      "comment_count": 4,
      "views": 0,
      "votes": 28
    },
    {
      "id": 307668,
      "title": "Does tracking help to boost private LB?  ",
      "comment_count": 8,
      "views": 0,
      "votes": 5
    },
    {
      "id": 307614,
      "title": "Turns out that LB is useless",
      "comment_count": 9,
      "views": 0,
      "votes": 21
    },
    {
      "id": 307753,
      "title": "12th place solution - YOLOv5 + Optical flow tracker",
      "comment_count": 5,
      "views": 0,
      "votes": 14
    },
    {
      "id": 307531,
      "title": "Good luck everyone! ",
      "comment_count": 5,
      "views": 0,
      "votes": 18
    },
    {
      "id": 307254,
      "title": "Are there any plan for publishing test data set after competition ended?",
      "comment_count": 5,
      "views": 0,
      "votes": 21
    },
    {
      "id": 307121,
      "title": "Image size greater than native resolution: theory?",
      "comment_count": 9,
      "views": 0,
      "votes": 28
    },
    {
      "id": 307640,
      "title": "Simple trick for AutoWBF",
      "comment_count": 1,
      "views": 0,
      "votes": 10
    },
    {
      "id": 307634,
      "title": "28th to 40th: EDA on Bboxes was useful",
      "comment_count": 0,
      "views": 0,
      "votes": 15
    },
    {
      "id": 307701,
      "title": "yolov5 in tensorflow ",
      "comment_count": 1,
      "views": 0,
      "votes": 12
    },
    {
      "id": 307600,
      "title": "Tensorflow2 inference time.",
      "comment_count": 6,
      "views": 0,
      "votes": 8
    },
    {
      "id": 308419,
      "title": "How to Submit",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 307876,
      "title": "Example Code of Seq-NMS",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 307598,
      "title": "🥇Top Solutions thread: Search for Golden Knowledge",
      "comment_count": 1,
      "views": 0,
      "votes": 10
    },
    {
      "id": 307891,
      "title": "Any Top Private Subs with Cascade RCNN",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 307617,
      "title": "115th (would have been ~48th ) - WBF ensemble (same model)",
      "comment_count": 5,
      "views": 0,
      "votes": 9
    },
    {
      "id": 307623,
      "title": "Efficientnet-b0 Stage 2 Post Process",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 307603,
      "title": "Sad...First competition on Kaggle and missed out on a silver medal...",
      "comment_count": 7,
      "views": 0,
      "votes": 8
    },
    {
      "id": 307588,
      "title": "How much LB did you gain after fixing the dataset ",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 306751,
      "title": "New release - Computer Vision: Algorithms and Applications, 2nd ed. ",
      "comment_count": 7,
      "views": 0,
      "votes": 27
    },
    {
      "id": 307295,
      "title": "If you want to improve the performance, try it.🔥🔥🔥",
      "comment_count": 7,
      "views": 0,
      "votes": 9
    },
    {
      "id": 290011,
      "title": "Object Detection Tutorials with YOLO | Free ",
      "comment_count": 4,
      "views": 0,
      "votes": 19
    },
    {
      "id": 307173,
      "title": "Visualized models with low scores revealed some labeling errors in the official test set",
      "comment_count": 7,
      "views": 0,
      "votes": 11
    },
    {
      "id": 307599,
      "title": "LB Earthquake !",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 307480,
      "title": "Will it Shake up like \"Jigsaw Rate Severity of Toxic Comments\"?",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 307928,
      "title": "Confidence in your models",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 307635,
      "title": "Shake Up is a question, how to estimate it?",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 307209,
      "title": "Which 4 submissions will you choose?",
      "comment_count": 8,
      "views": 0,
      "votes": 6
    },
    {
      "id": 307714,
      "title": "About when will the private leaderboard be finalized?",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 307572,
      "title": "Final: Public LB Score Distribution Plotted",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 307659,
      "title": "my forked unselected  notebook with 0.688, however",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 307908,
      "title": "Do you use mmdet?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 307664,
      "title": "Question about dataset",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 307174,
      "title": "New to Kaggle",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 305725,
      "title": "High score notebook (again!) in the lastest stage of the competition",
      "comment_count": 18,
      "views": 0,
      "votes": 45
    },
    {
      "id": 307579,
      "title": "Submission - 2 hours before the deadline",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 306125,
      "title": "Is there a way to beat 0.665 LB notebook?",
      "comment_count": 18,
      "views": 0,
      "votes": 6
    },
    {
      "id": 306922,
      "title": "Problem with norfair ! anyone facing a problem currently ? ",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 307485,
      "title": "Will the notebook be invalid if it runs past the deadline?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 307087,
      "title": "Can someone share some ressources where I can learn about this tracking business?",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 306129,
      "title": "What helped me in this competition in terms of the current leaderboard",
      "comment_count": 8,
      "views": 0,
      "votes": 18
    },
    {
      "id": 302586,
      "title": "SAHI - new tutorial and notebook - Slicing Aided Hyper Inference for Yolov5 and YoloX",
      "comment_count": 36,
      "views": 0,
      "votes": 68
    },
    {
      "id": 306143,
      "title": "About the Uncertainty of the Winning Model",
      "comment_count": 7,
      "views": 0,
      "votes": 9
    },
    {
      "id": 307107,
      "title": "No matching distribution found for setuptools>=42!!!",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 307128,
      "title": "So, will anyone win TF prize?",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 305557,
      "title": "Is there a way to beat the 0.65 F2 score on YOLOV5?",
      "comment_count": 6,
      "views": 0,
      "votes": 8
    },
    {
      "id": 304725,
      "title": "Something in the wrong way! High resolution and big model is not the key point.",
      "comment_count": 70,
      "views": 0,
      "votes": 50
    },
    {
      "id": 307297,
