{ "id": 437640, "title": "Test data is equal to train data ? ", "url": "/competitions/child-mind-institute-detect-sleep-states/discussion/437640", "author_name": "Diganta Kalita", "post_date": "2023-09-07T16:10:30.858000", "votes": 2, "comment_count": 8, "views": 0, "content": "

The test data given in test_series.parquet file looks exactly equal to the training data given in train_series.parquet file. Is this an error on my side downloading the data or this is how it is supposed to be?

\n

Even the timestamps and their values are equal. If this is the case then can't we make predictions using just the labels for the train file?

\n

\"\"

", "messages": [ { "id": 2428070, "postDate": "2023-09-07T16:20:15.423Z", "content": "

The test data in sample_submission.csv/test_series.parquet is simply a place-holder file; the real test data is hidden and is swapped in after submission (during the scoring phase, a process which takes around 40 minutes + notebook runtime).

\n

All the best,
\ncarl

", "rawMarkdown": "The test data in `sample_submission.csv`/`test_series.parquet` is simply a place-holder file; the real test data is hidden and is swapped in after submission (during the scoring phase, a process which takes around 40 minutes + notebook runtime).\n\nAll the best,\ncarl", "votes": 7, "replies": [ { "id": 2428163, "postDate": "2023-09-07T17:10:12.210Z", "rawMarkdown": "", "isDeleted": true }, { "id": 2432946, "postDate": "2023-09-11T08:26:12.290Z", "rawMarkdown": "", "isDeleted": true }, { "id": 2432949, "postDate": "2023-09-11T08:26:39.003Z", "rawMarkdown": "", "isDeleted": true } ] }, { "id": 2432951, "postDate": "2023-09-11T08:26:56.223Z", "content": "

This is a code competition. Thus, you do not have to make predictions in advance but provide a notebook which reads the test_series.parquet file and produces a submission file. The test_series.parquet file you can download is only there to test if your notebook can correctly read an process this file. After submitting your notebook, this file will be swapped with a unseen series file on which the competition is scored.

\n

Let me know if you need further clarification.

", "rawMarkdown": "This is a code competition. Thus, you do not have to make predictions in advance but provide a notebook which reads the test_series.parquet file and produces a submission file. The test_series.parquet file you can download is only there to test if your notebook can correctly read an process this file. After submitting your notebook, this file will be swapped with a unseen series file on which the competition is scored.\n\nLet me know if you need further clarification.", "votes": 4, "replies": [ { "id": 2438284, "postDate": "2023-09-14T07:16:18.057Z", "content": "

I see. Thank you very much,

", "rawMarkdown": "I see. Thank you very much," }, { "id": 2465967, "postDate": "2023-10-03T12:56:02.907Z", "content": "

Thanks. Very helpful.

", "rawMarkdown": "Thanks. Very helpful." } ] }, { "id": 2428055, "postDate": "2023-09-07T16:10:30.860Z", "content": "

The test data given in test_series.parquet file looks exactly equal to the training data given in train_series.parquet file. Is this an error on my side downloading the data or this is how it is supposed to be?

\n

Even the timestamps and their values are equal. If this is the case then can't we make predictions using just the labels for the train file?

\n

\"\"

", "rawMarkdown": "The test data given in test_series.parquet file looks exactly equal to the training data given in train_series.parquet file. Is this an error on my side downloading the data or this is how it is supposed to be? \n\nEven the timestamps and their values are equal. If this is the case then can't we make predictions using just the labels for the train file?\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4432605%2Ff13a234a39fa49d78189149459022d85%2FScreenshot%202023-09-07%20at%209.39.30%20PM.png?generation=1694103025036273&alt=media)", "votes": 2 }, { "id": 2428065, "postDate": "2023-09-07T16:16:38.297Z", "content": "

This test data is not from my splitting. Its from the test file provided with the dataset.

", "rawMarkdown": "This test data is not from my splitting. Its from the test file provided with the dataset." } ], "comments": [ { "id": 2428070, "author_name": "Carl McBride Ellis", "author_url": "", "post_date": "2023-09-07T16:20:15.423000", "content": "

The test data in sample_submission.csv/test_series.parquet is simply a place-holder file; the real test data is hidden and is swapped in after submission (during the scoring phase, a process which takes around 40 minutes + notebook runtime).

\n

All the best,
\ncarl

", "votes": 7, "replies": [ { "id": 2428163, "author_name": "", "author_url": "", "post_date": "2023-09-07T17:10:12.210000", "content": "", "votes": 0, "replies": [] }, { "id": 2432946, "author_name": "", "author_url": "", "post_date": "2023-09-11T08:26:12.290000", "content": "", "votes": 0, "replies": [] }, { "id": 2432949, "author_name": "", "author_url": "", "post_date": "2023-09-11T08:26:39.003000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 2432951, "author_name": "PatrickAIForFun", "author_url": "", "post_date": "2023-09-11T08:26:56.223000", "content": "

This is a code competition. Thus, you do not have to make predictions in advance but provide a notebook which reads the test_series.parquet file and produces a submission file. The test_series.parquet file you can download is only there to test if your notebook can correctly read an process this file. After submitting your notebook, this file will be swapped with a unseen series file on which the competition is scored.

\n

Let me know if you need further clarification.

", "votes": 4, "replies": [ { "id": 2438284, "author_name": "Diganta Kalita", "author_url": "", "post_date": "2023-09-14T07:16:18.057000", "content": "

I see. Thank you very much,

", "votes": 0, "replies": [] }, { "id": 2465967, "author_name": "", "author_url": "", "post_date": "2023-10-03T12:56:02.907000", "content": "

Thanks. Very helpful.

", "votes": 0, "replies": [] } ] }, { "id": 2428065, "author_name": "Diganta Kalita", "author_url": "", "post_date": "2023-09-07T16:16:38.297000", "content": "

This test data is not from my splitting. Its from the test file provided with the dataset.

", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "2428070": "The test data in `sample_submission.csv`/`test_series.parquet` is simply a place-holder file; the real test data is hidden and is swapped in after submission (during the scoring phase, a process which takes around 40 minutes + notebook runtime).\n\nAll the best,\ncarl", "2432951": "This is a code competition. Thus, you do not have to make predictions in advance but provide a notebook which reads the test_series.parquet file and produces a submission file. The test_series.parquet file you can download is only there to test if your notebook can correctly read an process this file. After submitting your notebook, this file will be swapped with a unseen series file on which the competition is scored.\n\nLet me know if you need further clarification.", "2428055": "The test data given in test_series.parquet file looks exactly equal to the training data given in train_series.parquet file. Is this an error on my side downloading the data or this is how it is supposed to be? \n\nEven the timestamps and their values are equal. If this is the case then can't we make predictions using just the labels for the train file?\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4432605%2Ff13a234a39fa49d78189149459022d85%2FScreenshot%202023-09-07%20at%209.39.30%20PM.png?generation=1694103025036273&alt=media)", "2428065": "This test data is not from my splitting. Its from the test file provided with the dataset." } }