{
  "total": 182,
  "topics": [
    {
      "id": 200506,
      "title": "Solutions and code of top-ranked teams",
      "comment_count": 1,
      "views": 0,
      "votes": 19
    },
    {
      "id": 172445,
      "title": "New to Machine Learning or to Kaggle? ",
      "comment_count": 13,
      "views": 0,
      "votes": 13
    },
    {
      "id": 188160,
      "title": "Congrats to the winners! Don't forget to submit your documentation!",
      "comment_count": 5,
      "views": 0,
      "votes": 9
    },
    {
      "id": 183733,
      "title": "ILR workshop at ECCV'20",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 172576,
      "title": "Host Baseline Example",
      "comment_count": 24,
      "views": 0,
      "votes": 50
    },
    {
      "id": 182188,
      "title": "Code competition debugging and errors",
      "comment_count": 1,
      "views": 0,
      "votes": 10
    },
    {
      "id": 172127,
      "title": "External Data Thread",
      "comment_count": 28,
      "views": 0,
      "votes": 6
    },
    {
      "id": 172128,
      "title": "Looking for a Team Megathread",
      "comment_count": 81,
      "views": 0,
      "votes": 12
    },
    {
      "id": 171157,
      "title": "Welcome to Landmark Recognition 2020",
      "comment_count": 21,
      "views": 0,
      "votes": 33
    },
    {
      "id": 176198,
      "title": "Reusing models from 2020 landmark retrieval challenge",
      "comment_count": 3,
      "views": 0,
      "votes": 16
    },
    {
      "id": 456225,
      "title": "Why are the target labels not in a sequence. ",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 187821,
      "title": "1st Place Solution",
      "comment_count": 93,
      "views": 0,
      "votes": 196
    },
    {
      "id": 187725,
      "title": "Congratulations!",
      "comment_count": 19,
      "views": 0,
      "votes": 33
    },
    {
      "id": 188299,
      "title": "2nd place solution",
      "comment_count": 23,
      "views": 0,
      "votes": 97
    },
    {
      "id": 187757,
      "title": "3rd Place Solution — A Pure Global Feature Approach",
      "comment_count": 43,
      "views": 0,
      "votes": 91
    },
    {
      "id": 173083,
      "title": "Global Average Precision (GAP) Estimation ",
      "comment_count": 4,
      "views": 0,
      "votes": 15
    },
    {
      "id": 173091,
      "title": "Some insights",
      "comment_count": 27,
      "views": 0,
      "votes": 40
    },
    {
      "id": 187694,
      "title": "Congrats Philipp and Christof!",
      "comment_count": 4,
      "views": 0,
      "votes": 31
    },
    {
      "id": 200006,
      "title": "Video material of 1st place and other top solutions",
      "comment_count": 2,
      "views": 0,
      "votes": 17
    },
    {
      "id": 190983,
      "title": "3rd Placd Solution Full Code Released",
      "comment_count": 1,
      "views": 0,
      "votes": 24
    },
    {
      "id": 194543,
      "title": "ArcFace and Image Embedding",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 192596,
      "title": "Notebook Exceeded Allowed Compute, Notebook Timeout",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 187894,
      "title": "7th place solution (with inference kernel + training code)",
      "comment_count": 7,
      "views": 0,
      "votes": 34
    },
    {
      "id": 187731,
      "title": "[22nd Place + Code] Margin Scheduling – First time on Kaggle :D",
      "comment_count": 5,
      "views": 0,
      "votes": 15
    },
    {
      "id": 187864,
      "title": "[21st place] - My solution to 2 silver medals in GL 2020",
      "comment_count": 14,
      "views": 0,
      "votes": 10
    },
    {
      "id": 188543,
      "title": "Post-competition LB Shake-up report",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 187712,
      "title": "My first silver medal! :)",
      "comment_count": 12,
      "views": 0,
      "votes": 10
    },
    {
      "id": 188732,
      "title": "Request to winners to please share either PyTorch/Tensorflow Training+Inference Pipeline",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 188268,
      "title": "31st place solution - single DELG model.",
      "comment_count": 2,
      "views": 0,
      "votes": 11
    },
    {
      "id": 187961,
      "title": "6th place solution",
      "comment_count": 4,
      "views": 0,
      "votes": 18
    },
    {
      "id": 187714,
      "title": "(Public>0.54) Provisional 36th solution and kernel",
      "comment_count": 2,
      "views": 0,
      "votes": 16
    },
    {
      "id": 187732,
      "title": "PyTorch DELG",
      "comment_count": 14,
      "views": 0,
      "votes": 11
    },
    {
      "id": 187756,
      "title": "When you risk everything or the most epic failure in kaggle competitions (Fron top 14% to to 72%)",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 187225,
      "title": "Fake accounts in Silver medals? (zero submissions, registered <14 days ago)",
      "comment_count": 4,
      "views": 0,
      "votes": 18
    },
    {
      "id": 187990,
      "title": "Those who are in need of GLDv2 Full DataSet",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 187720,
      "title": "My engineering approach to the problem. #49",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 187708,
      "title": "Public #87 --> Private #219 Shaken but not stirred",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 187328,
      "title": "Last day and all the best!",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 186299,
      "title": "Submission CSV Not Found",
      "comment_count": 12,
      "views": 0,
      "votes": 17
    },
    {
      "id": 187344,
      "title": "Last-minute question: local descriptor keypoints in the DELG paper?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 186653,
      "title": "Google Colab",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 186487,