      "title": "Submission Scoring Error. Support team required help to submit code for competition. ",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 306697,
      "title": "Will \"Submit Queued\" affect the notebook run-time?",
      "comment_count": 5,
      "views": 0,
      "votes": 5
    },
    {
      "id": 305792,
      "title": "Some annotated COTs from internet (429 images, 2021 labels)",
      "comment_count": 7,
      "views": 0,
      "votes": 30
    },
    {
      "id": 307260,
      "title": "Vote if you have used Fasterrcnn. And comment your acc",
      "comment_count": 8,
      "views": 0,
      "votes": -10
    },
    {
      "id": 306412,
      "title": "Has ever anyone tried this one? cropped cots",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 306647,
      "title": "Need Help with my first competition",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 302057,
      "title": "LB probing results",
      "comment_count": 19,
      "views": 0,
      "votes": 89
    },
    {
      "id": 306573,
      "title": "Can I use an imported model to make predictions in a submission entry?",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 306178,
      "title": "Homemade Yolo like model",
      "comment_count": 5,
      "views": 0,
      "votes": 4
    },
    {
      "id": 306211,
      "title": "YOLOv5 useful datasets for offline inference",
      "comment_count": 1,
      "views": 0,
      "votes": 10
    },
    {
      "id": 306772,
      "title": "Do not set category_id in COCO to 0",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 305626,
      "title": "are we overfitting the LB?",
      "comment_count": 30,
      "views": 0,
      "votes": 22
    },
    {
      "id": 306039,
      "title": "public score vs private score🔥🔥🔥",
      "comment_count": 8,
      "views": 0,
      "votes": 9
    },
    {
      "id": 306091,
      "title": "[Product Update Clarification] Run-time Increased??",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 306405,
      "title": "what would be the winning solutions?",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 305367,
      "title": "Made a dataset of just the cropped COTS images 👍",
      "comment_count": 7,
      "views": 0,
      "votes": 24
    },
    {
      "id": 305156,
      "title": "Code competition is quite hard...",
      "comment_count": 16,
      "views": 0,
      "votes": 6
    },
    {
      "id": 303988,
      "title": "Generate your own unlimited COTS - GAN Data Augumentation",
      "comment_count": 17,
      "views": 0,
      "votes": 24
    },
    {
      "id": 306175,
      "title": "Clarification of Rule: Is Accessing submission.csv is Officially Prohibited?",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 306443,
      "title": "Submission taking over 2 hours?!",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 304945,
      "title": "Training with Google Colab Overnight",
      "comment_count": 12,
      "views": 0,
      "votes": 13
    },
    {
      "id": 305694,
      "title": "I didn't understand how kaggle works. I can't submit my results.",
      "comment_count": 7,
      "views": 0,
      "votes": 4
    },
    {
      "id": 306083,
      "title": "Has the current Yolo series models come to the end of improving PB?",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 304342,
      "title": "Does anyone use TF and have standing in a medal zone?",
      "comment_count": 8,
      "views": 0,
      "votes": 11
    },
    {
      "id": 305384,
      "title": "Albumentations (Augmentation) don't work with YOLOV5 on Colab (by default)",
      "comment_count": 5,
      "views": 0,
      "votes": 7
    },
    {
      "id": 305831,
      "title": "Submission Query",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 305838,
      "title": "Tuple Index out of Range for non_max_suppression Function",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 306206,
      "title": "Can anyone please help me submitting . It just tell limit exceed",
      "comment_count": 11,
      "views": 0,
      "votes": -2
    },
    {
      "id": 305532,
      "title": "Which bboxes_type for YOLOR?",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 306195,
      "title": "Error while importing tensorflow today",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 302241,
      "title": "Logging F2 score while training YOLOv5",
      "comment_count": 28,
      "views": 0,
      "votes": 79
    },
    {
      "id": 303861,
      "title": "Higher Image Resolution CV-LB Unstable Correlation",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 305801,
      "title": "Imperfect annotation - the primary reason why some ideas (augmentations/big model/higher resolution) does not work?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 305355,
      "title": "Why is my code running multiple times locally with very different results？",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 301086,
      "title": "Situation on Leaderboard🤔!!",
      "comment_count": 16,
      "views": 0,
      "votes": 16
    },
    {
      "id": 305449,
      "title": "Question about localization ?",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 304923,
      "title": "How to reproduce the best model? (YOLOv5 0.631)",
      "comment_count": 7,
      "views": 0,
      "votes": 12
    },
    {
      "id": 303793,
      "title": "0.628 @ 6000 img_size with yolov5s",
      "comment_count": 10,
      "views": 0,
      "votes": 13
    },
    {
      "id": 304993,
      "title": "GAN Training Procedure - Make unlimited COTS",
      "comment_count": 3,
      "views": 0,
      "votes": 8
    },
    {
      "id": 304104,
      "title": "Yolov5 feature maps - see how network understand our starfishes …",
      "comment_count": 8,
      "views": 0,
      "votes": 38
    },
    {
      "id": 305154,
      "title": "problem for submission again?",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 305526,
      "title": "How to add appropriate tags in the notebook while making it public.",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 304367,
      "title": "fyi: underwater object detection algorithm contest",
      "comment_count": 4,
      "views": 0,
      "votes": 10
    },
    {
      "id": 304354,
      "title": "Are COTS edible .. by humans?",
      "comment_count": 5,
      "views": 0,
      "votes": 8
    },
    {
      "id": 304235,
      "title": "How to get sequence nr. of test set using Kaggle API?",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 304917,
      "title": "TIDE: A General Toolbox for Identifying Object Detection Errors",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 304641,
