{
  "id": 218867,
  "title": "I am getting \"Notebook exceeded allowed compute\" error on submission.",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/218867",
  "author_name": "",
  "post_date": "2021-02-12T11:18:13.312355Z",
  "votes": 1,
  "comment_count": 5,
  "views": 0,
  "content": "<p>I figure the maybe the RAM is getting full.<br>\nplease have a look at my code <a href=\"https://www.kaggle.com/mohneesh7/new-submission-notebook-cassava\" target=\"_blank\">HERE</a></p>\n<p>I am really stuck please help me.</p>",
  "messages": [
    {
      "id": "1197731",
      "postDate": "02/12/2021 11:18:13",
      "content": "<p>I figure the maybe the RAM is getting full.<br>\nplease have a look at my code <a href=\"https://www.kaggle.com/mohneesh7/new-submission-notebook-cassava\" target=\"_blank\">HERE</a></p>\n<p>I am really stuck please help me.</p>",
      "rawMarkdown": "I figure the maybe the RAM is getting full.\nplease have a look at my code [HERE](https://www.kaggle.com/mohneesh7/new-submission-notebook-cassava)\n\nI am really stuck please help me.",
      "votes": null
    },
    {
      "id": "1197734",
      "postDate": "02/12/2021 11:21:11",
      "content": "<p>Use K.clear_session() after each loop. Use model.predict_on_batch() instead on what you are using and convert every image to a tensor using tf.convert_to_tensor() </p>",
      "rawMarkdown": "Use K.clear_session() after each loop. Use model.predict_on_batch() instead on what you are using and convert every image to a tensor using tf.convert_to_tensor()",
      "votes": null
    },
    {
      "id": "1197760",
      "postDate": "02/12/2021 11:58:34",
      "content": "<p>thank you for the reply, I am trying out using clearing session, will let you know if it works.</p>",
      "rawMarkdown": "thank you for the reply, I am trying out using clearing session, will let you know if it works.",
      "votes": null
    },
    {
      "id": "1198650",
      "postDate": "02/13/2021 07:31:02",
      "content": "<p><a href=\"https://www.kaggle.com/mo312057\" target=\"_blank\">@mo312057</a> did it work ?</p>",
      "rawMarkdown": "mo312057 did it work ?",
      "votes": null
    },
    {
      "id": "1198871",
      "postDate": "02/13/2021 10:56:37",
      "content": "<p>It did work actually, the  clear session did the trick. Thank you very much</p>",
      "rawMarkdown": "It did work actually, the  clear session did the trick. Thank you very much",
      "votes": null
    },
    {
      "id": "1198882",
      "postDate": "02/13/2021 11:08:10",
      "content": "<p><a href=\"https://www.kaggle.com/mohneesh7\" target=\"_blank\">@mohneesh7</a> glad it worked </p>",
      "rawMarkdown": "mohneesh7 glad it worked",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1197734,
      "author_name": "mithilsalunkhe",
      "author_url": "",
      "post_date": "02/12/2021 11:21:11",
      "content": "<p>Use K.clear_session() after each loop. Use model.predict_on_batch() instead on what you are using and convert every image to a tensor using tf.convert_to_tensor() </p>",
      "votes": null,
      "replies": [
        {
          "id": 1197760,
          "author_name": "mohneesh7",
          "author_url": "",
          "post_date": "02/12/2021 11:58:34",
          "content": "<p>thank you for the reply, I am trying out using clearing session, will let you know if it works.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1198650,
          "author_name": "mithilsalunkhe",
          "author_url": "",
          "post_date": "02/13/2021 07:31:02",
          "content": "<p><a href=\"https://www.kaggle.com/mo312057\" target=\"_blank\">@mo312057</a> did it work ?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1198871,
          "author_name": "mohneesh7",
          "author_url": "",
          "post_date": "02/13/2021 10:56:37",
          "content": "<p>It did work actually, the  clear session did the trick. Thank you very much</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1198882,
          "author_name": "mithilsalunkhe",
          "author_url": "",
          "post_date": "02/13/2021 11:08:10",
          "content": "<p><a href=\"https://www.kaggle.com/mohneesh7\" target=\"_blank\">@mohneesh7</a> glad it worked </p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1197731": "I figure the maybe the RAM is getting full.\nplease have a look at my code [HERE](https://www.kaggle.com/mohneesh7/new-submission-notebook-cassava)\n\nI am really stuck please help me.",
    "1197734": "Use K.clear_session() after each loop. Use model.predict_on_batch() instead on what you are using and convert every image to a tensor using tf.convert_to_tensor()",
    "1197760": "thank you for the reply, I am trying out using clearing session, will let you know if it works.",
    "1198650": "mo312057 did it work ?",
    "1198871": "It did work actually, the  clear session did the trick. Thank you very much",
    "1198882": "mohneesh7 glad it worked"
  },
  "source": "meta"
}