{
  "total": 122,
  "topics": [
    {
      "id": 134816,
      "title": "Welcome to Plant Pathology 2020 Challenge!",
      "comment_count": 17,
      "views": 0,
      "votes": 27
    },
    {
      "id": 406482,
      "title": "Evaluation Metrics",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 336934,
      "title": "Detection of multiple disease on single leaf",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 154056,
      "title": "1st place solution(single model)",
      "comment_count": 45,
      "views": 0,
      "votes": 146
    },
    {
      "id": 234432,
      "title": "Does multiple diseases =1 means there is both rust and scab ",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 141549,
      "title": "Kernel run out of RAM/ How to pass data in batches?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 192675,
      "title": "Trying a block pre-processor ",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 135070,
      "title": "looking for a team!",
      "comment_count": 21,
      "views": 0,
      "votes": 7
    },
    {
      "id": 154033,
      "title": "[placeholder] (8th->5th) place solution",
      "comment_count": 10,
      "views": 0,
      "votes": 18
    },
    {
      "id": 144700,
      "title": "image eda tool",
      "comment_count": 4,
      "views": 0,
      "votes": 7
    },
    {
      "id": 164815,
      "title": "Problems with target and input size error",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 154183,
      "title": "[Summary] Top Place Solutions(updated)",
      "comment_count": 2,
      "views": 0,
      "votes": 32
    },
    {
      "id": 154473,
      "title": "very simple 46th solution + potential 17th solution",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 155929,
      "title": "The 2nd Solutions(Single fold and single model)",
      "comment_count": 4,
      "views": 0,
      "votes": 8
    },
    {
      "id": 161873,
      "title": "Powerful Image Processing cum EDA",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 162723,
      "title": "A very Powerful [EDA + Preprocessing] + [Deep Learning throttle] ",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 161989,
      "title": "rank not on leaderboard",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 160506,
      "title": "I couldnot find myself in leaderboard????",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 154159,
      "title": "0.98454(>1st) solution (RegNetY)",
      "comment_count": 7,
      "views": 0,
      "votes": 8
    },
    {
      "id": 157749,
      "title": "3rd Place Score Model Summary",
      "comment_count": 1,
      "views": 0,
      "votes": 7
    },
    {
      "id": 159297,
      "title": "When the accuracy is reduced, the loss is also reduced.Is it overfitting?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 154098,
      "title": "10th place - missed 30+ better submissions :)",
      "comment_count": 8,
      "views": 0,
      "votes": 16
    },
    {
      "id": 154034,
      "title": "[place holder] Original 5th place solution",
      "comment_count": 7,
      "views": 0,
      "votes": 23
    },
    {
      "id": 154035,
      "title": "[place holder]12th solution ",
      "comment_count": 2,
      "views": 0,
      "votes": 11
    },
    {
      "id": 154086,
      "title": "7th place solution",
      "comment_count": 7,
      "views": 0,
      "votes": 14
    },
    {
      "id": 154068,
      "title": "Real Test distribution and Shakeup explanation",
      "comment_count": 0,
      "views": 0,
      "votes": 15
    },
    {
      "id": 154045,
      "title": "4th place solution",
      "comment_count": 3,
      "views": 0,
      "votes": 9
    },
    {
      "id": 154444,
      "title": "TOP 4%",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 154128,
      "title": "20th Place: Bilinear CNN with FL + LS",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 153839,
      "title": "Single model scoring higher than 0.98?",
      "comment_count": 8,
      "views": 0,
      "votes": 3
    },
    {
      "id": 154126,
      "title": "[placeholder] 49th solution + 0.981 solution",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 154038,
      "title": "What is your highest private score? Did you choose it?",
      "comment_count": 14,
      "views": 0,
      "votes": 3
    },
    {
      "id": 149558,
      "title": "loss becomes nan while training on TPU",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 154059,
      "title": "33rd->27th place solution",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 154606,
      "title": "Private Leaderboard Finalised",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 154060,
      "title": "😍Could high scorers make public single model training code for us to learn",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 154132,
      "title": "Top/High scoring solutions using Fastai",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 152313,
      "title": "Google Colab TPU : Training stuck 'after' 1st epoch",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 153549,
      "title": "MIXUP logic in this competition",
      "comment_count": 1,
      "views": 0,
      "votes": 0
    },
    {
      "id": 153552,
