{
  "fetched_at": "2026-10-19T10:05:17Z",
  "source_url": "https://www.kaggle.com/competitions/aitxdil-project-jupyter/data",
  "pages": [
    {
      "name": "data-description",
      "content": "The data given is a set of images consisting of text with background noise. The task is to use these images and denoise them using autoencoders or deep learning models of your preference. The dataset is divided into the following:\nInput - Noisy Images with multiple resolutions\nOutput - Denoised Images with original resolutions\nSubmission - Unpacked individual pixel values with ID to correlate pixel locations.\n\n**1. Training_Data.zip** - Set of 100 noisy images to be used as part of model training\n**2. Cleaned_Training_Data.zip** - Set of 100 cleaned images corresponding to images in Training Dataset\n**3. Test_Data.zip** - Set of 44 images to be cleaned for scoring\n**4. sampleSubmission.csv** - Sample of submission file with id and value columns only",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Introducing the annual INR 25,000 event, Project Jupyter, hosted by the Army Institute of Technology, Pune, India. Where participants will be faced with a deep learning problem statement and present their solutions. This event is designed to challenge participants' knowledge and skills in Artificial Intelligence and Deep Learning. Participants will have the opportunity to apply various AI/DL techniques to build a model that can accurately solve the given problem. \nThis year’s event is powered by Deep Info Lab, an IT services company based in New Delhi, India that uses the uncharted potential of Deep Learning and Artificial Intelligence to provide smart and customized services to individuals and businesses. They also create products that streamline the digital workspace of professionals.\nThis challenge will involve a deep learning task with images as data. The task is to use these images and denoise them using autoencoders or deep learning models of your preference. The dataset is divided into the following:\n**Input** - Noisy Images with multiple resolutions\n**Output** - Denoised Images with original resolutions\n**Submission** - Unpacked individual pixel values with ID to correlate pixel locations.\n\n\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The competition leaderboard will use the Root Mean Square Error Metric (RMSE) for purpose of score calculation.\n\n$$\nRMSE = \\sqrt{\\frac{1}{n}\\Sigma\\_{i=1}^{n}{\\Big(\\frac{d\\_i -f\\_i}{\\sigma\\_i}\\Big)^2}}\n$$\n\nThis metric will be used per pixel value of each image and each image will be compared to ground truth images. These can be seen in the train set as cleaned training data. The images are all grayscale and thus all pixel values are to be calculated by finding a float value within a scale of 0 to 1. \n\nThe output is to be submitted in a CSV as given in the sample submission. Each pixel value becomes a new row. For example, a pixel at position (50,100) of image 120 with value 0.965 will be represented as:\n\n| Display Type | ID | Prediction |\n| --- | --- | --- |\n| Row Template | imgno_x_y | value |\n| Real | 120_50_100 | 0.965 |\n\n\nThe scores will be the summation of all the RMSE values of each pixel. The lower the score, the higher the leaderboard ranking.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "1.\tOne account per participant : You cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.\n2.\tNo sharing of code : Participants are not allowed to share their code with anyone else.\n3.\tIndividual participation\n4.\tCompetition Timelines : 15/04/23 11:00:00 to 23/04/23 11:00:00\n5.\tEligibility : \n       a.\tA registered account holder at Kaggle.com\n       b.\tThe participant must have registered for “PROJECT JUPYTER” on Kaggle.\n6.\tCOMPETITION DATA : \"Competition Data\" means the data or datasets available from the Competition Website for the purpose of use in the Competition, including any prototype or executable code provided on the Competition Website. The Competition Data will contain private and public test sets. Which data belongs to which set will not be made available to participants.\n8.\tDISQUALIFICATION : If a participant is found undermining the legitimate operation of the Competition by cheating, deception, or other unfair playing practices or abuses, threatens or harasses any other participants, then the organizing team holds all the rights to disqualify the participant.\n\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}