{
  "id": 240694,
  "title": "The loss curve of validation set fluctuates severely",
  "url": "/competitions/plant-pathology-2021-fgvc8/discussion/240694",
  "author_name": "Allan",
  "post_date": "2021-05-21T02:55:48.497000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>I trained serveral models, the loss curve on training set converged smoothly,   but the loss curve on validation set fluctuates severely. <br>\nI am kind of worried about the model tends to get over-fitting.</p>\n<p>Is there any solution for this?</p>",
  "messages": [
    {
      "id": 1316929,
      "postDate": "2021-05-21T02:55:48.497Z",
      "content": "<p>I trained serveral models, the loss curve on training set converged smoothly,   but the loss curve on validation set fluctuates severely. <br>\nI am kind of worried about the model tends to get over-fitting.</p>\n<p>Is there any solution for this?</p>",
      "rawMarkdown": "I trained serveral models, the loss curve on training set converged smoothly,   but the loss curve on validation set fluctuates severely. \nI am kind of worried about the model tends to get over-fitting.\n\nIs there any solution for this?"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1316929": "I trained serveral models, the loss curve on training set converged smoothly,   but the loss curve on validation set fluctuates severely. \nI am kind of worried about the model tends to get over-fitting.\n\nIs there any solution for this?"
  }
}