{
  "total": 283,
  "topics": [
    {
      "id": 299036,
      "title": "Competition is Finalized - Congrats to our Winners; Takeaways",
      "comment_count": 4,
      "views": 0,
      "votes": 12
    },
    {
      "id": 301578,
      "title": "Dear Kagglers, thank you!",
      "comment_count": 0,
      "views": 0,
      "votes": 11
    },
    {
      "id": 279425,
      "title": "Welcome to the Sartorius Cell Instance Segmentation competition!",
      "comment_count": 7,
      "views": 0,
      "votes": 47
    },
    {
      "id": 278584,
      "title": "New to Kaggle or Machine Learning? Come Say Hi!",
      "comment_count": 23,
      "views": 0,
      "votes": 11
    },
    {
      "id": 278585,
      "title": "Looking for a Team Megathread",
      "comment_count": 133,
      "views": 0,
      "votes": 19
    },
    {
      "id": 297985,
      "title": "9 Place solution: pretrain & semi-supervised & postprocessing & Trust both cv and lb",
      "comment_count": 26,
      "views": 0,
      "votes": 44
    },
    {
      "id": 279281,
      "title": "Papers on Neuronal Cells and Machine Learning",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 280137,
      "title": "Tutorial: detectron 2",
      "comment_count": 9,
      "views": 0,
      "votes": 83
    },
    {
      "id": 298869,
      "title": "1st place solution",
      "comment_count": 23,
      "views": 0,
      "votes": 101
    },
    {
      "id": 297988,
      "title": "2nd place solution",
      "comment_count": 77,
      "views": 0,
      "votes": 146
    },
    {
      "id": 278716,
      "title": "Papers on Cell Instance Segmentation 📝👍",
      "comment_count": 9,
      "views": 0,
      "votes": 84
    },
    {
      "id": 298021,
      "title": "3rd place solution",
      "comment_count": 13,
      "views": 0,
      "votes": 54
    },
    {
      "id": 334061,
      "title": "Looking forward to more games like this",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 285447,
      "title": "Unsupervised Instance Segmentation ??",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 298030,
      "title": "22th place solution / for English edition and 中文版本",
      "comment_count": 10,
      "views": 0,
      "votes": 20
    },
    {
      "id": 296148,
      "title": "Could not find provided output file submission.csv",
      "comment_count": 8,
      "views": 0,
      "votes": 5
    },
    {
      "id": 297986,
      "title": "6th place solution. Higher resolution is all you need.",
      "comment_count": 56,
      "views": 0,
      "votes": 72
    },
    {
      "id": 290402,
      "title": "New Data Augmentation Method for Instance Segmentation",
      "comment_count": 23,
      "views": 0,
      "votes": 44
    },
    {
      "id": 298038,
      "title": "11th place solution",
      "comment_count": 30,
      "views": 0,
      "votes": 77
    },
    {
      "id": 298081,
      "title": "5th place solution, maskrcnn, pseudo label and cellpose",
      "comment_count": 15,
      "views": 0,
      "votes": 84
    },
    {
      "id": 299228,
      "title": "(Losing) my first bronze :\"(",
      "comment_count": 4,
      "views": 0,
      "votes": 18
    },
    {
      "id": 298146,
      "title": "4th place solution",
      "comment_count": 14,
      "views": 0,
      "votes": 61
    },
    {
      "id": 297984,
      "title": "Go with the flow (3rd place)",
      "comment_count": 38,
      "views": 0,
      "votes": 124
    },
    {
      "id": 298090,
      "title": "Who else has tried U-Net in this competion? ",
      "comment_count": 5,
      "views": 0,
      "votes": 7
    },
    {
      "id": 289552,
      "title": "What manual manipulation of train data is allowed?",
      "comment_count": 19,
      "views": 0,
      "votes": 17
    },
    {
      "id": 298820,
      "title": "Evaluation : Is there a way to get scores of Classification of type of Cell with detectron2",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 298002,
      "title": "[Viettel.DGD] Train4Ever 7th place Solution",
      "comment_count": 20,
      "views": 0,
      "votes": 38
    },
    {
      "id": 297998,
      "title": "8th place solution",
      "comment_count": 7,
      "views": 0,
      "votes": 54
    },
    {
      "id": 298073,
      "title": "Cellpose origin",
      "comment_count": 3,
      "views": 0,
      "votes": 22
    },
    {
      "id": 297983,
      "title": "Happy Holidays! 🎅",
      "comment_count": 2,
      "views": 0,
      "votes": 22
    },
    {
      "id": 298022,
      "title": "Ensemble Techniques for Instance Segmentation",
      "comment_count": 5,
      "views": 0,
      "votes": 20
    },
    {
      "id": 297713,
      "title": "Shake up estimation",
      "comment_count": 19,
      "views": 0,
      "votes": 15
    },
    {
      "id": 298098,
      "title": "Instance segmentation app on Hugging Face Spaces",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 298005,