      "title": "<3 Does data enhancement work? ",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 301569,
      "title": "Great Barrier Reef with CenterNet2",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 305347,
      "title": "How to preserve files in the working directory? (I'm having trouble)",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 304749,
      "title": "Missing Annotations in train images csv",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 304954,
      "title": "Issues with MMDet and Sahi",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 304338,
      "title": "motion vector for free",
      "comment_count": 1,
      "views": 0,
      "votes": 12
    },
    {
      "id": 301528,
      "title": "Is this competition now Horse Run /GPU war?",
      "comment_count": 19,
      "views": 0,
      "votes": 23
    },
    {
      "id": 303021,
      "title": "How to reset test env",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 303723,
      "title": "SNIPER for Large Scale Training",
      "comment_count": 5,
      "views": 0,
      "votes": 13
    },
    {
      "id": 303930,
      "title": "any one try multi-fame detection method yet?",
      "comment_count": 6,
      "views": 0,
      "votes": 16
    },
    {
      "id": 304563,
      "title": "How important is it to participate in competitions for a successful career as a data scientist?",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 303839,
      "title": "What is your best ensembling score?",
      "comment_count": 20,
      "views": 0,
      "votes": null
    },
    {
      "id": 304165,
      "title": "Masks dataset for unique COTS",
      "comment_count": 3,
      "views": 0,
      "votes": 10
    },
    {
      "id": 305019,
      "title": "Can't Ensemble Yolor and Yolov5",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 303791,
      "title": "Style Transfer - Pytorch",
      "comment_count": 6,
      "views": 0,
      "votes": 10
    },
    {
      "id": 304414,
      "title": "Vision Transformers \"promising for future uses in object detection\" NeurIPS21GoogleBrainTeam",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 301728,
      "title": "What is the best solution for yolov5?",
      "comment_count": 11,
      "views": 0,
      "votes": 2
    },
    {
      "id": 300405,
      "title": "[placeholder] how to get lb 0.560 with single model",
      "comment_count": 55,
      "views": 0,
      "votes": 152
    },
    {
      "id": 303868,
      "title": "Help needed for understanding confidence score",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 301210,
      "title": "These COTS do not exist",
      "comment_count": 26,
      "views": 0,
      "votes": 43
    },
    {
      "id": 304293,
      "title": "yolov5 map vs pycocotools map difference",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 304067,
      "title": "TensorFlow video creation notebook",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 304721,
      "title": "How to use SAHI for image larger than 1280? ",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 303766,
      "title": "Why Yolov5s6 gives the best result in this game",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 303466,
      "title": "Submission format",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 303984,
      "title": "Train vs Inference notebooks",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 303532,
      "title": "Augmentations with YOLOV5",
      "comment_count": 13,
      "views": 0,
      "votes": 4
    },
    {
      "id": 303740,
      "title": "How many Grandmasters can one competition have?!!?",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 302965,
      "title": "YOLOv5 and YOLOR realizations are licensed by GPLv3 . Is this a problem?",
      "comment_count": 9,
      "views": 0,
      "votes": 21
    },
    {
      "id": 304143,
      "title": "Difference Between AP and this Competition's Evaluation Metrics",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 303919,
      "title": "Cots don't move. The key to tracking?",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 304331,
      "title": "About the real-time aspect of the competition",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 303565,
      "title": "Any other way to avoid missing cots rather than high resolution?",
      "comment_count": 4,
      "views": 0,
      "votes": 8
    },
    {
      "id": 304155,
      "title": "Public LB and Private LB",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 304250,
      "title": "Weird CV and LB",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 302632,
      "title": "[Solved] Something Odd in the LB Evaluation - 10K of Additional FPs Affect Little on the LB Score",
      "comment_count": 29,
      "views": 0,
      "votes": 46
    },
    {
      "id": 304058,
      "title": "Yellowish video bug",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 301703,
      "title": "what resolution did you use？",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 303788,
      "title": "how to predict with yolov4 weights...?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 304219,
      "title": "Is it OK to use notebook as dataset during inference?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 302985,
      "title": "How to handle bounding boxes overlapping with edges when using tiling/cropping?",
      "comment_count": 13,
      "views": 0,
      "votes": 7
    },
    {
      "id": 303690,
      "title": "Starfish annotation visualizer",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 304148,
      "title": "File conversion without data compression",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 303567,
      "title": "Darknet and models.models not found on my notebook!!!",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 303438,
      "title": "TF-OD Export Inference graph and model inferencing.",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 303958,
      "title": "non_max_suppression() Got an Unexpected Keyword Argument 'merge'",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 303996,
      "title": "I have one doubt. Is there a pretrained model available with faster-rcnn with resnet50 fpn trained on imagenet ",
      "comment_count": 3,
      "views": 0,
      "votes": -6
    },
    {
      "id": 303634,
      "title": "Yolov5 zero score?",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 303872,
      "title": "Download Test Images",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 301287,
      "title": "Alternatives to google colab pro",
      "comment_count": 12,
      "views": 0,
      "votes": 6
    },
    {
