{
  "id": 547464,
  "title": "Cannot Import local notebook",
  "url": "/competitions/child-mind-institute-problematic-internet-use/discussion/547464",
  "author_name": "",
  "post_date": "2024-11-21T18:26:10.470248900Z",
  "votes": 1,
  "comment_count": 4,
  "views": 0,
  "content": "<p>22 days ago I was able to import a local notebook and run it.</p>\n<p>Today the import no longer works.  The file appears to be uploaded as json like rather than notebook.  I am using VS for my local, but saved the file from anaconda jupyter editor with the same result.</p>\n<p>When I attempt to import the same file from 22 days ago it works correctly.   So I asssume the issue is with VS rather than kaggle, but clueless on how to fix this.</p>\n<p>Any ideas?</p>",
  "messages": [
    {
      "id": "3051893",
      "postDate": "11/21/2024 18:26:10",
      "content": "<p>22 days ago I was able to import a local notebook and run it.</p>\n<p>Today the import no longer works.  The file appears to be uploaded as json like rather than notebook.  I am using VS for my local, but saved the file from anaconda jupyter editor with the same result.</p>\n<p>When I attempt to import the same file from 22 days ago it works correctly.   So I asssume the issue is with VS rather than kaggle, but clueless on how to fix this.</p>\n<p>Any ideas?</p>",
      "rawMarkdown": "22 days ago I was able to import a local notebook and run it.\n\nToday the import no longer works.  The file appears to be uploaded as json like rather than notebook.  I am using VS for my local, but saved the file from anaconda jupyter editor with the same result.\n\nWhen I attempt to import the same file from 22 days ago it works correctly.   So I asssume the issue is with VS rather than kaggle, but clueless on how to fix this.\n\nAny ideas?",
      "votes": null
    },
    {
      "id": "3051908",
      "postDate": "11/21/2024 18:41:06",
      "content": "<p>Issue for sure in VS - I resaved the 22 day old file and it no longer imports correctly.</p>\n<p>Do notice that the notebook size of files that I can import are in the 15 mb size, while the newly saved files are much smaller at 150 kb size.</p>",
      "rawMarkdown": "Issue for sure in VS - I resaved the 22 day old file and it no longer imports correctly.\n\nDo notice that the notebook size of files that I can import are in the 15 mb size, while the newly saved files are much smaller at 150 kb size.",
      "votes": null
    },
    {
      "id": "3055474",
      "postDate": "11/25/2024 19:39:43",
      "content": "<p>I'm not sure what error you're seeing, but have you tried to convert your notebook into a .py (text) file and upload the result? There won't be cells anymore but at least you could use the code for a submission.</p>",
      "rawMarkdown": "I'm not sure what error you're seeing, but have you tried to convert your notebook into a .py (text) file and upload the result? There won't be cells anymore but at least you could use the code for a submission.",
      "votes": null
    },
    {
      "id": "3056260",
      "postDate": "11/26/2024 18:33:07",
      "content": "<p>Have not tried that yet - still hoping to understand why the VS starting saving notebooks in format not compatible with Kaggle import.  It seems that the format change is not \"undone\" if I open the notebook with anaconda-navigator Jupyter and save a new version there.    With the holidays this week I am probably just going to wait for the monthly update to VS to see if they identified and fixed this.  </p>\n<p>I have another machine that still running a little older VS version that I can use to get a submission.</p>",
      "rawMarkdown": "Have not tried that yet - still hoping to understand why the VS starting saving notebooks in format not compatible with Kaggle import.  It seems that the format change is not \"undone\" if I open the notebook with anaconda-navigator Jupyter and save a new version there.    With the holidays this week I am probably just going to wait for the monthly update to VS to see if they identified and fixed this.  \n\nI have another machine that still running a little older VS version that I can use to get a submission.",
