{ "id": 510982, "title": "Submition is wrong, test dataset undetected (?)", "url": "/competitions/birdclef-2024/discussion/510982", "author_name": "", "post_date": "2024-06-08T15:36:06.445380300Z", "votes": 2, "comment_count": 9, "views": 0, "content": "

Hi everyone,
\nI'm stuck, I'm trying to submit my notebook which works perfectly using the unlabeled soundscapes. However, when I submit, I expect the path '/kaggle/input/birdclef-2024/test_soundscapes/' to be filled with audio. Here is my code:

\n
test_audio_dir = \n\nogg_file_paths = [test_audio_dir+f  f  (os.listdir(test_audio_dir))]\n\n (ogg_file_paths)==:\n    ogg_file_paths = [+f  f  (os.listdir())[:]]\n
\n

When I submit using this, the file detected is still the readme. It is also the 1st competition I'm trying to submit to. Does anyone know how to unblock me? I will be extremely grateful.

\n

Best regards,

\n

Arnaud

", "messages": [ { "id": "2862142", "postDate": "06/08/2024 15:36:06", "content": "

Hi everyone,
\nI'm stuck, I'm trying to submit my notebook which works perfectly using the unlabeled soundscapes. However, when I submit, I expect the path '/kaggle/input/birdclef-2024/test_soundscapes/' to be filled with audio. Here is my code:

\n
test_audio_dir = \n\nogg_file_paths = [test_audio_dir+f  f  (os.listdir(test_audio_dir))]\n\n (ogg_file_paths)==:\n    ogg_file_paths = [+f  f  (os.listdir())[:]]\n
\n

When I submit using this, the file detected is still the readme. It is also the 1st competition I'm trying to submit to. Does anyone know how to unblock me? I will be extremely grateful.

\n

Best regards,

\n

Arnaud

", "rawMarkdown": "Hi everyone, \nI'm stuck, I'm trying to submit my notebook which works perfectly using the unlabeled soundscapes. However, when I submit, I expect the path '/kaggle/input/birdclef-2024/test_soundscapes/' to be filled with audio. Here is my code: \n\n```python\ntest_audio_dir = '/kaggle/input/birdclef-2024/test_soundscapes/'\n\nogg_file_paths = [test_audio_dir+f for f in sorted(os.listdir(test_audio_dir))]\n\nif len(ogg_file_paths)==1:\n ogg_file_paths = [f'{config.DATA_ROOT}/unlabeled_soundscapes/'+f for f in sorted(os.listdir(f'{config.DATA_ROOT}/unlabeled_soundscapes/'))[:10]]\n```\n\nWhen I submit using this, the file detected is still the readme. It is also the 1st competition I'm trying to submit to. Does anyone know how to unblock me? I will be extremely grateful. \n\nBest regards, \n\nArnaud", "votes": null }, { "id": "2862318", "postDate": "06/08/2024 17:29:27", "content": "

I'm using the following code with glob library, which I've initially taken from the public notebook.
\nYour code looks ok, but you have to filter out everything except oog files. For example, if readme file exists in the directory, and you try to process it like ogg file, you get the error.

\n
import \n\nROOT_FOLDER = \n\n (.()) > :\n    ogg_file_paths = .()\n:\n    ogg_file_paths = sorted(.())[:]\n
", "rawMarkdown": "I'm using the following code with glob library, which I've initially taken from [the public notebook](https://www.kaggle.com/code/markwijkhuizen/birdclef-2024-efficientvit-inference).\nYour code looks ok, but you have to filter out everything except oog files. For example, if readme file exists in the directory, and you try to process it like ogg file, you get the error.\n\n```\nimport glob\n\nROOT_FOLDER = '/kaggle/input/birdclef-2024'\n\nif len(glob.glob(f'{ROOT_FOLDER}/test_soundscapes/*.ogg')) > 0:\n ogg_file_paths = glob.glob(f'{ROOT_FOLDER}/test_soundscapes/*.ogg')\nelse:\n ogg_file_paths = sorted(glob.glob(f'{ROOT_FOLDER}/unlabeled_soundscapes/*.ogg'))[:10]\n```", "votes": null }, { "id": "2862638", "postDate": "06/08/2024 20:37:18", "content": "

