{ "id": 355355, "title": "Submission scoring error", "url": "/competitions/rsna-2022-cervical-spine-fracture-detection/discussion/355355", "author_name": "Collide_Conquer_19", "post_date": "2022-09-26T12:08:02.385000", "votes": 1, "comment_count": 3, "views": 0, "content": "

Hello @solverworld and all kagglers @vslaykovsky have completed the infernece phase of vit3d and also generated the submission file as required and when I have submitted the submission file it's throwing an error of submissions score error.

\n

Please have a look at my code and let me know the solution or problems which are attached to it.

\n

Thanking you in advance.
\nConquer_19.

", "messages": [ { "id": 1956357, "postDate": "2022-09-26T12:08:02.387Z", "content": "

Hello @solverworld and all kagglers @vslaykovsky have completed the infernece phase of vit3d and also generated the submission file as required and when I have submitted the submission file it's throwing an error of submissions score error.

\n

Please have a look at my code and let me know the solution or problems which are attached to it.

\n

Thanking you in advance.
\nConquer_19.

", "rawMarkdown": "Hello @solverworld and all kagglers @vslaykovsky have completed the infernece phase of vit3d and also generated the submission file as required and when I have submitted the submission file it's throwing an error of submissions score error.\n\nPlease have a look at my [code](https://www.kaggle.com/code/robber19/vit3d-infer/notebook) and let me know the solution or problems which are attached to it.\n\nThanking you in advance.\nConquer_19.", "votes": 1 }, { "id": 1957378, "postDate": "2022-09-26T23:33:54.160Z", "content": "

This isn't the source of an Exception, but one mistake I see is that you sort the slice files by their name. This will be an alphabetic sort and not a numerical sort. It will put 21.dcm before 3.dcm, for example. There are public notebooks that have example of the correct sort, something like:

\n
        slice_files = glob.glob('path_for_train_files/*.dcm')\n        slice_files.sort(key=lambda x: int(os.path.splitext(os.path.basename(x))[0]))\n
", "rawMarkdown": "This isn't the source of an Exception, but one mistake I see is that you sort the slice files by their name. This will be an alphabetic sort and not a numerical sort. It will put 21.dcm before 3.dcm, for example. There are public notebooks that have example of the correct sort, something like:\n```\n slice_files = glob.glob('path_for_train_files/*.dcm')\n slice_files.sort(key=lambda x: int(os.path.splitext(os.path.basename(x))[0]))\n```" }, { "id": 1956482, "postDate": "2022-09-26T13:27:47.637Z", "content": "

Hi,
\nAs far as I understand, this is the notebook you are directly submitting, right? If so, there is a problem with this line:

\n
df['row_id'] = df['row_id'].replace(['1.2.826.0.1.3680043.22327_C8','1.2.826.0.1.3680043.25399_C8','1.2.826.0.1.3680043.5876_C8'],['1.2.826.0.1.3680043.22327_patient_overall','1.2.826.0.1.3680043.25399_patient_overall','1.2.826.0.1.3680043.5876_patient_overall'])\n
\n

As this line only works for the three examples in the public /test folder because unseen samples have different SOPInstanceUIDs. I would suggest using something like this:

\n
df['row_id'].replace(to_replace=r'_C8', value='_patient_overall', regex=True, inplace=True)\n
\n

Hope this solves the issue.

", "rawMarkdown": "Hi,\nAs far as I understand, this is the notebook you are directly submitting, right? If so, there is a problem with this line:\n\n\n```\ndf['row_id'] = df['row_id'].replace(['1.2.826.0.1.3680043.22327_C8','1.2.826.0.1.3680043.25399_C8','1.2.826.0.1.3680043.5876_C8'],['1.2.826.0.1.3680043.22327_patient_overall','1.2.826.0.1.3680043.25399_patient_overall','1.2.826.0.1.3680043.5876_patient_overall'])\n```\n\nAs this line only works for the three examples in the public `/test` folder because unseen samples have different *SOPInstanceUIDs*. I would suggest using something like this:\n\n```\ndf['row_id'].replace(to_replace=r'_C8', value='_patient_overall', regex=True, inplace=True)\n```\n\nHope this solves the issue.", "replies": [ { "id": 1956529, "postDate": "2022-09-26T13:49:40.490Z", "content": "

Hope this helps and I'll try your input and report the same.

