{
  "id": 420081,
  "title": "Question regarding on the test set during submission",
  "url": "/competitions/hubmap-hacking-the-human-vasculature/discussion/420081",
  "author_name": "",
  "post_date": "2023-06-29T05:51:43.222035600Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I am new to the Kaggle code competition and this is the first time that participates in a code competition. During my submission, I am facing an issue where there is only 1 image in the test set. I am not quite sure which part is done wrong in my case, and I would appreciate it if someone can give me some forms of tips/info if I have done it correctly.</p>\n<p>In my steps:</p>\n<ol>\n<li>Disable Internet connection.</li>\n<li>Have all the Python packages offline and install them at the initial stage of the submission.</li>\n<li>Run the inference code and include a try exception block to bypass if there is any error using my trained model (I am using a YOLOv8).</li>\n<li>In the logging of the output for submission, there is only 1 image in the test set and the notebook inference was completed with only 1 image.</li>\n</ol>\n<p>My code is as below:</p>\n<pre><code>image_dir = \nimage_id = (f)\nimage_id =   f  image_id]\n\n</code></pre>\n<p>Output is as below:</p>\n<pre><code>dir: /kaggle/input/hubmap-hacking-the-human-vasculature/test, Number of test samples in test set: 1\n</code></pre>\n<p>I have been browsing through some of the inference codes in the Code section, however, I do not find any differences between them. Appreciated it if anyone could give me some forms of tips here.</p>",
  "messages": [
    {
      "id": "2322162",
      "postDate": "06/29/2023 05:51:43",
      "content": "<p>I am new to the Kaggle code competition and this is the first time that participates in a code competition. During my submission, I am facing an issue where there is only 1 image in the test set. I am not quite sure which part is done wrong in my case, and I would appreciate it if someone can give me some forms of tips/info if I have done it correctly.</p>\n<p>In my steps:</p>\n<ol>\n<li>Disable Internet connection.</li>\n<li>Have all the Python packages offline and install them at the initial stage of the submission.</li>\n<li>Run the inference code and include a try exception block to bypass if there is any error using my trained model (I am using a YOLOv8).</li>\n<li>In the logging of the output for submission, there is only 1 image in the test set and the notebook inference was completed with only 1 image.</li>\n</ol>\n<p>My code is as below:</p>\n<pre><code>image_dir = \nimage_id = (f)\nimage_id =   f  image_id]\n\n</code></pre>\n<p>Output is as below:</p>\n<pre><code>dir: /kaggle/input/hubmap-hacking-the-human-vasculature/test, Number of test samples in test set: 1\n</code></pre>\n<p>I have been browsing through some of the inference codes in the Code section, however, I do not find any differences between them. Appreciated it if anyone could give me some forms of tips here.</p>",
      "rawMarkdown": "I am new to the Kaggle code competition and this is the first time that participates in a code competition. During my submission, I am facing an issue where there is only 1 image in the test set. I am not quite sure which part is done wrong in my case, and I would appreciate it if someone can give me some forms of tips/info if I have done it correctly.\n\nIn my steps:\n1. Disable Internet connection.\n2. Have all the Python packages offline and install them at the initial stage of the submission.\n3. Run the inference code and include a try exception block to bypass if there is any error using my trained model (I am using a YOLOv8).\n4. In the logging of the output for submission, there is only 1 image in the test set and the notebook inference was completed with only 1 image.\n\nMy code is as below:\n```\nimage_dir = '/kaggle/input/hubmap-hacking-the-human-vasculature/test'\nimage_id = glob(f'{image_dir}/*.tif')\nimage_id = [f.split('/')[-1][:-4] for f in image_id]\nprint(f\"Test dir: {image_dir}, Number of test samples in test set: {len(image_id)}\" )\n```\nOutput is as below:\n```\nTest dir: /kaggle/input/hubmap-hacking-the-human-vasculature/test, Number of test samples in test set: 1\n```\nI have been browsing through some of the inference codes in the Code section, however, I do not find any differences between them. Appreciated it if anyone could give me some forms of tips here.",
      "votes": null
    },
    {
      "id": "2322359",
      "postDate": "06/29/2023 08:27:06",
      "content": "<p>Hi. You may find the answer to your question here.</p>\n<p><a href=\"https://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/discussion/418368\" target=\"_blank\">https://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/discussion/418368</a></p>",
      "rawMarkdown": "Hi. You may find the answer to your question here.\n\nhttps://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/discussion/418368",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2322359,
      "author_name": "mitsuyasuhoshino",
      "author_url": "",
      "post_date": "06/29/2023 08:27:06",
      "content": "<p>Hi. You may find the answer to your question here.</p>\n<p><a href=\"https://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/discussion/418368\" target=\"_blank\">https://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/discussion/418368</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2322162": "I am new to the Kaggle code competition and this is the first time that participates in a code competition. During my submission, I am facing an issue where there is only 1 image in the test set. I am not quite sure which part is done wrong in my case, and I would appreciate it if someone can give me some forms of tips/info if I have done it correctly.\n\nIn my steps:\n1. Disable Internet connection.\n2. Have all the Python packages offline and install them at the initial stage of the submission.\n3. Run the inference code and include a try exception block to bypass if there is any error using my trained model (I am using a YOLOv8).\n4. In the logging of the output for submission, there is only 1 image in the test set and the notebook inference was completed with only 1 image.\n\nMy code is as below:\n```\nimage_dir = '/kaggle/input/hubmap-hacking-the-human-vasculature/test'\nimage_id = glob(f'{image_dir}/*.tif')\nimage_id = [f.split('/')[-1][:-4] for f in image_id]\nprint(f\"Test dir: {image_dir}, Number of test samples in test set: {len(image_id)}\" )\n```\nOutput is as below:\n```\nTest dir: /kaggle/input/hubmap-hacking-the-human-vasculature/test, Number of test samples in test set: 1\n```\nI have been browsing through some of the inference codes in the Code section, however, I do not find any differences between them. Appreciated it if anyone could give me some forms of tips here.",
    "2322359": "Hi. You may find the answer to your question here.\n\nhttps://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/discussion/418368"
  },
  "source": "meta"
}