Hi All,
\nWhen running my notebook locally, I have no issues. I can restart the notebook, press run all, and have everything run fine. I have internet off when submitting, and every time I submit about 30 seconds in, the notebook gives me this error: \"Your notebook hit an unhandled error while rerunning your code. Note that the hidden dataset can be larger/smaller/different than the public dataset See more debugging tips\".
\nThe notebook itself however, successfully runs. Here are the logs:
\nSuccessfully ran in 261.6s
\nAccelerator
\nGPU P100
Environment
\nLatest Container Image
Output
\n21 B
Time
\n#
\nLog Message
\n12.0s 1 /opt/conda/lib/python3.10/site-packages/scipy/init.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.5\n12.0s 2 warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"
\n178.0s 3 538
\n178.0s 4 538
\n178.1s 5 (538, 224, 224, 3)
\n178.1s 6 (538,)
\n186.9s 7 /opt/conda/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
\n186.9s 8 warnings.warn(
\n186.9s 9 /opt/conda/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or None for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing weights=None.
\n186.9s 10 warnings.warn(msg)
\n253.0s 11 Predicted class: 1
\n253.0s 12 Output probabilities: tensor([0.0073, 0.8386, 0.1164, 0.0114, 0.0263])
\n253.0s 13 Predicted class name: EC
\n258.5s 14 /opt/conda/lib/python3.10/site-packages/traitlets/traitlets.py:2930: FutureWarning: --Exporter.preprocessors=[\"remove_papermill_header.RemovePapermillHeader\"] for containers is deprecated in traitlets 5.0. You can pass --Exporter.preprocessors item … multiple times to add items to a list.
\n258.5s 15 warn(
\n258.5s 16 [NbConvertApp] WARNING | Config option kernel_spec_manager_class not recognized by NbConvertApp.
\n258.5s 17 [NbConvertApp] Converting notebook notebook.ipynb to notebook
\n258.9s 18 [NbConvertApp] Writing 765566 bytes to notebook.ipynb
\n260.5s 19 /opt/conda/lib/python3.10/site-packages/traitlets/traitlets.py:2930: FutureWarning: --Exporter.preprocessors=[\"nbconvert.preprocessors.ExtractOutputPreprocessor\"] for containers is deprecated in traitlets 5.0. You can pass --Exporter.preprocessors item … multiple times to add items to a list.
\n260.5s 20 warn(
\n260.5s 21 [NbConvertApp] WARNING | Config option kernel_spec_manager_class not recognized by NbConvertApp.
\n260.6s 22 [NbConvertApp] Converting notebook notebook.ipynb to html
\n261.5s 23 [NbConvertApp] Support files will be in results___files/\n261.5s 24 [NbConvertApp] Making directory __results___files\n261.5s 25 [NbConvertApp] Making directory __results___files\n261.5s 26 [NbConvertApp] Making directory __results___files\n261.5s 27 [NbConvertApp] Making directory __results___files\n261.5s 28 [NbConvertApp] Writing 317193 bytes to __results.html
Any idea whats going on?
", "messages": [ { "id": 2479170, "postDate": "2023-10-12T13:26:32.607Z", "content": "Hi All,
\nWhen running my notebook locally, I have no issues. I can restart the notebook, press run all, and have everything run fine. I have internet off when submitting, and every time I submit about 30 seconds in, the notebook gives me this error: \"Your notebook hit an unhandled error while rerunning your code. Note that the hidden dataset can be larger/smaller/different than the public dataset See more debugging tips\".
\nThe notebook itself however, successfully runs. Here are the logs:
\nSuccessfully ran in 261.6s
\nAccelerator
\nGPU P100
Environment
\nLatest Container Image
Output
\n21 B
Time
\n#
\nLog Message
\n12.0s 1 /opt/conda/lib/python3.10/site-packages/scipy/init.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.5\n12.0s 2 warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"
\n178.0s 3 538
\n178.0s 4 538
\n178.1s 5 (538, 224, 224, 3)
\n178.1s 6 (538,)
\n186.9s 7 /opt/conda/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
\n186.9s 8 warnings.warn(
\n186.9s 9 /opt/conda/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or None for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing weights=None.