Hey Konstantin,
\nThanks for your reply. I tried this code before and now again. However, the code still produces the outcome with the unlabeled sounds and not with the test ones (csv of 10x182). I think it is the way I submit which might be wrong…
\nDo you have any further idea ?
\nBest,
\nArnaud

", "rawMarkdown": "Hey Konstantin, \nThanks for your reply. I tried this code before and now again. However, the code still produces the outcome with the unlabeled sounds and not with the test ones (csv of 10x182). I think it is the way I submit which might be wrong... \nDo you have any further idea ? \nBest, \nArnaud", "votes": null }, { "id": "2862985", "postDate": "06/09/2024 06:55:32", "content": "

You may be confusing saving and submitting.

\n

You cannot see what happens when you submit.

\n

You can see what happens when you save your notebook, before submititng it.

\n

Have you submitted at all?

", "rawMarkdown": "You may be confusing saving and submitting.\n\nYou cannot see what happens when you submit.\n\nYou can see what happens when you save your notebook, before submititng it.\n\nHave you submitted at all?", "votes": null }, { "id": "2863071", "postDate": "06/09/2024 08:04:03", "content": "

Hey CPMP,

\n

Thanks for the answer, I really appreciate it.

\n

From my understanding, I submitted a notebook now 14 times.
\nI always get this message:
\n\"Submission Scoring Error
\nYour notebook generated a submission file with incorrect format. Some examples causing this are: wrong number of rows or columns, empty values, an incorrect data type for a value, or invalid submission values from what is expected. See more debugging tips\"
\nI will attach my input (ckpt and whl for einops), as well as the logs of my submissions.

\n

Any further help would be much appreciated.

\n

Thanks a lot for your time,

\n

Best,
\nArnaud

", "rawMarkdown": "Hey CPMP, \n\nThanks for the answer, I really appreciate it. \n\nFrom my understanding, I submitted a notebook now 14 times. \nI always get this message: \n\"Submission Scoring Error\nYour notebook generated a submission file with incorrect format. Some examples causing this are: wrong number of rows or columns, empty values, an incorrect data type for a value, or invalid submission values from what is expected. See more debugging tips\"\nI will attach my input (ckpt and whl for einops), as well as the logs of my submissions. \n\nAny further help would be much appreciated. \n\nThanks a lot for your time, \n\nBest, \nArnaud", "votes": null }, { "id": "2863140", "postDate": "06/09/2024 08:40:57", "content": "

I am also unable to submit to competition.it saying your notebook is exceeding allowed runtime.how can I speed up my processing to avoid this and successfully submit my notebook

", "rawMarkdown": "I am also unable to submit to competition.it saying your notebook is exceeding allowed runtime.how can I speed up my processing to avoid this and successfully submit my notebook", "votes": null }, { "id": "2863340", "postDate": "06/09/2024 10:34:22", "content": "

OK, you have errors in submitting,, but that's not what you posted about. You have a topic with title: \"Can't submit, test dataset undetected (?)\"

\n

II replied to the issue you describe in your original post.

", "rawMarkdown": "OK, you have errors in submitting,, but that's not what you posted about. You have a topic with title: \"Can't submit, test dataset undetected (?)\"\n\nII replied to the issue you describe in your original post.", "votes": null }, { "id": "2863346", "postDate": "06/09/2024 10:42:01", "content": "

Thank you please I am waiting for you guidance

", "rawMarkdown": "Thank you please I am waiting for you guidance", "votes": null }, { "id": "2863631", "postDate": "06/09/2024 14:54:43", "content": "

I took a public notebook code to format my solution file.