      "votes": null
    },
    {
      "id": "3056264",
      "postDate": "11/26/2024 18:38:40",
      "content": "<p>This is the what the notebook looks like (first few lines) when imported into kaggle.</p>\n<p>{<br>\n \"cells\": [<br>\n  {<br>\n   \"cell_type\": \"markdown\",<br>\n   \"id\": \"3d48ebd0\",<br>\n   \"metadata\": {},<br>\n   \"source\": [<br>\n    \"* R012 - first model has valid sii only.  Counts as average\",<br>\n    \"* R013 - all models have valid sii\"<br>\n   ]<br>\n  },<br>\n  {<br>\n   \"cell_type\": \"markdown\",<br>\n   \"id\": \"390ea556\",<br>\n   \"metadata\": {},<br>\n   \"source\": [<br>\n    \"* use python 3.11.7\"<br>\n   ]<br>\n  },<br>\n  {<br>\n   \"cell_type\": \"code\",<br>\n   \"execution_count\": 3,<br>\n   \"id\": \"d4cb4f74\",<br>\n   \"metadata\": {},<br>\n   \"outputs\": [],<br>\n   \"source\": [<br>\n    \"# ! pip install optuna\",<br>\n    \"# ! pip install catboost\",<br>\n    \"# ! pip install optuna-integration\",<br>\n    \"# ! conda install dask --yes\",<br>\n    \"# ! pip uninstall lightgbm --yes\",<br>\n    \"# ! pip install lightgbm\",<br>\n    \"# ! pip uninstall pyarrow --yes\",<br>\n    \"# ! pip install pyarrow.\",<br>\n    \"# ! pip install optuna-integration[tfkeras]\",<br>\n    \"\"<br>\n   ]</p>",
      "rawMarkdown": "This is the what the notebook looks like (first few lines) when imported into kaggle.\n\n{\n \"cells\": [\n  {\n   \"cell_type\": \"markdown\",\n   \"id\": \"3d48ebd0\",\n   \"metadata\": {},\n   \"source\": [\n    \"* R012 - first model has valid sii only.  Counts as average\",\n    \"* R013 - all models have valid sii\"\n   ]\n  },\n  {\n   \"cell_type\": \"markdown\",\n   \"id\": \"390ea556\",\n   \"metadata\": {},\n   \"source\": [\n    \"* use python 3.11.7\"\n   ]\n  },\n  {\n   \"cell_type\": \"code\",\n   \"execution_count\": 3,\n   \"id\": \"d4cb4f74\",\n   \"metadata\": {},\n   \"outputs\": [],\n   \"source\": [\n    \"# ! pip install optuna\",\n    \"# ! pip install catboost\",\n    \"# ! pip install optuna-integration\",\n    \"# ! conda install dask --yes\",\n    \"# ! pip uninstall lightgbm --yes\",\n    \"# ! pip install lightgbm\",\n    \"# ! pip uninstall pyarrow --yes\",\n    \"# ! pip install pyarrow.\",\n    \"# ! pip install optuna-integration[tfkeras]\",\n    \"\"\n   ]",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3051908,
      "author_name": "pcjimmmy",
      "author_url": "",
      "post_date": "11/21/2024 18:41:06",
      "content": "<p>Issue for sure in VS - I resaved the 22 day old file and it no longer imports correctly.</p>\n<p>Do notice that the notebook size of files that I can import are in the 15 mb size, while the newly saved files are much smaller at 150 kb size.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 3055474,
      "author_name": "josephmarturano",
      "author_url": "",
      "post_date": "11/25/2024 19:39:43",
      "content": "<p>I'm not sure what error you're seeing, but have you tried to convert your notebook into a .py (text) file and upload the result? There won't be cells anymore but at least you could use the code for a submission.</p>",
      "votes": null,
      "replies": [
        {
          "id": 3056260,
          "author_name": "pcjimmmy",
          "author_url": "",
          "post_date": "11/26/2024 18:33:07",
          "content": "<p>Have not tried that yet - still hoping to understand why the VS starting saving notebooks in format not compatible with Kaggle import.  It seems that the format change is not \"undone\" if I open the notebook with anaconda-navigator Jupyter and save a new version there.    With the holidays this week I am probably just going to wait for the monthly update to VS to see if they identified and fixed this.  </p>\n<p>I have another machine that still running a little older VS version that I can use to get a submission.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 3056264,
