{
  "total": 537,
  "topics": [
    {
      "id": 136762,
      "title": "Top student team from Bangladesh",
      "comment_count": 6,
      "views": 0,
      "votes": 12
    },
    {
      "id": 136769,
      "title": "Competition Recap - Congrats to our Winners!",
      "comment_count": 5,
      "views": 0,
      "votes": 41
    },
    {
      "id": 122398,
      "title": "New to Machine Learning or Kaggle?",
      "comment_count": 29,
      "views": 0,
      "votes": 8
    },
    {
      "id": 122405,
      "title": "স্বাগতম (Welcome) from Bengali.AI",
      "comment_count": 8,
      "views": 0,
      "votes": 14
    },
    {
      "id": 122395,
      "title": "Welcome!",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 563095,
      "title": "can anyone help me with using cuda?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 137033,
      "title": "Bengali 2D UMAP Embedding",
      "comment_count": 5,
      "views": 0,
      "votes": 4
    },
    {
      "id": 127149,
      "title": "Visualize your results",
      "comment_count": 17,
      "views": 0,
      "votes": 73
    },
    {
      "id": 130539,
      "title": "Dynamic visualization of pytorch loss and metrics with Visdom ",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 128914,
      "title": "[pytorch] Tricks for Image Classification......",
      "comment_count": 17,
      "views": 0,
      "votes": 61
    },
    {
      "id": 134925,
      "title": "Your notebook tried to allocate more memory than is available. It has restarted.",
      "comment_count": 6,
      "views": 0,
      "votes": 7
    },
    {
      "id": 208876,
      "title": "handrwitten word extraction from json file ",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 129818,
      "title": "Class Activation Maps for all 3 symbols",
      "comment_count": 7,
      "views": 0,
      "votes": 10
    },
    {
      "id": 135966,
      "title": "2nd place solution",
      "comment_count": 60,
      "views": 0,
      "votes": 151
    },
    {
      "id": 132228,
      "title": "Which kind of Augmentation is useful in this competition？",
      "comment_count": 45,
      "views": 0,
      "votes": 19
    },
    {
      "id": 137024,
      "title": "Debrief - Thoughts on computer vision with limited hardware",
      "comment_count": 12,
      "views": 0,
      "votes": 29
    },
    {
      "id": 133119,
      "title": "Google Colab and pytorch - CUDA out of memory",
      "comment_count": 15,
      "views": 0,
      "votes": 1
    },
    {
      "id": 132894,
      "title": "[Keras Users] EfficientNet Noisy Student weights released",
      "comment_count": 15,
      "views": 0,
      "votes": 89
    },
    {
      "id": 122623,
      "title": "What is a parquet file?",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 128115,
      "title": "mixup/cutmix with label smoothing",
      "comment_count": 5,
      "views": 0,
      "votes": 34
    },
    {
      "id": 127486,
      "title": "Cutmix-mixup instability--shakup?",
      "comment_count": 7,
      "views": 0,
      "votes": 5
    },
    {
      "id": 130811,
      "title": "[pytorch] LSR implementation",
      "comment_count": 4,
      "views": 0,
      "votes": 9
    },
    {
      "id": 132319,
      "title": "Using fastai2 on Kaggle Kernels",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 133766,
      "title": "Cant get tensorboard to work",
      "comment_count": 11,
      "views": 0,
      "votes": 12
    },
    {
      "id": 135984,
      "title": "1st place solution /w code",
      "comment_count": 136,
      "views": 0,
      "votes": 488
    },
    {
      "id": 136021,
      "title": "14th Place - Hacking Macro Recall - Chris Writeup",
      "comment_count": 70,
      "views": 0,
      "votes": 157
    },
    {
      "id": 123198,
      "title": "Best single model",
      "comment_count": 408,
      "views": 0,
      "votes": 148
    },
    {
      "id": 157902,
      "title": "Late Submission - CNN Data Pipeline + Problem Solving Tutorial",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 134929,
      "title": "Worth Seeing Skills(Classification)",
      "comment_count": 5,
      "views": 0,
      "votes": 15
    },
    {
      "id": 134633,
      "title": "who trained GhostNet? 5 models inference ~6min",
      "comment_count": 8,
      "views": 0,
      "votes": 8
    },
    {
      "id": 132664,
      "title": "A New Lightweight CNN Architecture: GhostNet",
      "comment_count": 18,
      "views": 0,
      "votes": 26
    },
    {
      "id": 136102,
      "title": "🏅Gold Medal Solutions🏅",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 151355,
      "title": "Can't see API command for download",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 133857,
      "title": "For those using Colab (Pro): Preventing Session from disconnecting due to inactivity",
      "comment_count": 32,
      "views": 0,
      "votes": 34
    },
    {
      "id": 133160,
      "title": "How Image size effects pretrained model(Ex ResNet) performance?",
      "comment_count": 5,
      "views": 0,
      "votes": 4
    },
    {
      "id": 124646,
      "title": "Training on Colab?",
      "comment_count": 32,
      "views": 0,
      "votes": 6
    },
    {
      "id": 135750,
      "title": "Shocked before Shakeup",
      "comment_count": 25,
      "views": 0,
      "votes": 48
    },
    {
      "id": 149891,
      "title": "Explanation of Top Solutions and Comp Summary",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 130895,
      "title": "Training on Google Colab Pro: Issues faced",
      "comment_count": 11,
      "views": 0,
      "votes": 1
    },
    {
      "id": 129323,
      "title": "Retraining checkpoints increases train and validation loss in pytorch",
      "comment_count": 16,
      "views": 0,
      "votes": 11
    },
    {
      "id": 147602,
      "title": "Cite and Reference",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 126025,
      "title": "Problems with Colab",
      "comment_count": 19,
      "views": 0,
      "votes": 2
    },
    {
      "id": 122397,
      "title": "Looking for a Team Thread",
      "comment_count": 121,
      "views": 0,
      "votes": 11
    },
    {
      "id": 144549,
      "title": "167th Place Solution with code",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 131554,
      "title": "Notebooks with tqdm progress bar gives error when committed.",
      "comment_count": 8,
      "views": 0,
      "votes": 4
    },
    {
      "id": 134952,
      "title": "ValueError: Unknown activation function:swish",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 132517,
      "title": "Weight decay in Adam optimizer",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 137356,
      "title": "Disappointment",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 133811,
      "title": "Unable to submit",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 141437,
      "title": "I'm sharing My tips for this competition in Kaggler TV.",
      "comment_count": 4,
      "views": 0,
      "votes": 19
    },
    {
      "id": 122738,
      "title": "Notebook Timeout Error",
      "comment_count": 8,
      "views": 0,
      "votes": 4
    },
    {
      "id": 132937,
      "title": "Tensorflow version",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 136815,
      "title": "10th Place: Single Model Using Snapshot Ensemble w/ code",
      "comment_count": 31,
      "views": 0,
      "votes": 83
    },
    {
      "id": 135982,
      "title": "3rd place solution",
      "comment_count": 25,
      "views": 0,
      "votes": 85
    },
    {
      "id": 135990,
      "title": "8th Place Solution /w Code (minimal ver.)",
      "comment_count": 70,
      "views": 0,
      "votes": 173
    },
    {
      "id": 137768,
      "title": "Our simple approach 😮 and Gold Solutions Summary 💪",
      "comment_count": 8,
      "views": 0,
      "votes": 12
    },
    {
      "id": 135960,
      "title": "7th Place Data-based Generalization Solution",
      "comment_count": 35,
      "views": 0,
      "votes": 95
    },
    {
      "id": 136056,
      "title": "21st Solution using only Kaggle Kernels",
      "comment_count": 19,
      "views": 0,
      "votes": 58
    },
    {
      "id": 136129,
      "title": "5th place solution",
      "comment_count": 29,
      "views": 0,
      "votes": 90
    },
    {
      "id": 136116,
      "title": "13th place brief writeup (public 107th) ",
      "comment_count": 28,
      "views": 0,
      "votes": 29
    },
    {
      "id": 136027,
      "title": "what to do after competition?",
      "comment_count": 21,
      "views": 0,
      "votes": 79
    },
    {
      "id": 135985,
      "title": "9th Solution",
      "comment_count": 14,
      "views": 0,
      "votes": 45
    },
    {
      "id": 135811,
      "title": "A new 3x grandmaster??",
      "comment_count": 12,
      "views": 0,
      "votes": 12
    },
    {
      "id": 137167,
      "title": "1416th place solution + alpha",
      "comment_count": 13,
      "views": 0,
      "votes": 38
    },
    {
