{
  "id": 230141,
  "title": "How to speedup tuning hyper-parameters?",
  "url": "/competitions/indoor-location-navigation/discussion/230141",
  "author_name": "SuryaJR_Rafl",
  "post_date": "2021-04-02T07:27:27.820000",
  "votes": 2,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi,</p>\n<p>Based on devin's wonderful notebooks - <a href=\"https://www.kaggle.com/devinanzelmo/wifi-features\" target=\"_blank\">wifi features</a> and  <a href=\"https://www.kaggle.com/devinanzelmo/wifi-features-lightgbm-starter\" target=\"_blank\">lightgbm starter</a>, I tried tuning lightgbm parameters using libraries like hyperopt and flaml in <a href=\"https://www.kaggle.com/suryajrrafl/lightgbm-tuning-hyperopt-flaml\" target=\"_blank\">my notebook</a>. In kaggle kernel, the CPU usage is 400% but the RAM usage is still at 1.4GB. I am pretty new to these memory optimisation techniques, If somebody could point out how to speed up tuning it would be great.</p>\n<p>Thanks in advance.</p>",
  "messages": [
    {
      "id": 1260513,
      "postDate": "2021-04-02T07:27:27.820Z",
      "content": "<p>Hi,</p>\n<p>Based on devin's wonderful notebooks - <a href=\"https://www.kaggle.com/devinanzelmo/wifi-features\" target=\"_blank\">wifi features</a> and  <a href=\"https://www.kaggle.com/devinanzelmo/wifi-features-lightgbm-starter\" target=\"_blank\">lightgbm starter</a>, I tried tuning lightgbm parameters using libraries like hyperopt and flaml in <a href=\"https://www.kaggle.com/suryajrrafl/lightgbm-tuning-hyperopt-flaml\" target=\"_blank\">my notebook</a>. In kaggle kernel, the CPU usage is 400% but the RAM usage is still at 1.4GB. I am pretty new to these memory optimisation techniques, If somebody could point out how to speed up tuning it would be great.</p>\n<p>Thanks in advance.</p>",
      "rawMarkdown": "Hi,\n\nBased on devin's wonderful notebooks - [wifi features](https://www.kaggle.com/devinanzelmo/wifi-features) and  [lightgbm starter](https://www.kaggle.com/devinanzelmo/wifi-features-lightgbm-starter), I tried tuning lightgbm parameters using libraries like hyperopt and flaml in [my notebook](https://www.kaggle.com/suryajrrafl/lightgbm-tuning-hyperopt-flaml). In kaggle kernel, the CPU usage is 400% but the RAM usage is still at 1.4GB. I am pretty new to these memory optimisation techniques, If somebody could point out how to speed up tuning it would be great.\n\nThanks in advance.",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1260513": "Hi,\n\nBased on devin's wonderful notebooks - [wifi features](https://www.kaggle.com/devinanzelmo/wifi-features) and  [lightgbm starter](https://www.kaggle.com/devinanzelmo/wifi-features-lightgbm-starter), I tried tuning lightgbm parameters using libraries like hyperopt and flaml in [my notebook](https://www.kaggle.com/suryajrrafl/lightgbm-tuning-hyperopt-flaml). In kaggle kernel, the CPU usage is 400% but the RAM usage is still at 1.4GB. I am pretty new to these memory optimisation techniques, If somebody could point out how to speed up tuning it would be great.\n\nThanks in advance."
  }
}