      "id": 302761,
      "title": "How to implement your own Yolo by using SAHI",
      "comment_count": 5,
      "views": 0,
      "votes": 16
    },
    {
      "id": 303412,
      "title": "Could sb. help me? I can't make submissions",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 303344,
      "title": "Predicting from the saved model.",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 302544,
      "title": "Compare Yolov5, YoloX, YolorR.",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 303239,
      "title": "TensorFlow Learning helper Kit.",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 303234,
      "title": "COTSBOT was getting 99.4% accuracy.",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 302937,
      "title": "Can somone give feedback about my code?",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 303232,
      "title": "\"GPU Notebook <= 9 hours run-time\" meaning? on public / private?",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 301694,
      "title": "FFCV - a library to train fast! ⚡",
      "comment_count": 3,
      "views": 0,
      "votes": 14
    },
    {
      "id": 303289,
      "title": "Resources for learning about different versions of YOLO and RCNN",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 302130,
      "title": "Here is the magic: Estimating recall and precision from your public LB score",
      "comment_count": 11,
      "views": 0,
      "votes": 23
    },
    {
      "id": 302605,
      "title": "LB probing result: GT Labels per Frame in the Public LB",
      "comment_count": 3,
      "views": 0,
      "votes": 14
    },
    {
      "id": 302384,
      "title": "How many epochs should we train with?",
      "comment_count": 10,
      "views": 0,
      "votes": 6
    },
    {
      "id": 302779,
      "title": "Leveraging temporal information",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 302750,
      "title": "Pytorch added FCOS support!",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 301959,
      "title": "LB (TOP20) solutions SPEED 💥💥  analysis ",
      "comment_count": 17,
      "views": 0,
      "votes": 49
    },
    {
      "id": 302741,
      "title": "Yolov5 with higher size being timed out",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 300960,
      "title": "How to Handle When Learning Curve Become Unstable?",
      "comment_count": 16,
      "views": 0,
      "votes": 6
    },
    {
      "id": 302443,
      "title": "how is the batch size related to LB using yolov5?",
      "comment_count": 4,
      "views": 0,
      "votes": 7
    },
    {
      "id": 301217,
      "title": "failed to submit problem",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 302577,
      "title": "Magic Three: Rough Estimate of The Number of Tiny COTS in the Public LB",
      "comment_count": 4,
      "views": 0,
      "votes": 10
    },
    {
      "id": 301014,
      "title": "LB above 0.7! Wow!",
      "comment_count": 21,
      "views": 0,
      "votes": 42
    },
    {
      "id": 302636,
      "title": "[YOLOX] Installation package of latest YOLOX(0.2.0)  for competition environment",
      "comment_count": 5,
      "views": 0,
      "votes": 6
    },
    {
      "id": 302718,
      "title": "Processing time",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 298046,
      "title": "Happy New Year  starfish seekers and see you in .... 2022 ...",
      "comment_count": 15,
      "views": 0,
      "votes": 22
    },
    {
      "id": 301847,
      "title": "Does kaggle GPU is enough to get top 100? ",
      "comment_count": 13,
      "views": 0,
      "votes": 6
    },
    {
      "id": 302156,
      "title": "Magic Two: We Can Estimate Number of GT Labels on the Public Test Data",
      "comment_count": 8,
      "views": 0,
      "votes": 7
    },
    {
      "id": 301336,
      "title": "Will splitting image into a grid, give us higher score?",
      "comment_count": 11,
      "views": 0,
      "votes": 9
    },
    {
      "id": 302411,
      "title": "Do we need sliding window ???",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 302770,
      "title": "Has anyone tried Aux Loss ? ",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 302624,
      "title": "Why does a higher resolution get a higher score？？？",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 302740,
      "title": "Object Detection Intro",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 299976,
      "title": "there is one more image you can use for training",
      "comment_count": 26,
      "views": 0,
      "votes": 51
    },
    {
      "id": 300491,
      "title": "I am confused about submission process",
      "comment_count": 10,
      "views": 0,
      "votes": 1
    },
    {
      "id": 297774,
      "title": "Tips about \"Submission Scoring Error\"",
      "comment_count": 23,
      "views": 0,
      "votes": 4
    },
    {
      "id": 302198,
      "title": "Does augmentation and higher resolution really improve LB score? 🤓",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 302491,
      "title": "AttributeError: 'NoneType' object has no attribute 'python_exit_status' [YOLOv5]",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 296721,
      "title": "Competition logbook - ideas to IMPROVE LB score 💪💥",
      "comment_count": 26,
      "views": 0,
      "votes": 129
    },
    {
      "id": 301452,
      "title": "About the Training Reproducibility of LB: .579",
      "comment_count": 7,
      "views": 0,
      "votes": 25
    },
    {
      "id": 301494,
      "title": "Pacific Triton snail, the natural predator",
      "comment_count": 5,
      "views": 0,
      "votes": 17
    },
    {
      "id": 302441,
      "title": "Anyone Solved this problem??",
      "comment_count": 6,
      "views": 0,
      "votes": 0
    },
    {
      "id": 300768,
      "title": "Use timm pretrained backbones for you FasterRCNN!",
      "comment_count": 25,
      "views": 0,
      "votes": 22
    },
    {
      "id": 293723,
      "title": "Be CAREFUL with your TRAIN/VALID splits and avoid LB GAPS",
      "comment_count": 41,
      "views": 0,
      "votes": 128
    },
    {
      "id": 301271,
      "title": "Question about the metric",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 301956,
      "title": "Whats your best model except the YOLO family(v5/x/r etc)?",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 302394,
      "title": "Missing Annotations ",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 296905,
      "title": "Which Colab for competition ",
      "comment_count": 39,
      "views": 0,
      "votes": 37
    },
    {
      "id": 302231,