      "id": 136982,
      "title": "4th place solution",
      "comment_count": 11,
      "views": 0,
      "votes": 38
    },
    {
      "id": 136011,
      "title": "6th place solution",
      "comment_count": 17,
      "views": 0,
      "votes": 53
    },
    {
      "id": 136030,
      "title": "🏅Summary of Top Solutions🏅",
      "comment_count": 9,
      "views": 0,
      "votes": 84
    },
    {
      "id": 135963,
      "title": "Congrats to winners",
      "comment_count": 13,
      "views": 0,
      "votes": 23
    },
    {
      "id": 137552,
      "title": "Key of the 10th solution? : Where sSE Block looks",
      "comment_count": 0,
      "views": 0,
      "votes": 8
    },
    {
      "id": 137831,
      "title": "BengaliAi Shakeup Plot",
      "comment_count": 1,
      "views": 0,
      "votes": 8
    },
    {
      "id": 136075,
      "title": "Solution 38, some lessons learned",
      "comment_count": 11,
      "views": 0,
      "votes": 42
    },
    {
      "id": 137353,
      "title": "Potential 6th place solution with a single model ! [all about layers]",
      "comment_count": 0,
      "views": 0,
      "votes": 12
    },
    {
      "id": 137652,
      "title": "Kaggle Competition Experience(Team: Learning at snail pace) | Rank 1473",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 136795,
      "title": "I got a medal without doing any work",
      "comment_count": 13,
      "views": 0,
      "votes": 19
    },
    {
      "id": 135998,
      "title": "Gold medal - 12 place solution",
      "comment_count": 17,
      "views": 0,
      "votes": 40
    },
    {
      "id": 136791,
      "title": "14th Place Solution",
      "comment_count": 5,
      "views": 0,
      "votes": 23
    },
    {
      "id": 136005,
      "title": "Special Thank you to Heng and Qishen Ha",
      "comment_count": 9,
      "views": 0,
      "votes": 22
    },
    {
      "id": 136106,
      "title": "Selecting submission",
      "comment_count": 3,
      "views": 0,
      "votes": 32
    },
    {
      "id": 137950,
      "title": "My overview solution 0.9361 with code",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 135959,
      "title": "Overview of my solution",
      "comment_count": 17,
      "views": 0,
      "votes": 24
    },
    {
      "id": 135961,
      "title": "Wrong Submission - Lost 34th place ",
      "comment_count": 26,
      "views": 0,
      "votes": 12
    },
    {
      "id": 136054,
      "title": "Public LB vs Private LB",
      "comment_count": 1,
      "views": 0,
      "votes": 22
    },
    {
      "id": 137153,
      "title": "Unaccounted competition points & learned lessons",
      "comment_count": 3,
      "views": 0,
      "votes": 7
    },
    {
      "id": 136025,
      "title": "CAM CutMix Augmentation",
      "comment_count": 6,
      "views": 0,
      "votes": 25
    },
    {
      "id": 134697,
      "title": "In the interest of full disclosure.",
      "comment_count": 13,
      "views": 0,
      "votes": 22
    },
    {
      "id": 135834,
      "title": "Competition not over yet",
      "comment_count": 8,
      "views": 0,
      "votes": 25
    },
    {
      "id": 135962,
      "title": "Shocking at the shakeup!!!",
      "comment_count": 24,
      "views": 0,
      "votes": 9
    },
    {
      "id": 137029,
      "title": "CutMix/MixUp is **Not** All You Need?",
      "comment_count": 9,
      "views": 0,
      "votes": 9
    },
    {
      "id": 136870,
      "title": "22 place Solution",
      "comment_count": 7,
      "views": 0,
      "votes": 7
    },
    {
      "id": 136064,
      "title": "24th place solution - Post processing(+0.013 Private LB)",
      "comment_count": 1,
      "views": 0,
      "votes": 15
    },
    {
      "id": 135634,
      "title": "We don't believe in magic, but how do you do it?",
      "comment_count": 22,
      "views": 0,
      "votes": 14
    },
    {
      "id": 135785,
      "title": "Bengali Top 25 Animation",
      "comment_count": 0,
      "views": 0,
      "votes": 29
    },
    {
      "id": 137016,
      "title": "Interesting finding",
      "comment_count": 5,
      "views": 0,
      "votes": 5
    },
    {
      "id": 136111,
      "title": "An Important Lesson (for procrastinators like me)",
      "comment_count": 6,
      "views": 0,
      "votes": 4
    },
    {
      "id": 136877,
      "title": "257th Place Solution and Observations (public LB 55)",
      "comment_count": 4,
      "views": 0,
      "votes": 7
    },
    {
      "id": 136761,
      "title": "My Solution Writeup + Code",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 136799,
      "title": "55th place solution - A frugal approach",
      "comment_count": 3,
      "views": 0,
      "votes": 6
    },
    {
      "id": 135974,
      "title": "Is se_resnext weird?",
      "comment_count": 16,
      "views": 0,
      "votes": 3
    },
    {
      "id": 136024,
      "title": "This shake up makes me scream so hard",
      "comment_count": 7,
      "views": 0,
      "votes": 6
    },
    {
      "id": 136084,
      "title": "[RECAP] 41st place solution with code",
      "comment_count": 5,
      "views": 0,
      "votes": 14
    },
    {
      "id": 136132,
      "title": "Checking R,C,V LB scores",
      "comment_count": 13,
      "views": 0,
      "votes": 5
    },
    {
      "id": 136780,
      "title": "38th Place - ZoneMix - A domain specific augmentation",
      "comment_count": 8,
      "views": 0,
      "votes": 8
    },
    {
      "id": 135900,
      "title": "Different Data Distribution between Train/Test?",
      "comment_count": 10,
      "views": 0,
      "votes": 4
    },
    {
      "id": 135973,
      "title": "What I learnt as a beginner",
      "comment_count": 7,
      "views": 0,
      "votes": 6
    },
    {
      "id": 124539,
      "title": "Notebook Commits",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 136062,
      "title": "Is this much shake up normal?",
      "comment_count": 5,
      "views": 0,
      "votes": 3
    },
    {
      "id": 134161,
      "title": "Bengali on TPUs [1 model 1 fold -> 0.9708]",
      "comment_count": 43,
      "views": 0,
      "votes": 83
    },
    {
      "id": 135317,
      "title": "[pytorch] how to do OHEM?  (loss = loss_focal_ohem*a + loss_softmax*b + loss_triplet*c ？)",
      "comment_count": 8,
      "views": 0,
      "votes": 13
    },
    {
      "id": 135977,
      "title": "Biggest Shake in All Computer Vision Competitions?",
      "comment_count": 7,
      "views": 0,
      "votes": 3
    },
    {
      "id": 134844,
      "title": "Anyone managed to make your CV go above 0.99 with only cutout or no aug at all?",
      "comment_count": 20,
      "views": 0,
      "votes": 4
    },
    {
      "id": 136999,
      "title": "Anyone going to participate in Abstraction and Reasoning Challenge",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 136115,
      "title": "Congrats To All Winners",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 136152,
      "title": "Deleted notebook",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134651,
      "title": "Why cropping black pixels does not work",
      "comment_count": 19,
      "views": 0,
      "votes": 23
    },
    {
      "id": 135930,
      "title": "Newbie looking forward to the drama",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 135987,
      "title": "Submission CSV Not Found is better than real",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 136014,
      "title": "Thank you to all the masters and experts in the discussion forum",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 136902,
      "title": "External Data Graphemes",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 136886,
      "title": "Conclusion from Top Solutions: Solutiosn for Unseen Graphemes",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 135957,
      "title": "This competition is too fierce to fall asleep",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 135969,
      "title": "About the test data set.",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 135907,
      "title": "Absolute Madlad!",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 136832,
      "title": "lb disappear, help",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 136006,
      "title": "All Combinations 168 x 11 x 7 + Edge Cases",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 135763,
      "title": "Buying TPU credits with Kaggle Points",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 136040,
      "title": "A great Final Ending",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 135780,
      "title": "I wish that the error submission does not count as 1 of the 5 daily submissions",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 135979,
      "title": "Jumped 1200 positions on leader-board! ",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 135725,
      "title": "Shakeup",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 136072,
      "title": "Notebook Threw Exception error",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 135917,
