{
  "id": 298738,
  "title": "How to configure machine setup to handle ML models?",
  "url": "/competitions/tensorflow-great-barrier-reef/discussion/298738",
  "author_name": "Bala Baskar",
  "post_date": "2022-01-04T10:51:19.323000",
  "votes": 3,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi Kagglers,</p>\n<p>I have a fair understanding of how MLalgorithm and deep learning works. But, I have very little idea when it comes to the computational load that takes for a particular dataset or for a particular ML algorithm, or a combination of both (say larger dataset with slower or time taking algorithm) in building any ML model exercise.</p>\n<p>I really want to understand how the machine configuration is decided/computed as adequate, to solve problems for a given dataset (say, tabular or unstructured like image, audio, etc) and for any given algorithm (say, bagging, boosting, deep learning, MLP optimization, etc )</p>\n<p>Knowing this will definitely improve the efficiency of one's approach to arrive at solving a particular complex ML problem or to engage effectively in Kaggle competitions. I really need to understand and learn how this is done. Need some expert advice here and possibly some references.</p>\n<p>And, thanks in advance.</p>",
  "messages": [
    {
      "id": 1637985,
      "postDate": "2022-01-04T10:51:19.323Z",
      "content": "<p>Hi Kagglers,</p>\n<p>I have a fair understanding of how MLalgorithm and deep learning works. But, I have very little idea when it comes to the computational load that takes for a particular dataset or for a particular ML algorithm, or a combination of both (say larger dataset with slower or time taking algorithm) in building any ML model exercise.</p>\n<p>I really want to understand how the machine configuration is decided/computed as adequate, to solve problems for a given dataset (say, tabular or unstructured like image, audio, etc) and for any given algorithm (say, bagging, boosting, deep learning, MLP optimization, etc )</p>\n<p>Knowing this will definitely improve the efficiency of one's approach to arrive at solving a particular complex ML problem or to engage effectively in Kaggle competitions. I really need to understand and learn how this is done. Need some expert advice here and possibly some references.</p>\n<p>And, thanks in advance.</p>",
      "rawMarkdown": "Hi Kagglers,\n\nI have a fair understanding of how MLalgorithm and deep learning works. But, I have very little idea when it comes to the computational load that takes for a particular dataset or for a particular ML algorithm, or a combination of both (say larger dataset with slower or time taking algorithm) in building any ML model exercise.\n\nI really want to understand how the machine configuration is decided/computed as adequate, to solve problems for a given dataset (say, tabular or unstructured like image, audio, etc) and for any given algorithm (say, bagging, boosting, deep learning, MLP optimization, etc )\n\nKnowing this will definitely improve the efficiency of one's approach to arrive at solving a particular complex ML problem or to engage effectively in Kaggle competitions. I really need to understand and learn how this is done. Need some expert advice here and possibly some references.\n\nAnd, thanks in advance.",
      "votes": 3
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1637985": "Hi Kagglers,\n\nI have a fair understanding of how MLalgorithm and deep learning works. But, I have very little idea when it comes to the computational load that takes for a particular dataset or for a particular ML algorithm, or a combination of both (say larger dataset with slower or time taking algorithm) in building any ML model exercise.\n\nI really want to understand how the machine configuration is decided/computed as adequate, to solve problems for a given dataset (say, tabular or unstructured like image, audio, etc) and for any given algorithm (say, bagging, boosting, deep learning, MLP optimization, etc )\n\nKnowing this will definitely improve the efficiency of one's approach to arrive at solving a particular complex ML problem or to engage effectively in Kaggle competitions. I really need to understand and learn how this is done. Need some expert advice here and possibly some references.\n\nAnd, thanks in advance."
  }
}