{
  "id": 209959,
  "title": "Need help in TPU parallel training: Pytorch ",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/209959",
  "author_name": "",
  "post_date": "2021-01-09T07:53:02.855807400Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I hope someone can help, as I am stuck for a long time on this problem. </p>\n<p>I am trying to do KFold parallel training using multiple TPUs available. I am using the joblib library to do parallel training. However, I see that training happens on one core at a time. Here is the notebook.</p>\n<p>I am unable to find why training does not happen in multiple cores parallelly.</p>\n<p>Here is my notebook.<br>\n<a href=\"https://www.kaggle.com/krisho007/kfold-parallel-pytorch-lightning\" target=\"_blank\">https://www.kaggle.com/krisho007/kfold-parallel-pytorch-lightning</a> </p>\n<p>Any comment is useful. </p>",
  "messages": [
    {
      "id": "1145523",
      "postDate": "01/09/2021 07:53:02",
      "content": "<p>I hope someone can help, as I am stuck for a long time on this problem. </p>\n<p>I am trying to do KFold parallel training using multiple TPUs available. I am using the joblib library to do parallel training. However, I see that training happens on one core at a time. Here is the notebook.</p>\n<p>I am unable to find why training does not happen in multiple cores parallelly.</p>\n<p>Here is my notebook.<br>\n<a href=\"https://www.kaggle.com/krisho007/kfold-parallel-pytorch-lightning\" target=\"_blank\">https://www.kaggle.com/krisho007/kfold-parallel-pytorch-lightning</a> </p>\n<p>Any comment is useful. </p>",
      "rawMarkdown": "I hope someone can help, as I am stuck for a long time on this problem. \n\nI am trying to do KFold parallel training using multiple TPUs available. I am using the joblib library to do parallel training. However, I see that training happens on one core at a time. Here is the notebook.\n\nI am unable to find why training does not happen in multiple cores parallelly.\n\nHere is my notebook.\nhttps://www.kaggle.com/krisho007/kfold-parallel-pytorch-lightning \n\nAny comment is useful.",
      "votes": null
    },
    {
      "id": "1145573",
      "postDate": "01/09/2021 08:34:58",
      "content": "<p>in this notebook <a href=\"https://www.kaggle.com/mobassir/vit-pytorch-xla-tpu-for-leaf-disease\" target=\"_blank\">ViT - Pytorch xla (TPU) for leaf disease</a> and in this notebook <a href=\"https://www.kaggle.com/mobassir/faster-torch-tpu-baseline-for-leaf-disease\" target=\"_blank\">Faster torch tpu baseline for leaf disease</a> i tried to explain step by step procedure to train models on torch tpu 8 cores using this competition data,if you follow them carefully i believe you will be able to solve your issue,thank you</p>",
      "rawMarkdown": "in this notebook [ViT - Pytorch xla (TPU) for leaf disease](https://www.kaggle.com/mobassir/vit-pytorch-xla-tpu-for-leaf-disease) and in this notebook [Faster torch tpu baseline for leaf disease](https://www.kaggle.com/mobassir/faster-torch-tpu-baseline-for-leaf-disease) i tried to explain step by step procedure to train models on torch tpu 8 cores using this competition data,if you follow them carefully i believe you will be able to solve your issue,thank you",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1145573,
      "author_name": "mobassir",
      "author_url": "",
      "post_date": "01/09/2021 08:34:58",
      "content": "<p>in this notebook <a href=\"https://www.kaggle.com/mobassir/vit-pytorch-xla-tpu-for-leaf-disease\" target=\"_blank\">ViT - Pytorch xla (TPU) for leaf disease</a> and in this notebook <a href=\"https://www.kaggle.com/mobassir/faster-torch-tpu-baseline-for-leaf-disease\" target=\"_blank\">Faster torch tpu baseline for leaf disease</a> i tried to explain step by step procedure to train models on torch tpu 8 cores using this competition data,if you follow them carefully i believe you will be able to solve your issue,thank you</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1145523": "I hope someone can help, as I am stuck for a long time on this problem. \n\nI am trying to do KFold parallel training using multiple TPUs available. I am using the joblib library to do parallel training. However, I see that training happens on one core at a time. Here is the notebook.\n\nI am unable to find why training does not happen in multiple cores parallelly.\n\nHere is my notebook.\nhttps://www.kaggle.com/krisho007/kfold-parallel-pytorch-lightning \n\nAny comment is useful.",
    "1145573": "in this notebook [ViT - Pytorch xla (TPU) for leaf disease](https://www.kaggle.com/mobassir/vit-pytorch-xla-tpu-for-leaf-disease) and in this notebook [Faster torch tpu baseline for leaf disease](https://www.kaggle.com/mobassir/faster-torch-tpu-baseline-for-leaf-disease) i tried to explain step by step procedure to train models on torch tpu 8 cores using this competition data,if you follow them carefully i believe you will be able to solve your issue,thank you"
  },
  "source": "meta"
}