      "title": "top solution compilation",
      "comment_count": 2,
      "views": 0,
      "votes": 13
    },
    {
      "id": 298019,
      "title": "An interesting finding!",
      "comment_count": 5,
      "views": 0,
      "votes": 9
    },
    {
      "id": 297908,
      "title": "An easy way saving GPU time when submitting",
      "comment_count": 5,
      "views": 0,
      "votes": 8
    },
    {
      "id": 298620,
      "title": "How to debug : AssertionError: Torch not compiled with CUDA enabled",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 298001,
      "title": "Some curiosities after competition ends",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 297141,
      "title": "Does Copy paste augmentation work effectively?",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 296361,
      "title": "Warnings! Data leakage in the provided LIVECell_dataset_2021.",
      "comment_count": 5,
      "views": 0,
      "votes": 19
    },
    {
      "id": 297710,
      "title": "Inference Time Sharing?",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 293026,
      "title": "How to do transfer learning and get 0.3 score",
      "comment_count": 24,
      "views": 0,
      "votes": 46
    },
    {
      "id": 297841,
      "title": "Can you make a submission that will run on GPU if you finished your GPU quota?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 297800,
      "title": "Has anyone tried creating 'fake' train images?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 287163,
      "title": "Understanding COCO metrics",
      "comment_count": 3,
      "views": 0,
      "votes": 50
    },
    {
      "id": 296803,
      "title": "Similar challenges to continue?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 295539,
      "title": "Semi-supervised Experiments",
      "comment_count": 15,
      "views": 0,
      "votes": 11
    },
    {
      "id": 297582,
      "title": "mAP/CV/LB correlation?",
      "comment_count": 1,
      "views": 0,
      "votes": -1
    },
    {
      "id": 296221,
      "title": "*Notebook threw exception*,help ༼ つ ◕_◕ ༽つ",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 294742,
      "title": "different fold models giving very different results?",
      "comment_count": 12,
      "views": 0,
      "votes": 10
    },
    {
      "id": 294795,
      "title": "Anyone using cellpose notice that diameter has so much impact on LB score?",
      "comment_count": 12,
      "views": 0,
      "votes": 11
    },
    {
      "id": 296955,
      "title": "Some question on how to train mask-rcnn if any expert can help",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 297167,
      "title": "How to train semi supervised with detectron2 or mmdet?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 293033,
      "title": "Let Me Simplify The Problem Statement & Provide Solutions to You😊",
      "comment_count": 1,
      "views": 0,
      "votes": 9
    },
    {
      "id": 296913,
      "title": "There is a small but important problem confused me. ",
      "comment_count": 3,
      "views": 0,
      "votes": 0
    },
    {
      "id": 296408,
      "title": "Annotation bugs? Which annotations should I trust?",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 295603,
      "title": "RLE to mask bug",
      "comment_count": 4,
      "views": 0,
      "votes": 16
    },
    {
      "id": 294823,
      "title": "Model sharing dicussion. E.g., the upper limit of Mask R-CNN in this competition.",
      "comment_count": 14,
      "views": 0,
      "votes": 19
    },
    {
      "id": 295724,
      "title": "Huge difference among class-wise APs",
      "comment_count": 10,
      "views": 0,
      "votes": 9
    },
    {
      "id": 296132,
      "title": "Unable to add detectron 0.5 dataset to my kernel",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 295589,
      "title": "Last Monetary Prizes for 2021",
      "comment_count": 4,
      "views": 0,
      "votes": 26
    },
    {
      "id": 296513,
      "title": "Failed to merge livecell dataset into the competition dataset",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 289456,
      "title": "CV/LB correlation?",
      "comment_count": 37,
      "views": 0,
      "votes": 23
    },
    {
      "id": 295689,
      "title": "Can anyone tell me the situation you met when \"Submission Scoring Error\" occurred?",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 294672,
      "title": "Cellpose is outdated? EMBEDSEG",
      "comment_count": 10,
      "views": 0,
      "votes": 18
    },
    {
      "id": 295020,
      "title": "Discussion on data augmentation.",
      "comment_count": 3,
      "views": 0,
      "votes": 12
    },
    {
      "id": 289033,
      "title": "Best Single Model",
      "comment_count": 85,
      "views": 0,
      "votes": 50
    },
    {
      "id": 295846,
