{
  "id": 369531,
  "title": "How to submit?",
  "url": "/competitions/rsna-breast-cancer-detection/discussion/369531",
  "author_name": "Vovinsa",
  "post_date": "2022-11-30T11:37:05.604000",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>To submit predictions I need to turn off internet, and I cant't download model from torchvision. Also all my saved weights are utilised when I make submission</p>",
  "messages": [
    {
      "id": 2049855,
      "postDate": "2022-11-30T11:37:05.603Z",
      "content": "<p>To submit predictions I need to turn off internet, and I cant't download model from torchvision. Also all my saved weights are utilised when I make submission</p>",
      "rawMarkdown": "To submit predictions I need to turn off internet, and I cant't download model from torchvision. Also all my saved weights are utilised when I make submission",
      "votes": 1
    },
    {
      "id": 2050009,
      "postDate": "2022-11-30T13:32:15.827Z",
      "content": "<p>You'll need to load any dependencies and models into datasets and include them into your infer notebook. Here's a nice dataset for pylibjpeg an GDCM (for exporting DICOM files) that I use -&gt;  <a href=\"https://www.kaggle.com/code/vslaykovsky/rsna-2022-whl\" target=\"_blank\">https://www.kaggle.com/code/vslaykovsky/rsna-2022-whl</a></p>\n<p>Include this dataset, then add some code like this at the top to copy/install. </p>\n<pre><code>:\n     pylibjpeg\n:\n    !mkdir -p /root/.cache/torch/hub/checkpoints/\n    !pip install /kaggle//rsna--whl/{pydicom--py3-none-.whl,pylibjpeg--py3-none-.whl,python_gdcm--cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl}\n    !pip install /kaggle//rsna--whl/{torch--cp37-cp37m-manylinux1_x86_64.whl,torchvision--cp37-cp37m-manylinux1_x86_64.whl}\n</code></pre>\n<p>Then load your pretrained models into a dataset as well and reference them in your code.</p>",
      "rawMarkdown": "You'll need to load any dependencies and models into datasets and include them into your infer notebook. Here's a nice dataset for pylibjpeg an GDCM (for exporting DICOM files) that I use ->  https://www.kaggle.com/code/vslaykovsky/rsna-2022-whl\n\nInclude this dataset, then add some code like this at the top to copy/install. \n\n```python\ntry:\n    import pylibjpeg\nexcept:\n    !mkdir -p /root/.cache/torch/hub/checkpoints/\n    !pip install /kaggle/input/rsna-2022-whl/{pydicom-2.3.0-py3-none-any.whl,pylibjpeg-1.4.0-py3-none-any.whl,python_gdcm-3.0.15-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl}\n    !pip install /kaggle/input/rsna-2022-whl/{torch-1.12.1-cp37-cp37m-manylinux1_x86_64.whl,torchvision-0.13.1-cp37-cp37m-manylinux1_x86_64.whl}\n```\n\nThen load your pretrained models into a dataset as well and reference them in your code.",
      "votes": 2,
      "replies": [
        {
          "id": 2050786,
          "postDate": "2022-12-01T02:08:23.413Z",
          "content": "<p>thank you <a href=\"https://www.kaggle.com/davidbroberts\" target=\"_blank\">@davidbroberts</a> will give this method a try! 🙏</p>",
          "rawMarkdown": "thank you @davidbroberts will give this method a try! 🙏"
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 2050009,
      "author_name": "David Roberts",
      "author_url": "",
      "post_date": "2022-11-30T13:32:15.827000",
      "content": "<p>You'll need to load any dependencies and models into datasets and include them into your infer notebook. Here's a nice dataset for pylibjpeg an GDCM (for exporting DICOM files) that I use -&gt;  <a href=\"https://www.kaggle.com/code/vslaykovsky/rsna-2022-whl\" target=\"_blank\">https://www.kaggle.com/code/vslaykovsky/rsna-2022-whl</a></p>\n<p>Include this dataset, then add some code like this at the top to copy/install. </p>\n<pre><code>:\n     pylibjpeg\n:\n    !mkdir -p /root/.cache/torch/hub/checkpoints/\n    !pip install /kaggle//rsna--whl/{pydicom--py3-none-.whl,pylibjpeg--py3-none-.whl,python_gdcm--cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl}\n    !pip install /kaggle//rsna--whl/{torch--cp37-cp37m-manylinux1_x86_64.whl,torchvision--cp37-cp37m-manylinux1_x86_64.whl}\n</code></pre>\n<p>Then load your pretrained models into a dataset as well and reference them in your code.</p>",
      "votes": 2,
      "replies": [
        {
          "id": 2050786,
          "author_name": "Radek Osmulski",
          "author_url": "",
          "post_date": "2022-12-01T02:08:23.413000",
          "content": "<p>thank you <a href=\"https://www.kaggle.com/davidbroberts\" target=\"_blank\">@davidbroberts</a> will give this method a try! 🙏</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2049855": "To submit predictions I need to turn off internet, and I cant't download model from torchvision. Also all my saved weights are utilised when I make submission",
    "2050009": "You'll need to load any dependencies and models into datasets and include them into your infer notebook. Here's a nice dataset for pylibjpeg an GDCM (for exporting DICOM files) that I use ->  https://www.kaggle.com/code/vslaykovsky/rsna-2022-whl\n\nInclude this dataset, then add some code like this at the top to copy/install. \n\n```python\ntry:\n    import pylibjpeg\nexcept:\n    !mkdir -p /root/.cache/torch/hub/checkpoints/\n    !pip install /kaggle/input/rsna-2022-whl/{pydicom-2.3.0-py3-none-any.whl,pylibjpeg-1.4.0-py3-none-any.whl,python_gdcm-3.0.15-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl}\n    !pip install /kaggle/input/rsna-2022-whl/{torch-1.12.1-cp37-cp37m-manylinux1_x86_64.whl,torchvision-0.13.1-cp37-cp37m-manylinux1_x86_64.whl}\n```\n\nThen load your pretrained models into a dataset as well and reference them in your code."
  }
}