{
  "fetched_at": "2026-10-19T09:44:33Z",
  "source_url": "https://www.kaggle.com/competitions/day-3-kaggle-competition/data",
  "pages": [
    {
      "name": "data-description",
      "content": "<p>The data we've provided contains a training set and a testing set. The training directory contains 6485 images in total and has five folders each belonging to one of the five classes: Bed, chair, sofa, swivelchair and table. </p>\n<p>The objects have been mapped to different numerical values as mentioned below along with the number of files.</p> \n<p> -Bed(0) folder contains 1,100 images.</p>\n<p> -Chair(1) folder contains 1,627 images</p>\n<p> -Sofa(2) folder contains 1,500 images. </p>\n<p> -Swivelchair(3) folder contains 952 images. </p>\n<p> -Table(4) folder contains 1296 images. </p>\n<p></p>\n<p> The test set contains 2,164 images, that has mixed images of all the five categories.</p>\n<h4>Important:</h4>\n<p> Upload the submission file as a .csv file under the Submissions tab. </p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<p>This competition is the final stage of the 2-day workshop on \"Introduction to DL and CNN's through PyTorch\" held by IECSE and Vision! By now you must've gained some interesting knowledge on CNN's and the best way to put it to use is by participating in competitions and familiarising yourself with the applications. </p>\n<p>The most important part of any house interiors is the presence of furniture. It enhances and beautifies the house. The variety of furniture like chairs, tables, beds and their further type divisions bring out the overall grace of your house. </p>\n<p> What does this have have do with AI? Robots are an essential part of the future. We've heard news that some robots were made/in the process of being made, to help in household works . In the recent years, a lot of models for detection and recognition of objects have been developed, commonly working with 3D information, which gives to the robots more details and accuracy. This makes it possible for the robots to identify and classify objects. With recent advancements in the field of Deep Learning based on Computer Vision, the development of Image classification models are very much necessary and are also easier to develop than before.</p>\n<p> In this competition, we provide you with a dataset that contains various 3-D images of five types of household furniture taken in multiple angles. The dataset contains five labels: </p>\n<p> Bed, swivel-chair, chair, sofa and table. </p>\n<h3> Objectives of this competition are: </h3>\n<p> 1. Apply the knowledge and skills you've gained from this 2-day workshop to build a model.</p>\n<p>2. To train your model using images of the given training dataset and get the highest accuracy.</p>\n<p>3. Accurately classify a given image from testing dataset into it's respective type of furniture.</p>\n\n\n<p>All the best folks!!</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<p>Analyse your data and tune your hyperparamters in the best possible way. Do not submit your notebook as others will be able to view it. \nYou will be evaluated based on the metric \"Categorization Accuracy\". \n</p><h3> Note </h3>\n<p>Please upload your solution as a .csv file under the \"Submissions\" tab.</p>\n<p> Even though in the workshop you were taught implementation using PyTorch, you are free to use any framework of your choice. </p>\n<p> Timeline: 6th September, 5:00 PM - 7th September, 11:59 PM. </p>\n<p>The csv file should contain two columns: First one being the \"Image\" column that consists your image numbers, second one is the \"Target\" column that consists your image labels. (Make sure you've matched the furniture with the correct labels as mentioned in the \"Data\" tab)</p>\n<p> The top three contestants on the Leaderboard have to submit their notebook for evaluation AFTER the deadline. After evaluation, they will be given recognition on our social media accounts and of course an additional plus point in their CV.</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "\n<p>Please follow the following rules:</p>\n<p>1. Do not use any dishonest methods to predict the images.</p>\n<p>2. Don't share your notebook here as everyone can view it. Don't share it with another competitor.</p>\n<p>3. Make sure that your columns match with that of the sample file.</p>\n<p>4. You cannot sign into Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.</p>\n<p>5. Have fun!</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}