      "title": "20GB GLD dataset/tfrecords, to train on colab",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 187491,
      "title": "Submission Errors",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 186830,
      "title": "Can anyone could fix this issue",
      "comment_count": 1,
      "views": 0,
      "votes": 0
    },
    {
      "id": 186978,
      "title": "please ficx me this issue",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 186998,
      "title": "Notebook Timeout Error",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 180921,
      "title": " The world of RANSAC",
      "comment_count": 1,
      "views": 0,
      "votes": 16
    },
    {
      "id": 186654,
      "title": "ImageDataGenerator",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 185072,
      "title": "Notebook Timeout Error",
      "comment_count": 3,
      "views": 0,
      "votes": 8
    },
    {
      "id": 183323,
      "title": "Approach to tune RANSAC hyperparamers?",
      "comment_count": 15,
      "views": 0,
      "votes": 7
    },
    {
      "id": 185365,
      "title": "Efficient ways to train models with the Kaggle's GPU and TPU.",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 183629,
      "title": "EfficientNet Imagenet vs Noisy Student",
      "comment_count": 4,
      "views": 0,
      "votes": 9
    },
    {
      "id": 185294,
      "title": "Really dumb question: how to attach the full train set?",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 180090,
      "title": "\"Late Submission Disabled\" for the Retrieval track?",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 185691,
      "title": "pip installation consumes RAM",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 184040,
      "title": "Can we train on TPUs without making a TFRecord?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 183610,
      "title": "Just an idea",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 182643,
      "title": "Score of 0.0000 on a trained model with high validation accuracy",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 178698,
      "title": "How to deal with large amount of data with only 13 GB RAM ??",
      "comment_count": 10,
      "views": 0,
      "votes": 3
    },
    {
      "id": 180056,
      "title": "TF Record Dataset and Own Baseline",
      "comment_count": 6,
      "views": 0,
      "votes": 26
    },
    {
      "id": 183289,
      "title": "Submission without generating output",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 183303,
      "title": " Why when I use Host Baseline part of Submissions get Notebook Timeout error, but others with about the same execution time work well?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 183158,
      "title": "Thread for summarizing the results of previous year's winning strategy on present data.",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 182313,
      "title": "Private datasets need to be made public for submission?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 182515,
      "title": "Cloud TPU vs Colab TPU? Also, any tricks to speed up TPU training?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 181392,
      "title": "A trick to improve result",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 179975,
      "title": "Pretrained model in baseline",
      "comment_count": 13,
      "views": 0,
      "votes": 2
    },
    {
      "id": 182636,
      "title": "No module named 'evaluations'",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 182166,
      "title": "Zero Submission Score",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 182472,
      "title": "SOS!The installed package cannot be used when submitting code",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 182494,
      "title": "Google Cloud Dataset Retrieving Pytorch",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 178103,
      "title": "Few questions on training and loading model",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 182154,
      "title": "TPUs running slower than GPUs! Reason?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 182064,
      "title": "'EfficientNet' object has no attribute '_swish'",
      "comment_count": 6,
      "views": 0,
      "votes": 0
    },
    {
      "id": 180783,
      "title": "How to use ImageDataGenerator",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 181328,
      "title": "Global Average Precision",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 179472,
      "title": "Is changing RANSAC parameters the only way?",
      "comment_count": 13,
      "views": 0,
      "votes": 13
    },
    {
      "id": 181003,
      "title": "Batch inference in DELG",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 180720,
      "title": "Some good background information about RANSAC",
      "comment_count": 5,
      "views": 0,
      "votes": 10
    },
    {
      "id": 181915,
      "title": "Output Not Loading",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 181434,
      "title": "TPU error in Kernel",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 181156,
      "title": "PROVIDE BETTER RERUN ERROR STATEMENTS/LOGS.Submissions with errors are being considered within daily limits. CSV errors.",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 180535,
      "title": "Submission CSV Not Found : On Sample Submission only from Baseline code",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 181300,
      "title": "A notebook to run for more than 9 hours",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 181555,
      "title": "How to use only a subset of training image?",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 180947,
      "title": "CUDA : out of memory",
      "comment_count": 4,
      "views": 0,
      "votes": 3
    },
    {
      "id": 180865,