      "title": "Testing the limits of YOLO V1's performance",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 290005,
      "title": "🔥🔥Research Trends in Object Detection ",
      "comment_count": 20,
      "views": 0,
      "votes": 81
    },
    {
      "id": 302187,
      "title": "LB gap,   new model or  engineering features and ensemble?",
      "comment_count": 3,
      "views": 0,
      "votes": -1
    },
    {
      "id": 291541,
      "title": "Tensorflow Bonus Prize clarification",
      "comment_count": 18,
      "views": 0,
      "votes": 40
    },
    {
      "id": 299804,
      "title": "Why relation between CV and LB is not linear ? ",
      "comment_count": 6,
      "views": 0,
      "votes": 9
    },
    {
      "id": 290757,
      "title": "Best Single Model CV-LB",
      "comment_count": 207,
      "views": 0,
      "votes": 109
    },
    {
      "id": 301514,
      "title": "Will there be a big shakeup?",
      "comment_count": 4,
      "views": 0,
      "votes": 7
    },
    {
      "id": 301883,
      "title": "Very low score in submission",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 299870,
      "title": "[Kaggle Error] Account Bans & All Submissions throwing Error",
      "comment_count": 22,
      "views": 0,
      "votes": 17
    },
    {
      "id": 301295,
      "title": "How long do you train? - Discussion on epoch number (or training resource for poor)",
      "comment_count": 15,
      "views": 0,
      "votes": 8
    },
    {
      "id": 301664,
      "title": "Trade off inference time",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 301668,
      "title": "StylEx: A New Approach for Visual Explanation of Classifiers",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 301921,
      "title": "【solved-submission score error】%cd /kaggle/working before 'inference'",
      "comment_count": 4,
      "views": 0,
      "votes": -1
    },
    {
      "id": 301132,
      "title": "Training High Res ✅ Inference High Res ⚠️",
      "comment_count": 5,
      "views": 0,
      "votes": 16
    },
    {
      "id": 302162,
      "title": "How to achieve cross validation in object detection?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 302135,
      "title": "how can i infer the testimgs,i dont know where they are!!!",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 301748,
      "title": "[Why?]hubconf.py is not loaded. ",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 301766,
      "title": "16G GPU RAM & 9 hours inference Resource Allocation Competition",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 301923,
      "title": "How should add the no-bbox data to the validation data?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 301577,
      "title": "anyone gets good LB  using yolox_x model?",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 298514,
      "title": "No ensemble No knowledge distillation, what's your LB score?",
      "comment_count": 49,
      "views": 0,
      "votes": 29
    },
    {
      "id": 302054,
      "title": "Lot of LB movements",
      "comment_count": 1,
      "views": 0,
      "votes": -3
    },
    {
      "id": 302029,
      "title": "Train from scratch?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 296528,
      "title": "Submission scoring error",
      "comment_count": 14,
      "views": 0,
      "votes": 1
    },
    {
      "id": 301854,
      "title": "How long approximately takes to get a public score after the submission?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 301857,
      "title": "Is GPU used during submission testing?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 301541,
      "title": "Generate External Datasets With 3D model",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 301157,
      "title": "TransMix:Attend to Mix for VIT",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 297696,
      "title": "Let me introduce YoloR … :) my next yolo notebook ❤️",
      "comment_count": 21,
      "views": 0,
      "votes": 39
    },
    {
      "id": 301239,
      "title": "yolov5 training, the output of training info not complete problem",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 301493,
      "title": "How  to use unlabeled data?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 300959,
      "title": "Why do we have 4 final subs?",
      "comment_count": 9,
      "views": 0,
      "votes": 4
    },
    {
      "id": 300808,
      "title": "Happy to get Gold Medal 🥇 on My Notebook. 🎉🎊🥳",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 300601,
      "title": "copy kaggle dataset to google drive",
      "comment_count": 4,
      "views": 0,
      "votes": 9
    },
    {
      "id": 301177,
      "title": "get output of another notebook",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 289994,
      "title": "Looking for a Team Megathread",
      "comment_count": 266,
      "views": 0,
      "votes": 41
    },
    {
      "id": 301020,
      "title": "Does 9 hour limit also applies to private set?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 301265,
      "title": "Tensorflow Best Model CV-LB (on GPU)",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 301045,
      "title": "Unable to access data uploaded in \"/kaggle/working\" from another notebook",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 299861,
      "title": "kaggle error?",
      "comment_count": 11,
      "views": 0,
      "votes": 3
    },
    {
      "id": 300574,
      "title": "google colab : kaggle api: competition download does not contain all",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 298427,
      "title": "Why higher mAP leads to a significantly lower public score?",
      "comment_count": 33,
      "views": 0,
      "votes": 26
    },
    {
      "id": 300918,
      "title": "YOLOv5 on Google Colaboratory Issue and error handling",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 290345,
      "title": "Submission Error",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 290235,
      "title": "A cross-validation strategy: subsequences",
      "comment_count": 11,
      "views": 0,
      "votes": 37
    },
    {
      "id": 300647,
      "title": "merge team problem.",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 300087,
      "title": "Error when running my model on the test data",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 300307,
      "title": "problem editing notebook",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 299356,
      "title": "track id for train dataset!",
      "comment_count": 6,
      "views": 0,
      "votes": 21
    },
    {
      "id": 298498,
      "title": "How to add FPN to any timm backbone model?",
