{
  "fetched_at": "2026-10-19T14:00:38Z",
  "source_url": "https://www.kaggle.com/competitions/comp-4318-5318-2024-s-2-a-1/data",
  "pages": [
    {
      "name": "data-description",
      "content": "**IMPORTANT**: For assignment 1, please note that we provide only PART of the original Fashion-MNIST, you must use the GIVEN train.csv (not the original dataset from the official website) for training.\n\nParticipants should be able to answer these types of questions after reading the data description:\n\n```\n1. Q: What files do I need?\n   A: You need to upload your solution onto Kaggle. Your solution file is used to calculate the accuracy, which can be found in the Leaderboard after you upload it. \n\n2. Q: What am I predicting?\n   A: You are predicting the labels of each sample given the latent features.\n\n3. Q: What should I expect the data format to be?\n   A: The format of the solution file is '.csv'. \n\n4. Q: Do I have a sample solution?\n   A: Yes, a template solution is provided - \"sample.csv\".\n```\n\n**Files**\n\n- `train.csv `- the training set\n- `sample.csv `- a sample submission file in the correct format\n\n**Columns**\n\n- v1, …, v784 - the latent features\n- label - the label of a training sample",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "This competition is part of the COMP5318: Machine Learning and Data Mining course and centers on building a high-performance classifier. Participants will develop a model to classify 28x28 grayscale images into predefined categories. Due to the large dimensionality of the data, effective preprocessing and method selection will be critical to success.\n\nThe competition can be undertaken individually or in teams of two. All teams must register a Kaggle account and submit their solutions using their ***Canvas Group ID*** as the ***team name***. The leaderboard will track each model's performance on the test set, with part of the assignment grade based on these results.\n\n**[Data Input](https://www.kaggle.com/competitions/comp-4318-5318-2024-s-2-a-1/data)**: The dataset desc, required submission format are specified.\n**[Testing-Learderboard](https://www.kaggle.com/competitions/comp-4318-5318-2024-s-2-a-1/leaderboard)**: You can find how to test your model and how to submit your results via Kaggle Leaderboard.\n**Assignment 1 due**: Sunday week 7 - September 15th 2024, 11:59PM.\n\nThis is the overview of Kaggle Leaderboard, go to Canvas -> Assignment -> Assignment 1_2024.ipynb for full details of the assignment.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The performance of your classifier will be evaluated in terms of the top-1 accuracy metric, i.e.\n\n**Accuracy**:\n\n$$\n\\text{Accuracy} = \\frac{\\text{Number of Correct Predictions}}{\\text{Total Number of Predictions}}\n$$\n\n",
      "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 - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "- **Team name**: **Canvas Group ID**.\n- **Submission Deadline**: **11:59 pm September 15th, 2024**.\n\n- ***Accuracy on private test set ≥ 95% gets 5 points.***\n- ***95%>Accuracy on private test set ≥ 91% gets 4 points.***\n- ***91%>Accuracy on private test set ≥ 80% gets 3 points..***\n- ***80%>Accuracy on private test set gets 2 points.***\n\n*Bonus* : Top 5% gets a 5-point bonus",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Leaderboard Submission Format",
      "content": "You are required to submit a `test_output.csv`. The submission file should contain 5,000 testing samples with two columns: `id`, `label`.\n\nThe file should contain a header and have the following format:\n```\nid,label\n0,6\n1,2\n2,4\n3,0\n```\n**IMPORTANT**: the template - `sample.csv` can be found in [**Data**](https://www.kaggle.com/competitions/comp-4318-5318-2024-s-2-a-1/data). The python code of outputting `test_output.csv` can be found in` Assignment 1_2024.ipynb` Part 3.2.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "IMPORTANT NOTES",
      "content": "Assignment 1 Leaderboard Rules\n\n1) You **MUST** read and follow the instruction of Canvas -> Assignment -> Assignment 1_2024.ipynb.\n\n2) You **MUST NOT** submit the prediction results to the leaderboard that is not from your code.\n\n3) You **MUST** use the given template - sample.csv for leaderboard submissions.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}