\n

Make sure that the number of columns and their names are the same as in the sample submission file.

", "rawMarkdown": "I took a public notebook code to format my solution file.\n\nMake sure that the number of columns and their names are the same as in the sample submission file.", "votes": null }, { "id": "2863911", "postDate": "06/09/2024 19:18:45", "content": "

Still when I submit the output follows the same shape as the unlabeled audio that I extracted (10x183).
\nI thought that by submitting, the test path would be filled.

", "rawMarkdown": "Still when I submit the output follows the same shape as the unlabeled audio that I extracted (10x183). \nI thought that by submitting, the test path would be filled.", "votes": null } ], "comments": [ { "id": 2862318, "author_name": "kdmitrie", "author_url": "", "post_date": "06/08/2024 17:29:27", "content": "

I'm using the following code with glob library, which I've initially taken from the public notebook.
\nYour code looks ok, but you have to filter out everything except oog files. For example, if readme file exists in the directory, and you try to process it like ogg file, you get the error.

\n
import \n\nROOT_FOLDER = \n\n (.()) > :\n    ogg_file_paths = .()\n:\n    ogg_file_paths = sorted(.())[:]\n
", "votes": null, "replies": [ { "id": 2862638, "author_name": "arnaudchampetier", "author_url": "", "post_date": "06/08/2024 20:37:18", "content": "

Hey Konstantin,
\nThanks for your reply. I tried this code before and now again. However, the code still produces the outcome with the unlabeled sounds and not with the test ones (csv of 10x182). I think it is the way I submit which might be wrong…
\nDo you have any further idea ?
\nBest,
\nArnaud

", "votes": null, "replies": [] } ] }, { "id": 2862985, "author_name": "cpmpml", "author_url": "", "post_date": "06/09/2024 06:55:32", "content": "

You may be confusing saving and submitting.

\n

You cannot see what happens when you submit.

\n

You can see what happens when you save your notebook, before submititng it.

\n

Have you submitted at all?

", "votes": null, "replies": [ { "id": 2863071, "author_name": "arnaudchampetier", "author_url": "", "post_date": "06/09/2024 08:04:03", "content": "

Hey CPMP,

\n

Thanks for the answer, I really appreciate it.

\n

From my understanding, I submitted a notebook now 14 times.
\nI always get this message:
\n\"Submission Scoring Error
\nYour notebook generated a submission file with incorrect format. Some examples causing this are: wrong number of rows or columns, empty values, an incorrect data type for a value, or invalid submission values from what is expected. See more debugging tips\"
\nI will attach my input (ckpt and whl for einops), as well as the logs of my submissions.

\n

Any further help would be much appreciated.

\n

Thanks a lot for your time,

\n

Best,
\nArnaud

", "votes": null, "replies": [ { "id": 2863340, "author_name": "cpmpml", "author_url": "", "post_date": "06/09/2024 10:34:22", "content": "

OK, you have errors in submitting,, but that's not what you posted about. You have a topic with title: \"Can't submit, test dataset undetected (?)\"

\n

II replied to the issue you describe in your original post.

", "votes": null, "replies": [ { "id": 2863346, "author_name": "ahmedkhawaja", "author_url": "", "post_date": "06/09/2024 10:42:01", "content": "

Thank you please I am waiting for you guidance

", "votes": null, "replies": [ { "id": 2863631, "author_name": "cpmpml", "author_url": "", "post_date": "06/09/2024 14:54:43", "content": "

I took a public notebook code to format my solution file.

\n

Make sure that the number of columns and their names are the same as in the sample submission file.

", "votes": null, "replies": [ { "id": 2863911, "author_name": "arnaudchampetier", "author_url": "", "post_date": "06/09/2024 19:18:45", "content": "

Still when I submit the output follows the same shape as the unlabeled audio that I extracted (10x183).
\nI thought that by submitting, the test path would be filled.

