{
  "id": 183769,
  "title": "Validation scheme",
  "url": "/competitions/rsna-str-pulmonary-embolism-detection/discussion/183769",
  "author_name": "Konrad Banachewicz",
  "post_date": "2020-09-18T00:24:17.569000",
  "votes": 5,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Quick examination shows that there is no overlap between train and test on either series or study variables, so I think it's useful to keep that in mind with the validation scheme we use - for your convenience, see the kernel below:</p>\n<p><a href=\"https://www.kaggle.com/konradb/validation-scheme\" target=\"_blank\">https://www.kaggle.com/konradb/validation-scheme</a></p>\n<p>Based on your heart's desire, you can split either on series or study.</p>",
  "messages": [
    {
      "id": 1015111,
      "postDate": "2020-09-18T00:24:17.570Z",
      "content": "<p>Quick examination shows that there is no overlap between train and test on either series or study variables, so I think it's useful to keep that in mind with the validation scheme we use - for your convenience, see the kernel below:</p>\n<p><a href=\"https://www.kaggle.com/konradb/validation-scheme\" target=\"_blank\">https://www.kaggle.com/konradb/validation-scheme</a></p>\n<p>Based on your heart's desire, you can split either on series or study.</p>",
      "rawMarkdown": "Quick examination shows that there is no overlap between train and test on either series or study variables, so I think it's useful to keep that in mind with the validation scheme we use - for your convenience, see the kernel below:\n\nhttps://www.kaggle.com/konradb/validation-scheme\n\nBased on your heart's desire, you can split either on series or study.",
      "votes": 5
    },
    {
      "id": 1016574,
      "postDate": "2020-09-19T05:11:17.760Z",
      "content": "<p>Thank you for sharing!</p>\n<p>Is your intention to perform cross-validation so that the same patient data is not mixed in each fold? (Like Group Kflod)</p>\n<p>xfolds ['StudyInstanceUID']. Nunique () and xfolds ['SeriesInstanceUID'] .nunique () are 7279 respectively, which should be a one-to-one correspondence, but why you created xfolds ['fold_st'] and xfolds ['fold_sr'] individually？</p>\n<p>I would appreciate it if you could reply me.</p>",
      "rawMarkdown": "Thank you for sharing!\n\nIs your intention to perform cross-validation so that the same patient data is not mixed in each fold? (Like Group Kflod)\n\nxfolds ['StudyInstanceUID']. Nunique () and xfolds ['SeriesInstanceUID'] .nunique () are 7279 respectively, which should be a one-to-one correspondence, but why you created xfolds ['fold_st'] and xfolds ['fold_sr'] individually？\n\n\nI would appreciate it if you could reply me."
    }
  ],
  "comments": [
    {
      "id": 1016574,
      "author_name": "patriot",
      "author_url": "",
      "post_date": "2020-09-19T05:11:17.760000",
      "content": "<p>Thank you for sharing!</p>\n<p>Is your intention to perform cross-validation so that the same patient data is not mixed in each fold? (Like Group Kflod)</p>\n<p>xfolds ['StudyInstanceUID']. Nunique () and xfolds ['SeriesInstanceUID'] .nunique () are 7279 respectively, which should be a one-to-one correspondence, but why you created xfolds ['fold_st'] and xfolds ['fold_sr'] individually？</p>\n<p>I would appreciate it if you could reply me.</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1015111": "Quick examination shows that there is no overlap between train and test on either series or study variables, so I think it's useful to keep that in mind with the validation scheme we use - for your convenience, see the kernel below:\n\nhttps://www.kaggle.com/konradb/validation-scheme\n\nBased on your heart's desire, you can split either on series or study.",
    "1016574": "Thank you for sharing!\n\nIs your intention to perform cross-validation so that the same patient data is not mixed in each fold? (Like Group Kflod)\n\nxfolds ['StudyInstanceUID']. Nunique () and xfolds ['SeriesInstanceUID'] .nunique () are 7279 respectively, which should be a one-to-one correspondence, but why you created xfolds ['fold_st'] and xfolds ['fold_sr'] individually？\n\n\nI would appreciate it if you could reply me."
  }
}