{
  "id": 370680,
  "title": "How to add labels to hugging face dataset?",
  "url": "/competitions/rsna-breast-cancer-detection/discussion/370680",
  "author_name": "Ben Lebovitz",
  "post_date": "2022-12-05T21:14:13.959000",
  "votes": 2,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi, I was planning to use hugging face ViT for feature extraction. I'm really new to HF and I can't figure out how to add the training labels to the dataset. You can do it if it's in the same directory, but I can't figure out how to just point it to a CSV in another directory. (Google is no help, ChatGPT gives some hilariously wrong, but confident answers).</p>\n<p>let me know if this isn't possible, and I guess I'll switch to hacking through fast.ai. But it seems like this must be an easy thing to do that I just can't find the answer to…</p>\n<p>Maybe this whole idea is dumb though… my current thought to approach this contest is to simply do feature extraction, and load it into a SVM and GBDT models. Not sure if I'm on the right track at all here. Any advice is welcome!</p>",
  "messages": [
    {
      "id": 2056209,
      "postDate": "2022-12-05T21:14:13.960Z",
      "content": "<p>Hi, I was planning to use hugging face ViT for feature extraction. I'm really new to HF and I can't figure out how to add the training labels to the dataset. You can do it if it's in the same directory, but I can't figure out how to just point it to a CSV in another directory. (Google is no help, ChatGPT gives some hilariously wrong, but confident answers).</p>\n<p>let me know if this isn't possible, and I guess I'll switch to hacking through fast.ai. But it seems like this must be an easy thing to do that I just can't find the answer to…</p>\n<p>Maybe this whole idea is dumb though… my current thought to approach this contest is to simply do feature extraction, and load it into a SVM and GBDT models. Not sure if I'm on the right track at all here. Any advice is welcome!</p>",
      "rawMarkdown": "Hi, I was planning to use hugging face ViT for feature extraction. I'm really new to HF and I can't figure out how to add the training labels to the dataset. You can do it if it's in the same directory, but I can't figure out how to just point it to a CSV in another directory. (Google is no help, ChatGPT gives some hilariously wrong, but confident answers).\n\nlet me know if this isn't possible, and I guess I'll switch to hacking through fast.ai. But it seems like this must be an easy thing to do that I just can't find the answer to...\n\nMaybe this whole idea is dumb though... my current thought to approach this contest is to simply do feature extraction, and load it into a SVM and GBDT models. Not sure if I'm on the right track at all here. Any advice is welcome!",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2056209": "Hi, I was planning to use hugging face ViT for feature extraction. I'm really new to HF and I can't figure out how to add the training labels to the dataset. You can do it if it's in the same directory, but I can't figure out how to just point it to a CSV in another directory. (Google is no help, ChatGPT gives some hilariously wrong, but confident answers).\n\nlet me know if this isn't possible, and I guess I'll switch to hacking through fast.ai. But it seems like this must be an easy thing to do that I just can't find the answer to...\n\nMaybe this whole idea is dumb though... my current thought to approach this contest is to simply do feature extraction, and load it into a SVM and GBDT models. Not sure if I'm on the right track at all here. Any advice is welcome!"
  }
}