{
  "id": 523078,
  "title": "training accuracy is higher but predict isn't 训练正确率很高但预测正确率不到0.3",
  "url": "/competitions/digit-recognizer/discussion/523078",
  "author_name": "",
  "post_date": "2024-07-30T05:41:12.477000",
  "votes": 0,
  "comment_count": 1,
  "views": null,
  "content": "<p>In my notebook called ** Digit Number** , the accuracy on training can above 9.7 and higher ,but when I use the model to predict , the result isn't well only can above 0.1 Why?</p>\n<p><em>Maybe my code has any bugs?\nOr the net params is wrong?</em><br>\nIf you want to see my notebook go to the code my notebook is shared called ** Digit Number**</p>\n<p>在我的notebook<strong>Digit Number</strong>中在训练事正确率很高，但是在预测时正确率只有0.1这是为什么？</p>\n<p><em>我的代码有问题？还是模型的参数未加载正确</em><br>\n如果你想看我的代码我的代码已经分享了叫<strong>Digit Number</strong>在code里面<br>\n<a href=\"url\" target=\"_blank\"></a><a href=\"https://www.kaggle.com/code/wzyzozf/digit-number\" target=\"_blank\">https://www.kaggle.com/code/wzyzozf/digit-number</a> </p>",
  "messages": [
    {
      "id": 2974674,
      "postDate": "2024-08-30T23:54:06.023Z",
      "content": "<p>典型的过拟合。建议先学习ML基础。</p>",
      "rawMarkdown": "典型的过拟合。建议先学习ML基础。"
    },
    {
      "id": 2940463,
      "postDate": "2024-07-30T05:41:12.477Z",
      "content": "<p>In my notebook called ** Digit Number** , the accuracy on training can above 9.7 and higher ,but when I use the model to predict , the result isn't well only can above 0.1 Why?</p>\n<p><em>Maybe my code has any bugs?\nOr the net params is wrong?</em><br>\nIf you want to see my notebook go to the code my notebook is shared called ** Digit Number**</p>\n<p>在我的notebook<strong>Digit Number</strong>中在训练事正确率很高，但是在预测时正确率只有0.1这是为什么？</p>\n<p><em>我的代码有问题？还是模型的参数未加载正确</em><br>\n如果你想看我的代码我的代码已经分享了叫<strong>Digit Number</strong>在code里面<br>\n<a href=\"url\" target=\"_blank\"></a><a href=\"https://www.kaggle.com/code/wzyzozf/digit-number\" target=\"_blank\">https://www.kaggle.com/code/wzyzozf/digit-number</a> </p>",
      "rawMarkdown": "In my notebook called ** Digit Number** , the accuracy on training can above 9.7 and higher ,but when I use the model to predict , the result isn't well only can above 0.1 Why?\n\n*Maybe my code has any bugs?\nOr the net params is wrong?*\nIf you want to see my notebook go to the code my notebook is shared called ** Digit Number**\n\n在我的notebook**Digit Number**中在训练事正确率很高，但是在预测时正确率只有0.1这是为什么？\n\n*我的代码有问题？还是模型的参数未加载正确*\n如果你想看我的代码我的代码已经分享了叫**Digit Number**在code里面\n[https://www.kaggle.com/code/wzyzozf/digit-number ](url)"
    }
  ],
  "comments": [
    {
      "id": 2974674,
      "author_name": "SeaYellow",
      "author_url": "",
      "post_date": "2024-08-30T23:54:06.023000",
      "content": "<p>典型的过拟合。建议先学习ML基础。</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2974674": "典型的过拟合。建议先学习ML基础。",
    "2940463": "In my notebook called ** Digit Number** , the accuracy on training can above 9.7 and higher ,but when I use the model to predict , the result isn't well only can above 0.1 Why?\n\n*Maybe my code has any bugs?\nOr the net params is wrong?*\nIf you want to see my notebook go to the code my notebook is shared called ** Digit Number**\n\n在我的notebook**Digit Number**中在训练事正确率很高，但是在预测时正确率只有0.1这是为什么？\n\n*我的代码有问题？还是模型的参数未加载正确*\n如果你想看我的代码我的代码已经分享了叫**Digit Number**在code里面\n[https://www.kaggle.com/code/wzyzozf/digit-number ](url)"
  }
}