      "comment_count": 9,
      "views": 0,
      "votes": 4
    },
    {
      "id": 299930,
      "title": "❗Warning❗ Don't run your notebook now!",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 293812,
      "title": "Papers on Video Object Detection",
      "comment_count": 17,
      "views": 0,
      "votes": 75
    },
    {
      "id": 300129,
      "title": "how to union  the detected bboxes  based on IoU?",
      "comment_count": 10,
      "views": 0,
      "votes": 1
    },
    {
      "id": 290795,
      "title": "Notebook Exceeded Allowed Compute",
      "comment_count": 7,
      "views": 0,
      "votes": 4
    },
    {
      "id": 299409,
      "title": "Yolov5 model performance is not up to mark.",
      "comment_count": 15,
      "views": 0,
      "votes": 2
    },
    {
      "id": 299156,
      "title": "New Year plans?",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 299001,
      "title": "YoloR on Kaggle - TRAIN + INFER",
      "comment_count": 9,
      "views": 0,
      "votes": 19
    },
    {
      "id": 296788,
      "title": "Question to the guys at top of the leaderboard",
      "comment_count": 5,
      "views": 0,
      "votes": 14
    },
    {
      "id": 300090,
      "title": "What do you think we be the cutoff for gold at the end of the competition? (Public Score)",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 299154,
      "title": "🔥🔥🔥Compilation of all the ways to work with Huge Datasets",
      "comment_count": 3,
      "views": 0,
      "votes": 11
    },
    {
      "id": 294205,
      "title": "Journey from LB 0.038 ---> 0.502 ---> 0.515 🖤 🔥",
      "comment_count": 44,
      "views": 0,
      "votes": 55
    },
    {
      "id": 299721,
      "title": "Yolo Weights in Keras TensorFlow",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 299457,
      "title": "Yolox tables and charts",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 296505,
      "title": "can I train my model on my own server?",
      "comment_count": 11,
      "views": 0,
      "votes": 2
    },
    {
      "id": 299598,
      "title": "YOLOX Output Download Failing",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 298588,
      "title": "[solved] Submission Scoring Error - we've tried everything in the other threads",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 299472,
      "title": "Help: Error when submitting notebook",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 299143,
      "title": "GPU Traffic ?!",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 291663,
      "title": "YoloX Kaggle notebook - I made it! ❤️",
      "comment_count": 31,
      "views": 0,
      "votes": 64
    },
    {
      "id": 298877,
      "title": "conf = 0.13 too small?",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 292807,
      "title": "Is This False Positive?",
      "comment_count": 6,
      "views": 0,
      "votes": 23
    },
    {
      "id": 297875,
      "title": "Concealed Object Detection",
      "comment_count": 1,
      "views": 0,
      "votes": 44
    },
    {
      "id": 299116,
      "title": "Why can't I use GPU?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 297779,
      "title": "Image vs Bounding Box Augmentation",
      "comment_count": 6,
      "views": 0,
      "votes": 24
    },
    {
      "id": 299108,
      "title": "problem for downloading output file?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 298738,
      "title": "How to configure machine setup to handle ML models?",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 298535,
      "title": "Data-centric: have you noticed unlabeled objects in training set?",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 298559,
      "title": "test dataset size?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 297865,
      "title": "Notebook score ... my thoughts about voting and forking",
      "comment_count": 14,
      "views": 0,
      "votes": 18
    },
    {
      "id": 298972,
      "title": "Your notebook tried to allocate more memory than is available. It has restarted.",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 291021,
      "title": "One ethical question, does Starfishes would be killed after Our code finds them?",
      "comment_count": 19,
      "views": 0,
      "votes": 51
    },
    {
      "id": 290472,
      "title": "🔥🔥 Congratulations to Awsaf for becoming Notebooks GrandMaster",
      "comment_count": 40,
      "views": 0,
      "votes": 37
    },
    {
      "id": 298340,
      "title": "[solved] Help! yet another \"Submission Scoring Error\"",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 290042,
      "title": "YOLOv5: Here we go again | LB: 0.45+",
      "comment_count": 23,
      "views": 0,
      "votes": 29
    },
    {
      "id": 290152,
      "title": "🔥🔥 Congratulations to DrHB for becoming Discussion GrandMaster",
      "comment_count": 28,
      "views": 0,
      "votes": 41
    },
    {
      "id": 294767,
      "title": "Help needed with Tensorflow Object Detection API",
      "comment_count": 9,
      "views": 0,
      "votes": 1
    },
    {
      "id": 295597,
      "title": "How are the false positives measured in the metric?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 298503,
      "title": "Are there only three images in the public test set?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 298110,
      "title": "Happy New Year 2022 ",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 294854,
      "title": "COCO style competition metric",
      "comment_count": 10,
      "views": 0,
      "votes": 24
    },
    {
      "id": 298538,
      "title": "Scaled YOLOv4 and submission error?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 297994,
      "title": "Video Object Detection Models",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 296903,
      "title": "Submission Scoring Error",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 298274,
      "title": "Could not list notebooks: Failed to fetch",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 297213,
      "title": "Why internet option needs to keep off during inference on kaggle?",
      "comment_count": 7,
      "views": 0,
      "votes": 4
    },
    {
      "id": 298106,
      "title": "Training EfficientDet with EfficientNetv2 as backbone, Blog Article.",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 296171,
      "title": "Anyone encountered mAP&F2-score drop problem?",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 297534,
      "title": "YOLOX resources for beginners",
      "comment_count": 2,
      "views": 0,
      "votes": 16
    },
    {
      "id": 290083,
