{
  "id": 107076,
  "title": "Can we get NVIDIA Apex in docker",
  "url": "/competitions/aptos2019-blindness-detection/discussion/107076",
  "author_name": "",
  "post_date": "2019-09-02T05:34:55.060810700Z",
  "votes": 4,
  "comment_count": 6,
  "views": 0,
  "content": "<p>Installing apex from public dataset during each kernel run is taking up a lot of time. Can we get apex added to the docker?</p>",
  "messages": [
    {
      "id": "615565",
      "postDate": "09/02/2019 05:34:55",
      "content": "<p>Installing apex from public dataset during each kernel run is taking up a lot of time. Can we get apex added to the docker?</p>",
      "rawMarkdown": "Installing apex from public dataset during each kernel run is taking up a lot of time. Can we get apex added to the docker?",
      "votes": null
    },
    {
      "id": "615594",
      "postDate": "09/02/2019 06:22:51",
      "content": "<p>Yes you can! just like this <a href=\"https://www.kaggle.com/chanhu/eye-efficientnet-pytorch-lb-0-777\">kernel</a></p>",
      "rawMarkdown": "Yes you can! just like this [kernel](https://www.kaggle.com/chanhu/eye-efficientnet-pytorch-lb-0-777)",
      "votes": null
    },
    {
      "id": "615625",
      "postDate": "09/02/2019 07:30:31",
      "content": "<p>I know I can add it as external dataset and install it myself. What I mean is for Kaggle to add it into the docker so we don't have to waste time installing it our self.</p>",
      "rawMarkdown": "I know I can add it as external dataset and install it myself. What I mean is for Kaggle to add it into the docker so we don't have to waste time installing it our self.",
      "votes": null
    },
    {
      "id": "616010",
      "postDate": "09/02/2019 15:43:43",
      "content": "<p>You're welcome to submit a PR: <a href=\"https://github.com/Kaggle/docker-python#requesting-new-packages\">https://github.com/Kaggle/docker-python#requesting-new-packages</a></p>",
      "rawMarkdown": "You're welcome to submit a PR: https://github.com/Kaggle/docker-python#requesting-new-packages",
      "votes": null
    },
    {
      "id": "616076",
      "postDate": "09/02/2019 16:59:41",
      "content": "<p>I checked just now, there is another PR active regarding APEX: <a href=\"https://github.com/Kaggle/docker-python/pull/369/files\">https://github.com/Kaggle/docker-python/pull/369/files</a> <br>\nThat PR might need some changes. It's different to the apex repository instructions on adding the package to docker. Will take a look later. </p>",
      "rawMarkdown": "I checked just now, there is another PR active regarding APEX: https://github.com/Kaggle/docker-python/pull/369/files  \nThat PR might need some changes. It's different to the apex repository instructions on adding the package to docker. Will take a look later.",
      "votes": null
    },
    {
      "id": "616122",
      "postDate": "09/02/2019 17:58:34",
      "content": "<p>Thank you!</p>",
      "rawMarkdown": "Thank you!",
      "votes": null
    },
    {
      "id": "633277",
      "postDate": "09/24/2019 16:57:06",
      "content": "<p>Hi <a href=\"/sohier\">@sohier</a> I just sent a PR with the addition of apex into the Kaggle docker.\n<a href=\"https://github.com/Kaggle/docker-python/pull/617\">https://github.com/Kaggle/docker-python/pull/617</a></p>",
      "rawMarkdown": "Hi @sohier I just sent a PR with the addition of apex into the Kaggle docker.\nhttps://github.com/Kaggle/docker-python/pull/617",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 615594,
      "author_name": "oceanwong",
      "author_url": "",
      "post_date": "09/02/2019 06:22:51",
      "content": "<p>Yes you can! just like this <a href=\"https://www.kaggle.com/chanhu/eye-efficientnet-pytorch-lb-0-777\">kernel</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 615625,
          "author_name": "convexproblems",
          "author_url": "",
          "post_date": "09/02/2019 07:30:31",
          "content": "<p>I know I can add it as external dataset and install it myself. What I mean is for Kaggle to add it into the docker so we don't have to waste time installing it our self.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 616010,
      "author_name": "sohier",
      "author_url": "",
      "post_date": "09/02/2019 15:43:43",
      "content": "<p>You're welcome to submit a PR: <a href=\"https://github.com/Kaggle/docker-python#requesting-new-packages\">https://github.com/Kaggle/docker-python#requesting-new-packages</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 616076,
          "author_name": "convexproblems",
          "author_url": "",
          "post_date": "09/02/2019 16:59:41",
          "content": "<p>I checked just now, there is another PR active regarding APEX: <a href=\"https://github.com/Kaggle/docker-python/pull/369/files\">https://github.com/Kaggle/docker-python/pull/369/files</a> <br>\nThat PR might need some changes. It's different to the apex repository instructions on adding the package to docker. Will take a look later. </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 616122,
          "author_name": "sohier",
          "author_url": "",
          "post_date": "09/02/2019 17:58:34",
          "content": "<p>Thank you!</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 633277,
          "author_name": "convexproblems",
          "author_url": "",
          "post_date": "09/24/2019 16:57:06",
          "content": "<p>Hi <a href=\"/sohier\">@sohier</a> I just sent a PR with the addition of apex into the Kaggle docker.\n<a href=\"https://github.com/Kaggle/docker-python/pull/617\">https://github.com/Kaggle/docker-python/pull/617</a></p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "615565": "Installing apex from public dataset during each kernel run is taking up a lot of time. Can we get apex added to the docker?",
    "615594": "Yes you can! just like this [kernel](https://www.kaggle.com/chanhu/eye-efficientnet-pytorch-lb-0-777)",
    "615625": "I know I can add it as external dataset and install it myself. What I mean is for Kaggle to add it into the docker so we don't have to waste time installing it our self.",
    "616010": "You're welcome to submit a PR: https://github.com/Kaggle/docker-python#requesting-new-packages",
    "616076": "I checked just now, there is another PR active regarding APEX: https://github.com/Kaggle/docker-python/pull/369/files  \nThat PR might need some changes. It's different to the apex repository instructions on adding the package to docker. Will take a look later.",
    "616122": "Thank you!",
    "633277": "Hi @sohier I just sent a PR with the addition of apex into the Kaggle docker.\nhttps://github.com/Kaggle/docker-python/pull/617"
  },
  "source": "meta"
}