{
  "id": 237354,
  "title": "Submission Scoring Error",
  "url": "/competitions/plant-pathology-2021-fgvc8/discussion/237354",
  "author_name": "",
  "post_date": "2021-05-08T10:34:17.236873200Z",
  "votes": 2,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I have worked on problem in my local and upload there my model and .csv file. After that, I put them in the output folder in the notebook and submitted the notebook. It gives me a \"Submission Scoring Error\", anyone has any idea?</p>",
  "messages": [
    {
      "id": "1297803",
      "postDate": "05/08/2021 10:34:17",
      "content": "<p>I have worked on problem in my local and upload there my model and .csv file. After that, I put them in the output folder in the notebook and submitted the notebook. It gives me a \"Submission Scoring Error\", anyone has any idea?</p>",
      "rawMarkdown": "I have worked on problem in my local and upload there my model and .csv file. After that, I put them in the output folder in the notebook and submitted the notebook. It gives me a \"Submission Scoring Error\", anyone has any idea?",
      "votes": null
    },
    {
      "id": "1297966",
      "postDate": "05/08/2021 13:14:09",
      "content": "<p>This means that your submission file has an incorrect format or it's corrupted.</p>",
      "rawMarkdown": "This means that your submission file has an incorrect format or it's corrupted.",
      "votes": null
    },
    {
      "id": "1298079",
      "postDate": "05/08/2021 14:38:06",
      "content": "<p>You have to submit an inference notebook in which your model will predict all the images on the test_images path and write results for all images into submission.csv file. If your notebook contains only the locally created csv file then it gives error since it does not include predictions for the private test set. You can do the following to submit predictions:</p>\n<p>Upload your locally trained model as a dataset, add that dataset into your notebook. Use the model in the dataset to predict all the images in the test_images folder and create the csv file as submission.csv. </p>",
      "rawMarkdown": "You have to submit an inference notebook in which your model will predict all the images on the test_images path and write results for all images into submission.csv file. If your notebook contains only the locally created csv file then it gives error since it does not include predictions for the private test set. You can do the following to submit predictions:\n\nUpload your locally trained model as a dataset, add that dataset into your notebook. Use the model in the dataset to predict all the images in the test_images folder and create the csv file as submission.csv.",
      "votes": null
    },
    {
      "id": "1322727",
      "postDate": "05/25/2021 16:13:51",
      "content": "<p>Hi, even I got the same error when I directly read the filenames from test directory to the dataframe (when we do this, the files order are getting changed). So I tried sorting the filenames using below code, then everything worked fine.<br>\nimport os<br>\nfiles = os.listdir(test_dir)<br>\nsorted_files = sorted(files)<br>\nsorted_files</p>",
      "rawMarkdown": "Hi, even I got the same error when I directly read the filenames from test directory to the dataframe (when we do this, the files order are getting changed). So I tried sorting the filenames using below code, then everything worked fine.\nimport os\nfiles = os.listdir(test_dir)\nsorted_files = sorted(files)\nsorted_files",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1297966,
      "author_name": "atamazian",
      "author_url": "",
      "post_date": "05/08/2021 13:14:09",
      "content": "<p>This means that your submission file has an incorrect format or it's corrupted.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1298079,
      "author_name": "burakbekci",
      "author_url": "",
      "post_date": "05/08/2021 14:38:06",
      "content": "<p>You have to submit an inference notebook in which your model will predict all the images on the test_images path and write results for all images into submission.csv file. If your notebook contains only the locally created csv file then it gives error since it does not include predictions for the private test set. You can do the following to submit predictions:</p>\n<p>Upload your locally trained model as a dataset, add that dataset into your notebook. Use the model in the dataset to predict all the images in the test_images folder and create the csv file as submission.csv. </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1322727,
      "author_name": "srikanth46v",
      "author_url": "",
      "post_date": "05/25/2021 16:13:51",
      "content": "<p>Hi, even I got the same error when I directly read the filenames from test directory to the dataframe (when we do this, the files order are getting changed). So I tried sorting the filenames using below code, then everything worked fine.<br>\nimport os<br>\nfiles = os.listdir(test_dir)<br>\nsorted_files = sorted(files)<br>\nsorted_files</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1297803": "I have worked on problem in my local and upload there my model and .csv file. After that, I put them in the output folder in the notebook and submitted the notebook. It gives me a \"Submission Scoring Error\", anyone has any idea?",
    "1297966": "This means that your submission file has an incorrect format or it's corrupted.",
    "1298079": "You have to submit an inference notebook in which your model will predict all the images on the test_images path and write results for all images into submission.csv file. If your notebook contains only the locally created csv file then it gives error since it does not include predictions for the private test set. You can do the following to submit predictions:\n\nUpload your locally trained model as a dataset, add that dataset into your notebook. Use the model in the dataset to predict all the images in the test_images folder and create the csv file as submission.csv.",
    "1322727": "Hi, even I got the same error when I directly read the filenames from test directory to the dataframe (when we do this, the files order are getting changed). So I tried sorting the filenames using below code, then everything worked fine.\nimport os\nfiles = os.listdir(test_dir)\nsorted_files = sorted(files)\nsorted_files"
  },
  "source": "meta"
}