\n

Thanks,
\nConquer _19.

", "rawMarkdown": "Hope this helps and I'll try your input and report the same.\n\nThanks, \nConquer _19." } ] } ], "comments": [ { "id": 1957378, "author_name": "SolverWorld", "author_url": "", "post_date": "2022-09-26T23:33:54.160000", "content": "

This isn't the source of an Exception, but one mistake I see is that you sort the slice files by their name. This will be an alphabetic sort and not a numerical sort. It will put 21.dcm before 3.dcm, for example. There are public notebooks that have example of the correct sort, something like:

\n
        slice_files = glob.glob('path_for_train_files/*.dcm')\n        slice_files.sort(key=lambda x: int(os.path.splitext(os.path.basename(x))[0]))\n
", "votes": 0, "replies": [] }, { "id": 1956482, "author_name": "Bardia Khosravi", "author_url": "", "post_date": "2022-09-26T13:27:47.637000", "content": "

Hi,
\nAs far as I understand, this is the notebook you are directly submitting, right? If so, there is a problem with this line:

\n
df['row_id'] = df['row_id'].replace(['1.2.826.0.1.3680043.22327_C8','1.2.826.0.1.3680043.25399_C8','1.2.826.0.1.3680043.5876_C8'],['1.2.826.0.1.3680043.22327_patient_overall','1.2.826.0.1.3680043.25399_patient_overall','1.2.826.0.1.3680043.5876_patient_overall'])\n
\n

As this line only works for the three examples in the public /test folder because unseen samples have different SOPInstanceUIDs. I would suggest using something like this:

\n
df['row_id'].replace(to_replace=r'_C8', value='_patient_overall', regex=True, inplace=True)\n
\n

Hope this solves the issue.

", "votes": 0, "replies": [ { "id": 1956529, "author_name": "Collide_Conquer_19", "author_url": "", "post_date": "2022-09-26T13:49:40.490000", "content": "

Hope this helps and I'll try your input and report the same.

\n

Thanks,
\nConquer _19.

", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "1956357": "Hello @solverworld and all kagglers @vslaykovsky have completed the infernece phase of vit3d and also generated the submission file as required and when I have submitted the submission file it's throwing an error of submissions score error.\n\nPlease have a look at my [code](https://www.kaggle.com/code/robber19/vit3d-infer/notebook) and let me know the solution or problems which are attached to it.\n\nThanking you in advance.\nConquer_19.", "1957378": "This isn't the source of an Exception, but one mistake I see is that you sort the slice files by their name. This will be an alphabetic sort and not a numerical sort. It will put 21.dcm before 3.dcm, for example. There are public notebooks that have example of the correct sort, something like:\n```\n slice_files = glob.glob('path_for_train_files/*.dcm')\n slice_files.sort(key=lambda x: int(os.path.splitext(os.path.basename(x))[0]))\n```", "1956482": "Hi,\nAs far as I understand, this is the notebook you are directly submitting, right? If so, there is a problem with this line:\n\n\n```\ndf['row_id'] = df['row_id'].replace(['1.2.826.0.1.3680043.22327_C8','1.2.826.0.1.3680043.25399_C8','1.2.826.0.1.3680043.5876_C8'],['1.2.826.0.1.3680043.22327_patient_overall','1.2.826.0.1.3680043.25399_patient_overall','1.2.826.0.1.3680043.5876_patient_overall'])\n```\n\nAs this line only works for the three examples in the public `/test` folder because unseen samples have different *SOPInstanceUIDs*. I would suggest using something like this:\n\n```\ndf['row_id'].replace(to_replace=r'_C8', value='_patient_overall', regex=True, inplace=True)\n```\n\nHope this solves the issue." } }