{
  "id": 674468,
  "title": "line_start and line_end in test.csv",
  "url": "/competitions/deep-past-initiative-machine-translation/discussion/674468",
  "author_name": "",
  "post_date": "2026-02-20T16:08:00.825074100Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Probably a newbie question: What are the columns \"line_start\" and \"line_end\" in the file \"test.csv\"?</p>",
  "messages": [
    {
      "id": "3408436",
      "postDate": "02/20/2026 16:08:00",
      "content": "<p>Probably a newbie question: What are the columns \"line_start\" and \"line_end\" in the file \"test.csv\"?</p>",
      "rawMarkdown": "Probably a newbie question: What are the columns \"line_start\" and \"line_end\" in the file \"test.csv\"?",
      "votes": null
    },
    {
      "id": "3408447",
      "postDate": "02/20/2026 16:31:15",
      "content": "<p>They delineate the sentence boundaries. Scholars usually determine a sentence by marking the number of lines that compose a sentence, e.g. 1-3) … 4-7) … 8-10) … . Those can be useful for determining in the transliteration how to go from lines of the text to sentences.</p>",
      "rawMarkdown": "They delineate the sentence boundaries. Scholars usually determine a sentence by marking the number of lines that compose a sentence, e.g. 1-3) ... 4-7) ... 8-10) ... . Those can be useful for determining in the transliteration how to go from lines of the text to sentences.",
      "votes": null
    },
    {
      "id": "3408795",
      "postDate": "02/21/2026 10:22:15",
      "content": "<p>Just to clarify, here is a screenshot of \"test.csv\":</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F470785%2Fc1d794312c2301557164833a13da73c2%2Fquestion_about_test_data.png?generation=1771668049357344&amp;alt=media\" alt=\"\"></p>\n<p>For example: For id=0, we have line_start=1 and line_end=7. How does that relate to the \"transliteration\" column?</p>\n<p>I would think it refers to the original cuneiform, but am not sure.</p>\n<p>Also confusing is that I don't find \"line_start\" and \"line_end\" in the training data or in \"Sentences_Oare_FirstWord_LinNum.csv\".</p>",
      "rawMarkdown": "Just to clarify, here is a screenshot of \"test.csv\":\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F470785%2Fc1d794312c2301557164833a13da73c2%2Fquestion_about_test_data.png?generation=1771668049357344&alt=media)\n\nFor example: For id=0, we have line_start=1 and line_end=7. How does that relate to the \"transliteration\" column?\n\nI would think it refers to the original cuneiform, but am not sure.\n\nAlso confusing is that I don't find \"line_start\" and \"line_end\" in the training data or in \"Sentences_Oare_FirstWord_LinNum.csv\".",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3408447,
      "author_name": "deeppast",
      "author_url": "",
      "post_date": "02/20/2026 16:31:15",
      "content": "<p>They delineate the sentence boundaries. Scholars usually determine a sentence by marking the number of lines that compose a sentence, e.g. 1-3) … 4-7) … 8-10) … . Those can be useful for determining in the transliteration how to go from lines of the text to sentences.</p>",
      "votes": null,
      "replies": [
        {
          "id": 3408795,
          "author_name": "garbo999",
          "author_url": "",
          "post_date": "02/21/2026 10:22:15",
          "content": "<p>Just to clarify, here is a screenshot of \"test.csv\":</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F470785%2Fc1d794312c2301557164833a13da73c2%2Fquestion_about_test_data.png?generation=1771668049357344&amp;alt=media\" alt=\"\"></p>\n<p>For example: For id=0, we have line_start=1 and line_end=7. How does that relate to the \"transliteration\" column?</p>\n<p>I would think it refers to the original cuneiform, but am not sure.</p>\n<p>Also confusing is that I don't find \"line_start\" and \"line_end\" in the training data or in \"Sentences_Oare_FirstWord_LinNum.csv\".</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3408436": "Probably a newbie question: What are the columns \"line_start\" and \"line_end\" in the file \"test.csv\"?",
    "3408447": "They delineate the sentence boundaries. Scholars usually determine a sentence by marking the number of lines that compose a sentence, e.g. 1-3) ... 4-7) ... 8-10) ... . Those can be useful for determining in the transliteration how to go from lines of the text to sentences.",
    "3408795": "Just to clarify, here is a screenshot of \"test.csv\":\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F470785%2Fc1d794312c2301557164833a13da73c2%2Fquestion_about_test_data.png?generation=1771668049357344&alt=media)\n\nFor example: For id=0, we have line_start=1 and line_end=7. How does that relate to the \"transliteration\" column?\n\nI would think it refers to the original cuneiform, but am not sure.\n\nAlso confusing is that I don't find \"line_start\" and \"line_end\" in the training data or in \"Sentences_Oare_FirstWord_LinNum.csv\"."
  },
  "source": "meta"
}