{
  "id": 211311,
  "title": "Are you facing problem in submission? output file not found?",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/211311",
  "author_name": "",
  "post_date": "2021-01-14T15:40:09.023187200Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>To do submission in this competition you need to save your trained model and then use the same model to predict the test data.<br>\nI will show the way in Pytorch.<br>\n1: Training notebook:<br>\nIn this notebook, your real model will be present and after training the model you need to save the model.</p>\n<h1>Specify a path</h1>\n<p>PATH = './model_save.pth'  #model_save is name</p>\n<h1>Save</h1>\n<p>torch.save(model.state_dict(), PATH)#saving the state_dict instead of whole model</p>\n<h1>download the model_save.pth file from the <em>/kaggle/working</em> directory.</h1>\n<p>2: Create another notebook for submission.<br>\nClick on the Add data and upload the model_save.pth file.<br>\nNow you are ready to use the model for prediction and submission.<br>\ncreate a dummy of your model.<br>\nmodel = MODEL() # It's like a dummy model where we will update the parameters from model_save.pth file<br>\nmodel.load_state_dict(torch.load(PATH))</p>\n<p>now you are ready to go!!</p>\n<p>Feel free to post any comment </p>",
  "messages": [
    {
      "id": "1153043",
      "postDate": "01/14/2021 15:40:09",
      "content": "<p>To do submission in this competition you need to save your trained model and then use the same model to predict the test data.<br>\nI will show the way in Pytorch.<br>\n1: Training notebook:<br>\nIn this notebook, your real model will be present and after training the model you need to save the model.</p>\n<h1>Specify a path</h1>\n<p>PATH = './model_save.pth'  #model_save is name</p>\n<h1>Save</h1>\n<p>torch.save(model.state_dict(), PATH)#saving the state_dict instead of whole model</p>\n<h1>download the model_save.pth file from the <em>/kaggle/working</em> directory.</h1>\n<p>2: Create another notebook for submission.<br>\nClick on the Add data and upload the model_save.pth file.<br>\nNow you are ready to use the model for prediction and submission.<br>\ncreate a dummy of your model.<br>\nmodel = MODEL() # It's like a dummy model where we will update the parameters from model_save.pth file<br>\nmodel.load_state_dict(torch.load(PATH))</p>\n<p>now you are ready to go!!</p>\n<p>Feel free to post any comment </p>",
      "rawMarkdown": "To do submission in this competition you need to save your trained model and then use the same model to predict the test data.\nI will show the way in Pytorch.\n1: Training notebook:\nIn this notebook, your real model will be present and after training the model you need to save the model.\n# Specify a path\nPATH = './model_save.pth'  #model_save is name\n# Save\ntorch.save(model.state_dict(), PATH)#saving the state_dict instead of whole model\n#download the model_save.pth file from the */kaggle/working* directory.\n2: Create another notebook for submission.\nClick on the Add data and upload the model_save.pth file.\nNow you are ready to use the model for prediction and submission.\ncreate a dummy of your model.\nmodel = MODEL() # It's like a dummy model where we will update the parameters from model_save.pth file\nmodel.load_state_dict(torch.load(PATH))\n\nnow you are ready to go!!\n\nFeel free to post any comment",
      "votes": null
    },
    {
      "id": "1161227",
      "postDate": "01/20/2021 12:27:52",
      "content": "<p>can this method solve the problem that submission not found?</p>",
      "rawMarkdown": "can this method solve the problem that submission not found?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1161227,
      "author_name": "allenggle",
      "author_url": "",
      "post_date": "01/20/2021 12:27:52",
      "content": "<p>can this method solve the problem that submission not found?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1153043": "To do submission in this competition you need to save your trained model and then use the same model to predict the test data.\nI will show the way in Pytorch.\n1: Training notebook:\nIn this notebook, your real model will be present and after training the model you need to save the model.\n# Specify a path\nPATH = './model_save.pth'  #model_save is name\n# Save\ntorch.save(model.state_dict(), PATH)#saving the state_dict instead of whole model\n#download the model_save.pth file from the */kaggle/working* directory.\n2: Create another notebook for submission.\nClick on the Add data and upload the model_save.pth file.\nNow you are ready to use the model for prediction and submission.\ncreate a dummy of your model.\nmodel = MODEL() # It's like a dummy model where we will update the parameters from model_save.pth file\nmodel.load_state_dict(torch.load(PATH))\n\nnow you are ready to go!!\n\nFeel free to post any comment",
    "1161227": "can this method solve the problem that submission not found?"
  },
  "source": "meta"
}