{
  "id": 146177,
  "title": "RAM issues",
  "url": "/competitions/prostate-cancer-grade-assessment/discussion/146177",
  "author_name": "Digvijay Yadav",
  "post_date": "2020-04-26T07:34:43.690000",
  "votes": -1,
  "comment_count": 7,
  "views": 0,
  "content": "<p>My RAM is getting full what should I o?</p>",
  "messages": [
    {
      "id": 821594,
      "postDate": "2020-04-26T09:21:56.760Z",
      "content": "<p>Hi, <a href=\"/digvijayyadav\">@digvijayyadav</a> I know it's a bit frustrating. But please provide enough information to get help from others. Like when you say, I am out of memory, what does it mean to anyone? Are you playing PUBG on your CPU PC or what? 😄 </p>\n\n<p>IMHO, to get help, please provide what you are up to. What you are doing and what is your environmental status, like are you using Colab, Kaggle Kernel, or local PC or anything basically that relates to it. I saw downvote of your issue but I upvote it, just to give you encouragement to ask the question when you in need. Its kaggle not stack-overflow. 👍 </p>",
      "rawMarkdown": "Hi, @digvijayyadav I know it's a bit frustrating. But please provide enough information to get help from others. Like when you say, I am out of memory, what does it mean to anyone? Are you playing PUBG on your CPU PC or what? 😄 \n\nIMHO, to get help, please provide what you are up to. What you are doing and what is your environmental status, like are you using Colab, Kaggle Kernel, or local PC or anything basically that relates to it. I saw downvote of your issue but I upvote it, just to give you encouragement to ask the question when you in need. Its kaggle not stack-overflow. 👍 ",
      "votes": 6,
      "replies": [
        {
          "id": 821597,
          "postDate": "2020-04-26T09:24:41.103Z",
          "content": "<p><a href=\"/ipythonx\">@ipythonx</a> Thanks for reaching out i am using kaggle kernal I will posting my doubts in a moment</p>",
          "rawMarkdown": "@ipythonx Thanks for reaching out i am using kaggle kernal I will posting my doubts in a moment\n"
        }
      ]
    },
    {
      "id": 821920,
      "postDate": "2020-04-26T14:04:35.547Z",
      "content": "<p>First, you should definitely not running experiments locally.\nWhen you run experiments which creates big neural network models it takes lots of memory.\nTherefore, you need to consider running experiments on cloud environments.</p>\n\n<p>For this purpose, I'd like to recommend to use data-science platform.\nPersonally, I'm using cnvrg.io platform for versioning my data, run experiments with big data sets and deploy models to production.</p>\n\n<p>By the way - they have launched free version of their platform this month.</p>\n\n<p><a href=\"https://cnvrg.io/platform/core/\">https://cnvrg.io/platform/core/</a></p>",
      "rawMarkdown": "First, you should definitely not running experiments locally.\nWhen you run experiments which creates big neural network models it takes lots of memory.\nTherefore, you need to consider running experiments on cloud environments.\n\nFor this purpose, I'd like to recommend to use data-science platform.\nPersonally, I'm using cnvrg.io platform for versioning my data, run experiments with big data sets and deploy models to production.\n\nBy the way - they have launched free version of their platform this month.\n\nhttps://cnvrg.io/platform/core/",
      "votes": 1,
      "replies": [
        {
          "id": 822771,
          "postDate": "2020-04-27T05:54:30.903Z",
          "content": "<p><a href=\"/omerliberman\">@omerliberman</a> thanks a lot!!</p>",
          "rawMarkdown": "@omerliberman thanks a lot!!"
