I am not much familiar notebook style submission, and am having difficulties. I submitted a notebook with sample submissions a number of times, and I got 0.5 as expected; however, my test_soundscapes folder is never populated. I guess I need the test audio files to make the predictions and submit. Am I missing something like a specific parameter setting, or doing something wrong, any thoughts?
", "messages": [ { "id": "2789440", "postDate": "05/02/2024 17:07:55", "content": "I am not much familiar notebook style submission, and am having difficulties. I submitted a notebook with sample submissions a number of times, and I got 0.5 as expected; however, my test_soundscapes folder is never populated. I guess I need the test audio files to make the predictions and submit. Am I missing something like a specific parameter setting, or doing something wrong, any thoughts?
", "rawMarkdown": "I am not much familiar notebook style submission, and am having difficulties. I submitted a notebook with sample submissions a number of times, and I got 0.5 as expected; however, my test_soundscapes folder is never populated. I guess I need the test audio files to make the predictions and submit. Am I missing something like a specific parameter setting, or doing something wrong, any thoughts?", "votes": null }, { "id": "2789928", "postDate": "05/02/2024 22:48:57", "content": "Hello!
\nThe test files are only visible to the notebook running /during submission/ - they are not observable by humans such as (hopefully) yourself.
", "rawMarkdown": "Hello! \n\nThe test files are only visible to the notebook running /during submission/ - they are not observable by humans such as (hopefully) yourself.", "votes": null }, { "id": "2790915", "postDate": "05/03/2024 11:41:48", "content": "On what dataset can I do the inference for submissions, considering the test data is unreachable? I tried unlabeled dataset but then the submission fails. This is I find hard to understand.
", "rawMarkdown": "On what dataset can I do the inference for submissions, considering the test data is unreachable? I tried unlabeled dataset but then the submission fails. This is I find hard to understand.", "votes": null }, { "id": "2791505", "postDate": "05/03/2024 17:31:15", "content": "Hi; check the 'Data' description in the competition description. While the submission notebook is running, it will have access to the test files.
\nIt will also be helpful to start from one of the basic submission notebooks, like this one:
\nhttps://www.kaggle.com/code/myso1987/how-to-submit-to-birdclef-2024
\nAdapting this code will make it easier to focus on modeling, rather than spending a lot of time getting the submission format right.
Hello, I apologize for jumping into the thread.
\nI copied your code above, and submit to verify the \"test_soundscapes\" whether it works or not.
\nbut, displayed ”Number of test soundscapes: 0”.
In \"Session options\" , I toggled \"Internet\" and turn of the Internet.
\nThen, in \"Submit to competition\", I pressed the \"submit\" button.
\nWhat should I do anything else?
\nCould you tell me how to submit in this competion?
If you submit you can't see any prints. Maybe you saved your notebook, and did not submit it?
", "rawMarkdown": "If you submit you can't see any prints. Maybe you saved your notebook, and did not submit it?", "votes": null }, { "id": "2823803", "postDate": "05/19/2024 12:29:48", "content": "Thank you for your responce.
\nI pressed the \"submit\" button, followed by \"Save and Run all\" was done.
\nSo, I was aware of \"Submission was done\".
\nBut Notebooks log opened by \"Open logs in viewer\" displayed \"Number of test soundscapes: 0\".
\nIs this method incorrect?
\"Save and Run all\" is saving your notebook, not submitting it.
", "rawMarkdown": "\"Save and Run all\" is saving your notebook, not submitting it.", "votes": null }, { "id": "2824581", "postDate": "05/19/2024 21:22:56", "content": "Thank you for your responce.
\nI'm sorry for my poor explanation.
\nWhen the ”Submit” button was pressed, the notebook was run all at the same time.
\nBut you meant that it was not inference, didn't you?
If you don't mind, could you tell me how to submit in your way?
", "rawMarkdown": "Thank you for your responce.\n\nI'm sorry for my poor explanation.\nWhen the ”Submit” button was pressed, the notebook was run all at the same time.\nBut you meant that it was not inference, didn't you?\n\nIf you don't mind, could you tell me how to submit in your way?", "votes": null }, { "id": "2824677", "postDate": "05/20/2024 01:36:27", "content": "You cannot see the log of what happens when you submit.
", "rawMarkdown": "You cannot see the log of what happens when you submit.", "votes": null }, { "id": "2825395", "postDate": "05/20/2024 11:27:01", "content": "Thank you for your responce in many times.
\nIn Competition top page, and \"Submission\" tab, we can select submitted notebook for scoring.
\nAnd we can open notebook preview in inference.
\nThe logs for inference is displayed, is it correct?
\nFor collecting test data, the code in my notebook was below.
\nROOT_FOLDER = '/kaggle/input/birdclef-2024'
\ntest_audio_dir = '../input/birdclef-2024/test_soundscapes/'
\naudio_folder = True
\nogg_file_paths = [f for f in sorted(os.listdir(test_audio_dir))]
\nogg_file_paths = [file.split('.')[0] for file in ogg_file_paths if file.endswith('.ogg')]
\nif len(ogg_file_paths) == 0:
\nogg_file_paths = sorted(glob.glob(f'{ROOT_FOLDER}/unlabeled_soundscapes/*.ogg'))[:10]
\naudio_folder = False
\nprint(audio_folder)
\nprint(ogg_file_paths[:5])
And displayed like below.
