{
  "id": 87180,
  "title": "What About using K Fold cross validation while training the neural Network ?",
  "url": "/competitions/iwildcam-2019-fgvc6/discussion/87180",
  "author_name": "",
  "post_date": "2019-03-29T11:02:22.986429800Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hello , I am being an active participant and from some other competitions I can observe that using stratified k fold cross validation strategy we can validate our model more robustly , so can we use something like that , if someone can provide framework and code samples to do that k-fold cross validation for this competition.</p>",
  "messages": [
    {
      "id": "503000",
      "postDate": "03/29/2019 11:02:22",
      "content": "<p>Hello , I am being an active participant and from some other competitions I can observe that using stratified k fold cross validation strategy we can validate our model more robustly , so can we use something like that , if someone can provide framework and code samples to do that k-fold cross validation for this competition.</p>",
      "rawMarkdown": "Hello , I am being an active participant and from some other competitions I can observe that using stratified k fold cross validation strategy we can validate our model more robustly , so can we use something like that , if someone can provide framework and code samples to do that k-fold cross validation for this competition.",
      "votes": null
    },
    {
      "id": "541732",
      "postDate": "06/03/2019 01:01:48",
      "content": "<p>You can split the dataset to train and val by sklean.preprocess. And you should feed the train and val data with index to your NN model. The code will be different if you use different DL lib. But I can give some suggestion, you will find some useful kernel in some tabular data competition of thie website.</p>\n\n<blockquote>\n  <p><strong>Athar Sayed wrote</strong></p>\n  \n  <blockquote>\n    <p>Hello , I am being an active participant and from some other competitions I can observe that using stratified k fold cross validation strategy we can validate our model more robustly , so can we use something like that , if someone can provide framework and code samples to do that k-fold cross validation for this competition.</p>\n  </blockquote>\n</blockquote>",
      "rawMarkdown": "You can split the dataset to train and val by sklean.preprocess. And you should feed the train and val data with index to your NN model. The code will be different if you use different DL lib. But I can give some suggestion, you will find some useful kernel in some tabular data competition of thie website.\n\n&gt; **Athar Sayed wrote**\n&gt; \n&gt; &gt; Hello , I am being an active participant and from some other competitions I can observe that using stratified k fold cross validation strategy we can validate our model more robustly , so can we use something like that , if someone can provide framework and code samples to do that k-fold cross validation for this competition.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 541732,
      "author_name": "tenffe",
      "author_url": "",
      "post_date": "06/03/2019 01:01:48",
      "content": "<p>You can split the dataset to train and val by sklean.preprocess. And you should feed the train and val data with index to your NN model. The code will be different if you use different DL lib. But I can give some suggestion, you will find some useful kernel in some tabular data competition of thie website.</p>\n\n<blockquote>\n  <p><strong>Athar Sayed wrote</strong></p>\n  \n  <blockquote>\n    <p>Hello , I am being an active participant and from some other competitions I can observe that using stratified k fold cross validation strategy we can validate our model more robustly , so can we use something like that , if someone can provide framework and code samples to do that k-fold cross validation for this competition.</p>\n  </blockquote>\n</blockquote>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "503000": "Hello , I am being an active participant and from some other competitions I can observe that using stratified k fold cross validation strategy we can validate our model more robustly , so can we use something like that , if someone can provide framework and code samples to do that k-fold cross validation for this competition.",
    "541732": "You can split the dataset to train and val by sklean.preprocess. And you should feed the train and val data with index to your NN model. The code will be different if you use different DL lib. But I can give some suggestion, you will find some useful kernel in some tabular data competition of thie website.\n\n&gt; **Athar Sayed wrote**\n&gt; \n&gt; &gt; Hello , I am being an active participant and from some other competitions I can observe that using stratified k fold cross validation strategy we can validate our model more robustly , so can we use something like that , if someone can provide framework and code samples to do that k-fold cross validation for this competition."
  },
  "source": "meta"
}