      "title": "Why the MIN_PIXELS = {0: 75, 1:75, 2:150} ? Has anyone done experiment on it?",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 295799,
      "title": "ZeroCostDL4Mic | 38 Notebooks for Common Deep Learning Approaches for Microscopy Imaging.",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 295237,
      "title": "Key factor would be detect center",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 295978,
      "title": "I'm sorry, I'm not sure.",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 295977,
      "title": "How do I save it?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 286368,
      "title": "[Up-to-date Summary] We passed 30% of the competition, These are the stories up to this point!",
      "comment_count": 11,
      "views": 0,
      "votes": 154
    },
    {
      "id": 285516,
      "title": "[placeholder] my approach and results",
      "comment_count": 55,
      "views": 0,
      "votes": 110
    },
    {
      "id": 295294,
      "title": "Cellpose logging",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 292415,
      "title": "Submission taking a long time to scoring the results.🙄",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 291371,
      "title": "The clean astro mask are here!!!",
      "comment_count": 22,
      "views": 0,
      "votes": 125
    },
    {
      "id": 291552,
      "title": "Announcement on Public Notebook Sharing",
      "comment_count": 2,
      "views": 0,
      "votes": 25
    },
    {
      "id": 294560,
      "title": "Submission Scoring Error - for some weights file - detectron2",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 294834,
      "title": "Has anyone tried semi-training? Does that improve your score?",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 295537,
      "title": "how to get clean_masks?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 295528,
      "title": "What information do the 3 channels represent?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 295057,
      "title": "Labkit: Annotation tool for microscopy",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 292852,
      "title": "Evaluation Script for MMDetection",
      "comment_count": 4,
      "views": 0,
      "votes": 12
    },
    {
      "id": 294755,
      "title": "Quick review: EMBEDSEG",
      "comment_count": 0,
      "views": 0,
      "votes": 11
    },
    {
      "id": 294094,
      "title": "just to confirm:  predictions are not allowed to overlap, but are holes allowed?",
      "comment_count": 2,
      "views": 0,
      "votes": 15
    },
    {
      "id": 293159,
      "title": "UNet strikes... a pose!",
      "comment_count": 20,
      "views": 0,
      "votes": 55
    },
    {
      "id": 294590,
      "title": "There are 16 Object detection papers in NeurIPS  2021 | here with code and abstract",
      "comment_count": 1,
      "views": 0,
      "votes": 12
    },
    {
      "id": 287023,
      "title": "Saving the Best checkpoint in Detectron2",
      "comment_count": 17,
      "views": 0,
      "votes": 86
    },
    {
      "id": 294176,
      "title": "Quick review: Cellpose",
      "comment_count": 1,
      "views": 0,
      "votes": 19
    },
    {
      "id": 294898,
      "title": "Accept uncertainty of labels: Bayesian inference",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 293874,
      "title": "What I tried and did not work?",
      "comment_count": 4,
      "views": 0,
      "votes": 31
    },
    {
      "id": 293894,
      "title": "Inference every epoch is not suitable (and how to avoid CUDA/CPU OOM)",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 293799,
      "title": "Mask2Former: Masked-attention Mask Transformer for Universal Image Segmentation",
      "comment_count": 4,
      "views": 0,
      "votes": 11
    },
    {
      "id": 294424,
      "title": "Nuclei Segmentation in Histopathology Images using  Deep Learning with Local and Global Views",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 294391,
      "title": "Is it possible to ensemble Unets with different backbones?",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 291639,
      "title": "First experiment with cleaned astro masks",
      "comment_count": 16,
      "views": 0,
      "votes": 55
    },
    {
      "id": 293302,
      "title": "Soft NMS or NMS?",
      "comment_count": 2,
      "views": 0,
      "votes": 9
    },
    {
      "id": 294006,
      "title": "Augmentation in Detectron2",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 290617,
      "title": "IoU-based loss functions for bounding box regression",
      "comment_count": 6,
      "views": 0,
      "votes": 14
    },
    {
      "id": 294011,
      "title": "Hybrid Task Cascade",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 293794,
      "title": "ensemble of  predicted mask?",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 292754,
      "title": "MMDetection V_2.18.0 offline installation ",
      "comment_count": 2,