      "title": "f2 score implementation",
      "comment_count": 7,
      "views": 0,
      "votes": 22
    },
    {
      "id": 297930,
      "title": "Yolo V5 (Head layer)",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 297135,
      "title": "Faced the following error while training FasterRCNN ",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 293601,
      "title": "[Notebook Threw Exception] Solve",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 297429,
      "title": "Consused about yolox output bbox coordinates.",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 297747,
      "title": "Inference Tips for YOLOv5 beginner?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 296533,
      "title": "Problem of the Ground truth bboxes in train dataset images",
      "comment_count": 5,
      "views": 0,
      "votes": 14
    },
    {
      "id": 294782,
      "title": "Submission CSV not found",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 296170,
      "title": "How to handle unlabeled COTS data ? ",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 290584,
      "title": "Fast underwater image enhancement for Improved Visual Perception [github+Paper]",
      "comment_count": 28,
      "views": 0,
      "votes": 89
    },
    {
      "id": 297290,
      "title": "Submission Time",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 296531,
      "title": "Recommended methods for exploratory data analysis on this dataset?",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 295782,
      "title": "Advice for low computing resources",
      "comment_count": 8,
      "views": 0,
      "votes": 8
    },
    {
      "id": 296464,
      "title": "NN Easily Fooled by Strange Poses. Not Madonna. But Inception, Yolo and Coco.",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 296024,
      "title": "Anyone succeeds with sequence_frame?",
      "comment_count": 7,
      "views": 0,
      "votes": 10
    },
    {
      "id": 291547,
      "title": "Error while importing greatbarrierreef",
      "comment_count": 13,
      "views": 0,
      "votes": 2
    },
    {
      "id": 296119,
      "title": "TF submission API image array format",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 296065,
      "title": "is the test dataset dynamically changing?　　",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 296434,
      "title": "YOLOX training",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 295400,
      "title": "Saving and Loading Trained Models for Submission",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 296257,
      "title": "Is there a way to continue training in YOLOx?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 296312,
      "title": "Trouble Saving Persistent Data",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 295885,
      "title": "why  I got   -1.00 for recall and precision?",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 295275,
      "title": "Error with installing Tensorflow with GPU",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 295218,
      "title": "How is bad performance model by human eyes scores much better on LB",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 294489,
      "title": "Any solution to Yolo \"Exception: Cache may be out of date, try `force_reload=True`\"",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 295288,
      "title": "Object Detection in TF2.0",
      "comment_count": 1,
      "views": 0,
      "votes": 8
    },
    {
      "id": 295630,
      "title": "TensorFlow OD API Pipeline [0.474]",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 294175,
      "title": "NVIDIA TensorRT | Speeding Up Inference | 6X faster | TensorFlow & PyTorch | One line of code",
      "comment_count": 2,
      "views": 0,
      "votes": 15
    },
    {
      "id": 293719,
      "title": "Starting Tips Required",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 294869,
      "title": "Better model but lower score",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 293305,
      "title": "Configs - TensorFlow OD API",
      "comment_count": 3,
      "views": 0,
      "votes": 16
    },
    {
      "id": 294794,
      "title": "Using previous frame's annotations (Kalman filter)?",
      "comment_count": 1,
      "views": 0,
      "votes": 7
    },
    {
      "id": 295352,
      "title": "anyone tried TTA(Test Time Augmentation)?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 295442,
      "title": "anyone uses extra-dataset? is It allowed or not?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 295430,
      "title": "Submission CSV Not Found!!!!!!!",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 295568,
      "title": "Looking for Data scientists & Data architects for Marine Conservation",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 293642,
      "title": "Pip error: could not create 'build': Read-only file system",
      "comment_count": 7,
      "views": 0,
      "votes": 8
    },
    {
      "id": 295494,
      "title": "TensorFlow tutorial for beginners. | Best tutorial content that I've found so far. 💯✔️",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 294989,
      "title": "Question on Competition API",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 294084,
      "title": "Balanced Fold Splitting",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 294589,
      "title": "16 Object detection papers in NeurIPS 2021 | You can find them here | with code and abstract",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 290353,
      "title": "Useful External Datasets, Models and  Resources ",
      "comment_count": 3,
      "views": 0,
      "votes": 90
    },
    {
      "id": 291261,
      "title": "SOTA model: Swin Transformer V2",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 294540,
      "title": "Train then Infer Notebook",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 293415,
      "title": "How can I make an Object Detection model on TF?",
      "comment_count": 1,
      "views": 0,
      "votes": 11
    },
    {
      "id": 294221,
      "title": "Articles about Small Object Detection",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 294389,
      "title": "Does the test set contain images without targets？",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 294013,
      "title": "Leading SOTA Object Detection Algorithms?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 294410,
      "title": "Looking for a team",
      "comment_count": 2,
      "views": 0,
      "votes": -2
    },
    {
      "id": 294199,
      "title": "Kaggle Output download UI problem",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 291645,