      "title": "Applying TTA on multiple_diseases only improves my LB score",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 149845,
      "title": "Smoothing helped me improve my Public LB rank from 102 to 59",
      "comment_count": 17,
      "views": 0,
      "votes": 17
    },
    {
      "id": 135065,
      "title": "Leafsnap: A dataset of leaf images for classification",
      "comment_count": 1,
      "views": 0,
      "votes": 10
    },
    {
      "id": 147804,
      "title": "Deadline of Plant Pathology 2020 - FGVC7 competition is extended!",
      "comment_count": 18,
      "views": 0,
      "votes": 5
    },
    {
      "id": 149554,
      "title": "SHAKE SHAKE SHAKE by KC & THE SUNSHINE BAND",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 152027,
      "title": "Imbalanced Data:Plant Pathology 2020 - FGVC7",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 144632,
      "title": "\"Your notebook tried to allocate more memory than is available. It has restarted.\" ERROR",
      "comment_count": 10,
      "views": 0,
      "votes": -2
    },
    {
      "id": 149808,
      "title": "Ensembling strategy",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 150523,
      "title": "Try Pseudolabeling",
      "comment_count": 8,
      "views": 0,
      "votes": 3
    },
    {
      "id": 147793,
      "title": "Why the Leaderboard lies - and why I need to apologize",
      "comment_count": 8,
      "views": 0,
      "votes": 23
    },
    {
      "id": 148253,
      "title": "Issues with loading the dataset in google colab.",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 151843,
      "title": "Does one need extra data to get a high LB score?",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 151484,
      "title": "Would you like to make a team?",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 151269,
      "title": "Can anyone provide me a guide to using TPU with Pytorch?",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 147889,
      "title": "Dealing with multiple_diseases class",
      "comment_count": 13,
      "views": 0,
      "votes": 3
    },
    {
      "id": 150974,
      "title": "EfficientNet Not installing when committing notebook",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 150041,
      "title": "Score not improving ! ",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 149930,
      "title": "TPU Training using PyTorch XLA parellely",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 149400,
      "title": "Extra Healthy/Diseased Plant Leaf Datasets",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 149729,
      "title": "Dataset Fidelity - Duplicate/Similar Images",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 149686,
      "title": "Simple TPU kernel with few models ensemble (LB score = 0.978)",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 149609,
      "title": "What LR schedulers do you use?",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 149523,
      "title": "message\"We can't find that page.\"",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 148938,
      "title": "Looking for guidance from top scorers",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 149388,
      "title": "some questions about how to improve accuracy",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 149381,
      "title": "Transfer Learning - Freezing strategy",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 148631,
      "title": "Val 98% but LB 97%",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 140014,
      "title": "Best single model.",
      "comment_count": 62,
      "views": 0,
      "votes": 25
    },
    {
      "id": 145816,
      "title": "[Share] Original + Plant Village Dataset",
      "comment_count": 17,
      "views": 0,
      "votes": 12
    },
    {
      "id": 148857,
      "title": "How much highest score achieved using GPU not TPU",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 147232,
      "title": "404We can't find that page. You can search Kaggle above or visit our homepage.",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 148495,
      "title": "Error while loading Keras Model to re-train it",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 145378,
      "title": "Cropping images to reduce noise",
      "comment_count": 4,
      "views": 0,
      "votes": 11
    },
    {
      "id": 147759,
      "title": "The effect of precision on the submission file",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 146993,
      "title": "Help with Dealing with Class Imbalance",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 147860,
      "title": "How can I use Colab for this project?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 147888,
      "title": "pytorch-lightning+resnet50 baseline",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 145656,
      "title": "Submission file",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 145969,
      "title": "Someone in top is cheating",
      "comment_count": 7,
      "views": 0,
      "votes": 10
    },
    {
      "id": 146287,
      "title": "Looking for a team mate!",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 146679,