      "views": 0,
      "votes": 13
    },
    {
      "id": 285623,
      "title": "Getting per pixel scores from the detectron",
      "comment_count": 26,
      "views": 0,
      "votes": 53
    },
    {
      "id": 294216,
      "title": "submission.csv not found while using ensemble inference",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 284694,
      "title": "LIVECELL_dataset_2021 annotationfiles are corrupt",
      "comment_count": 10,
      "views": 0,
      "votes": 14
    },
    {
      "id": 293175,
      "title": "Cuda out of Memory",
      "comment_count": 5,
      "views": 0,
      "votes": 5
    },
    {
      "id": 284215,
      "title": "Threshold by class",
      "comment_count": 19,
      "views": 0,
      "votes": 85
    },
    {
      "id": 291930,
      "title": "Does anyone try to train separate model for each class?",
      "comment_count": 4,
      "views": 0,
      "votes": 8
    },
    {
      "id": 292560,
      "title": "【Valid MAP IOU 0.26】Experiments of using EfficientnetV2 in FPN Mask RCNN ",
      "comment_count": 7,
      "views": 0,
      "votes": 11
    },
    {
      "id": 278936,
      "title": "What is Run Length Encoding (RLE)?",
      "comment_count": 8,
      "views": 0,
      "votes": 31
    },
    {
      "id": 281205,
      "title": "Annotations Are too Noisy for the Metric",
      "comment_count": 18,
      "views": 0,
      "votes": 174
    },
    {
      "id": 293069,
      "title": "Is K fold Not working at LB",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 290309,
      "title": "How to make a Model Ensemble in this Competition？",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 292094,
      "title": "A way to visualise predictions",
      "comment_count": 2,
      "views": 0,
      "votes": 28
    },
    {
      "id": 287774,
      "title": "Let's talk about TTA and Ensembling...",
      "comment_count": 15,
      "views": 0,
      "votes": 37
    },
    {
      "id": 279790,
      "title": "Tips in submission and baseline (in the beginning of the competition)",
      "comment_count": 39,
      "views": 0,
      "votes": 205
    },
    {
      "id": 285384,
      "title": "LIVECell paper summarized",
      "comment_count": 9,
      "views": 0,
      "votes": 63
    },
    {
      "id": 291773,
      "title": "What is \"This training is CPU bounding\" meaning ?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 291089,
      "title": "Submission Scoring Error",
      "comment_count": 9,
      "views": 0,
      "votes": 11
    },
    {
      "id": 291425,
      "title": "Mosaic Augmentation",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 288677,
      "title": "Deep Double Decent: More epoch More better?",
      "comment_count": 11,
      "views": 0,
      "votes": 7
    },
    {
      "id": 290446,
      "title": "Error while reading annotation file livecell_coco_train.json using pycocotools",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 290409,
      "title": "Custom backbone for Detectron2?",
      "comment_count": 3,
      "views": 0,
      "votes": 8
    },
    {
      "id": 292869,
      "title": "U2net output+notebook threw error",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 291940,
      "title": "Submission error, What’s happening?",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 292849,
      "title": "Experimental GAN approach, model too cautious, ideas?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 291127,
      "title": "How to use MMdetection to view CV score",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 291532,
      "title": "Simple network for semantic segmentation",
      "comment_count": 0,
      "views": 0,
      "votes": 8
    },
    {
      "id": 280250,
      "title": "Overlaps - ambiguous or not?",
      "comment_count": 13,
      "views": 0,
      "votes": 47
    },
    {
      "id": 291627,
      "title": "Getting stuck with zero LB score",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 291311,
      "title": "How to get the contours of a binary mask?",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 292600,
      "title": "Manually flaged broken masks (\"isbroken\" column in the \"train.csv\")",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 292511,
      "title": "What‘s the meaning of annotation",
      "comment_count": 3,
      "views": 0,
      "votes": -4
    },
    {
      "id": 292162,
      "title": "train.csv rows",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 284062,
      "title": "Have you tried detectron2-ResNeSt?",
      "comment_count": 13,
      "views": 0,
      "votes": 13
    },
    {
      "id": 291733,
      "title": "Semisupervised data",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 291333,
      "title": "From mask to bounding boxes ",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 285046,
      "title": "Looks good by running notebook but gives error for submission",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 278951,
