{
  "id": 364119,
  "title": "Looking for Teammate {LightGBM, XGBoost, Catboost}",
  "url": "/competitions/open-problems-multimodal/discussion/364119",
  "author_name": "Balaji Selvaraj",
  "post_date": "2022-11-04T16:48:35.227000",
  "votes": -1,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi Kagglers,</p>\n<p>We are a team of 4 members currently making use of MLP based pipeline for Multiome &amp; Citeseq.<br>\nSo far we have done the following</p>\n<ol>\n<li>Preprocessing of data </li>\n<li>Dimensionality reduction</li>\n<li>{Torch &amp; Keras} Pipeline for Multiome, CiteSeq </li>\n<li>Most of our models are NN</li>\n</ol>\n<p>As of now, our final ensemble is providing a score of 0.813 {rank ~41 }<br>\nThere are a few more experiments that are in progress. Hoping to touch 0.814 soon.</p>\n<p>We are looking for a teammate with a good score using at least one of the following </p>\n<ol>\n<li>LightGBM</li>\n<li>XGBoost</li>\n<li>CatBoost</li>\n</ol>\n<p>We will be able to share the best strategies that we found during MLP model development, which can improve the performance of the above algorithms.</p>\n<p>All the best to all Kagglers !!!!!!!!</p>",
  "messages": [
    {
      "id": 2017264,
      "postDate": "2022-11-04T16:48:35.227Z",
      "content": "<p>Hi Kagglers,</p>\n<p>We are a team of 4 members currently making use of MLP based pipeline for Multiome &amp; Citeseq.<br>\nSo far we have done the following</p>\n<ol>\n<li>Preprocessing of data </li>\n<li>Dimensionality reduction</li>\n<li>{Torch &amp; Keras} Pipeline for Multiome, CiteSeq </li>\n<li>Most of our models are NN</li>\n</ol>\n<p>As of now, our final ensemble is providing a score of 0.813 {rank ~41 }<br>\nThere are a few more experiments that are in progress. Hoping to touch 0.814 soon.</p>\n<p>We are looking for a teammate with a good score using at least one of the following </p>\n<ol>\n<li>LightGBM</li>\n<li>XGBoost</li>\n<li>CatBoost</li>\n</ol>\n<p>We will be able to share the best strategies that we found during MLP model development, which can improve the performance of the above algorithms.</p>\n<p>All the best to all Kagglers !!!!!!!!</p>",
      "rawMarkdown": "Hi Kagglers,\n\nWe are a team of 4 members currently making use of MLP based pipeline for Multiome & Citeseq.\nSo far we have done the following\n\n1. Preprocessing of data \n2. Dimensionality reduction\n3. {Torch & Keras} Pipeline for Multiome, CiteSeq \n4. Most of our models are NN\n\nAs of now, our final ensemble is providing a score of 0.813 {rank ~41 }\nThere are a few more experiments that are in progress. Hoping to touch 0.814 soon.\n\nWe are looking for a teammate with a good score using at least one of the following \n1. LightGBM\n2. XGBoost\n3. CatBoost\n\nWe will be able to share the best strategies that we found during MLP model development, which can improve the performance of the above algorithms.\n\nAll the best to all Kagglers !!!!!!!!",
      "votes": -1
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2017264": "Hi Kagglers,\n\nWe are a team of 4 members currently making use of MLP based pipeline for Multiome & Citeseq.\nSo far we have done the following\n\n1. Preprocessing of data \n2. Dimensionality reduction\n3. {Torch & Keras} Pipeline for Multiome, CiteSeq \n4. Most of our models are NN\n\nAs of now, our final ensemble is providing a score of 0.813 {rank ~41 }\nThere are a few more experiments that are in progress. Hoping to touch 0.814 soon.\n\nWe are looking for a teammate with a good score using at least one of the following \n1. LightGBM\n2. XGBoost\n3. CatBoost\n\nWe will be able to share the best strategies that we found during MLP model development, which can improve the performance of the above algorithms.\n\nAll the best to all Kagglers !!!!!!!!"
  }
}