{
  "id": 430023,
  "title": "Using Pretrained Models",
  "url": "/competitions/rsna-2023-abdominal-trauma-detection/discussion/430023",
  "author_name": "Anurag Kumar",
  "post_date": "2023-08-08T02:40:58.792000",
  "votes": 4,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hello! This is my first real competition on Kaggle so I wanted to start off by using transfer learning with a pre-trained model. I noticed that in the code requirements, it says that you can use a pre-trained model but you cannot connect to the internet. I tried using ResNet50, but that requires loading it from the internet, something that is not allowed in the competition rules. Am I missing something?</p>",
  "messages": [
    {
      "id": 2379127,
      "postDate": "2023-08-08T02:40:58.793Z",
      "content": "<p>Hello! This is my first real competition on Kaggle so I wanted to start off by using transfer learning with a pre-trained model. I noticed that in the code requirements, it says that you can use a pre-trained model but you cannot connect to the internet. I tried using ResNet50, but that requires loading it from the internet, something that is not allowed in the competition rules. Am I missing something?</p>",
      "rawMarkdown": "Hello! This is my first real competition on Kaggle so I wanted to start off by using transfer learning with a pre-trained model. I noticed that in the code requirements, it says that you can use a pre-trained model but you cannot connect to the internet. I tried using ResNet50, but that requires loading it from the internet, something that is not allowed in the competition rules. Am I missing something?",
      "votes": 4
    },
    {
      "id": 2380194,
      "postDate": "2023-08-08T13:13:38.987Z",
      "content": "<p>While you can't have the internet available you can save all sorts of libraries, etc in a Data Set and load that data set into your notebook.</p>\n<p>Lots of examples of how this is done on shared notebooks in this competition and many others in the past.</p>\n<p>The notebook I am currently playing with shows loading example in the \"Install Libaries\"<br>\n<a href=\"https://www.kaggle.com/code/awsaf49/rsna-atd-cnn-tpu-infer\" target=\"_blank\">RSNA-ATD: CNN [TPU][Infer]</a></p>",
      "rawMarkdown": "While you can't have the internet available you can save all sorts of libraries, etc in a Data Set and load that data set into your notebook.\n\nLots of examples of how this is done on shared notebooks in this competition and many others in the past.\n\nThe notebook I am currently playing with shows loading example in the \"Install Libaries\"\n[RSNA-ATD: CNN [TPU][Infer]](https://www.kaggle.com/code/awsaf49/rsna-atd-cnn-tpu-infer)",
      "votes": 2
    }
  ],
  "comments": [
    {
      "id": 2380194,
      "author_name": "PC Jimmmy",
      "author_url": "",
      "post_date": "2023-08-08T13:13:38.987000",
      "content": "<p>While you can't have the internet available you can save all sorts of libraries, etc in a Data Set and load that data set into your notebook.</p>\n<p>Lots of examples of how this is done on shared notebooks in this competition and many others in the past.</p>\n<p>The notebook I am currently playing with shows loading example in the \"Install Libaries\"<br>\n<a href=\"https://www.kaggle.com/code/awsaf49/rsna-atd-cnn-tpu-infer\" target=\"_blank\">RSNA-ATD: CNN [TPU][Infer]</a></p>",
      "votes": 2,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2379127": "Hello! This is my first real competition on Kaggle so I wanted to start off by using transfer learning with a pre-trained model. I noticed that in the code requirements, it says that you can use a pre-trained model but you cannot connect to the internet. I tried using ResNet50, but that requires loading it from the internet, something that is not allowed in the competition rules. Am I missing something?",
    "2380194": "While you can't have the internet available you can save all sorts of libraries, etc in a Data Set and load that data set into your notebook.\n\nLots of examples of how this is done on shared notebooks in this competition and many others in the past.\n\nThe notebook I am currently playing with shows loading example in the \"Install Libaries\"\n[RSNA-ATD: CNN [TPU][Infer]](https://www.kaggle.com/code/awsaf49/rsna-atd-cnn-tpu-infer)"
  }
}