      "title": "common tricks in kaggle image segmentation problems ",
      "comment_count": 2,
      "views": 0,
      "votes": 20
    },
    {
      "id": 290603,
      "title": "color augmentation",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 290441,
      "title": "Did you notice duplicates with contradicting annotations?",
      "comment_count": 9,
      "views": 0,
      "votes": 14
    },
    {
      "id": 290214,
      "title": "Handling Broken Masks",
      "comment_count": 2,
      "views": 0,
      "votes": 13
    },
    {
      "id": 288376,
      "title": "Can someone tell me how to generate the right format of submission?",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 291050,
      "title": "Why I submission fail?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 289895,
      "title": "Several backbone for instance segmentation",
      "comment_count": 2,
      "views": 0,
      "votes": 10
    },
    {
      "id": 287930,
      "title": "Is the RLE 0 indexed? ",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 290391,
      "title": "Why only detectron2 works so far",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 288464,
      "title": "U-net, Mask R-CNN and RetinaNet illustrations ...and more ",
      "comment_count": 3,
      "views": 0,
      "votes": 15
    },
    {
      "id": 288564,
      "title": "Mask Scoring R-CNN",
      "comment_count": 8,
      "views": 0,
      "votes": 10
    },
    {
      "id": 289449,
      "title": "why my notebook script background becomes black?",
      "comment_count": 6,
      "views": 0,
      "votes": 9
    },
    {
      "id": 290384,
      "title": "Some questions on train.csv",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 289702,
      "title": "problem with saving notebook result",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 289262,
      "title": "Kaggling on Colab: How to set everything up",
      "comment_count": 4,
      "views": 0,
      "votes": 13
    },
    {
      "id": 289030,
      "title": "Some questions in the `LIVECell_dataset_2021` datasets",
      "comment_count": 5,
      "views": 0,
      "votes": 7
    },
    {
      "id": 289501,
      "title": "A Survey on Instance Segmentation: State of the art",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 289261,
      "title": "Training CNN Runtime comparison Kaggle vs. Google Colab\t",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 288792,
      "title": "Some of your uploads failed. Please try again.",
      "comment_count": 10,
      "views": 0,
      "votes": 12
    },
    {
      "id": 280130,
      "title": "Humbled to become a Kaggle Notebooks Master",
      "comment_count": 29,
      "views": 0,
      "votes": 50
    },
    {
      "id": 288718,
      "title": "Paper: A Survey of Visual Transformers",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 287487,
      "title": "Are we predicting cells or a mask?",
      "comment_count": 11,
      "views": 0,
      "votes": 18
    },
    {
      "id": 284999,
      "title": "why x101  is not as good as R-50 in detectron2?",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 289026,
      "title": "How long does it take to inference?(Take Mask RCNN as example)",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 280172,
      "title": "Top GitHub Source Codes on Cell Instance Segmentation",
      "comment_count": 7,
      "views": 0,
      "votes": 28
    },
    {
      "id": 288762,
      "title": "Order of instances in metric / submission",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 287523,
      "title": "How To Transform *.csv to coco Dataset annotations?",
      "comment_count": 3,
      "views": 0,
      "votes": -2
    },
    {
      "id": 284185,
      "title": "Notebook Threw Exception ! ",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 279995,
      "title": "No Overlap Issue: Quick Fix",
      "comment_count": 5,
      "views": 0,
      "votes": 61
    },
    {
      "id": 289037,
      "title": "Which bbox is right?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 287758,
      "title": "Different image sizes",
      "comment_count": 4,
      "views": 0,
      "votes": 8
    },
    {
      "id": 287988,
      "title": "Pixel Overlap Across instances, how does it affects model ?",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 285546,
      "title": "Best way to split folds for cross-validation",
      "comment_count": 6,
      "views": 0,
      "votes": 55
    },
    {
      "id": 280793,
      "title": "Pytorch Lightning but 0.000 Score",
      "comment_count": 8,
      "views": 0,
      "votes": 8
    },
    {
      "id": 284057,
      "title": "I visualized every training image so you don't have to",
      "comment_count": 3,
      "views": 0,
      "votes": 83
    },
    {
      "id": 288205,