      "title": "Converting prediction scores to one-hot leads to poor score on LB",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 145789,
      "title": "How to use EfficientNet + TPU for Pytorch?",
      "comment_count": 11,
      "views": 0,
      "votes": 2
    },
    {
      "id": 138223,
      "title": "Which techniques worked and which didn't ? ",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 142278,
      "title": "Problem with transfering data to Google Colaboratory",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 137972,
      "title": "Is there any correlation between your CV and LB?",
      "comment_count": 18,
      "views": 0,
      "votes": 14
    },
    {
      "id": 143098,
      "title": "Model Stacking",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 145949,
      "title": "Can random seed influence LB score?",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 140640,
      "title": "how about growing apple tree as train data",
      "comment_count": 9,
      "views": 0,
      "votes": 40
    },
    {
      "id": 145290,
      "title": "Tiers will motivate beginners",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 142883,
      "title": "Huge gap between CV & LB. Any help is welcome",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 144640,
      "title": " How to use label smoothing on tpu, I failed, but gpu can run",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 144956,
      "title": "CutOut or Random Erasing in TensorFlow Datasets",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 142451,
      "title": "Multi-stage pipeline?",
      "comment_count": 7,
      "views": 0,
      "votes": 6
    },
    {
      "id": 144718,
      "title": "tf或者keras怎么实现mixup",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 144310,
      "title": "What kind of image processing works for you !",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 144287,
      "title": "where is my submission.csv? :)",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 143974,
      "title": "why tf.keras.metric.AUC() not work?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 143684,
      "title": "Problem when downloading data via the API",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 142049,
      "title": "Images that have fingers.",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 143895,
      "title": "problems with training",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 143680,
      "title": "TTA with TPU and tensorflow dataset HELP !",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 141015,
      "title": "Another dataset: Plant Village",
      "comment_count": 4,
      "views": 0,
      "votes": 12
    },
    {
      "id": 142793,
      "title": "How to calculate F1_score efficiently?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 142381,
      "title": "How to analyze this competition? What I am missing?",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 139862,
      "title": "About noise label and how to solve it.",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 142231,
      "title": "Can't understand the submission.csv",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 135475,
      "title": "Why there is no medal awrad in this Resarch competition",
      "comment_count": 6,
      "views": 0,
      "votes": 13
    },
    {
      "id": 135051,
      "title": "Duplicated images",
      "comment_count": 9,
      "views": 0,
      "votes": 35
    },
    {
      "id": 140013,
      "title": "Useful Augmentations.",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 140766,
      "title": "Use 5K-fold Efficientnet B7 ",
      "comment_count": 2,
      "views": 0,
      "votes": 6
    },
    {
      "id": 141580,
      "title": "Error when training EfficientNet",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 140342,
      "title": " End to End Fastai tutorial for image classification",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 140463,
      "title": "Load Augmented Data to GCS",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 138257,
      "title": "Teaching the Network to Crop the Image",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 139915,
      "title": "Can't commit.",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 138430,
      "title": "Are there bad labels?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 135703,
      "title": "A similar competition in FGVC6 which may be helpful",
      "comment_count": 1,
      "views": 0,
      "votes": 11
    },
    {
      "id": 135085,
      "title": "This doesn't appear to be a multi-label problem",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 137953,
      "title": "Can I solve this as a regression problem?",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 137861,
      "title": "Looking for Merge",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 134899,
      "title": "prize as described in website",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 135392,
      "title": "Starter kernel: EDA and Models",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 134932,
      "title": "Will the  CVPR 2020 be canceled?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    }
  ],
  "errors": []
}