      "title": "Why same annotation a lot?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 291694,
      "title": "Customize your Faster RCNN",
      "comment_count": 4,
      "views": 0,
      "votes": 20
    },
    {
      "id": 292369,
      "title": "A CRISPR gene drive could make cots extinct",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 291063,
      "title": " 8 Methods on Underwater Image Enhancement and Color Restoration, With Code",
      "comment_count": 3,
      "views": 0,
      "votes": 52
    },
    {
      "id": 293796,
      "title": "Mobile Data Usage While Training Models ?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 293607,
      "title": "I am Interested to join team",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 291520,
      "title": "Future frames are available in Time-series API",
      "comment_count": 3,
      "views": 0,
      "votes": 32
    },
    {
      "id": 292223,
      "title": "Interesting Video",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 290651,
      "title": "What's wrong with Kaggle's Hotness Algorithm?",
      "comment_count": 12,
      "views": 0,
      "votes": 21
    },
    {
      "id": 290350,
      "title": "Problem statement is not only Object detection it's \"Video\" Object Detection, given data is in a video sequence ",
      "comment_count": 14,
      "views": 0,
      "votes": 33
    },
    {
      "id": 290675,
      "title": "Bounding box exceeding image height",
      "comment_count": 3,
      "views": 0,
      "votes": 39
    },
    {
      "id": 291674,
      "title": "Offline python module installation ",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 290411,
      "title": "Generating additional labeling data",
      "comment_count": 2,
      "views": 0,
      "votes": 15
    },
    {
      "id": 290216,
      "title": "New to ML and object detection",
      "comment_count": 5,
      "views": 0,
      "votes": 10
    },
    {
      "id": 290062,
      "title": "The Great Barrier Reef and COTS (Crown-of-thorns starfish)",
      "comment_count": 9,
      "views": 0,
      "votes": 35
    },
    {
      "id": 292366,
      "title": "failed to create dataset",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 291348,
      "title": "Problem with f2_score if len(pred_bboxes) > len(gt_boxes)?",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 291902,
      "title": "Will the test set images have the \"occasional gaps\" as in the training set?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 290961,
      "title": "Missing Annotations",
      "comment_count": 1,
      "views": 0,
      "votes": 11
    },
    {
      "id": 291828,
      "title": "can't quick save my interactive notebook result?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 291416,
      "title": "PIP installation problem",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 291047,
      "title": "Object/Frame-level Temporal Aggregation for Better Video Object Detection [Github awesome list] ",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 290992,
      "title": "EfficientDet",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 291014,
      "title": "PyTorch update for internet off kernel",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 290145,
      "title": "🎥 We use continuous image. Is this useful??",
      "comment_count": 5,
      "views": 0,
      "votes": 7
    },
    {
      "id": 290049,
      "title": "Minor Correction To Data Documentation",
      "comment_count": 2,
      "views": 0,
      "votes": 18
    },
    {
      "id": 291075,
      "title": "[Idea] I wish there are EDA notebooks prizes in each Kaggle competition.",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 290677,
      "title": "External training allowed?",
      "comment_count": 4,
      "views": 0,
      "votes": 8
    },
    {
      "id": 291210,
      "title": "any TPU training starter notebook?",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 291474,
      "title": "Confusion about the number of test images",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 290632,
      "title": "Papers on Similar topic: Object Detection + Crown of Throns Starfish Detection",
      "comment_count": 0,
      "views": 0,
      "votes": 21
    },
    {
      "id": 291489,
      "title": "Anyone Develop on an iPad?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 291123,
      "title": "How to convert pytorch to tensorflow ",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 291038,
      "title": "[Survey]Crown-of-thorns starfish",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 290136,
      "title": "Understanding the Metric : F2 score",
      "comment_count": 2,
      "views": 0,
      "votes": 43
    },
    {
      "id": 290487,
      "title": "Kagglers and Tensorflow Detect. They Control: COTS Control Forum Abstracts 2021.",
      "comment_count": 0,
      "views": 0,
      "votes": 18
    },
    {
      "id": 290057,
      "title": "Pure Pytorch FasterR-CNN Starter [LB=0.416]",
      "comment_count": 7,
      "views": 0,
      "votes": 24
    },
    {
      "id": 290456,
      "title": "Scale Invariance and \"Perspective Scaling\"",
      "comment_count": 2,
      "views": 0,
      "votes": 11
    },
    {
      "id": 290910,
      "title": "F2 /Fbeta implementation as a metric like class",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 290320,
      "title": "It would be much more fun if we have performance leaderboard",
      "comment_count": 2,
      "views": 0,
      "votes": 18
    },
    {
      "id": 290656,
      "title": "Thoughts on Conv3D vs TimeDistributed(Conv2D)? ",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 290228,
      "title": "👍💥 Notebooks of Object detection",
      "comment_count": 2,
      "views": 0,
      "votes": 12
    },
    {
      "id": 290139,
      "title": "Have you heard about F-ß Score? ",
      "comment_count": 0,
      "views": 0,
      "votes": 17
    },
    {
      "id": 290339,
      "title": "Do we must use Tensorflow in this competition?",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 290006,
      "title": "Object Detection by Andrew NG | Free | DeepLearningAI 💯✔️",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 290242,
      "title": "🔥🔥 Tutorial and Paper on Object Detection-(github link)",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 290008,
      "title": "Introduce yourself and Share your idea!",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 290154,
      "title": "How do I detect objects with TensorFlow?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    }
  ],
  "errors": []
}