      "title": "Ensembling with keras",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 136043,
      "title": "Collecting Top Solutions",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 135552,
      "title": "Slide Algorithm - might be a new era in deep learning",
      "comment_count": 1,
      "views": 0,
      "votes": 13
    },
    {
      "id": 135691,
      "title": "5 submissions left!",
      "comment_count": 1,
      "views": 0,
      "votes": 6
    },
    {
      "id": 136066,
      "title": "O God, only if I chose the right kernel",
      "comment_count": 2,
      "views": 0,
      "votes": -7
    },
    {
      "id": 135944,
      "title": "Tensorflow/Keras BUG: Repeatedly calling model.predict(...) results in memory leak #13118",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134434,
      "title": " Have a Try on Validating With Unseen?",
      "comment_count": 20,
      "views": 0,
      "votes": 34
    },
    {
      "id": 134589,
      "title": "how to visualize effects epoch, learning rate, optimizer ",
      "comment_count": 8,
      "views": 0,
      "votes": 28
    },
    {
      "id": 134905,
      "title": "GeM pooling convergence slow?",
      "comment_count": 17,
      "views": 0,
      "votes": 8
    },
    {
      "id": 135829,
      "title": "Fastai batch predictions from image array",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 135474,
      "title": "Val loss and Val accuracy both increasing with training, how is this possible?",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 135824,
      "title": "Submission 404 error",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 135715,
      "title": "Competition deadline",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 135698,
      "title": "When will submission limits reset? Which time zone?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 135559,
      "title": "FastAi unable to get scored",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 135487,
      "title": "Scoring error after submission ... :(",
      "comment_count": 10,
      "views": 0,
      "votes": null
    },
    {
      "id": 134472,
      "title": "Multi-Output on TPUs",
      "comment_count": 8,
      "views": 0,
      "votes": 8
    },
    {
      "id": 135734,
      "title": "Test Time Augmentation?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 135363,
      "title": "Anyone else waiting for GPU/TPU quota reset for the last try ?",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 135563,
      "title": "Notebook: Reading Parquet Files - RAM/CPU Optimization",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 135321,
      "title": "About K-Fold in Pytorch with images",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 133219,
      "title": "Some thoughts",
      "comment_count": 28,
      "views": 0,
      "votes": 56
    },
    {
      "id": 134851,
      "title": "Huge difference between CV and LB",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 132644,
      "title": "[solved]fastai cutout can't work",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 135165,
      "title": "Robust Learning with Jacobian Regularization",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 133746,
      "title": "I made a mistake...",
      "comment_count": 33,
      "views": 0,
      "votes": 27
    },
    {
      "id": 134612,
      "title": "[HELP]: What is TTA and Label Smoothing? How to do it?",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 134808,
      "title": "Cannot load data/train on colab",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134035,
      "title": "Are your models robust enough for unseen graphemes?",
      "comment_count": 16,
      "views": 0,
      "votes": 24
    },
    {
      "id": 133066,
      "title": "so I am totally lost",
      "comment_count": 26,
      "views": 0,
      "votes": 18
    },
    {
      "id": 135108,
      "title": "Equalization Loss for Long-Tailed Object Recognition",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 135222,
      "title": "Notebook Exceeded Allowed Compute",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 135132,
      "title": "Be care about 'shuffle'!",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134988,
      "title": "NaN loss",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134943,
      "title": "efficientnet b6 implementation",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 135164,
      "title": "\"Less Label, More Learning\"",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 135158,
      "title": "Model / Augmentation /Hyperparameter exploration",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 134601,
      "title": "Confusing CV and LB",
      "comment_count": 12,
      "views": 0,
      "votes": 4
    },
    {
      "id": 134933,
      "title": "Notebook taking too much time to submit",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 135003,
      "title": "I got error while finding public score",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 134602,
      "title": "Inference speed issue related to pytorch",
      "comment_count": 5,
      "views": 0,
      "votes": 14
    },
    {
      "id": 134199,
      "title": "Need some advice",
      "comment_count": 13,
      "views": 0,
      "votes": 6
    },
    {
      "id": 128161,
      "title": "GridMask implementation",
      "comment_count": 20,
      "views": 0,
      "votes": 75
    },
    {
      "id": 134738,
      "title": "A generalized handwriting model",
      "comment_count": 1,
      "views": 0,
      "votes": 7
    },
    {
      "id": 135044,
      "title": "What is Weights in an Ensemble?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 134055,
      "title": "Submission CSV not found / Submission scoring error",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 134976,
      "title": "Submission Script/Kernel",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134739,
      "title": "Need help about data",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 134781,
      "title": "Notebook cannot use internet access",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 135298,
      "title": "ZZz",
      "comment_count": 0,
      "views": 0,
      "votes": -8
    },
    {
      "id": 134900,
      "title": "How to compute the class_weight to address imbalance issue?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 133321,
      "title": "strange!!! ensemble of lb0.9828 and lb0.9832 is lb0.9795",
      "comment_count": 9,
      "views": 0,
      "votes": 6
    },
    {
      "id": 134258,
      "title": "Handling multi consonant graphemes",
      "comment_count": 6,
      "views": 0,
      "votes": 14
    },
    {
      "id": 134760,
      "title": "How Can  I Slove 'Submission Scoring Error'?",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 134470,
      "title": "Notebook Threw Exception",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 133609,
      "title": "I really suck at image competitions",
      "comment_count": 21,
      "views": 0,
      "votes": 17
    },
    {
      "id": 133735,
      "title": "Bengali AI Dataset - EDA Grapheme Combinations",
      "comment_count": 9,
      "views": 0,
      "votes": 9
    },
    {
      "id": 134757,
      "title": "Need Mentor",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 132898,
      "title": "report on experiments on  \"No Augmentation\"",
      "comment_count": 51,
      "views": 0,
      "votes": 43
    },
    {
      "id": 133322,
      "title": "FMix: FMix improves performance over MixUp and CutMix for a number of state-of-the- art models across a range of data sets and problem settings",
      "comment_count": 6,
      "views": 0,
      "votes": 27
    },
    {
      "id": 122604,
      "title": "Bengali Data Sets & Publications!  ",
      "comment_count": 11,
      "views": 0,
      "votes": 35
    },
    {
      "id": 134619,
      "title": "Why does my inference have submission scoring error?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134618,
      "title": "Finally solved submission errors",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 132642,
      "title": "Useful Baseline Data Augmentations?",
      "comment_count": 24,
      "views": 0,
      "votes": 18
    },
    {
      "id": 134064,
      "title": "How to normalize training images?",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 134493,
      "title": "Image IO performance",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 133568,
      "title": "What do you think of the shake up?",
      "comment_count": 10,
      "views": 0,
      "votes": 14
    },
    {
      "id": 132492,
      "title": "list of all mixup variant",
      "comment_count": 11,
      "views": 0,
      "votes": 32
    },
    {
      "id": 134380,
      "title": "Unicode Visualization of the Bengali Alphabet",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 133760,
      "title": "Any other model except se50? ",
      "comment_count": 15,
      "views": 0,
      "votes": 6