      "title": "Outsourcing computations into private notebook pipeline",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 286693,
      "title": "Trust CV or LB?",
      "comment_count": 13,
      "views": 0,
      "votes": 10
    },
    {
      "id": 286995,
      "title": "[Question to Kaggle Team] GPU Session cap reached, but there no Active Events ",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 287308,
      "title": "Using the dataset for research",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 286195,
      "title": "0.000 submission score",
      "comment_count": 20,
      "views": 0,
      "votes": 4
    },
    {
      "id": 286843,
      "title": "MMDetection's vs Detectron2 ",
      "comment_count": 8,
      "views": 0,
      "votes": 19
    },
    {
      "id": 287638,
      "title": "Why Unet is suitable for this competition?",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 287728,
      "title": "Sartorius Detectron - Training + W&B",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 284135,
      "title": "Summary of every SOTA instance segmentation paper of the past 2 years ",
      "comment_count": 4,
      "views": 0,
      "votes": 75
    },
    {
      "id": 287029,
      "title": "Overlap Fractions - Outliers - Duplicates",
      "comment_count": 0,
      "views": 0,
      "votes": 27
    },
    {
      "id": 286553,
      "title": "UNet Strikes Back 🔥 | LB: 0.155+",
      "comment_count": 14,
      "views": 0,
      "votes": 27
    },
    {
      "id": 284357,
      "title": "Ensembling Semantic Segmentation models",
      "comment_count": 3,
      "views": 0,
      "votes": 14
    },
    {
      "id": 287230,
      "title": "Is this annotation correct?",
      "comment_count": 1,
      "views": 0,
      "votes": 10
    },
    {
      "id": 282542,
      "title": "Can I train a model on the local machine?",
      "comment_count": 10,
      "views": 0,
      "votes": 8
    },
    {
      "id": 284283,
      "title": "Lets talk Augmentations!",
      "comment_count": 3,
      "views": 0,
      "votes": 28
    },
    {
      "id": 285422,
      "title": "Train Attention based Residual Unet with watershed algorithm for instance segmentation",
      "comment_count": 8,
      "views": 0,
      "votes": 16
    },
    {
      "id": 285351,
      "title": "CV LB-Correlation and Best Single Model",
      "comment_count": 11,
      "views": 0,
      "votes": 4
    },
    {
      "id": 286696,
      "title": "Internet access disabled",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 285694,
      "title": "CLAHE Preprocessing",
      "comment_count": 7,
      "views": 0,
      "votes": 6
    },
    {
      "id": 279488,
      "title": "broken mask repro, should be an easy fix",
      "comment_count": 21,
      "views": 0,
      "votes": 54
    },
    {
      "id": 286096,
      "title": "no free colab anymore?",
      "comment_count": 8,
      "views": 0,
      "votes": 2
    },
    {
      "id": 278663,
      "title": "Refactoring Inversion's classical RLE code",
      "comment_count": 7,
      "views": 0,
      "votes": 45
    },
    {
      "id": 285927,
      "title": "Reproducing Sartorius LIVECell paper results",
      "comment_count": 3,
      "views": 0,
      "votes": 9
    },
    {
      "id": 278706,
      "title": "Competition deadline",
      "comment_count": 15,
      "views": 0,
      "votes": 69
    },
    {
      "id": 285968,
      "title": "Library or pure torch?",
      "comment_count": 4,
      "views": 0,
      "votes": 7
    },
    {
      "id": 286058,
      "title": "Evaluating 26 models to split the problem.",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 286003,
      "title": "deepflash2 (a deep-learning pipeline for segmentation of ambiguous microscopic images)",
      "comment_count": 3,
      "views": 0,
      "votes": 9
    },
    {
      "id": 286684,
      "title": "Create new kernel enviroment get error ImportError: /usr/lib/x86_64-linux-gnu/libstdc++.so.6",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 285507,
      "title": "Detectron2 and reproducability ",
      "comment_count": 6,
      "views": 0,
      "votes": 9
    },
    {
      "id": 285799,
      "title": "KAGGLE UI: notebook output dataset  create/update options",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 285825,
      "title": "Question about metric at test time",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 285321,
      "title": "What does it mean by: \" LIVECell is the predecessor dataset to this competition\" ?",
      "comment_count": 10,
      "views": 0,
      "votes": 14
    },
    {
      "id": 286059,
      "title": "Python Lists",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 278883,
      "title": "[Info] Instance Segmentation Models (quick list)",
      "comment_count": 9,
      "views": 0,
      "votes": 63
    },
    {
      "id": 285519,
      "title": "Efficient conversion to COCO Dataset",
      "comment_count": 3,
