{
  "id": 397470,
  "title": "Tentative solution",
  "url": "/competitions/asl-signs/discussion/397470",
  "author_name": "",
  "post_date": "2023-03-25T19:17:54.203401300Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p><em>Loss Function</em>: Since it is a multi-class classification, Sparse Categorical Loss can help<br>\n<em>Optimozer</em>: RMSProp with hyperparameter tuning like learning rate, momentum<br>\n<em>Model</em>: Conv2D+ LSTM, Attention Layer, Using Crop Layer for getting rid of redundant black part of the images</p>",
  "messages": [
    {
      "id": "2196993",
      "postDate": "03/25/2023 19:17:54",
      "content": "<p><em>Loss Function</em>: Since it is a multi-class classification, Sparse Categorical Loss can help<br>\n<em>Optimozer</em>: RMSProp with hyperparameter tuning like learning rate, momentum<br>\n<em>Model</em>: Conv2D+ LSTM, Attention Layer, Using Crop Layer for getting rid of redundant black part of the images</p>",
      "rawMarkdown": "*Loss Function*: Since it is a multi-class classification, Sparse Categorical Loss can help\n*Optimozer*: RMSProp with hyperparameter tuning like learning rate, momentum\n*Model*: Conv2D+ LSTM, Attention Layer, Using Crop Layer for getting rid of redundant black part of the images",
      "votes": null
    },
    {
      "id": "2197018",
      "postDate": "03/25/2023 19:56:13",
      "content": "<p>Black images? Do you mean filtering missed hands/faces? <br>\nWould you be kind as to share what CV do you obtain? On what KFOLD split do you prefer?</p>",
      "rawMarkdown": "Black images? Do you mean filtering missed hands/faces? \nWould you be kind as to share what CV do you obtain? On what KFOLD split do you prefer?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2197018,
      "author_name": "meowmeowmeowmeowmeow",
      "author_url": "",
      "post_date": "03/25/2023 19:56:13",
      "content": "<p>Black images? Do you mean filtering missed hands/faces? <br>\nWould you be kind as to share what CV do you obtain? On what KFOLD split do you prefer?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2196993": "*Loss Function*: Since it is a multi-class classification, Sparse Categorical Loss can help\n*Optimozer*: RMSProp with hyperparameter tuning like learning rate, momentum\n*Model*: Conv2D+ LSTM, Attention Layer, Using Crop Layer for getting rid of redundant black part of the images",
    "2197018": "Black images? Do you mean filtering missed hands/faces? \nWould you be kind as to share what CV do you obtain? On what KFOLD split do you prefer?"
  },
  "source": "meta"
}