    },
    {
      "id": 133203,
      "title": "How many folds do you use in cross validation?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134219,
      "title": "Submission Scoring Error !!!",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 134330,
      "title": "Problem of imbalanced categorie....",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 132118,
      "title": "Things that work (Or did not work?) (For me)",
      "comment_count": 35,
      "views": 0,
      "votes": 43
    },
    {
      "id": 134279,
      "title": "Model Errors when submitting.",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 134176,
      "title": "how about sigmoid instead of softmax - curiosity",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 134089,
      "title": "an idea from quantization network training",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 132838,
      "title": "Densenet Trials",
      "comment_count": 21,
      "views": 0,
      "votes": 6
    },
    {
      "id": 134108,
      "title": "Notebook Timeout",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 134211,
      "title": "Notebook Exceeded Allowed Compute",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 132977,
      "title": "Image resolution beyond 137x236?",
      "comment_count": 12,
      "views": 0,
      "votes": 13
    },
    {
      "id": 134070,
      "title": "Training with Square VS Non-Square Image!",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 133868,
      "title": "A different method",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134056,
      "title": "[Beginner doubt]",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 134125,
      "title": "How to TTA?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 134119,
      "title": "Unable to submit",
      "comment_count": 7,
      "views": 0,
      "votes": 0
    },
    {
      "id": 129225,
      "title": "study on autoagument",
      "comment_count": 8,
      "views": 0,
      "votes": 26
    },
    {
      "id": 133897,
      "title": "Submission in status 'running' for scores more than 12 hours",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 134042,
      "title": "What is your strategy of tuning data augmentation parameters?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 133910,
      "title": "Resizing Parquet to Image : Non-Square Image",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 133612,
      "title": "How many models can be run in the submission kernel?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 133217,
      "title": "CUTMIX: Train Loss and Validation Loss Not Improving.  ",
      "comment_count": 10,
      "views": 0,
      "votes": 3
    },
    {
      "id": 133671,
      "title": "load weights ",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 133685,
      "title": "Inference time",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 128368,
      "title": "Approach for 0.97 with 64x64x1 input",
      "comment_count": 34,
      "views": 0,
      "votes": 62
    },
    {
      "id": 133729,
      "title": "Accuracy Secret Sauce: 100% on Validation and 95% on Test",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 133507,
      "title": "Submission Scoring Error",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 128637,
      "title": "[pytorch] ohem loss  implementation",
      "comment_count": 26,
      "views": 0,
      "votes": 72
    },
    {
      "id": 132976,
      "title": "How to use 3 channels?",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 128457,
      "title": "A list of submission errors and how to resolve",
      "comment_count": 19,
      "views": 0,
      "votes": 24
    },
    {
      "id": 133369,
      "title": "Stratification",
      "comment_count": 0,
      "views": 0,
      "votes": 7
    },
    {
      "id": 130099,
      "title": "\"Notebook Exceeded Allowed Compute\" cause",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 133386,
      "title": "why Submission Scoring Error  ",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 133402,
      "title": "What is needed in Notebook submission?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 133358,
      "title": "Unable to create new Dataset- Kaggle gets stuck at Bundling",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 133182,
      "title": "Grapheme CV score not improving",
      "comment_count": 1,
      "views": 0,
      "votes": 8
    },
    {
      "id": 131550,
      "title": "Submission csv not found",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 133305,
      "title": "Can anyone guide me? This is my first competition",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 131883,
      "title": "Bengali.AI: Handwritten Grapheme Classification Using PyTorch (Part-1 & 2)",
      "comment_count": 5,
      "views": 0,
      "votes": 7
    },
    {
      "id": 127938,
      "title": "how to display graphemes correctly",
      "comment_count": 3,
      "views": 0,
      "votes": 14
    },
    {
      "id": 133393,
      "title": "Help needed with Normalize",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 131844,
      "title": "Notebook Exceeded Allowed Compute..",
      "comment_count": 8,
      "views": 0,
      "votes": null
    },
    {
      "id": 133332,
      "title": "Why did I get a score of 0.0614 whatever I submitted in Handwritten Grapheme Classification",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 133234,
      "title": "[Local loss] may be a great idea for this competition",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 133292,
      "title": "Cutout Hyper Parameter Selection?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 132722,
      "title": "Cutmix/Mixup Reports",
      "comment_count": 2,
      "views": 0,
      "votes": 17
    },
    {
      "id": 133286,
      "title": "could denoising treatement enhance computer vision performance ?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 133239,
      "title": "Submission time out",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 131358,
      "title": "It feels like I'm doing some basic thing very wrong",
      "comment_count": 64,
      "views": 0,
      "votes": 44
    },
    {
      "id": 130097,
      "title": "How do you cope with overfitting?",
      "comment_count": 10,
      "views": 0,
      "votes": 11
    },
    {
      "id": 132453,
      "title": "Dataset Independent Augmentation Methods (My Result)",
      "comment_count": 8,
      "views": 0,
      "votes": 17
    },
    {
      "id": 133133,
      "title": "Notebook Exceeded Allowed Compute",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 132026,
      "title": "Confusion Matrix",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 133070,
      "title": "How to load a model made by a pip installed module",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 132006,
      "title": "Cutmix-Mixup Callback in fast.ai",
      "comment_count": 16,
      "views": 0,
      "votes": 16
    },
    {
      "id": 133078,
      "title": "Quick questions about kernels submitted",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 133079,
      "title": "bug in submission",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 131694,
      "title": "Large gap between LB/CV",
      "comment_count": 17,
      "views": 0,
      "votes": 2
    },
    {
      "id": 126504,
      "title": "mixup/cutmix is all you need",
      "comment_count": 57,
      "views": 0,
      "votes": 134
    },
    {
      "id": 127976,
      "title": "Worth Seeing Posts and Notebooks",
      "comment_count": 28,
      "views": 0,
      "votes": 204
    },
    {
      "id": 132652,
      "title": "LB/CV gap due to models/augumentations/hyperparams?",
      "comment_count": 8,
      "views": 0,
      "votes": 2
    },
    {
      "id": 131886,
      "title": "is 168 classes for grapheme root enough?",
      "comment_count": 2,
      "views": 0,
      "votes": 8
    },
    {
      "id": 133399,
      "title": "100% ACC",
      "comment_count": 2,
      "views": 0,
      "votes": -17
    },
    {
      "id": 130518,
      "title": "TPU",
      "comment_count": 5,
      "views": 0,
      "votes": 4
    },
    {
      "id": 130678,
      "title": "How to create resized image dataset and save as kernels output for faster training? ",
      "comment_count": 10,
      "views": 0,
      "votes": 25
    },
    {
      "id": 131394,
      "title": "What's your training time?",
      "comment_count": 56,
      "views": 0,
      "votes": 13
    },
    {
      "id": 128159,
      "title": "Notebook Exceeded Allowed Compute",
      "comment_count": 10,
      "views": 0,
      "votes": 3
    },
    {
      "id": 132514,
      "title": "my problems about fastai ImageList",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 132724,
      "title": "Which is better, 3 channel image or 1 channel image?",
      "comment_count": 5,
      "views": 0,
      "votes": 0
    },
    {
      "id": 132012,
      "title": "Notebook Timeout",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 130311,