        }
      ]
    },
    {
      "id": 821551,
      "postDate": "2020-04-26T08:46:35.063Z",
      "content": "<p>Use smaller images.</p>",
      "rawMarkdown": "Use smaller images.",
      "votes": 1
    },
    {
      "id": 821613,
      "postDate": "2020-04-26T09:44:19.153Z",
      "content": "<p>Are you using skimage by any chance? If you want to sample efficiently from the slide without loading the whole image, you can take a look at the OpenSlide examples. OpenSlide can read directly from the file, instead of loading it first into memory.</p>",
      "rawMarkdown": "Are you using skimage by any chance? If you want to sample efficiently from the slide without loading the whole image, you can take a look at the OpenSlide examples. OpenSlide can read directly from the file, instead of loading it first into memory.",
      "votes": 2,
      "replies": [
        {
          "id": 821753,
          "postDate": "2020-04-26T11:52:53.873Z",
          "content": "<p>I am using openslide and i think becuase of them it is happening. <a href=\"/wouterbulten\">@wouterbulten</a> Thanks for suggesting skiimage I ill try that</p>",
          "rawMarkdown": "I am using openslide and i think becuase of them it is happening. @wouterbulten Thanks for suggesting skiimage I ill try that"
        }
      ]
    },
    {
      "id": 821465,
      "postDate": "2020-04-26T07:34:43.690Z",
      "content": "<p>My RAM is getting full what should I o?</p>",
      "rawMarkdown": "My RAM is getting full what should I o?",
      "votes": -1
    }
  ],
  "comments": [
    {
      "id": 821594,
      "author_name": "Innat",
      "author_url": "",
      "post_date": "2020-04-26T09:21:56.760000",
      "content": "<p>Hi, <a href=\"/digvijayyadav\">@digvijayyadav</a> I know it's a bit frustrating. But please provide enough information to get help from others. Like when you say, I am out of memory, what does it mean to anyone? Are you playing PUBG on your CPU PC or what? 😄 </p>\n\n<p>IMHO, to get help, please provide what you are up to. What you are doing and what is your environmental status, like are you using Colab, Kaggle Kernel, or local PC or anything basically that relates to it. I saw downvote of your issue but I upvote it, just to give you encouragement to ask the question when you in need. Its kaggle not stack-overflow. 👍 </p>",
      "votes": 6,
      "replies": [
        {
          "id": 821597,
          "author_name": "Digvijay Yadav",
          "author_url": "",
          "post_date": "2020-04-26T09:24:41.103000",
          "content": "<p><a href=\"/ipythonx\">@ipythonx</a> Thanks for reaching out i am using kaggle kernal I will posting my doubts in a moment</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 821920,
      "author_name": "Omer Liberman",
      "author_url": "",
      "post_date": "2020-04-26T14:04:35.547000",
      "content": "<p>First, you should definitely not running experiments locally.\nWhen you run experiments which creates big neural network models it takes lots of memory.\nTherefore, you need to consider running experiments on cloud environments.</p>\n\n<p>For this purpose, I'd like to recommend to use data-science platform.\nPersonally, I'm using cnvrg.io platform for versioning my data, run experiments with big data sets and deploy models to production.</p>\n\n<p>By the way - they have launched free version of their platform this month.</p>\n\n<p><a href=\"https://cnvrg.io/platform/core/\">https://cnvrg.io/platform/core/</a></p>",
      "votes": 1,
      "replies": [
        {
          "id": 822771,
          "author_name": "Digvijay Yadav",
          "author_url": "",
          "post_date": "2020-04-27T05:54:30.903000",
          "content": "<p><a href=\"/omerliberman\">@omerliberman</a> thanks a lot!!</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 821551,
      "author_name": "Matt",
      "author_url": "",
      "post_date": "2020-04-26T08:46:35.063000",
      "content": "<p>Use smaller images.</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 821613,
      "author_name": "Wouter Bulten",
      "author_url": "",
      "post_date": "2020-04-26T09:44:19.153000",
      "content": "<p>Are you using skimage by any chance? If you want to sample efficiently from the slide without loading the whole image, you can take a look at the OpenSlide examples. OpenSlide can read directly from the file, instead of loading it first into memory.</p>",
      "votes": 2,
      "replies": [
        {
          "id": 821753,
          "author_name": "Digvijay Yadav",
          "author_url": "",
          "post_date": "2020-04-26T11:52:53.873000",
          "content": "<p>I am using openslide and i think becuase of them it is happening. <a href=\"/wouterbulten\">@wouterbulten</a> Thanks for suggesting skiimage I ill try that</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "821594": "Hi, @digvijayyadav I know it's a bit frustrating. But please provide enough information to get help from others. Like when you say, I am out of memory, what does it mean to anyone? Are you playing PUBG on your CPU PC or what? 😄 \n\nIMHO, to get help, please provide what you are up to. What you are doing and what is your environmental status, like are you using Colab, Kaggle Kernel, or local PC or anything basically that relates to it. I saw downvote of your issue but I upvote it, just to give you encouragement to ask the question when you in need. Its kaggle not stack-overflow. 👍 ",
    "821920": "First, you should definitely not running experiments locally.\nWhen you run experiments which creates big neural network models it takes lots of memory.\nTherefore, you need to consider running experiments on cloud environments.\n\nFor this purpose, I'd like to recommend to use data-science platform.\nPersonally, I'm using cnvrg.io platform for versioning my data, run experiments with big data sets and deploy models to production.\n\nBy the way - they have launched free version of their platform this month.\n\nhttps://cnvrg.io/platform/core/",
    "821551": "Use smaller images.",
    "821613": "Are you using skimage by any chance? If you want to sample efficiently from the slide without loading the whole image, you can take a look at the OpenSlide examples. OpenSlide can read directly from the file, instead of loading it first into memory.",
    "821465": "My RAM is getting full what should I o?"
  }
}