          "author_name": "pcjimmmy",
          "author_url": "",
          "post_date": "11/26/2024 18:38:40",
          "content": "<p>This is the what the notebook looks like (first few lines) when imported into kaggle.</p>\n<p>{<br>\n \"cells\": [<br>\n  {<br>\n   \"cell_type\": \"markdown\",<br>\n   \"id\": \"3d48ebd0\",<br>\n   \"metadata\": {},<br>\n   \"source\": [<br>\n    \"* R012 - first model has valid sii only.  Counts as average\",<br>\n    \"* R013 - all models have valid sii\"<br>\n   ]<br>\n  },<br>\n  {<br>\n   \"cell_type\": \"markdown\",<br>\n   \"id\": \"390ea556\",<br>\n   \"metadata\": {},<br>\n   \"source\": [<br>\n    \"* use python 3.11.7\"<br>\n   ]<br>\n  },<br>\n  {<br>\n   \"cell_type\": \"code\",<br>\n   \"execution_count\": 3,<br>\n   \"id\": \"d4cb4f74\",<br>\n   \"metadata\": {},<br>\n   \"outputs\": [],<br>\n   \"source\": [<br>\n    \"# ! pip install optuna\",<br>\n    \"# ! pip install catboost\",<br>\n    \"# ! pip install optuna-integration\",<br>\n    \"# ! conda install dask --yes\",<br>\n    \"# ! pip uninstall lightgbm --yes\",<br>\n    \"# ! pip install lightgbm\",<br>\n    \"# ! pip uninstall pyarrow --yes\",<br>\n    \"# ! pip install pyarrow.\",<br>\n    \"# ! pip install optuna-integration[tfkeras]\",<br>\n    \"\"<br>\n   ]</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3051893": "22 days ago I was able to import a local notebook and run it.\n\nToday the import no longer works.  The file appears to be uploaded as json like rather than notebook.  I am using VS for my local, but saved the file from anaconda jupyter editor with the same result.\n\nWhen I attempt to import the same file from 22 days ago it works correctly.   So I asssume the issue is with VS rather than kaggle, but clueless on how to fix this.\n\nAny ideas?",
    "3051908": "Issue for sure in VS - I resaved the 22 day old file and it no longer imports correctly.\n\nDo notice that the notebook size of files that I can import are in the 15 mb size, while the newly saved files are much smaller at 150 kb size.",
    "3055474": "I'm not sure what error you're seeing, but have you tried to convert your notebook into a .py (text) file and upload the result? There won't be cells anymore but at least you could use the code for a submission.",
    "3056260": "Have not tried that yet - still hoping to understand why the VS starting saving notebooks in format not compatible with Kaggle import.  It seems that the format change is not \"undone\" if I open the notebook with anaconda-navigator Jupyter and save a new version there.    With the holidays this week I am probably just going to wait for the monthly update to VS to see if they identified and fixed this.  \n\nI have another machine that still running a little older VS version that I can use to get a submission.",
    "3056264": "This is the what the notebook looks like (first few lines) when imported into kaggle.\n\n{\n \"cells\": [\n  {\n   \"cell_type\": \"markdown\",\n   \"id\": \"3d48ebd0\",\n   \"metadata\": {},\n   \"source\": [\n    \"* R012 - first model has valid sii only.  Counts as average\",\n    \"* R013 - all models have valid sii\"\n   ]\n  },\n  {\n   \"cell_type\": \"markdown\",\n   \"id\": \"390ea556\",\n   \"metadata\": {},\n   \"source\": [\n    \"* use python 3.11.7\"\n   ]\n  },\n  {\n   \"cell_type\": \"code\",\n   \"execution_count\": 3,\n   \"id\": \"d4cb4f74\",\n   \"metadata\": {},\n   \"outputs\": [],\n   \"source\": [\n    \"# ! pip install optuna\",\n    \"# ! pip install catboost\",\n    \"# ! pip install optuna-integration\",\n    \"# ! conda install dask --yes\",\n    \"# ! pip uninstall lightgbm --yes\",\n    \"# ! pip install lightgbm\",\n    \"# ! pip uninstall pyarrow --yes\",\n    \"# ! pip install pyarrow.\",\n    \"# ! pip install optuna-integration[tfkeras]\",\n    \"\"\n   ]"
  },
  "source": "meta"
}