{
  "id": 157899,
  "title": "Pytorch Error and Notebook Update",
  "url": "/competitions/jigsaw-multilingual-toxic-comment-classification/discussion/157899",
  "author_name": "",
  "post_date": "2020-06-12T12:53:57.902680100Z",
  "votes": 4,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I am facing the following behavior : I trained a pytorch model on TPU and then make inference on GPU but I am having the following error message : </p>\n\n<p>*<strong>RuntimeError: version_ &lt;= kMaxSupportedFileFormatVersion INTERNAL ASSERT FAILED at /opt/conda/conda-bld/pytorch_1579022034529/work/caffe2/serialize/inline_container.cc:132, please report a bug to PyTorch. Attempted to read a PyTorch file with version 4, but the maximum supported version for reading is 2. Your PyTorch installation may be too old. (init at /opt/conda/conda-bld/pytorch_1579022034529/work/caffe2/serialize/inline_container.cc:132)\nframe #0: c10::Error::Error(c10::SourceLocation, std::string const&amp;) + 0x47 (0x7fe51bc67627 in /opt/conda/lib/python3.6/site-packages/torch/lib/libc10.so)\nframe #1: caffe2::serialize::PyTorchStreamReader::init() + 0x1f5b (0x7fe524fa19ab in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch.so)\nframe #2: caffe2::serialize::PyTorchStreamReader::PyTorchStreamReader(std::string const&amp;) + 0x64 (0x7fe524fa2bc4 in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch.so)\n......*</strong></p>\n\n<p>I used this approach from the first start of the competition and it works fine, unfortunately starting from yesterday i am keeping getting this error.</p>\n\n<p>I can see @tanulsingh077  and @rftexas reported same behavior in tweet-sentiment-extraction competition. </p>",
  "messages": [
    {
      "id": "883201",
      "postDate": "06/12/2020 12:53:57",
      "content": "<p>I am facing the following behavior : I trained a pytorch model on TPU and then make inference on GPU but I am having the following error message : </p>\n\n<p>*<strong>RuntimeError: version_ &lt;= kMaxSupportedFileFormatVersion INTERNAL ASSERT FAILED at /opt/conda/conda-bld/pytorch_1579022034529/work/caffe2/serialize/inline_container.cc:132, please report a bug to PyTorch. Attempted to read a PyTorch file with version 4, but the maximum supported version for reading is 2. Your PyTorch installation may be too old. (init at /opt/conda/conda-bld/pytorch_1579022034529/work/caffe2/serialize/inline_container.cc:132)\nframe #0: c10::Error::Error(c10::SourceLocation, std::string const&amp;) + 0x47 (0x7fe51bc67627 in /opt/conda/lib/python3.6/site-packages/torch/lib/libc10.so)\nframe #1: caffe2::serialize::PyTorchStreamReader::init() + 0x1f5b (0x7fe524fa19ab in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch.so)\nframe #2: caffe2::serialize::PyTorchStreamReader::PyTorchStreamReader(std::string const&amp;) + 0x64 (0x7fe524fa2bc4 in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch.so)\n......*</strong></p>\n\n<p>I used this approach from the first start of the competition and it works fine, unfortunately starting from yesterday i am keeping getting this error.</p>\n\n<p>I can see @tanulsingh077  and @rftexas reported same behavior in tweet-sentiment-extraction competition. </p>",
      "rawMarkdown": "I am facing the following behavior : I trained a pytorch model on TPU and then make inference on GPU but I am having the following error message : \n\n***RuntimeError: version_ &lt;= kMaxSupportedFileFormatVersion INTERNAL ASSERT FAILED at /opt/conda/conda-bld/pytorch_1579022034529/work/caffe2/serialize/inline_container.cc:132, please report a bug to PyTorch. Attempted to read a PyTorch file with version 4, but the maximum supported version for reading is 2. Your PyTorch installation may be too old. (init at /opt/conda/conda-bld/pytorch_1579022034529/work/caffe2/serialize/inline_container.cc:132)\nframe #0: c10::Error::Error(c10::SourceLocation, std::string const&amp;) + 0x47 (0x7fe51bc67627 in /opt/conda/lib/python3.6/site-packages/torch/lib/libc10.so)\nframe #1: caffe2::serialize::PyTorchStreamReader::init() + 0x1f5b (0x7fe524fa19ab in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch.so)\nframe #2: caffe2::serialize::PyTorchStreamReader::PyTorchStreamReader(std::string const&amp;) + 0x64 (0x7fe524fa2bc4 in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch.so)\n......