{
  "id": 683749,
  "title": "Notebook Threw Exception during submission but runs fine on public test set",
  "url": "/competitions/stanford-rna-3d-folding-2/discussion/683749",
  "author_name": "Jiacheng Ma",
  "post_date": "2026-03-22T00:57:10.168000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>My notebook runs perfectly on the public test_sequences.csv (28 sequences) and completes the \"Commit\" process without any errors. However, every time I submit, I get a \"Notebook Threw Exception\" and no score is calculated.\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F26940609%2F1cb6347661ebf83a0152891c6f8620b1%2F2026-03-22%20085028.png?generation=1774141010788195&amp;alt=media\" alt=\"\">\nI suspect it might be related to the hidden private test set. Has anyone encountered similar issues? I am specifically concerned about:\nOOM on long sequences: Does the private set contain much longer sequences that might crash a Transformer-based model?\nMissing MSA files: Are there targets in the hidden set that lack corresponding MSA files in the input folder?\nID Formatting: I am using the target_id_resid format. Is there any edge case in stoichiometry (multi-chain) that I should be aware of?\nAny tips or common pitfalls for Part 2 would be greatly appreciated! Thanks in advance.</p>",
  "messages": [
    {
      "id": 3426156,
      "postDate": "2026-03-22T00:57:10.170Z",
      "content": "<p>My notebook runs perfectly on the public test_sequences.csv (28 sequences) and completes the \"Commit\" process without any errors. However, every time I submit, I get a \"Notebook Threw Exception\" and no score is calculated.\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F26940609%2F1cb6347661ebf83a0152891c6f8620b1%2F2026-03-22%20085028.png?generation=1774141010788195&amp;alt=media\" alt=\"\">\nI suspect it might be related to the hidden private test set. Has anyone encountered similar issues? I am specifically concerned about:\nOOM on long sequences: Does the private set contain much longer sequences that might crash a Transformer-based model?\nMissing MSA files: Are there targets in the hidden set that lack corresponding MSA files in the input folder?\nID Formatting: I am using the target_id_resid format. Is there any edge case in stoichiometry (multi-chain) that I should be aware of?\nAny tips or common pitfalls for Part 2 would be greatly appreciated! Thanks in advance.</p>",
      "rawMarkdown": "My notebook runs perfectly on the public test_sequences.csv (28 sequences) and completes the \"Commit\" process without any errors. However, every time I submit, I get a \"Notebook Threw Exception\" and no score is calculated.\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F26940609%2F1cb6347661ebf83a0152891c6f8620b1%2F2026-03-22%20085028.png?generation=1774141010788195&alt=media)\nI suspect it might be related to the hidden private test set. Has anyone encountered similar issues? I am specifically concerned about:\nOOM on long sequences: Does the private set contain much longer sequences that might crash a Transformer-based model?\nMissing MSA files: Are there targets in the hidden set that lack corresponding MSA files in the input folder?\nID Formatting: I am using the target_id_resid format. Is there any edge case in stoichiometry (multi-chain) that I should be aware of?\nAny tips or common pitfalls for Part 2 would be greatly appreciated! Thanks in advance.\n"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "3426156": "My notebook runs perfectly on the public test_sequences.csv (28 sequences) and completes the \"Commit\" process without any errors. However, every time I submit, I get a \"Notebook Threw Exception\" and no score is calculated.\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F26940609%2F1cb6347661ebf83a0152891c6f8620b1%2F2026-03-22%20085028.png?generation=1774141010788195&alt=media)\nI suspect it might be related to the hidden private test set. Has anyone encountered similar issues? I am specifically concerned about:\nOOM on long sequences: Does the private set contain much longer sequences that might crash a Transformer-based model?\nMissing MSA files: Are there targets in the hidden set that lack corresponding MSA files in the input folder?\nID Formatting: I am using the target_id_resid format. Is there any edge case in stoichiometry (multi-chain) that I should be aware of?\nAny tips or common pitfalls for Part 2 would be greatly appreciated! Thanks in advance.\n"
  }
}