{
  "id": 203439,
  "title": "TPU without internet",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/203439",
  "author_name": "",
  "post_date": "2020-12-15T09:58:30.139370700Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>How to save model trained on TPU for load model to run  without internet for competition<br>\nFor your information, I am a beginner</p>",
  "messages": [
    {
      "id": "1113223",
      "postDate": "12/15/2020 09:58:30",
      "content": "<p>How to save model trained on TPU for load model to run  without internet for competition<br>\nFor your information, I am a beginner</p>",
      "rawMarkdown": "How to save model trained on TPU for load model to run  without internet for competition\nFor your information, I am a beginner",
      "votes": null
    },
    {
      "id": "1113478",
      "postDate": "12/15/2020 13:28:54",
      "content": "<p>Make a new dataset from the output TPU model.</p>",
      "rawMarkdown": "Make a new dataset from the output TPU model.",
      "votes": null
    },
    {
      "id": "1116552",
      "postDate": "12/17/2020 09:07:10",
      "content": "<p>If you are using TF--&gt;<br>\nmodel.save_weights(model_path), you can use it.<br>\nfor more information, you can take a look at --&gt;<a href=\"https://www.kaggle.com/dimitreoliveira/cassava-leaf-disease-tpu-tensorflow-training\" target=\"_blank\">https://www.kaggle.com/dimitreoliveira/cassava-leaf-disease-tpu-tensorflow-training</a><br>\nits inference part is also given.</p>",
      "rawMarkdown": "If you are using TF-->\nmodel.save_weights(model_path), you can use it.\nfor more information, you can take a look at -->https://www.kaggle.com/dimitreoliveira/cassava-leaf-disease-tpu-tensorflow-training\nits inference part is also given.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1113478,
      "author_name": "aeryss",
      "author_url": "",
      "post_date": "12/15/2020 13:28:54",
      "content": "<p>Make a new dataset from the output TPU model.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1116552,
      "author_name": "shivam17818",
      "author_url": "",
      "post_date": "12/17/2020 09:07:10",
      "content": "<p>If you are using TF--&gt;<br>\nmodel.save_weights(model_path), you can use it.<br>\nfor more information, you can take a look at --&gt;<a href=\"https://www.kaggle.com/dimitreoliveira/cassava-leaf-disease-tpu-tensorflow-training\" target=\"_blank\">https://www.kaggle.com/dimitreoliveira/cassava-leaf-disease-tpu-tensorflow-training</a><br>\nits inference part is also given.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1113223": "How to save model trained on TPU for load model to run  without internet for competition\nFor your information, I am a beginner",
    "1113478": "Make a new dataset from the output TPU model.",
    "1116552": "If you are using TF-->\nmodel.save_weights(model_path), you can use it.\nfor more information, you can take a look at -->https://www.kaggle.com/dimitreoliveira/cassava-leaf-disease-tpu-tensorflow-training\nits inference part is also given."
  },
  "source": "meta"
}