{
  "id": 137138,
  "title": "Trouble with training on Colab",
  "url": "/competitions/deepfake-detection-challenge/discussion/137138",
  "author_name": "Atmanand Singh",
  "post_date": "2020-03-19T10:24:09.022000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>I am using fit_generator on Colab to train on a 42gb of data but doing that fills up the disk space for some reason. Why is that happening and what to do?</p>\n\n<p>I have tried setting the queue size as 0, turning multi_processing on/off, experimenting with the number of workers but to no avail.</p>",
  "messages": [
    {
      "id": 779416,
      "postDate": "2020-03-19T10:24:09.023Z",
      "content": "<p>I am using fit_generator on Colab to train on a 42gb of data but doing that fills up the disk space for some reason. Why is that happening and what to do?</p>\n\n<p>I have tried setting the queue size as 0, turning multi_processing on/off, experimenting with the number of workers but to no avail.</p>",
      "rawMarkdown": "I am using fit_generator on Colab to train on a 42gb of data but doing that fills up the disk space for some reason. Why is that happening and what to do?\n\nI have tried setting the queue size as 0, turning multi_processing on/off, experimenting with the number of workers but to no avail."
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "779416": "I am using fit_generator on Colab to train on a 42gb of data but doing that fills up the disk space for some reason. Why is that happening and what to do?\n\nI have tried setting the queue size as 0, turning multi_processing on/off, experimenting with the number of workers but to no avail."
  }
}