{
  "fetched_at": "2026-10-19T23:19:48Z",
  "source_url": "https://www.kaggle.com/competitions/11785-spring2021-rc0e-colab/data",
  "pages": [
    {
      "name": "data-description",
      "content": "## File Description \n* **confidential_intel.npy** - the results given by the magical machine. It's a 2-d array of shape (100000, 3). Each row corresponds to a person's info in the order of [P(A), P(B|A), P(B)]\n\n* **sample.csv** - a sample submission file in the correct format",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Notice: \n**This is only a toy example to help you get familiar with Kaggle+Colab. There is no Deep Learning concepts, and you should make sure you can pass it before tackling the real assignments.**\n\n## Does s/he like me?\n\nWhen you have a crush on someone, do you want to know how s/he feels? In this example, we will use **Bayes Theorem** to help 100,000 people who have this concern.\n\nOne of them, Bob, tells you that he has a crush on Alice. Even though Alice often smiles when she meets Bob, he still does not know whether Alice likes him back. He asks you whether he should propose to her when she smiles at him.\n\nYou happen to have a magical machine designed by Haoxuan Inc. that can tell\n\n* **P(A)**: Probability of Alice liking Bob\n* **P(B|A)**: Probability of Alice smiling at Bob given that she likes Bob\n* **P(B)**: Probability of Alice smiling at Bob\n\n(**A**: Alice likes Bob, **B**: Alice smiles at Bob)\n\nHowever, **the machine warns you that you shall never tell any of the data directly to Bob**. Hence, you decide to calculate the Posterior Probability *P(A|B)* (the probability of Alice liking Bob given that she smiles at him) and directly encourage him to propose if *P(A|B) > 80%*, and discourage him otherwise. You apply Bayes Theorem and told Bob the right thing to do, and he succeeded.\n\nAfter that, you become so famous that tons of people come to you for your advice. Since you are very busy working on IDL assignments, you hire IDL staff to pre-process the data for you. We have prepared all the data on Kaggle and now it's your time to download the data, process them, generate results and submit your predictions back to Kaggle!\n\n\nPlease follow the Colab Tutorial Notebook and the PDF that we have provided for more information.\n\nPortals:\n[Colab Tutorial Notebook](https://colab.research.google.com/drive/1J9myYPvoMu5JiPdoec5Zt4tteCx5dpti?usp=sharing)\n[Tutorial PDF](https://hxzhu9819.github.io/homepage/posts/11785_rc0e_colab)",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "You should get full marks (1.0000 in this case) if you implement everything correctly (since the problem is deterministic.) Your labels should either be 0 or 1.\n\nWe have also provided a sample submission csv file. Please notice that you should have the exact format as requested, or you will be in trouble.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}