{
  "id": 483982,
  "title": "Summary of the models and maximum number of features used in the competition",
  "url": "/competitions/home-credit-credit-risk-model-stability/discussion/483982",
  "author_name": "yunsuxiaozi",
  "post_date": "2024-03-15T01:37:29.242000",
  "votes": 5,
  "comment_count": 0,
  "views": 0,
  "content": "<p>In this competition, due to the large amount of hidden test data, the number of features that the competition model can use is limited, and accidentally \"out of memory\" may appear. Here, I hope to gather everyone's strength and summarize the experimental results we have done so far. This can not only help others reduce errors, but also reduce the number of errors we submitted.</p>\n<p>Participants who are willing to provide experimental data need to provide information on what model was used (e.g. lgb), how many folds of cross validation were used (e.g. 10), how much data was used (e.g. 800000), how many features were used (e.g. 200), whether training inference separation was used (e.g. Yes),whether the submission was successful (e.g. Yes), and any other details you would like to mention (e.g. None)</p>\n<table>\n<thead>\n<tr>\n<th>model</th>\n<th>num_folds</th>\n<th>num_datas</th>\n<th>num_features</th>\n<th>train_inference_separation</th>\n<th>submission successful</th>\n<th>other mention</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>LinearRegression</td>\n<td>20</td>\n<td>1526659(full)</td>\n<td>254</td>\n<td>No</td>\n<td>Yes</td>\n<td>reduce_memory</td>\n</tr>\n<tr>\n<td>lgb</td>\n<td>10</td>\n<td>1526659(full)</td>\n<td>220</td>\n<td>Yes</td>\n<td>Yes</td>\n<td>reduce_memory</td>\n</tr>\n</tbody>\n</table>",
  "messages": [
    {
      "id": 2697571,
      "postDate": "2024-03-15T01:37:29.243Z",
      "content": "<p>In this competition, due to the large amount of hidden test data, the number of features that the competition model can use is limited, and accidentally \"out of memory\" may appear. Here, I hope to gather everyone's strength and summarize the experimental results we have done so far. This can not only help others reduce errors, but also reduce the number of errors we submitted.</p>\n<p>Participants who are willing to provide experimental data need to provide information on what model was used (e.g. lgb), how many folds of cross validation were used (e.g. 10), how much data was used (e.g. 800000), how many features were used (e.g. 200), whether training inference separation was used (e.g. Yes),whether the submission was successful (e.g. Yes), and any other details you would like to mention (e.g. None)</p>\n<table>\n<thead>\n<tr>\n<th>model</th>\n<th>num_folds</th>\n<th>num_datas</th>\n<th>num_features</th>\n<th>train_inference_separation</th>\n<th>submission successful</th>\n<th>other mention</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>LinearRegression</td>\n<td>20</td>\n<td>1526659(full)</td>\n<td>254</td>\n<td>No</td>\n<td>Yes</td>\n<td>reduce_memory</td>\n</tr>\n<tr>\n<td>lgb</td>\n<td>10</td>\n<td>1526659(full)</td>\n<td>220</td>\n<td>Yes</td>\n<td>Yes</td>\n<td>reduce_memory</td>\n</tr>\n</tbody>\n</table>",
      "rawMarkdown": "In this competition, due to the large amount of hidden test data, the number of features that the competition model can use is limited, and accidentally \"out of memory\" may appear. Here, I hope to gather everyone's strength and summarize the experimental results we have done so far. This can not only help others reduce errors, but also reduce the number of errors we submitted.\n\nParticipants who are willing to provide experimental data need to provide information on what model was used (e.g. lgb), how many folds of cross validation were used (e.g. 10), how much data was used (e.g. 800000), how many features were used (e.g. 200), whether training inference separation was used (e.g. Yes),whether the submission was successful (e.g. Yes), and any other details you would like to mention (e.g. None)\n\n| model | num_folds | num_datas|num_features|train_inference_separation|submission successful |other mention|\n| --- | --- | --- | --- | --- | --- | --- |\n| LinearRegression | 20 | 1526659(full)|254|No|Yes|reduce_memory|\n| lgb | 10 | 1526659(full)|220|Yes|Yes|reduce_memory|\n",
      "votes": 5
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2697571": "In this competition, due to the large amount of hidden test data, the number of features that the competition model can use is limited, and accidentally \"out of memory\" may appear. Here, I hope to gather everyone's strength and summarize the experimental results we have done so far. This can not only help others reduce errors, but also reduce the number of errors we submitted.\n\nParticipants who are willing to provide experimental data need to provide information on what model was used (e.g. lgb), how many folds of cross validation were used (e.g. 10), how much data was used (e.g. 800000), how many features were used (e.g. 200), whether training inference separation was used (e.g. Yes),whether the submission was successful (e.g. Yes), and any other details you would like to mention (e.g. None)\n\n| model | num_folds | num_datas|num_features|train_inference_separation|submission successful |other mention|\n| --- | --- | --- | --- | --- | --- | --- |\n| LinearRegression | 20 | 1526659(full)|254|No|Yes|reduce_memory|\n| lgb | 10 | 1526659(full)|220|Yes|Yes|reduce_memory|\n"
  }
}