      "title": "7 things that did not worked",
      "comment_count": 52,
      "views": 0,
      "votes": 62
    },
    {
      "id": 132008,
      "title": "[Native Speaker please help] Component Combinations?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 132244,
      "title": "Training with fp16",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 131734,
      "title": "Other 7 things that did not worked (part 2)",
      "comment_count": 29,
      "views": 0,
      "votes": 15
    },
    {
      "id": 132158,
      "title": "Anybody doing inference on 224x224 sized images?",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 132473,
      "title": "num_workers speed tip",
      "comment_count": 0,
      "views": 0,
      "votes": 6
    },
    {
      "id": 132135,
      "title": "Going beyond .99?",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 132665,
      "title": "How much GPU Time is avaliable for submission?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 128059,
      "title": "Generating More data",
      "comment_count": 16,
      "views": 0,
      "votes": 38
    },
    {
      "id": 132654,
      "title": "Custom logging for Kaggle Kernels",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 128592,
      "title": "[pytorch] A simple implementation of mixup/cutout/Margin loss.....",
      "comment_count": 2,
      "views": 0,
      "votes": 31
    },
    {
      "id": 132537,
      "title": "Can I submit the trained model?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 131679,
      "title": "Today is International Mother Language Day",
      "comment_count": 15,
      "views": 0,
      "votes": 12
    },
    {
      "id": 131944,
      "title": "How to solve Bengali.AI challenge(Youtube videos)",
      "comment_count": 6,
      "views": 0,
      "votes": 6
    },
    {
      "id": 128739,
      "title": "Large LB/CV difference",
      "comment_count": 26,
      "views": 0,
      "votes": 2
    },
    {
      "id": 132288,
      "title": "EfficientNet in TF Keras with cutMix + MixUp + Augmix + GridMask",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 128665,
      "title": "[pytorch] focal loss + ohem implementation",
      "comment_count": 7,
      "views": 0,
      "votes": 23
    },
    {
      "id": 123757,
      "title": "[placeholder] lb0.963+ pytorch starter kit",
      "comment_count": 123,
      "views": 0,
      "votes": 173
    },
    {
      "id": 131861,
      "title": "One Epoch Is All You Need(?)",
      "comment_count": 2,
      "views": 0,
      "votes": 7
    },
    {
      "id": 132119,
      "title": "confusion about mixup/cutmix",
      "comment_count": 11,
      "views": 0,
      "votes": 1
    },
    {
      "id": 132171,
      "title": "CutMixUp",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 132327,
      "title": "Which learning rate and optimizer did you choose?",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 132189,
      "title": "Does Increasing Image size helps?",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 131078,
      "title": "Train Test Distribution",
      "comment_count": 18,
      "views": 0,
      "votes": 31
    },
    {
      "id": 132046,
      "title": "Error when committing new kernel",
      "comment_count": 11,
      "views": 0,
      "votes": null
    },
    {
      "id": 130819,
      "title": "use 3d printer to collect train data ....",
      "comment_count": 5,
      "views": 0,
      "votes": 13
    },
    {
      "id": 131244,
      "title": "fastai2 multi-head example - Question about public sharing",
      "comment_count": 5,
      "views": 0,
      "votes": 8
    },
    {
      "id": 131865,
      "title": "Cutmix for single channel images?",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 131985,
      "title": "I want to know whether I have some ways to use pre-trained model for SE ResNest in pytorch",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 131868,
      "title": "Anyone training on 224x224?",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 132090,
      "title": "How much is your computer's memory？",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 132107,
      "title": "Kernel running problem",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 130007,
      "title": "Optimizer choice",
      "comment_count": 28,
      "views": 0,
      "votes": 6
    },
    {
      "id": 131156,
      "title": "One classifier for all 3 outputs vs 3 distinct classifiers",
      "comment_count": 15,
      "views": 0,
      "votes": 4
    },
    {
      "id": 132036,
      "title": "Convert a grapheme to root, vowel and consonant",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 131832,
      "title": "How long will it take to submit the results?",
      "comment_count": 7,
      "views": 0,
      "votes": 0
    },
    {
      "id": 131891,
      "title": "example of a notebook that works a submission",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 131343,
      "title": "Dose the submit code will comsue my GPU time?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 127883,
      "title": "submission scoring error",
      "comment_count": 16,
      "views": 0,
      "votes": 6
    },
    {
      "id": 131248,
      "title": "Example Words",
      "comment_count": 4,
      "views": 0,
      "votes": 9
    },
    {
      "id": 124904,
      "title": "Solving grapheme_root is the key to 0.98?",
      "comment_count": 38,
      "views": 0,
      "votes": 55
    },
    {
      "id": 129790,
      "title": "se_resnext50_32x4d  vs se_resnext101_32x4d",
      "comment_count": 21,
      "views": 0,
      "votes": 27
    },
    {
      "id": 129697,
      "title": "AugMix implementation based on albumentations",
      "comment_count": 5,
      "views": 0,
      "votes": 32
    },
    {
      "id": 130503,
      "title": "Thinking out of the box",
      "comment_count": 16,
      "views": 0,
      "votes": 40
    },
    {
      "id": 131375,
      "title": "pre-processing images in the test images (private dataset)?",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 131786,
      "title": "Model upload takes forever",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 131428,
      "title": "process of submission",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 130911,
      "title": "Commit Error of New Notebook",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 131535,
      "title": "How to change the dataset from .parquet to .feather type?",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 131596,
      "title": "Dark theme for data scientists!❤️ ❤️",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 122993,
      "title": "Inference within 20 minutes",
      "comment_count": 22,
      "views": 0,
      "votes": 79
    },
    {
      "id": 131393,
      "title": "Saturation Point",
      "comment_count": 12,
      "views": 0,
      "votes": 2
    },
    {
      "id": 126574,
      "title": "Implementation Details of Cutmix & Mixup?",
      "comment_count": 1,
      "views": 0,
      "votes": 14
    },
    {
      "id": 131536,
      "title": "score of my submission ",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 126054,
      "title": "Optimization tips, eliminate Kaggle failures",
      "comment_count": 10,
      "views": 0,
      "votes": 44
    },
    {
      "id": 130286,
      "title": "Congrats to Abhishek for becoming the first 4x GM!",
      "comment_count": 6,
      "views": 0,
      "votes": 3
    },
    {
      "id": 131057,
      "title": "Same submission.csv, different LB score",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 129841,
      "title": "Storm Incoming?",
      "comment_count": 37,
      "views": 0,
      "votes": 21
    },
    {
      "id": 130980,
      "title": "Notebook Timeout",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 130991,
      "title": "What is AutoML?",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 131061,
      "title": "CPU inference kernel commit time 63 seconds, but submission timed out ",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 131321,
      "title": "Is there anything wrong with colab TF 2.1?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 131144,
      "title": "Python 3.5 Trainning Problem",
      "comment_count": 9,
      "views": 0,
      "votes": null
    },
    {
      "id": 128198,
      "title": "Morphological transformations as image augmentation",
      "comment_count": 5,
      "views": 0,
      "votes": 34
    },
    {
      "id": 130797,
      "title": "Label Smoothing Regularization(LSR)",
      "comment_count": 4,
      "views": 0,
      "votes": 6
    },
    {
      "id": 122396,
      "title": "External data disclosure thread",
      "comment_count": 76,
      "views": 0,
      "votes": 16
    },
    {
      "id": 130394,
      "title": "Early Stopping vs. Saving Based on Validation",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 125886,
      "title": "Getting started with fastai2",
      "comment_count": 6,
      "views": 0,
      "votes": 8
