{
  "id": 203206,
  "title": "why my network loss and acc always stay at same number?",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/203206",
  "author_name": "",
  "post_date": "2020-12-14T08:10:27.552591Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>i'm new here, for  this network, i create a notebook for leaf classify. but , however, my model always stay at same accury and loss in different epoch? i do some change to overcome this problem, such as make ever layers requires_grad = True, and make model as global variable, and so on, but not overcome this problem. can you help me to find out the problem i have? here is my notebook location<br>\n<a href=\"https://www.kaggle.com/buctl2013014093/train-leaf-classify\" target=\"_blank\">https://www.kaggle.com/buctl2013014093/train-leaf-classify</a></p>",
  "messages": [
    {
      "id": "1112025",
      "postDate": "12/14/2020 08:10:27",
      "content": "<p>i'm new here, for  this network, i create a notebook for leaf classify. but , however, my model always stay at same accury and loss in different epoch? i do some change to overcome this problem, such as make ever layers requires_grad = True, and make model as global variable, and so on, but not overcome this problem. can you help me to find out the problem i have? here is my notebook location<br>\n<a href=\"https://www.kaggle.com/buctl2013014093/train-leaf-classify\" target=\"_blank\">https://www.kaggle.com/buctl2013014093/train-leaf-classify</a></p>",
      "rawMarkdown": "i'm new here, for  this network, i create a notebook for leaf classify. but , however, my model always stay at same accury and loss in different epoch? i do some change to overcome this problem, such as make ever layers requires_grad = True, and make model as global variable, and so on, but not overcome this problem. can you help me to find out the problem i have? here is my notebook location\nhttps://www.kaggle.com/buctl2013014093/train-leaf-classify",
      "votes": null
    },
    {
      "id": "1112319",
      "postDate": "12/14/2020 13:25:54",
      "content": "<p>The notebook link you have given is not the right one ig. </p>",
      "rawMarkdown": "The notebook link you have given is not the right one ig.",
      "votes": null
    },
    {
      "id": "1112513",
      "postDate": "12/14/2020 17:07:11",
      "content": "<p>i change the url,  so you can find it correctly :), and can you give me some advice for my model?</p>",
      "rawMarkdown": "i change the url,  so you can find it correctly :), and can you give me some advice for my model?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1112319,
      "author_name": "shanmukh05",
      "author_url": "",
      "post_date": "12/14/2020 13:25:54",
      "content": "<p>The notebook link you have given is not the right one ig. </p>",
      "votes": null,
      "replies": [
        {
          "id": 1112513,
          "author_name": "buctl2013014093",
          "author_url": "",
          "post_date": "12/14/2020 17:07:11",
          "content": "<p>i change the url,  so you can find it correctly :), and can you give me some advice for my model?</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1112025": "i'm new here, for  this network, i create a notebook for leaf classify. but , however, my model always stay at same accury and loss in different epoch? i do some change to overcome this problem, such as make ever layers requires_grad = True, and make model as global variable, and so on, but not overcome this problem. can you help me to find out the problem i have? here is my notebook location\nhttps://www.kaggle.com/buctl2013014093/train-leaf-classify",
    "1112319": "The notebook link you have given is not the right one ig.",
    "1112513": "i change the url,  so you can find it correctly :), and can you give me some advice for my model?"
  },
  "source": "meta"
}