      "title": "Wikipedia Categories",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 181489,
      "title": "Submission score 0.0000",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 179964,
      "title": "How to make a simple CNN model for this dataset?",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 181260,
      "title": "Most test images do not have landmarks according to provided GT metadata...?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 181393,
      "title": "Efficient Diffusion on Region Manifolds: Recovering Small Objects with Compact CNN Representations",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 181043,
      "title": "AssertionError: Torch not compiled with CUDA enabled",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 181100,
      "title": "why my kaggle gpu is not being used?",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 179548,
      "title": "Zero scores",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 172470,
      "title": "Submission Scoring Error",
      "comment_count": 5,
      "views": 0,
      "votes": 6
    },
    {
      "id": 180278,
      "title": "has anybody encountered nan loss during training? any solution to this??? ",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 179904,
      "title": "Metric: Global Average Precision(GAP)",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 178501,
      "title": "GPU not working",
      "comment_count": 15,
      "views": 0,
      "votes": 1
    },
    {
      "id": 180718,
      "title": "how to prepare x_train and y_train?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 180545,
      "title": "CUDA error 59: Device-side assert triggered",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 177893,
      "title": "Complete TF-Record Dataset 90GB",
      "comment_count": 4,
      "views": 0,
      "votes": 17
    },
    {
      "id": 176744,
      "title": "TPU and GCS",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 178907,
      "title": "(Kaggle) Error on Every Submission?",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 179748,
      "title": "Question to organizers re: 100k data split",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 179945,
      "title": "Some private test dataset doubts. ",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 178171,
      "title": "Submission completion notifications",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 176697,
      "title": "Can we pre-compute embeddings? Are image ids the same in PB and PV sets?",
      "comment_count": 15,
      "views": 0,
      "votes": 6
    },
    {
      "id": 179575,
      "title": "submission scoring error - R",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 179502,
      "title": "Wow, winners from Google Landmark Retrieval are leading...",
      "comment_count": 1,
      "views": 0,
      "votes": -4
    },
    {
      "id": 178260,
      "title": "A question to organizers",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 178093,
      "title": "Same Kernel produces multiple LB Scores on submission. Any ideas why?",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 178788,
      "title": "How to use GLDv1",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 178539,
      "title": "Not able to reply people's comment",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 178320,
      "title": "Can anyone tell me the format of the output column?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 177817,
      "title": "Cutmix in Tensorflow",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 176883,
      "title": "What is \"Kaggle Error\"?",
      "comment_count": 5,
      "views": 0,
      "votes": 3
    },
    {
      "id": 178122,
      "title": "DELG Saved Models",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 172838,
      "title": "Getting started with landmark recognition?",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 177038,
      "title": "Nice to see Kaggle team in the competition :)",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 178037,
      "title": "Fast experimentation",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 177391,
      "title": "Format of saving model weights",
      "comment_count": 2,
      "views": 0,
      "votes": 0
    },
    {
      "id": 176988,
      "title": "Looking for a team!",
      "comment_count": 10,
      "views": 0,
      "votes": 1
    },
    {
      "id": 177017,
      "title": "How to store label in tfrecord.",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 176099,
      "title": "Small question about stopping gradients in DELF implementation",
      "comment_count": 9,
      "views": 0,
      "votes": 5
    },
    {
      "id": 176951,
      "title": "Best way to use cv2.imread for imeges from Google Cloud Storage?",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 174184,
      "title": "Training tf records available here for download !!!!!",
      "comment_count": 20,
      "views": 0,
      "votes": 17
    },
    {
      "id": 177412,
      "title": "Validation loss doesn't decrease as training loss does",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 177193,
      "title": "Understanding the tfrecord.",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 176392,
      "title": "Is it possible to make tf records for the private sets?",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 176319,
      "title": "How many unique landmarks are in the testing set?",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 177013,
      "title": "Sample Submission Format Query",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 175671,
      "title": "About the Artwork Recognition Challenge and Product Retrieval Challenge ",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 176487,
      "title": "Reflecting on last year edition",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 176244,
