{
  "fetched_at": "2026-10-19T14:25:56Z",
  "source_url": "https://www.kaggle.com/competitions/code-marathon-ml-2/data",
  "pages": [
    {
      "name": "rules",
      "content": "# Rules for Round 1 of CODEMARATHON 5.0 -  MACHINE LEARNING\n\n- The number of submissions is limited to 20 per day.\n- Top 10 participants on the leaderboard will have to submit their source files as well as model weights for verification.\n- Plagiarism is strictly prohibited. The source files, as well as the submission csv files of all leaderboard toppers, will be evaluated for detecting plagiarism.\n- In case plagiarism is detected, the participant will be immediately disqualified and the next participant on the leaderboard will be moved up.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "data-description",
      "content": "<p>The required train and test data has been provided in a zip file format (data.zip). The training and test data has been supplied in 'train.csv' and 'test.csv' respectively, the descriptions for both of which have been provided below.  </p>\n<p>For making a submission, please generate a CSV file in the same format as 'sample_submission.csv'. The 'identifier' field of each prediction must match with the corresponding 'identifier' of the test sample on which the prediction has been made.</p>\nThe following image contains the specific format that is to be followed while making submissions.\n\n![sample submission format][1]\n\n<h2>File descriptions</h2>\n<ul>\n<li><strong>train.csv</strong> - the training set, containing 300 training samples</li>\n<li><strong>test.csv</strong> - the test set, containing 20000 testing samples, for which the predictions must be made</li>\n<li><strong>sample_submission.csv</strong> - a sample submission file in the correct format</li>\n</ul>\n<h2>Data fields</h2>\n<ul>\n<li><strong>identifier</strong> - an index unique to each sample, required for generating predictions in the prescribed format</li>\n<li><strong>f0, f1, f2, ..., g149 </strong> - the 300 features for each sample</li>\n<li><strong>target (only present in 'train.csv')</strong> - the binary target variable (either 0 or 1) that must be predicted</li>\n</ul>\n\n\n  [1]: https://gdurl.com/I5FW",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "# This round is being scrapped for a while due to upload of wrong dataset. Please bear with us while we  update the dataset.\n<p>Welcome to CODEMARATHON 5.0 - MACHINE LEARNING (Round 2) which is being organized by CSES, Dept. of Computer Science and Engineering, IIT (ISM), Dhanbad.</p>\n## Problem Statement\nCongratulations on performing well in Round 1 of this competition. It's time that we give your heavy-duty hardware, that that can grind through all those images like a beast, some well-deserved rest with a small and easy (or is it?!) CSV based dataset.\n\nThis round is a **weird** one. The challenge is to **generalize to a test set of 20000 samples** using only a handful of training examples **(300,  to be precise)**. All you have to do is model your predictions on a 20,000x300 matrix of test samples using only 300 training samples... **without overfitting**!\n\nYour task is to predict a binary target variable based on 300 features.\n\nDo your best, model without overfitting and win! Your data science-y insights might help you with this one.\n\nPeople overfit like crazy, can you not?\n\n## Data\nThe train set contains 300 samples, each consisting of an identifier (indexes into the data), 300 features (based on which you have to predict the binary target) and the target itself (true binary label for the sample).\n\nYou have been provided with the relevant dataset for this competition in a zip file 'data.zip'. Please see the [data section][1]  for downloading the data. The dataset and submission formats have been described in detail in the data tab/section.\n\n## Competition Rules \nPlease have a look at the competition rules [here][2]. All rules are to be followed strictly. Any use of unfair means will result in immediate disqualification from the competition.\n\n## Competition Timeline\nStart: 8:00 pm, 19th March 2019\nEnd: 11:59 pm, 31st March 2019\n\n## Submisssion Deadline\n11:59 pm, 31st March 2019\n\n## Evaluation\nFor information related to evaluation, please see the [evaluation  section][3] (on the left).\n\n\n  [1]: https://www.kaggle.com/c/code-marathon-ml-2/data\n  [2]: https://www.kaggle.com/c/code-marathon-ml-2/rules\n  [3]: https://www.kaggle.com/c/code-marathon-ml-2#evaluation",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "#### The evaluation consists of two phases:\n 1. **Public Test:** This evaluation takes place (on a random fifth (20%) of the test samples) as soon as a submission is made and the participant acquires a position on the public leaderboard accordingly.\n 2. **Private Test:** This evaluation is performed (on the remaining four fifths (80%) of the test samples) only after the submission deadline ends (11:59 pm, 31st March 2019).\n\nThe final leaderboard will be decided on the basis of your performances in both the Public and Private Tests (Private test ties will be broken using public test performance).\n\n<hr>\n<p>The evaluation metric for this competition is <a href=\"https://en.wikipedia.org/wiki/Accuracy_and_precision#In_binary_classification\"> Categorization Accuracy</a>. It is given by:</p>\n<p>\\[ Categorization\\  Accuracy = \\frac{No.\\ of\\ Correct\\ Predictions}{Total\\ No.\\ of\\ Samples}\\]</p>\n<p>This means that the fraction of samples from the test set that you classify correctly will be you score on the leaderboard.</p>\n<h2>Submission Format</h2>\n<p><strong>For each sample in the test set</strong>, the submission file should contain two columns: identifier and target. The specified entries must be comma delimited.</p>\n<p>The submission csv file should contain a header and have the following format:</p>\n<pre>identifier,target<br>1,0<br><br>2,0<br><br>3,1<br><br>4,0<br><br>5,1<br></pre>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}