{
  "id": 417504,
  "title": "Will submission protein change?",
  "url": "/competitions/cafa-5-protein-function-prediction/discussion/417504",
  "author_name": "",
  "post_date": "2023-06-16T02:12:51.578808400Z",
  "votes": 3,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Will the protein change after the competition deadline?<br>\nin short,<br>\n<a href=\"https://www.kaggle.com/datasets/sergeifironov/t5embeds\" target=\"_blank\">https://www.kaggle.com/datasets/sergeifironov/t5embeds</a><br>\nWill precomputed embeddings such as become obsolete after the deadline?<br>\nAlso, regarding the file name,<br>\nWill \"testsuperset.fasta\" change to \"testset.fasta\" or something like that?<br>\nDo you re-run the notebook with those changes, or just re-evaluate the submitted TSV with different evaluation data?</p>",
  "messages": [
    {
      "id": "2304441",
      "postDate": "06/16/2023 02:12:51",
      "content": "<p>Will the protein change after the competition deadline?<br>\nin short,<br>\n<a href=\"https://www.kaggle.com/datasets/sergeifironov/t5embeds\" target=\"_blank\">https://www.kaggle.com/datasets/sergeifironov/t5embeds</a><br>\nWill precomputed embeddings such as become obsolete after the deadline?<br>\nAlso, regarding the file name,<br>\nWill \"testsuperset.fasta\" change to \"testset.fasta\" or something like that?<br>\nDo you re-run the notebook with those changes, or just re-evaluate the submitted TSV with different evaluation data?</p>",
      "rawMarkdown": "Will the protein change after the competition deadline?\nin short,\nhttps://www.kaggle.com/datasets/sergeifironov/t5embeds\nWill precomputed embeddings such as become obsolete after the deadline?\nAlso, regarding the file name,\nWill \"testsuperset.fasta\" change to \"testset.fasta\" or something like that?\nDo you re-run the notebook with those changes, or just re-evaluate the submitted TSV with different evaluation data?",
      "votes": null
    },
    {
      "id": "2305226",
      "postDate": "06/16/2023 14:17:14",
      "content": "<p>No, the precomputed embeddings won't become obsolete, since there won't be any addition to \"testsuperset.fasta\" and the link you've added has all the embeddings for the proteins in the \"testsuperset.fasta\" in the test_embed files. Only the proteins in the \"testsuperset.fasta\" will be used for evaluation, and I understand that is what your question is. </p>\n<p>Also, as far as my understanding goes, you do not have to rerun with a different file name. You make predictions for \"testsuperset.fasta\" and submit them. That is what I have done so far.</p>",
      "rawMarkdown": "No, the precomputed embeddings won't become obsolete, since there won't be any addition to \"testsuperset.fasta\" and the link you've added has all the embeddings for the proteins in the \"testsuperset.fasta\" in the test_embed files. Only the proteins in the \"testsuperset.fasta\" will be used for evaluation, and I understand that is what your question is. \n\nAlso, as far as my understanding goes, you do not have to rerun with a different file name. You make predictions for \"testsuperset.fasta\" and submit them. That is what I have done so far.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2305226,
      "author_name": "priyanagda",
      "author_url": "",
      "post_date": "06/16/2023 14:17:14",
      "content": "<p>No, the precomputed embeddings won't become obsolete, since there won't be any addition to \"testsuperset.fasta\" and the link you've added has all the embeddings for the proteins in the \"testsuperset.fasta\" in the test_embed files. Only the proteins in the \"testsuperset.fasta\" will be used for evaluation, and I understand that is what your question is. </p>\n<p>Also, as far as my understanding goes, you do not have to rerun with a different file name. You make predictions for \"testsuperset.fasta\" and submit them. That is what I have done so far.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2304441": "Will the protein change after the competition deadline?\nin short,\nhttps://www.kaggle.com/datasets/sergeifironov/t5embeds\nWill precomputed embeddings such as become obsolete after the deadline?\nAlso, regarding the file name,\nWill \"testsuperset.fasta\" change to \"testset.fasta\" or something like that?\nDo you re-run the notebook with those changes, or just re-evaluate the submitted TSV with different evaluation data?",
    "2305226": "No, the precomputed embeddings won't become obsolete, since there won't be any addition to \"testsuperset.fasta\" and the link you've added has all the embeddings for the proteins in the \"testsuperset.fasta\" in the test_embed files. Only the proteins in the \"testsuperset.fasta\" will be used for evaluation, and I understand that is what your question is. \n\nAlso, as far as my understanding goes, you do not have to rerun with a different file name. You make predictions for \"testsuperset.fasta\" and submit them. That is what I have done so far."
  },
  "source": "meta"
}