    },
    {
      "id": 130986,
      "title": "Has anyone tried Tf-Hub efficientnet? ",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 125221,
      "title": "Label Smoothing and TTA",
      "comment_count": 5,
      "views": 0,
      "votes": 3
    },
    {
      "id": 122421,
      "title": "What are Graphemes?",
      "comment_count": 37,
      "views": 0,
      "votes": 78
    },
    {
      "id": 130905,
      "title": "How to use GPU with Pytorch",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 123432,
      "title": "Some experiments with CNN tails ",
      "comment_count": 26,
      "views": 0,
      "votes": 70
    },
    {
      "id": 122924,
      "title": "Swag Prizes and Free Cloud Credits for Google AutoML",
      "comment_count": 33,
      "views": 0,
      "votes": 20
    },
    {
      "id": 130032,
      "title": "Getting “HTTP Error 403: Forbidden” in google colab",
      "comment_count": 17,
      "views": 0,
      "votes": 8
    },
    {
      "id": 123859,
      "title": "2 graphemes with the same combination of components?",
      "comment_count": 6,
      "views": 0,
      "votes": 30
    },
    {
      "id": 126761,
      "title": "Flipping properties of Bengali characters?",
      "comment_count": 13,
      "views": 0,
      "votes": 13
    },
    {
      "id": 123766,
      "title": "Requesting effective ways to reduce GPU usage to <=2 hrs",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 130179,
      "title": "submission problem",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 130120,
      "title": "Training using Google Cloud Platform? ",
      "comment_count": 5,
      "views": 0,
      "votes": 3
    },
    {
      "id": 128911,
      "title": "Efficientnet - Trials and Analysis",
      "comment_count": 20,
      "views": 0,
      "votes": 23
    },
    {
      "id": 129748,
      "title": "About Automated ML tools and Pretrained models",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 123093,
      "title": "Problem scoring my submission file ",
      "comment_count": 13,
      "views": 0,
      "votes": 2
    },
    {
      "id": 129948,
      "title": "Submission error",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 130256,
      "title": "Training on google colab: Getting error OSError: [Errno 5] Input/output error: '../input/grapheme-imgs-224x224'",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 129668,
      "title": "How I solved my Submission error..",
      "comment_count": 3,
      "views": 0,
      "votes": 4
    },
    {
      "id": 130019,
      "title": "Class Imbalance?",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 127660,
      "title": "175 people have scores of 0.9663; 1 person has a score of 0.9664 and 1 of 0.9662",
      "comment_count": 15,
      "views": 0,
      "votes": 22
    },
    {
      "id": 129270,
      "title": "How to speed up training phase?",
      "comment_count": 17,
      "views": 0,
      "votes": 7
    },
    {
      "id": 129752,
      "title": "Productivity issues on kaggle kernels 2: How to view actual runtime of kaggle kernels",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 129807,
      "title": "Image Shape..do you use square or rectangular?",
      "comment_count": 0,
      "views": 0,
      "votes": 3
    },
    {
      "id": 128765,
      "title": "[Keras] Earlystopping use recall score",
      "comment_count": 12,
      "views": 0,
      "votes": 8
    },
    {
      "id": 129939,
      "title": "GAIERROR",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 127902,
      "title": "why cutout/mixup works",
      "comment_count": 13,
      "views": 0,
      "votes": 35
    },
    {
      "id": 129273,
      "title": "[Question] about cutmix and mixup(help me....)",
      "comment_count": 9,
      "views": 0,
      "votes": 2
    },
    {
      "id": 129456,
      "title": "Strong correlation between val_grapheme_recall and LB",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 129743,
      "title": "Productivity issue on kaggle kernels 1: Tracing variables using ctrl-C ctrl-V takes a lot of time",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 129593,
      "title": "how to consider the correlation?",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 125991,
      "title": "Yet another Submission Scoring Error",
      "comment_count": 12,
      "views": 0,
      "votes": 2
    },
    {
      "id": 129659,
      "title": " 0.0614 constant score",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 129025,
      "title": "Question to moderators: Are we allowed to use external data to train our models?",
      "comment_count": 7,
      "views": 0,
      "votes": 2
    },
    {
      "id": 129038,
      "title": "Best score using only kaggle kernels",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 129352,
      "title": "Error in \"Competition Timeline\" start date.",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 129064,
      "title": "Fast.ai training issue using multiple notebooks? (100 epochs)",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 128222,
      "title": "Reading the Robot Mind",
      "comment_count": 1,
      "views": 0,
      "votes": 4
    },
    {
      "id": 124287,
      "title": "Notebook exceeded allowed compute",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 129133,
      "title": "kernel keeps searching online for pretrained model",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 128527,
      "title": "Is there any way to evaluate a model at an early stage?",
      "comment_count": 9,
      "views": 0,
      "votes": 3
    },
    {
      "id": 128912,
      "title": "About Augmentation ",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 129045,
      "title": "How long does it take you to load test_image_data_.parquet?",
      "comment_count": 16,
      "views": 0,
      "votes": 0
    },
    {
      "id": 126337,
      "title": "[placeholder] autoML results",
      "comment_count": 10,
      "views": 0,
      "votes": 21
    },
    {
      "id": 128898,
      "title": "Augmentation on Validation Set?",
      "comment_count": 5,
      "views": 0,
      "votes": 4
    },
    {
      "id": 126209,
      "title": "Strange tradeoff",
      "comment_count": 27,
      "views": 0,
      "votes": 6
    },
    {
      "id": 125513,
      "title": "TIME TAKEN TO GENERATE PUBLIC SCORE, HOW MUCH? ",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 122849,
      "title": "Multi-label stratified Kfold",
      "comment_count": 14,
      "views": 0,
      "votes": 54
    },
    {
      "id": 128973,
      "title": "Is it useful to flip horizontally ?",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 129131,
      "title": "API error while commiting my notebook",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 126553,
      "title": "[Curiously Solved]Submission Errors",
      "comment_count": 19,
      "views": 0,
      "votes": 3
    },
    {
      "id": 128986,
      "title": "How to use all of the available data in one epoch?",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 128962,
      "title": "Submission rules",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 126922,
      "title": " help!!! submit-to-competition always error Notebook Exceeded Allowed Compute",
      "comment_count": 8,
      "views": 0,
      "votes": 1
    },
    {
      "id": 127719,
      "title": "Papers worth reading?",
      "comment_count": 2,
      "views": 0,
      "votes": 22
    },
    {
      "id": 127973,
      "title": "Image size",
      "comment_count": 16,
      "views": 0,
      "votes": 1
    },
    {
      "id": 128254,
      "title": "Uniform Manifold Approximation and Projection (UMAP)",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 125186,
      "title": "Notebook Timeout Issue",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 128710,
      "title": "[keras] focal implementation",
      "comment_count": 0,
      "views": 0,
      "votes": 2
    },
    {
      "id": 128684,
      "title": "How to normalize the data when using pre-trained model",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 128743,
      "title": "Validation and Cutmix/Mixup",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 128434,
      "title": "What's your single best model's architecture like?",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 128196,
      "title": "Maximum batch size",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 128164,
      "title": "How many graphemes exist?",
      "comment_count": 4,
      "views": 0,
      "votes": 4
    },
    {
      "id": 128534,
      "title": "3 Channels",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 123632,
      "title": "How to resolve \"Submission Error\" in my case",
      "comment_count": 14,
      "views": 0,
      "votes": 32
    },
    {
      "id": 127958,
      "title": "Comparision of different model for bengali handwritten  recognition. ",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 125828,
      "title": "Approach for 0.97 and What's next?",
      "comment_count": 14,