      "title": "How can a beginner do this huge task?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 176047,
      "title": "Notebook timeout error",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 172231,
      "title": "TF records",
      "comment_count": 10,
      "views": 0,
      "votes": 25
    },
    {
      "id": 172132,
      "title": "Why this competition is so difficult to train?",
      "comment_count": 20,
      "views": 0,
      "votes": 3
    },
    {
      "id": 175261,
      "title": "Basic GAN Project",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 175951,
      "title": " Why are the scores different even in the same csv file?",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 176106,
      "title": "A crazy idea.",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 176105,
      "title": "A crazy idea.",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 171755,
      "title": "Relevant research papers",
      "comment_count": 4,
      "views": 0,
      "votes": 9
    },
    {
      "id": 174493,
      "title": "Processing time of Baseline submission",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 172907,
      "title": "Public and Private training data",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 172340,
      "title": "Looking for a teammate",
      "comment_count": 3,
      "views": 0,
      "votes": 0
    },
    {
      "id": 175171,
      "title": "Are GLRec and GLRet datasets identical?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 175290,
      "title": "Landmark Image Recognition with TensorFlow Hub DELF Module",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 173111,
      "title": "What exactly is the 100k private training set?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 174958,
      "title": "how to find image outliers.. ",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 174919,
      "title": "Can someone clarify the public and private data set for me.",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 174186,
      "title": "Validation tf records available here for download !!!",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 174916,
      "title": "Looking for team members",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 174722,
      "title": "'Submission CSV Not Found' error even I can get submission.csv in my output.",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 171289,
      "title": "file size problem for Google Colab",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 170989,
      "title": "Last year's top solutions",
      "comment_count": 17,
      "views": 0,
      "votes": 76
    },
    {
      "id": 172737,
      "title": "Landmark2019-1st-and-3rd-Place-Solution",
      "comment_count": 3,
      "views": 0,
      "votes": 10
    },
    {
      "id": 171562,
      "title": "Fast way of reading images",
      "comment_count": 13,
      "views": 0,
      "votes": 11
    },
    {
      "id": 174285,
      "title": "Please Ignore",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 173799,
      "title": "GLD-V2 explorer is neat!",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 173518,
      "title": "Any information from metadata?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 172955,
      "title": "How to handle such huge size dataset ??",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 173688,
      "title": "Submission CSV Not Found",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 172585,
      "title": "Too large file size",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 172586,
      "title": "Strange problem with submission score",
      "comment_count": 5,
      "views": 0,
      "votes": 5
    },
    {
      "id": 173421,
      "title": "Keras Memory Utilization",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 171051,
      "title": "Understanding dataset folder structure",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 172722,
      "title": "Dataset confusion",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 173017,
      "title": "Effect of randomness in ransac?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 172646,
      "title": "Images in the form of long strips",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 172439,
      "title": "Will there be distractors in the test set?",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 172717,
      "title": "Does query expansion help for recognition-via-retrieval?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 171742,
      "title": "Landmark or Non Landmark Identification in PyTorch",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 171291,
      "title": "Why less restrictions for recognition but not retrieval?",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 172048,
      "title": "About the label and categorization",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 171728,
      "title": "Train Data Contains Non-Landmark Images",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 171450,
      "title": "It is taking too much time just to predict features of training files to compare.",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 171434,
      "title": "Why is there a gap between landmark ids?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 171442,
      "title": "Increase LB precision",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 171429,
      "title": "sample_submission.csv VS all samples in the test folder",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 171515,
      "title": "If anyone wanna know what is micro averaged precision?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 171040,
      "title": "how to participate",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 171195,
      "title": "Share links and resources",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    }
  ],
  "errors": []
}