{
  "id": 329685,
  "title": "methods to load test data set",
  "url": "/competitions/amex-default-prediction/discussion/329685",
  "author_name": "",
  "post_date": "2022-06-08T05:03:38.326771Z",
  "votes": 9,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I mainly use the kaggle notebook for this competition. I am seeking efficient methods to load the test set. So far, I am doing:</p>\n<ol>\n<li>cut the test set into 5 subsets in a notebook and then load them into the model notebook</li>\n<li>use pd.read_csv(path, num_rows = test.shape[0]//n_fold)</li>\n</ol>",
  "messages": [
    {
      "id": "1814602",
      "postDate": "06/08/2022 05:03:38",
      "content": "<p>I mainly use the kaggle notebook for this competition. I am seeking efficient methods to load the test set. So far, I am doing:</p>\n<ol>\n<li>cut the test set into 5 subsets in a notebook and then load them into the model notebook</li>\n<li>use pd.read_csv(path, num_rows = test.shape[0]//n_fold)</li>\n</ol>",
      "rawMarkdown": "I mainly use the kaggle notebook for this competition. I am seeking efficient methods to load the test set. So far, I am doing:\n1. cut the test set into 5 subsets in a notebook and then load them into the model notebook\n2. use pd.read_csv(path, num_rows = test.shape[0]//n_fold)",
      "votes": null
    },
    {
      "id": "1815704",
      "postDate": "06/09/2022 10:36:14",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/konohayui\" target=\"_blank\">@konohayui</a> . To handle the test data I am using the dask_cudf </p>",
      "rawMarkdown": "Hi @konohayui . To handle the test data I am using the dask_cudf",
      "votes": null
    },
    {
      "id": "1815718",
      "postDate": "06/09/2022 11:03:24",
      "content": "<p>Have you tried decreasing the size of the dataset? This notebook explains something like that.. <a href=\"https://www.kaggle.com/competitions/amex-default-prediction/discussion/328054\" target=\"_blank\">https://www.kaggle.com/competitions/amex-default-prediction/discussion/328054</a></p>",
      "rawMarkdown": "Have you tried decreasing the size of the dataset? This notebook explains something like that.. https://www.kaggle.com/competitions/amex-default-prediction/discussion/328054",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1815704,
      "author_name": "paulojunqueira",
      "author_url": "",
      "post_date": "06/09/2022 10:36:14",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/konohayui\" target=\"_blank\">@konohayui</a> . To handle the test data I am using the dask_cudf </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1815718,
      "author_name": "leodaniel",
      "author_url": "",
      "post_date": "06/09/2022 11:03:24",
      "content": "<p>Have you tried decreasing the size of the dataset? This notebook explains something like that.. <a href=\"https://www.kaggle.com/competitions/amex-default-prediction/discussion/328054\" target=\"_blank\">https://www.kaggle.com/competitions/amex-default-prediction/discussion/328054</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1814602": "I mainly use the kaggle notebook for this competition. I am seeking efficient methods to load the test set. So far, I am doing:\n1. cut the test set into 5 subsets in a notebook and then load them into the model notebook\n2. use pd.read_csv(path, num_rows = test.shape[0]//n_fold)",
    "1815704": "Hi @konohayui . To handle the test data I am using the dask_cudf",
    "1815718": "Have you tried decreasing the size of the dataset? This notebook explains something like that.. https://www.kaggle.com/competitions/amex-default-prediction/discussion/328054"
  },
  "source": "meta"
}