***\n\nI used this approach from the first start of the competition and it works fine, unfortunately starting from yesterday i am keeping getting this error.\n\nI can see @tanulsingh077  and @rftexas reported same behavior in tweet-sentiment-extraction competition.",
      "votes": null
    },
    {
      "id": "887166",
      "postDate": "06/15/2020 14:12:19",
      "content": "<p>same error 😨 </p>",
      "rawMarkdown": "same error 😨",
      "votes": null
    },
    {
      "id": "890402",
      "postDate": "06/17/2020 13:44:43",
      "content": "<p>Hi! Thanks for your report. We are working on this - there are currently a couple of workarounds. One is shared here:</p>\n\n<p><a href=\"https://www.kaggle.com/c/tweet-sentiment-extraction/discussion/157695#886125\">https://www.kaggle.com/c/tweet-sentiment-extraction/discussion/157695#886125</a></p>",
      "rawMarkdown": "Hi! Thanks for your report. We are working on this - there are currently a couple of workarounds. One is shared here:\n\nhttps://www.kaggle.com/c/tweet-sentiment-extraction/discussion/157695#886125",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 887166,
      "author_name": "nxhong93",
      "author_url": "",
      "post_date": "06/15/2020 14:12:19",
      "content": "<p>same error 😨 </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 890402,
      "author_name": "philculliton",
      "author_url": "",
      "post_date": "06/17/2020 13:44:43",
      "content": "<p>Hi! Thanks for your report. We are working on this - there are currently a couple of workarounds. One is shared here:</p>\n\n<p><a href=\"https://www.kaggle.com/c/tweet-sentiment-extraction/discussion/157695#886125\">https://www.kaggle.com/c/tweet-sentiment-extraction/discussion/157695#886125</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "883201": "I am facing the following behavior : I trained a pytorch model on TPU and then make inference on GPU but I am having the following error message : \n\n***RuntimeError: version_ &lt;= kMaxSupportedFileFormatVersion INTERNAL ASSERT FAILED at /opt/conda/conda-bld/pytorch_1579022034529/work/caffe2/serialize/inline_container.cc:132, please report a bug to PyTorch. Attempted to read a PyTorch file with version 4, but the maximum supported version for reading is 2. Your PyTorch installation may be too old. (init at /opt/conda/conda-bld/pytorch_1579022034529/work/caffe2/serialize/inline_container.cc:132)\nframe #0: c10::Error::Error(c10::SourceLocation, std::string const&amp;) + 0x47 (0x7fe51bc67627 in /opt/conda/lib/python3.6/site-packages/torch/lib/libc10.so)\nframe #1: caffe2::serialize::PyTorchStreamReader::init() + 0x1f5b (0x7fe524fa19ab in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch.so)\nframe #2: caffe2::serialize::PyTorchStreamReader::PyTorchStreamReader(std::string const&amp;) + 0x64 (0x7fe524fa2bc4 in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch.so)\n......***\n\nI used this approach from the first start of the competition and it works fine, unfortunately starting from yesterday i am keeping getting this error.\n\nI can see @tanulsingh077  and @rftexas reported same behavior in tweet-sentiment-extraction competition.",
    "887166": "same error 😨",
    "890402": "Hi! Thanks for your report. We are working on this - there are currently a couple of workarounds. One is shared here:\n\nhttps://www.kaggle.com/c/tweet-sentiment-extraction/discussion/157695#886125"
  },
  "source": "meta"
}