{
  "id": 464891,
  "title": "about scoring",
  "url": "/competitions/blood-vessel-segmentation/discussion/464891",
  "author_name": "",
  "post_date": "2024-01-02T04:38:14.934960800Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>i'm almost new to code competition. so i don't know what kind of content should be included in the jupyter notebook to score properly. i simply created the submission.csv file in the correct directory, but it was not scored. what should be included in the jupyter notebook?</p>",
  "messages": [
    {
      "id": "2583112",
      "postDate": "01/02/2024 04:38:14",
      "content": "<p>i'm almost new to code competition. so i don't know what kind of content should be included in the jupyter notebook to score properly. i simply created the submission.csv file in the correct directory, but it was not scored. what should be included in the jupyter notebook?</p>",
      "rawMarkdown": "i'm almost new to code competition. so i don't know what kind of content should be included in the jupyter notebook to score properly. i simply created the submission.csv file in the correct directory, but it was not scored. what should be included in the jupyter notebook?",
      "votes": null
    },
    {
      "id": "2583474",
      "postDate": "01/02/2024 08:57:09",
      "content": "<p>The model inference code should be in your jupyter notebook that you'll submit to the competition.</p>\n<p>You could train in a separate notebook, save the model as a dataset on Kaggle and use it directly in your inference notebook to avoid training during the 9 hour inference notebook time limit.</p>\n<p>Just browse the code tab of the competition where you'll find several examples of inference notebooks that'll demonstrate the broad idea. All the best!</p>",
      "rawMarkdown": "The model inference code should be in your jupyter notebook that you'll submit to the competition.\n\nYou could train in a separate notebook, save the model as a dataset on Kaggle and use it directly in your inference notebook to avoid training during the 9 hour inference notebook time limit.\n\nJust browse the code tab of the competition where you'll find several examples of inference notebooks that'll demonstrate the broad idea. All the best!",
      "votes": null
    },
    {
      "id": "2584512",
      "postDate": "01/03/2024 02:17:36",
      "content": "<p>thank you for replying!!</p>\n<p>`<br>\nresult = {<br>\n    'id':['kidney_5_0000', 'kidney_5_0001', 'kidney_5_0002', 'kidney_6_0000', 'kidney_6_0001', 'kidney_6_0002'],<br>\n    'rle':[]<br>\n}</p>\n<p>paths = glob.glob('/kaggle/input/blood-vessel-segmentation/test/*<em>/</em>.tif', recursive=True)<br>\npaths.sort()<br>\nfor path in paths:<br>\n    image = cv2.imread(path)<br>\n    image = torchvision.transforms.ToTensor()(image)<br>\n    mask = inference(image)<br>\n    print(mask.shape)<br>\n    result['rle'].append(rle_encode(mask.detach()))<br>\nresult = pd.DataFrame(result)<br>\nresult.to_csv('/kaggle/working/submission.csv', index=False)<br>\n`</p>\n<p>i wrote the code like this. but when I submit the code, it says Notebook Threw Exception. mask.shape is [1303, 912] and mask only consists of 0 or 1. i am using rle_encode function code provided by the contest.</p>",
      "rawMarkdown": "thank you for replying!!\n\n`\nresult = {\n    'id':['kidney_5_0000', 'kidney_5_0001', 'kidney_5_0002', 'kidney_6_0000', 'kidney_6_0001', 'kidney_6_0002'],\n    'rle':[]\n}\n\npaths = glob.glob('/kaggle/input/blood-vessel-segmentation/test/**/*.tif', recursive=True)\npaths.sort()\nfor path in paths:\n    image = cv2.imread(path)\n    image = torchvision.transforms.ToTensor()(image)\n    mask = inference(image)\n    print(mask.shape)\n    result['rle'].append(rle_encode(mask.detach()))\nresult = pd.DataFrame(result)\nresult.to_csv('/kaggle/working/submission.csv', index=False)\n`\n\ni wrote the code like this. but when I submit the code, it says Notebook Threw Exception. mask.shape is [1303, 912] and mask only consists of 0 or 1. i am using rle_encode function code provided by the contest.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2583474,
      "author_name": "phanisrikanth",
      "author_url": "",
      "post_date": "01/02/2024 08:57:09",
      "content": "<p>The model inference code should be in your jupyter notebook that you'll submit to the competition.</p>\n<p>You could train in a separate notebook, save the model as a dataset on Kaggle and use it directly in your inference notebook to avoid training during the 9 hour inference notebook time limit.</p>\n<p>Just browse the code tab of the competition where you'll find several examples of inference notebooks that'll demonstrate the broad idea. All the best!</p>",
      "votes": null,
      "replies": [
        {
          "id": 2584512,
          "author_name": "seoul1988",
          "author_url": "",
          "post_date": "01/03/2024 02:17:36",
          "content": "<p>thank you for replying!!</p>\n<p>`<br>\nresult = {<br>\n    'id':['kidney_5_0000', 'kidney_5_0001', 'kidney_5_0002', 'kidney_6_0000', 'kidney_6_0001', 'kidney_6_0002'],<br>\n    'rle':[]<br>\n}</p>\n<p>paths = glob.glob('/kaggle/input/blood-vessel-segmentation/test/*<em>/</em>.tif', recursive=True)<br>\npaths.sort()<br>\nfor path in paths:<br>\n    image = cv2.imread(path)<br>\n    image = torchvision.transforms.ToTensor()(image)<br>\n    mask = inference(image)<br>\n    print(mask.shape)<br>\n    result['rle'].append(rle_encode(mask.detach()))<br>\nresult = pd.DataFrame(result)<br>\nresult.to_csv('/kaggle/working/submission.csv', index=False)<br>\n`</p>\n<p>i wrote the code like this. but when I submit the code, it says Notebook Threw Exception. mask.shape is [1303, 912] and mask only consists of 0 or 1. i am using rle_encode function code provided by the contest.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2583112": "i'm almost new to code competition. so i don't know what kind of content should be included in the jupyter notebook to score properly. i simply created the submission.csv file in the correct directory, but it was not scored. what should be included in the jupyter notebook?",
    "2583474": "The model inference code should be in your jupyter notebook that you'll submit to the competition.\n\nYou could train in a separate notebook, save the model as a dataset on Kaggle and use it directly in your inference notebook to avoid training during the 9 hour inference notebook time limit.\n\nJust browse the code tab of the competition where you'll find several examples of inference notebooks that'll demonstrate the broad idea. All the best!",
    "2584512": "thank you for replying!!\n\n`\nresult = {\n    'id':['kidney_5_0000', 'kidney_5_0001', 'kidney_5_0002', 'kidney_6_0000', 'kidney_6_0001', 'kidney_6_0002'],\n    'rle':[]\n}\n\npaths = glob.glob('/kaggle/input/blood-vessel-segmentation/test/**/*.tif', recursive=True)\npaths.sort()\nfor path in paths:\n    image = cv2.imread(path)\n    image = torchvision.transforms.ToTensor()(image)\n    mask = inference(image)\n    print(mask.shape)\n    result['rle'].append(rle_encode(mask.detach()))\nresult = pd.DataFrame(result)\nresult.to_csv('/kaggle/working/submission.csv', index=False)\n`\n\ni wrote the code like this. but when I submit the code, it says Notebook Threw Exception. mask.shape is [1303, 912] and mask only consists of 0 or 1. i am using rle_encode function code provided by the contest."
  },
  "source": "meta"
}