      "views": 0,
      "votes": 46
    },
    {
      "id": 128508,
      "title": "Keep getting 'No module named ...' error in kernel",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 128436,
      "title": "MixMatch SSL with CutMix and MixMatch using fai",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 128178,
      "title": "How to submit a notebook with GPU?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 128260,
      "title": "Local CV vs. Public LB?",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 128096,
      "title": "[Solved] Inference Issue - CSV Not Found",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 127513,
      "title": "help!, what is ttf format and how to convert data to that?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 128064,
      "title": "Difficult Classes to predict",
      "comment_count": 0,
      "views": 0,
      "votes": 10
    },
    {
      "id": 122980,
      "title": "Problem with inference",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 128155,
      "title": "How does the competetion ensure fair play",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 126833,
      "title": "Wrong Train Data Labels -Organizers plz clarify about Private Set",
      "comment_count": 14,
      "views": 0,
      "votes": 35
    },
    {
      "id": 128139,
      "title": "Generic vs Specific model",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 124221,
      "title": "Saving Memory few tips",
      "comment_count": 12,
      "views": 0,
      "votes": 24
    },
    {
      "id": 125689,
      "title": "Problems with kernel commiting",
      "comment_count": 9,
      "views": 0,
      "votes": 1
    },
    {
      "id": 127662,
      "title": "I'm confused by Bengali Grapheme",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 123135,
      "title": "”Submit to Competition“ Error occurred",
      "comment_count": 19,
      "views": 0,
      "votes": 4
    },
    {
      "id": 127793,
      "title": "BornoNet: Bangla Handwritten Characters Recognition Using Convolutional Neural Network",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 127622,
      "title": "GPU environment and installation",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 127763,
      "title": "[Domain knowledge needed] Error analysis",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 127748,
      "title": "Code - 0",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 123895,
      "title": "Issues with Parquet files and R",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 122509,
      "title": "Papers (some with code)",
      "comment_count": 5,
      "views": 0,
      "votes": 43
    },
    {
      "id": 127326,
      "title": "About private stage",
      "comment_count": 3,
      "views": 0,
      "votes": 1
    },
    {
      "id": 126828,
      "title": "In Childhood teachers always told to practice handwriting :(",
      "comment_count": 12,
      "views": 0,
      "votes": 6
    },
    {
      "id": 127053,
      "title": "Last place asking for help ;)",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 126973,
      "title": "How to train model with tf.GradientTape() ?",
      "comment_count": 0,
      "views": 0,
      "votes": 4
    },
    {
      "id": 126847,
      "title": "Is it allowed for this competition to use pretrained NN?",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 126781,
      "title": "Dude... where's my car...? (submission.csv)",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 127248,
      "title": "Something is not okay about inference",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 125450,
      "title": "[off-topic] Learn, Share, Compete but don't CHEAT",
      "comment_count": 8,
      "views": 0,
      "votes": 11
    },
    {
      "id": 126557,
      "title": "Isn't Cutmix a bit risky for this competition?",
      "comment_count": 4,
      "views": 0,
      "votes": 11
    },
    {
      "id": 126534,
      "title": "[Solved] My Every Submission got score 0.0614 ...",
      "comment_count": 19,
      "views": 0,
      "votes": 2
    },
    {
      "id": 126202,
      "title": "report alternative approach here",
      "comment_count": 16,
      "views": 0,
      "votes": 7
    },
    {
      "id": 126247,
      "title": "constant problems with submission script",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 126648,
      "title": "Help for TTA Solution ",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 122504,
      "title": "Another Issue with Score Metric",
      "comment_count": 3,
      "views": 0,
      "votes": 18
    },
    {
      "id": 126667,
      "title": "Keras data generator doesn't work in multiprocessing.",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 123002,
      "title": "Consonant Conjuncts vs Consonant Diacritics",
      "comment_count": 12,
      "views": 0,
      "votes": 28
    },
    {
      "id": 122758,
      "title": "Competition Close to My Heart",
      "comment_count": 5,
      "views": 0,
      "votes": 14
    },
    {
      "id": 126457,
      "title": "What is the advantage of using Parquet?",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 126505,
      "title": "A new image preprocessing method",
      "comment_count": 0,
      "views": 0,
      "votes": 5
    },
    {
      "id": 125567,
      "title": "The leaderboard is too fast",
      "comment_count": 15,
      "views": 0,
      "votes": 14
    },
    {
      "id": 125492,
      "title": "How Deep will you GO??",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 124243,
      "title": "Multi-Output or Multi-Models?",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 125940,
      "title": "Submission: Kaggle Error ",
      "comment_count": 10,
      "views": 0,
      "votes": 4
    },
    {
      "id": 123956,
      "title": "One model with 3 ouputs vs 3 different models",
      "comment_count": 11,
      "views": 0,
      "votes": 10
    },
    {
      "id": 124249,
      "title": "Shake-Shake ",
      "comment_count": 6,
      "views": 0,
      "votes": 9
    },
    {
      "id": 125639,
      "title": "Yet another \"Submission CSV Not Found\"",
      "comment_count": 10,
      "views": 0,
      "votes": 5
    },
    {
      "id": 125819,
      "title": "If you are using mxresnet, here is small boost",
      "comment_count": 3,
      "views": 0,
      "votes": 10
    },
    {
      "id": 126125,
      "title": "I'm looking at the submission csv outputs of two different notebooks I have, both are Identical but holds different scoring, is there something i'm missing here?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 125724,
      "title": "GPU Timer anomaly",
      "comment_count": 6,
      "views": 0,
      "votes": 2
    },
    {
      "id": 126048,
      "title": "Taking long time in making submissions?",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 125529,
      "title": "A way to debug memory problem",
      "comment_count": 5,
      "views": 0,
      "votes": 7
    },
    {
      "id": 125929,
      "title": "Question: Is it legal to save two notebooks for same submission?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 125046,
      "title": "Teacher-Student",
      "comment_count": 10,
      "views": 0,
      "votes": 5
    },
    {
      "id": 125536,
      "title": "Submission Scoring Error - Help",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 125494,
      "title": "Notebook Threw Exception",
      "comment_count": 5,
      "views": 0,
      "votes": 2
    },
    {
      "id": 125359,
      "title": "How to train for long hours in GPU in Kaggle kernels?",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 125486,
      "title": "fast.ai resnet34 not using GPU",
      "comment_count": 2,
      "views": 0,
      "votes": 3
    },
    {
      "id": 125502,
      "title": "Multi-class mixup recall evaluation, maybe you need",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 125196,
      "title": "What about PseudoLabelling?",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 125598,
      "title": "Kaldi",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 125580,
      "title": "Test set size",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 123709,
      "title": "Pytorch X Keras/TF",
      "comment_count": 5,
      "views": 0,
      "votes": 6
    },
    {
      "id": 125312,
      "title": "scoring Issue",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 125185,
      "title": "Ideas for Saving Memory",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 125384,
      "title": "Why is zip file added to kernel not automatically unzipped?",
      "comment_count": 0,
      "views": 0,
      "votes": 0
    },
    {
      "id": 124852,
      "title": "tf.keras model not working",
      "comment_count": 12,
      "views": 0,
      "votes": null
    },
    {
      "id": 125154,
      "title": "Data augmentation and Pre-Processing",
      "comment_count": 1,
      "views": 0,
      "votes": 1
    },
    {
      "id": 125014,