\n186.9s 10 warnings.warn(msg)
\n253.0s 11 Predicted class: 1
\n253.0s 12 Output probabilities: tensor([0.0073, 0.8386, 0.1164, 0.0114, 0.0263])
\n253.0s 13 Predicted class name: EC
\n258.5s 14 /opt/conda/lib/python3.10/site-packages/traitlets/traitlets.py:2930: FutureWarning: --Exporter.preprocessors=[\"remove_papermill_header.RemovePapermillHeader\"] for containers is deprecated in traitlets 5.0. You can pass --Exporter.preprocessors item … multiple times to add items to a list.
\n258.5s 15 warn(
\n258.5s 16 [NbConvertApp] WARNING | Config option kernel_spec_manager_class not recognized by NbConvertApp.
\n258.5s 17 [NbConvertApp] Converting notebook notebook.ipynb to notebook
\n258.9s 18 [NbConvertApp] Writing 765566 bytes to notebook.ipynb
\n260.5s 19 /opt/conda/lib/python3.10/site-packages/traitlets/traitlets.py:2930: FutureWarning: --Exporter.preprocessors=[\"nbconvert.preprocessors.ExtractOutputPreprocessor\"] for containers is deprecated in traitlets 5.0. You can pass --Exporter.preprocessors item … multiple times to add items to a list.
\n260.5s 20 warn(
\n260.5s 21 [NbConvertApp] WARNING | Config option kernel_spec_manager_class not recognized by NbConvertApp.
\n260.6s 22 [NbConvertApp] Converting notebook notebook.ipynb to html
\n261.5s 23 [NbConvertApp] Support files will be in results___files/\n261.5s 24 [NbConvertApp] Making directory __results___files\n261.5s 25 [NbConvertApp] Making directory __results___files\n261.5s 26 [NbConvertApp] Making directory __results___files\n261.5s 27 [NbConvertApp] Making directory __results___files\n261.5s 28 [NbConvertApp] Writing 317193 bytes to __results.html
Any idea whats going on?
", "rawMarkdown": "Hi All,\n\nWhen running my notebook locally, I have no issues. I can restart the notebook, press run all, and have everything run fine. I have internet off when submitting, and every time I submit about 30 seconds in, the notebook gives me this error: \"Your notebook hit an unhandled error while rerunning your code. Note that the hidden dataset can be larger/smaller/different than the public dataset See more debugging tips\". \n\nThe notebook itself however, successfully runs. Here are the logs:\n\nSuccessfully ran in 261.6s\nAccelerator\nGPU P100\n\nEnvironment\nLatest Container Image\n\nOutput\n21 B\n\n\nTime\n#\nLog Message\n12.0s\t1\t/opt/conda/lib/python3.10/site-packages/scipy/__init__.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.5\n12.0s\t2\t warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n178.0s\t3\t538\n178.0s\t4\t538\n178.1s\t5\t(538, 224, 224, 3)\n178.1s\t6\t(538,)\n186.9s\t7\t/opt/conda/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n186.9s\t8\t warnings.warn(\n186.9s\t9\t/opt/conda/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=None`.\n186.9s\t10\t warnings.warn(msg)\n253.0s\t11\tPredicted class: 1\n253.0s\t12\tOutput probabilities: tensor([0.0073, 0.8386, 0.1164, 0.0114, 0.0263])\n253.0s\t13\tPredicted class name: EC\n258.5s\t14\t/opt/conda/lib/python3.10/site-packages/traitlets/traitlets.py:2930: FutureWarning: --Exporter.preprocessors=[\"remove_papermill_header.RemovePapermillHeader\"] for containers is deprecated in traitlets 5.0. You can pass `--Exporter.preprocessors item` ... multiple times to add items to a list.\n258.5s\t15\t warn(\n258.5s\t16\t[NbConvertApp] WARNING | Config option `kernel_spec_manager_class` not recognized by `NbConvertApp`.