      "views": 0,
      "votes": 11
    },
    {
      "id": 286082,
      "title": "Scoring Error",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 286476,
      "title": " LB metric and test set labels",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 282788,
      "title": "Displaying validation score in detectron",
      "comment_count": 6,
      "views": 0,
      "votes": 32
    },
    {
      "id": 285609,
      "title": "Very slow evaluation",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 285504,
      "title": "Validation strategy",
      "comment_count": 4,
      "views": 0,
      "votes": 8
    },
    {
      "id": 285681,
      "title": "Missing cell in annotation?",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 286159,
      "title": "Filtering predictions",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 285732,
      "title": "pretrained weight dataset versions to use",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 286051,
      "title": "Python Masks",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 285345,
      "title": "Is thie Competition Resource intensive",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 285356,
      "title": "Private Dataset with model",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 285497,
      "title": "is the 3 downloaded test images included in the public LB score?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 284329,
      "title": "Dataset Explanation",
      "comment_count": 2,
      "views": 0,
      "votes": 9
    },
    {
      "id": 284619,
      "title": "Python Functions",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 285348,
      "title": "Talk of the town",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 284415,
      "title": "Winning solutions past Kaggle competition: Single Cell Classification ",
      "comment_count": 0,
      "views": 0,
      "votes": 13
    },
    {
      "id": 285149,
      "title": "Confused on 'predictions should not overlap'",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 284769,
      "title": "detectron2 training loss display problem,  Kaggle vs Colab",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 284285,
      "title": "[TL;DR list] - Every single Imagenet paper in the last 2 years summarized ",
      "comment_count": 0,
      "views": 0,
      "votes": 13
    },
    {
      "id": 283851,
      "title": "Top past solutions for instance segmentation problems",
      "comment_count": 2,
      "views": 0,
      "votes": 28
    },
    {
      "id": 284410,
      "title": "Winning solution of Kaggle open images 2019 - video presentation ",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 284513,
      "title": "TFRecord Dataset from Images and Their Masks (Faster training with TPUs)",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 284939,
      "title": "Deadline for competition",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 285116,
      "title": "Calculation of Mean Average Precision",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 284049,
      "title": "TPU problem?",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 279028,
      "title": "using GAN (generative adversarial net) ...",
      "comment_count": 1,
      "views": 0,
      "votes": 9
    },
    {
      "id": 282544,
      "title": "Understanding Cell Types",
      "comment_count": 4,
      "views": 0,
      "votes": 17
    },
    {
      "id": 284620,
      "title": "Python Functions",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 283999,
      "title": "Coco Json Format Files for Train Dataset",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 284059,
      "title": "Question: Challenge Metrics vs. (COCO) Object Detection Metric",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 284275,
      "title": "Semantic Segmentation or Instance Segmentation ?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 278989,
      "title": "Top Instance Segmentation Papers with Code",
      "comment_count": 1,
      "views": 0,
      "votes": 14
    },
    {
      "id": 283796,
      "title": "Good Papers and resources for beginners inInstance Segmentation???",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 279996,
      "title": "Mask-RCNN | MMDetection Starter - LB: 0.270",
      "comment_count": 16,
      "views": 0,
      "votes": 66
    },
    {
      "id": 281440,
      "title": "0.000 on LB",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 284169,
      "title": "Should we reconstruct a wholly-new model or architecture to solve this problem?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 281084,
      "title": "Analysis of Overlaps I.",
      "comment_count": 1,
      "views": 0,
      "votes": 36
    },
    {
      "id": 283818,
      "title": "about the submission.csv",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 281904,
      "title": "How does Evaluation work?",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 282015,