      "title": "How i can uplode trained model from colab to kaggle and make the final submission.. and also it gives me \"Submission CSV Not Found\" error.",
      "comment_count": 6,
      "views": 0,
      "votes": null
    },
    {
      "id": 122716,
      "title": "Why was parquet file format chosen for this competition?",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 123751,
      "title": "synthesizing data from Ekush Dataset?",
      "comment_count": 7,
      "views": 0,
      "votes": 15
    },
    {
      "id": 124927,
      "title": "Why time out when calculating the score?",
      "comment_count": 5,
      "views": 0,
      "votes": null
    },
    {
      "id": 124908,
      "title": "Missing Notebook",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 124179,
      "title": "Notebook Timeout Error",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 124454,
      "title": "Manual LB probing?",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 123025,
      "title": "fast.ai starter kernel now available [0.964 LB]",
      "comment_count": 10,
      "views": 0,
      "votes": 27
    },
    {
      "id": 124639,
      "title": "Problem with accuracy when uploading a trained model",
      "comment_count": 4,
      "views": 0,
      "votes": 0
    },
    {
      "id": 124325,
      "title": "Training models on Google Colab...",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 123980,
      "title": "High Scoring Public Kernels",
      "comment_count": 10,
      "views": 0,
      "votes": 14
    },
    {
      "id": 124516,
      "title": "Submission error",
      "comment_count": 2,
      "views": 0,
      "votes": 1
    },
    {
      "id": 124362,
      "title": "Usage of Class Map CSV file",
      "comment_count": 2,
      "views": 0,
      "votes": -1
    },
    {
      "id": 124528,
      "title": "How to overcome \"Notebook Exceeded Allowed computation time\" while submitting the answers",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 123212,
      "title": "Submission Scoring Error / Submission CSV Not Found",
      "comment_count": 6,
      "views": 0,
      "votes": 5
    },
    {
      "id": 124401,
      "title": "Wired thing happening",
      "comment_count": 8,
      "views": 0,
      "votes": -4
    },
    {
      "id": 123752,
      "title": "Recall Score For Keras Metrics",
      "comment_count": 16,
      "views": 0,
      "votes": 4
    },
    {
      "id": 124096,
      "title": "Shape of the image",
      "comment_count": 4,
      "views": 0,
      "votes": null
    },
    {
      "id": 123913,
      "title": "Unable to Submit",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 124204,
      "title": "Issue downloading a weights file from the internet",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 123376,
      "title": "Many pre-trained models (40+)",
      "comment_count": 2,
      "views": 0,
      "votes": 31
    },
    {
      "id": 123659,
      "title": "What is this error supposed to mean?",
      "comment_count": 3,
      "views": 0,
      "votes": 5
    },
    {
      "id": 123248,
      "title": "How long did your model train.",
      "comment_count": 11,
      "views": 0,
      "votes": 10
    },
    {
      "id": 122720,
      "title": "why is 0.0000?  my submit",
      "comment_count": 3,
      "views": 0,
      "votes": 3
    },
    {
      "id": 123658,
      "title": "Submission Score worst than expected",
      "comment_count": 5,
      "views": 0,
      "votes": 3
    },
    {
      "id": 123753,
      "title": "Individual loss or combined loss? ",
      "comment_count": 2,
      "views": 0,
      "votes": 2
    },
    {
      "id": 123680,
      "title": "Getting Error Submission CSV Not Found",
      "comment_count": 4,
      "views": 0,
      "votes": 2
    },
    {
      "id": 123877,
      "title": "Invalid Redis response payload: {\"Key\":\"timeout\",\"Result\":[\"\"]}",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 123544,
      "title": "Kernel restarting again and again",
      "comment_count": 7,
      "views": 0,
      "votes": 1
    },
    {
      "id": 123676,
      "title": "Beginner Notebook Submission Questions",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 123478,
      "title": "Can organizers provide more test images? Please!",
      "comment_count": 9,
      "views": 0,
      "votes": null
    },
    {
      "id": 122778,
      "title": "Using fastai to load parquet file",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 123801,
      "title": "How to efficiently read and resize paraquet files?",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 123782,
      "title": "Submission CSV Not Found Error",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 122921,
      "title": "Kaggle should have paid GPU limit extension",
      "comment_count": 4,
      "views": 0,
      "votes": 10
    },
    {
      "id": 123655,
      "title": "Total number of images in test set?",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 123417,
      "title": "RAM Usage -Small but important point",
      "comment_count": 0,
      "views": 0,
      "votes": 9
    },
    {
      "id": 123255,
      "title": "GCP Credits?",
      "comment_count": 2,
      "views": 0,
      "votes": 4
    },
    {
      "id": 123627,
      "title": "How to avoid 'Notebook threw exception' error",
      "comment_count": 0,
      "views": 0,
      "votes": 1
    },
    {
      "id": 123388,
      "title": "Kaggle Kernel CPU shooting up while loading parquet ?",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 123252,
      "title": "Overfitting on test set?",
      "comment_count": 5,
      "views": 0,
      "votes": 1
    },
    {
      "id": 122721,
      "title": "Phonemes and Graphemes",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 122467,
      "title": "Train image dataset (256px)",
      "comment_count": 12,
      "views": 0,
      "votes": 23
    },
    {
      "id": 122694,
      "title": "Loading weight in kernel",
      "comment_count": 1,
      "views": 0,
      "votes": null
    },
    {
      "id": 122581,
      "title": "Winning Notebooks from Similar Competitions",
      "comment_count": 0,
      "views": 0,
      "votes": 30
    },
    {
      "id": 122912,
      "title": "Struggling with getting images into a model",
      "comment_count": 9,
      "views": 0,
      "votes": 1
    },
    {
      "id": 122515,
      "title": "Baseline results",
      "comment_count": 6,
      "views": 0,
      "votes": 20
    },
    {
      "id": 122549,
      "title": "Top Great Articles About Bengali Language Recognition",
      "comment_count": 2,
      "views": 0,
      "votes": 22
    },
    {
      "id": 122474,
      "title": "Train a model without GPU takes about 48 hours per epoch",
      "comment_count": 10,
      "views": 0,
      "votes": 0
    },
    {
      "id": 122913,
      "title": "Private notebook rule",
      "comment_count": 6,
      "views": 0,
      "votes": 1
    },
    {
      "id": 122639,
      "title": "[Grapheme] LB LB 0.9259 Inference kernel and model kernels",
      "comment_count": 1,
      "views": 0,
      "votes": 12
    },
    {
      "id": 122700,
      "title": "Not the expected PublicScore",
      "comment_count": 4,
      "views": 0,
      "votes": 1
    },
    {
      "id": 122731,
      "title": "Images cropped to character & resized to 100x100 (Dataset)",
      "comment_count": 0,
      "views": 0,
      "votes": 16
    },
    {
      "id": 122925,
      "title": "Help with Inference scheme.. ",
      "comment_count": 2,
      "views": 0,
      "votes": null
    },
    {
      "id": 122420,
      "title": "Is the number of test images 12?",
      "comment_count": 4,
      "views": 0,
      "votes": 5
    },
    {
      "id": 122657,
      "title": "Facing problem when submitting ",
      "comment_count": 7,
      "views": 0,
      "votes": null
    },
    {
      "id": 122629,
      "title": "Unequal proportion of subclasses in graphemes",
      "comment_count": 1,
      "views": 0,
      "votes": 2
    },
    {
      "id": 122580,
      "title": "Understanding Graphemes { English }",
      "comment_count": 1,
      "views": 0,
      "votes": 5
    },
    {
      "id": 122425,
      "title": "little bug in evaluation metric",
      "comment_count": 2,
      "views": 0,
      "votes": 5
    },
    {
      "id": 122898,
      "title": "Find teammate",
      "comment_count": 0,
      "views": 0,
      "votes": -5
    },
    {
      "id": 122681,
      "title": "help!Who can tell me the processing flow of this image recognition?",
      "comment_count": 3,
      "views": 0,
      "votes": null
    },
    {
      "id": 122490,
      "title": "Query on uploading trained models",
      "comment_count": 3,
      "views": 0,
      "votes": 2
    },
    {
      "id": 122401,
      "title": "Learn Bengali Alphabets(Vowels)  in English",
      "comment_count": 1,
      "views": 0,
      "votes": 3
    },
    {
      "id": 122550,
      "title": "Hi everyone",
      "comment_count": 1,
      "views": 0,
      "votes": null
    }
  ],
  "errors": []
}