\n258.5s\t17\t[NbConvertApp] Converting notebook __notebook__.ipynb to notebook\n258.9s\t18\t[NbConvertApp] Writing 765566 bytes to __notebook__.ipynb\n260.5s\t19\t/opt/conda/lib/python3.10/site-packages/traitlets/traitlets.py:2930: FutureWarning: --Exporter.preprocessors=[\"nbconvert.preprocessors.ExtractOutputPreprocessor\"] for containers is deprecated in traitlets 5.0. You can pass `--Exporter.preprocessors item` ... multiple times to add items to a list.\n260.5s\t20\t warn(\n260.5s\t21\t[NbConvertApp] WARNING | Config option `kernel_spec_manager_class` not recognized by `NbConvertApp`.\n260.6s\t22\t[NbConvertApp] Converting notebook __notebook__.ipynb to html\n261.5s\t23\t[NbConvertApp] Support files will be in __results___files/\n261.5s\t24\t[NbConvertApp] Making directory __results___files\n261.5s\t25\t[NbConvertApp] Making directory __results___files\n261.5s\t26\t[NbConvertApp] Making directory __results___files\n261.5s\t27\t[NbConvertApp] Making directory __results___files\n261.5s\t28\t[NbConvertApp] Writing 317193 bytes to __results__.html\n\nAny idea whats going on?", "votes": 5 }, { "id": 2482467, "postDate": "2023-10-15T04:26:39.687Z", "content": "Were you able to get the cause. I am also facing the same.
", "rawMarkdown": "Were you able to get the cause. I am also facing the same.", "votes": 4, "replies": [ { "id": 2490401, "postDate": "2023-10-20T16:33:31.577Z", "content": "No :/ any update on your end?
", "rawMarkdown": "No :/ any update on your end?", "replies": [ { "id": 2496496, "postDate": "2023-10-24T05:10:30.990Z", "content": "I figured out that in my case the submission csv file was not getting created correctly. I hope you already know that during the notebook submission training data won't be available, so use 2 notebooks, one for training and another for inference. In inference notebook, check that the submission csv is generated correctly. One method i used to figure the issue out was by drawing inference on train data itself and checking that the submission csv is generated properly. Hope this helps you.
", "rawMarkdown": "I figured out that in my case the submission csv file was not getting created correctly. I hope you already know that during the notebook submission training data won't be available, so use 2 notebooks, one for training and another for inference. In inference notebook, check that the submission csv is generated correctly. One method i used to figure the issue out was by drawing inference on train data itself and checking that the submission csv is generated properly. Hope this helps you.", "votes": 5 } ] } ] }, { "id": 2490631, "postDate": "2023-10-20T21:22:09.400Z", "content": "Hey. I had a similar issue and by trial and error (and approx. 20 wasted submissions) I figured out that the problem was the following:
\nI use OpenCV to load the PNGs. Since the images are so huge, I got an exception. It seems that the one provided test image was small enough. To fix this, you have to set the limit to a very high number:
the import!\nos = (,).()\nimport cv2\nHope this helps someone!
", "rawMarkdown": "Hey. I had a similar issue and by trial and error (and approx. 20 wasted submissions) I figured out that the problem was the following:\nI use OpenCV to load the PNGs. Since the images are so huge, I got an exception. It seems that the one provided test image was small enough. To fix this, you have to set the limit to a very high number:\n\n```\n#Before the import!\nos.environ[\"OPENCV_IO_MAX_IMAGE_PIXELS\"] = pow(2,40).__str__()\nimport cv2\n```\nHope this helps someone!", "votes": 1 }, { "id": 2957854, "postDate": "2024-08-13T13:31:28.333Z", "content": "can you solve problem?