\nFalse
\n['/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000170626.ogg', '/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000308629.ogg', '/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000389428.ogg', '/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000424265.ogg', '/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000450112.ogg']
\nAnd \"Notebook Threw Exception\" error occured that is because unlabeled data were used for inference, then submission.csv was smaller.
\nI have no idea for correct submission…
You see the log of the run that happens when you save the notebook. When you submit a new run starts with test data. You cannot see the log of the submit run.
", "rawMarkdown": "You see the log of the run that happens when you save the notebook. When you submit a new run starts with test data. You cannot see the log of the submit run.", "votes": null }, { "id": "2825592", "postDate": "05/20/2024 13:40:29", "content": "Oh…
\nYou meant my notebook reached test data, but any other bug caused \"Notebook Threw Exception\" error?
\n\"Save and Run all\" was correctly done, then how do I solve this problem?
Hello!
\nThe test files are only visible to the notebook running /during submission/ - they are not observable by humans such as (hopefully) yourself.
", "votes": null, "replies": [ { "id": 2790915, "author_name": "gurselkaracor", "author_url": "", "post_date": "05/03/2024 11:41:48", "content": "On what dataset can I do the inference for submissions, considering the test data is unreachable? I tried unlabeled dataset but then the submission fails. This is I find hard to understand.
", "votes": null, "replies": [ { "id": 2791505, "author_name": "tomdenton", "author_url": "", "post_date": "05/03/2024 17:31:15", "content": "Hi; check the 'Data' description in the competition description. While the submission notebook is running, it will have access to the test files.
\nIt will also be helpful to start from one of the basic submission notebooks, like this one:
\nhttps://www.kaggle.com/code/myso1987/how-to-submit-to-birdclef-2024
\nAdapting this code will make it easier to focus on modeling, rather than spending a lot of time getting the submission format right.
Hello, I apologize for jumping into the thread.
\nI copied your code above, and submit to verify the \"test_soundscapes\" whether it works or not.
\nbut, displayed ”Number of test soundscapes: 0”.
In \"Session options\" , I toggled \"Internet\" and turn of the Internet.
\nThen, in \"Submit to competition\", I pressed the \"submit\" button.
\nWhat should I do anything else?
\nCould you tell me how to submit in this competion?
If you submit you can't see any prints. Maybe you saved your notebook, and did not submit it?
", "votes": null, "replies": [ { "id": 2823803, "author_name": "asdfgkaggle", "author_url": "", "post_date": "05/19/2024 12:29:48", "content": "Thank you for your responce.
\nI pressed the \"submit\" button, followed by \"Save and Run all\" was done.
\nSo, I was aware of \"Submission was done\".
\nBut Notebooks log opened by \"Open logs in viewer\" displayed \"Number of test soundscapes: 0\".
\nIs this method incorrect?
\"Save and Run all\" is saving your notebook, not submitting it.
", "votes": null, "replies": [ { "id": 2824581, "author_name": "asdfgkaggle", "author_url": "", "post_date": "05/19/2024 21:22:56", "content": "Thank you for your responce.
\nI'm sorry for my poor explanation.
\nWhen the ”Submit” button was pressed, the notebook was run all at the same time.
\nBut you meant that it was not inference, didn't you?
If you don't mind, could you tell me how to submit in your way?
", "votes": null, "replies": [ { "id": 2824677, "author_name": "cpmpml", "author_url": "", "post_date": "05/20/2024 01:36:27", "content": "You cannot see the log of what happens when you submit.
", "votes": null, "replies": [ { "id": 2825395, "author_name": "asdfgkaggle", "author_url": "", "post_date": "05/20/2024 11:27:01", "content": "Thank you for your responce in many times.
\nIn Competition top page, and \"Submission\" tab, we can select submitted notebook for scoring.
\nAnd we can open notebook preview in inference.
\nThe logs for inference is displayed, is it correct?
\nFor collecting test data, the code in my notebook was below.
\nROOT_FOLDER = '/kaggle/input/birdclef-2024'
\ntest_audio_dir = '../input/birdclef-2024/test_soundscapes/'
\naudio_folder = True
\nogg_file_paths = [f for f in sorted(os.listdir(test_audio_dir))]
\nogg_file_paths = [file.split('.')[0] for file in ogg_file_paths if file.endswith('.ogg')]
\nif len(ogg_file_paths) == 0:
\nogg_file_paths = sorted(glob.glob(f'{ROOT_FOLDER}/unlabeled_soundscapes/*.ogg'))[:10]
\naudio_folder = False
\nprint(audio_folder)
\nprint(ogg_file_paths[:5])
And displayed like below.
\nFalse
\n['/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000170626.ogg', '/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000308629.ogg', '/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000389428.ogg', '/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000424265.ogg', '/kaggle/input/birdclef-2024/unlabeled_soundscapes/1000450112.ogg']
\nAnd \"Notebook Threw Exception\" error occured that is because unlabeled data were used for inference, then submission.csv was smaller.
\nI have no idea for correct submission…
You see the log of the run that happens when you save the notebook. When you submit a new run starts with test data. You cannot see the log of the submit run.
", "votes": null, "replies": [ { "id": 2825592, "author_name": "asdfgkaggle", "author_url": "", "post_date": "05/20/2024 13:40:29", "content": "Oh…
\nYou meant my notebook reached test data, but any other bug caused \"Notebook Threw Exception\" error?
\n\"Save and Run all\" was correctly done, then how do I solve this problem?