", "votes": null, "replies": [] } ] } ] } ] } ] } ] }, { "id": 2863140, "author_name": "ahmedkhawaja", "author_url": "", "post_date": "06/09/2024 08:40:57", "content": "

I am also unable to submit to competition.it saying your notebook is exceeding allowed runtime.how can I speed up my processing to avoid this and successfully submit my notebook

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2862142": "Hi everyone, \nI'm stuck, I'm trying to submit my notebook which works perfectly using the unlabeled soundscapes. However, when I submit, I expect the path '/kaggle/input/birdclef-2024/test_soundscapes/' to be filled with audio. Here is my code: \n\n```python\ntest_audio_dir = '/kaggle/input/birdclef-2024/test_soundscapes/'\n\nogg_file_paths = [test_audio_dir+f for f in sorted(os.listdir(test_audio_dir))]\n\nif len(ogg_file_paths)==1:\n ogg_file_paths = [f'{config.DATA_ROOT}/unlabeled_soundscapes/'+f for f in sorted(os.listdir(f'{config.DATA_ROOT}/unlabeled_soundscapes/'))[:10]]\n```\n\nWhen I submit using this, the file detected is still the readme. It is also the 1st competition I'm trying to submit to. Does anyone know how to unblock me? I will be extremely grateful. \n\nBest regards, \n\nArnaud", "2862318": "I'm using the following code with glob library, which I've initially taken from [the public notebook](https://www.kaggle.com/code/markwijkhuizen/birdclef-2024-efficientvit-inference).\nYour code looks ok, but you have to filter out everything except oog files. For example, if readme file exists in the directory, and you try to process it like ogg file, you get the error.\n\n```\nimport glob\n\nROOT_FOLDER = '/kaggle/input/birdclef-2024'\n\nif len(glob.glob(f'{ROOT_FOLDER}/test_soundscapes/*.ogg')) > 0:\n ogg_file_paths = glob.glob(f'{ROOT_FOLDER}/test_soundscapes/*.ogg')\nelse:\n ogg_file_paths = sorted(glob.glob(f'{ROOT_FOLDER}/unlabeled_soundscapes/*.ogg'))[:10]\n```", "2862638": "Hey Konstantin, \nThanks for your reply. I tried this code before and now again. However, the code still produces the outcome with the unlabeled sounds and not with the test ones (csv of 10x182). I think it is the way I submit which might be wrong... \nDo you have any further idea ? \nBest, \nArnaud", "2862985": "You may be confusing saving and submitting.\n\nYou cannot see what happens when you submit.\n\nYou can see what happens when you save your notebook, before submititng it.\n\nHave you submitted at all?", "2863071": "Hey CPMP, \n\nThanks for the answer, I really appreciate it. \n\nFrom my understanding, I submitted a notebook now 14 times. \nI always get this message: \n\"Submission Scoring Error\nYour notebook generated a submission file with incorrect format. Some examples causing this are: wrong number of rows or columns, empty values, an incorrect data type for a value, or invalid submission values from what is expected. See more debugging tips\"\nI will attach my input (ckpt and whl for einops), as well as the logs of my submissions. \n\nAny further help would be much appreciated. \n\nThanks a lot for your time, \n\nBest, \nArnaud", "2863140": "I am also unable to submit to competition.it saying your notebook is exceeding allowed runtime.how can I speed up my processing to avoid this and successfully submit my notebook", "2863340": "OK, you have errors in submitting,, but that's not what you posted about. You have a topic with title: \"Can't submit, test dataset undetected (?)\"\n\nII replied to the issue you describe in your original post.", "2863346": "Thank you please I am waiting for you guidance", "2863631": "I took a public notebook code to format my solution file.\n\nMake sure that the number of columns and their names are the same as in the sample submission file.", "2863911": "Still when I submit the output follows the same shape as the unlabeled audio that I extracted (10x183). \nI thought that by submitting, the test path would be filled." }, "source": "meta" }