", "rawMarkdown": "can you solve problem?" }, { "id": 2522416, "postDate": "2023-11-12T16:39:14.177Z", "content": "same probleme any help
", "rawMarkdown": "same probleme any help", "replies": [ { "id": 2522607, "postDate": "2023-11-12T21:13:38.790Z", "content": "Have you tried splitting your notebook into training and inference (submit this one)? See Balmukund's reply below.
", "rawMarkdown": "Have you tried splitting your notebook into training and inference (submit this one)? See [Balmukund's reply below](https://www.kaggle.com/competitions/UBC-OCEAN/discussion/446642#2496496)." } ] }, { "id": 2488062, "postDate": "2023-10-19T02:07:35.017Z", "content": "I am also facing the same problem😭
", "rawMarkdown": "I am also facing the same problem😭", "replies": [ { "id": 2505223, "postDate": "2023-10-30T13:17:33.963Z", "content": "I have same problem.
\nif you solve it. can you let me know how to solve it?
does annyone solve the problème ?
", "rawMarkdown": "does annyone solve the problème ?\n" } ] } ], "comments": [ { "id": 2482467, "author_name": "Balmukund", "author_url": "", "post_date": "2023-10-15T04:26:39.687000", "content": "Were you able to get the cause. I am also facing the same.
", "votes": 4, "replies": [ { "id": 2490401, "author_name": "jdonnelly0804", "author_url": "", "post_date": "2023-10-20T16:33:31.577000", "content": "No :/ any update on your end?
", "votes": 0, "replies": [ { "id": 2496496, "author_name": "Balmukund", "author_url": "", "post_date": "2023-10-24T05:10:30.990000", "content": "I figured out that in my case the submission csv file was not getting created correctly. I hope you already know that during the notebook submission training data won't be available, so use 2 notebooks, one for training and another for inference. In inference notebook, check that the submission csv is generated correctly. One method i used to figure the issue out was by drawing inference on train data itself and checking that the submission csv is generated properly. Hope this helps you.
", "votes": 5, "replies": [] } ] } ] }, { "id": 2490631, "author_name": "Manuel K", "author_url": "", "post_date": "2023-10-20T21:22:09.400000", "content": "Hey. I had a similar issue and by trial and error (and approx. 20 wasted submissions) I figured out that the problem was the following:
\nI use OpenCV to load the PNGs. Since the images are so huge, I got an exception. It seems that the one provided test image was small enough. To fix this, you have to set the limit to a very high number:
the import!\nos = (,).()\nimport cv2\nHope this helps someone!
", "votes": 1, "replies": [] }, { "id": 2957854, "author_name": "Edna Mora", "author_url": "", "post_date": "2024-08-13T13:31:28.333000", "content": "can you solve problem?
", "votes": 0, "replies": [] }, { "id": 2522416, "author_name": "DIATTARA Amadou", "author_url": "", "post_date": "2023-11-12T16:39:14.177000", "content": "same probleme any help
", "votes": 0, "replies": [ { "id": 2522607, "author_name": "Russ Tokuyama", "author_url": "", "post_date": "2023-11-12T21:13:38.790000", "content": "Have you tried splitting your notebook into training and inference (submit this one)? See Balmukund's reply below.
", "votes": 0, "replies": [] } ] }, { "id": 2488062, "author_name": "Seeing Times", "author_url": "", "post_date": "2023-10-19T02:07:35.017000", "content": "I am also facing the same problem😭
", "votes": 0, "replies": [ { "id": 2505223, "author_name": "JoongSub Lim", "author_url": "", "post_date": "2023-10-30T13:17:33.963000", "content": "I have same problem.
\nif you solve it. can you let me know how to solve it?
does annyone solve the problème ?
", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "2479170": "Hi All,\n\nWhen running my notebook locally, I have no issues. I can restart the notebook, press run all, and have everything run fine. I have internet off when submitting, and every time I submit about 30 seconds in, the notebook gives me this error: \"Your notebook hit an unhandled error while rerunning your code. Note that the hidden dataset can be larger/smaller/different than the public dataset See more debugging tips\". \n\nThe notebook itself however, successfully runs. Here are the logs:\n\nSuccessfully ran in 261.6s\nAccelerator\nGPU P100\n\nEnvironment\nLatest Container Image\n\nOutput\n21 B\n\n\nTime\n#\nLog Message\n12.0s\t1\t/opt/conda/lib/python3.10/site-packages/scipy/__init__.py:146: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.23.5\n12.0s\t2\t warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n178.0s\t3\t538\n178.0s\t4\t538\n178.1s\t5\t(538, 224, 224, 3)\n178.1s\t6\t(538,)\n186.9s\t7\t/opt/conda/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n186.9s\t8\t warnings.warn(\n186.9s\t9\t/opt/conda/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=None`.\n186.9s\t10\t warnings.warn(msg)\n253.0s\t11\tPredicted class: 1\n253.0s\t12\tOutput probabilities: tensor([0.0073, 0.8386, 0.1164, 0.0114, 0.0263])\n253.0s\t13\tPredicted class name: EC\n258.5s\t14\t/opt/conda/lib/python3.10/site-packages/traitlets/traitlets.py:2930: FutureWarning: --Exporter.preprocessors=[\"remove_papermill_header.RemovePapermillHeader\"] for containers is deprecated in traitlets 5.0. You can pass `--Exporter.preprocessors item` ... multiple times to add items to a list.\n258.5s\t15\t warn(\n258.5s\t16\t[NbConvertApp] WARNING | Config option `kernel_spec_manager_class` not recognized by `NbConvertApp`.\n258.5s\t17\t[NbConvertApp] Converting notebook __notebook__.ipynb to notebook\n258.9s\t18\t[NbConvertApp] Writing 765566 bytes to __notebook__.ipynb\n260.5s\t19\t/opt/conda/lib/python3.10/site-packages/traitlets/traitlets.py:2930: FutureWarning: --Exporter.preprocessors=[\"nbconvert.preprocessors.ExtractOutputPreprocessor\"] for containers is deprecated in traitlets 5.0. You can pass `--Exporter.preprocessors item` ... multiple times to add items to a list.\n260.5s\t20\t warn(\n260.5s\t21\t[NbConvertApp] WARNING | Config option `kernel_spec_manager_class` not recognized by `NbConvertApp`.\n260.6s\t22\t[NbConvertApp] Converting notebook __notebook__.ipynb to html\n261.5s\t23\t[NbConvertApp] Support files will be in __results___files/\n261.5s\t24\t[NbConvertApp] Making directory __results___files\n261.5s\t25\t[NbConvertApp] Making directory __results___files\n261.5s\t26\t[NbConvertApp] Making directory __results___files\n261.5s\t27\t[NbConvertApp] Making directory __results___files\n261.5s\t28\t[NbConvertApp] Writing 317193 bytes to __results__.html\n\nAny idea whats going on?", "2482467": "Were you able to get the cause. I am also facing the same.", "2490631": "Hey. I had a similar issue and by trial and error (and approx. 20 wasted submissions) I figured out that the problem was the following:\nI use OpenCV to load the PNGs. Since the images are so huge, I got an exception. It seems that the one provided test image was small enough. To fix this, you have to set the limit to a very high number:\n\n```\n#Before the import!\nos.environ[\"OPENCV_IO_MAX_IMAGE_PIXELS\"] = pow(2,40).__str__()\nimport cv2\n```\nHope this helps someone!", "2957854": "can you solve problem?", "2522416": "same probleme any help", "2488062": "I am also facing the same problem😭" } }