      "title": "what exactly is problem statement?",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 280291,
      "title": "Relevant past Comps and its Top solutions + extra",
      "comment_count": 4,
      "views": 0,
      "votes": 17
    },
    {
      "id": 282116,
      "title": "error submission  if running out of gpu quote.",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 282270,
      "title": "Help for submission issue using Keras model ",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 280584,
      "title": "Can't downgrade from Tensorflow 2.6.0 to Tensorflow 2.2.0",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 281657,
      "title": "submission.csv not found",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 282653,
      "title": "The order of RLE index for submission",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 280742,
      "title": "Important Papers for Competition",
      "comment_count": 0,
      "views": 0,
      "votes": 14
    },
    {
      "id": 280583,
      "title": "One to many mapping",
      "comment_count": 0,
      "views": 0,
      "votes": 14
    },
    {
      "id": 280688,
      "title": "What should be segmented?",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 278801,
      "title": "🦠Broken masks?",
      "comment_count": 6,
      "views": 0,
      "votes": 37
    },
    {
      "id": 280740,
      "title": "What would be the range of best lb score",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 279923,
      "title": "Understanding Run Length Encoding",
      "comment_count": 2,
      "views": 0,
      "votes": 10
    },
    {
      "id": 279514,
      "title": "is it a good idea to work on local machine/Cloud and then submit/re-run on kaggle notebook?",
      "comment_count": 5,
      "views": 0,
      "votes": 8
    },
    {
      "id": 278735,
      "title": "Previous Competitions on Instance Segmentation",
      "comment_count": 2,
      "views": 0,
      "votes": 28
    },
    {
      "id": 279200,
      "title": "Is this a One-class Semantic Segmentation Competition?",
      "comment_count": 7,
      "views": 0,
      "votes": 18
    },
    {
      "id": 279352,
      "title": "0.000 score on leaderboard",
      "comment_count": 6,
      "views": 0,
      "votes": 11
    },
    {
      "id": 280294,
      "title": "1 Pixel Segmentations and Other Extremely Small Segmentations In Train Set???",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 278804,
      "title": "plate_time is equal to elapsed_timedelta?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 280751,
      "title": "Facing problem on custom training detectron2 on kaggle",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 279365,
      "title": "Simple baseline model with LB > 0 made public",
      "comment_count": 0,
      "views": 0,
      "votes": 17
    },
    {
      "id": 279305,
      "title": "Expoiting Rotational Symmetry in Cell Images - Histology insight ",
      "comment_count": 0,
      "views": 0,
      "votes": 15
    },
    {
      "id": 280018,
      "title": "Follow-up baseline after Inoichan's topic (Instance segmentation model)",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 279779,
      "title": "Number of classes",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 279413,
      "title": "[Which reshape ?] Does RLE in this competition really follow top -> bottom, then left to right ?",
      "comment_count": 1,
      "views": 0,
      "votes": 7
    },
    {
      "id": 279303,
      "title": "Previous Cell/Histopathology related competitions - ",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 279561,
      "title": "Discussion of Functions in Python",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 278976,
      "title": "Need clarification for the non-invasive method via light microscopy.",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 278878,
      "title": "Training Mask Dataset (NPZ)",
      "comment_count": 0,
      "views": 0,
      "votes": 13
    },
    {
      "id": 278717,
      "title": "Evaluation metrics: Intersection over union ",
      "comment_count": 1,
      "views": 0,
      "votes": 12
    },
    {
      "id": 278945,
      "title": "Difference between Object detection and Object Segmentation?",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 279478,
      "title": "Looking for a team  / Sartorious",
      "comment_count": 3,
      "views": 0,
      "votes": -4
    },
    {
      "id": 279029,
      "title": "Run length decoding Beginners",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 280103,
      "title": "Anyone wants to team up?",
      "comment_count": 1,
      "views": 0,
      "votes": -7
    },
    {
      "id": 279031,
      "title": "Lovasz loss, probability calibration and threshold",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    }
  ],
  "errors": []
}