{
  "id": 118265,
  "title": "Use the  tensorboard-2.0 in kaggle .it show 404:not Found ,why？",
  "url": "/competitions/tensorflow2-question-answering/discussion/118265",
  "author_name": "",
  "post_date": "2019-11-20T13:04:00.084031600Z",
  "votes": 3,
  "comment_count": 1,
  "views": 0,
  "content": "",
  "messages": [
    {
      "id": "677653",
      "postDate": "11/20/2019 13:04:00",
      "content": "",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "681931",
      "postDate": "11/26/2019 17:39:06",
      "content": "<p>Same problem here:</p>\n\n<ul>\n<li>Switching between CPU/GPU instance does not help</li>\n<li>Using lastest docker image does not help</li>\n<li>Testing with the demo kernel from <a href=\"https://www.kaggle.com/product-feedback/89671#latest-566465\">https://www.kaggle.com/product-feedback/89671#latest-566465</a> works but that's an image using Tensorflow 1.13 (and you cannot easily upgrade to 2.0.0 to test with this image)</li>\n<li>In the TF2.0 QA kernel, trying to force reinstall Tensorflow 2.0 (with <code>!pip install --upgrade --ignore-installed tensorflow==2.0.0</code>) also produce some dependencies errors with allennlp, ... however tensorflow and tensorboard installation looks to proceed ok ...  but this totally breaks the extension <code>%load_ext tensorboard.notebook</code> : you get a stacktrace instead of the jupyter 404 error...</li>\n</ul>\n\n<p>I guess that there might be an integration issue between the docker image used in this kernel and Tensorflow 2.0 / Tensorboard 2.0.x ?</p>",
      "rawMarkdown": "Same problem here:\n\n- Switching between CPU/GPU instance does not help\n- Using lastest docker image does not help\n- Testing with the demo kernel from https://www.kaggle.com/product-feedback/89671#latest-566465 works but that's an image using Tensorflow 1.13 (and you cannot easily upgrade to 2.0.0 to test with this image)\n- In the TF2.0 QA kernel, trying to force reinstall Tensorflow 2.0 (with `!pip install --upgrade --ignore-installed tensorflow==2.0.0`) also produce some dependencies errors with allennlp, ... however tensorflow and tensorboard installation looks to proceed ok ...  but this totally breaks the extension `%load_ext tensorboard.notebook` : you get a stacktrace instead of the jupyter 404 error...\n\nI guess that there might be an integration issue between the docker image used in this kernel and Tensorflow 2.0 / Tensorboard 2.0.x ?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 681931,
      "author_name": "fdasilva",
      "author_url": "",
      "post_date": "11/26/2019 17:39:06",
      "content": "<p>Same problem here:</p>\n\n<ul>\n<li>Switching between CPU/GPU instance does not help</li>\n<li>Using lastest docker image does not help</li>\n<li>Testing with the demo kernel from <a href=\"https://www.kaggle.com/product-feedback/89671#latest-566465\">https://www.kaggle.com/product-feedback/89671#latest-566465</a> works but that's an image using Tensorflow 1.13 (and you cannot easily upgrade to 2.0.0 to test with this image)</li>\n<li>In the TF2.0 QA kernel, trying to force reinstall Tensorflow 2.0 (with <code>!pip install --upgrade --ignore-installed tensorflow==2.0.0</code>) also produce some dependencies errors with allennlp, ... however tensorflow and tensorboard installation looks to proceed ok ...  but this totally breaks the extension <code>%load_ext tensorboard.notebook</code> : you get a stacktrace instead of the jupyter 404 error...</li>\n</ul>\n\n<p>I guess that there might be an integration issue between the docker image used in this kernel and Tensorflow 2.0 / Tensorboard 2.0.x ?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "677653": "",
    "681931": "Same problem here:\n\n- Switching between CPU/GPU instance does not help\n- Using lastest docker image does not help\n- Testing with the demo kernel from https://www.kaggle.com/product-feedback/89671#latest-566465 works but that's an image using Tensorflow 1.13 (and you cannot easily upgrade to 2.0.0 to test with this image)\n- In the TF2.0 QA kernel, trying to force reinstall Tensorflow 2.0 (with `!pip install --upgrade --ignore-installed tensorflow==2.0.0`) also produce some dependencies errors with allennlp, ... however tensorflow and tensorboard installation looks to proceed ok ...  but this totally breaks the extension `%load_ext tensorboard.notebook` : you get a stacktrace instead of the jupyter 404 error...\n\nI guess that there might be an integration issue between the docker image used in this kernel and Tensorflow 2.0 / Tensorboard 2.0.x ?"
  },
  "source": "meta"
}