{
  "id": 437313,
  "title": "how to deal with  layout_nlp_default dataset",
  "url": "/competitions/predict-ai-model-runtime/discussion/437313",
  "author_name": "MiHu",
  "post_date": "2023-09-06T09:46:21.300000",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Reading the layout_nlp_default dataset seems to consume a lot of memory. Do you have any good solutions?</p>",
  "messages": [
    {
      "id": 2425942,
      "postDate": "2023-09-06T09:46:21.300Z",
      "content": "<p>Reading the layout_nlp_default dataset seems to consume a lot of memory. Do you have any good solutions?</p>",
      "rawMarkdown": "Reading the layout_nlp_default dataset seems to consume a lot of memory. Do you have any good solutions?\n\n\n",
      "votes": 1
    },
    {
      "id": 2426072,
      "postDate": "2023-09-06T11:40:12.243Z",
      "content": "<p>I ran into the same problem when I first tried running layout_train.py. I managed to fix it by tweaking the _MAX_CONFIGS value in the train_args.py file located in the baselines/layout folder. However, now each epoch takes a very long time to complete. I haven't tried running it with the maximum number of configurations all at once, so I'm not sure if it would still be slow in that case. also instead of changing it in from the file you can use  layout_train.py .  .  .    --max_configs 100</p>",
      "rawMarkdown": "I ran into the same problem when I first tried running layout_train.py. I managed to fix it by tweaking the _MAX_CONFIGS value in the train_args.py file located in the baselines/layout folder. However, now each epoch takes a very long time to complete. I haven't tried running it with the maximum number of configurations all at once, so I'm not sure if it would still be slow in that case. also instead of changing it in from the file you can use  layout_train.py .  .  .    --max_configs 100"
    },
    {
      "id": 2427074,
      "postDate": "2023-09-07T04:38:01.420Z",
      "rawMarkdown": "",
      "votes": -1,
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 2426072,
      "author_name": "amirreza01110",
      "author_url": "",
      "post_date": "2023-09-06T11:40:12.243000",
      "content": "<p>I ran into the same problem when I first tried running layout_train.py. I managed to fix it by tweaking the _MAX_CONFIGS value in the train_args.py file located in the baselines/layout folder. However, now each epoch takes a very long time to complete. I haven't tried running it with the maximum number of configurations all at once, so I'm not sure if it would still be slow in that case. also instead of changing it in from the file you can use  layout_train.py .  .  .    --max_configs 100</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2427074,
      "author_name": "",
      "author_url": "",
      "post_date": "2023-09-07T04:38:01.420000",
      "content": "",
      "votes": -1,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2425942": "Reading the layout_nlp_default dataset seems to consume a lot of memory. Do you have any good solutions?\n\n\n",
    "2426072": "I ran into the same problem when I first tried running layout_train.py. I managed to fix it by tweaking the _MAX_CONFIGS value in the train_args.py file located in the baselines/layout folder. However, now each epoch takes a very long time to complete. I haven't tried running it with the maximum number of configurations all at once, so I'm not sure if it would still be slow in that case. also instead of changing it in from the file you can use  layout_train.py .  .  .    --max_configs 100",
    "2427074": ""
  }
}