{
  "fetched_at": "2026-10-19T09:08:15Z",
  "source_url": "https://www.kaggle.com/competitions/caltech-256-image-classification/data",
  "pages": [
    {
      "name": "data-description",
      "content": "## Context\nThe Caltech 256 is considered an improvement to its predecessor, the Caltech 101 dataset, with new features such as larger category sizes, new and larger clutter categories, and overall increased difficulty. This is a great dataset to train models for visual recognition: How can we recognize toasters, cell phones, skyscrapers, sail boats, speed boats and many other categories in cluttered pictures? How can we classify images in these categories?\n\n## Content\nThere are 30,607 images in this dataset spanning 256 object categories. Object categories are extremely diverse, ranging from vehicles to computer monitor to tuning fork to watch to skyscraper to guitar. \n\n## Acknowledgements\nOriginal data source and banner image: http://www.vision.caltech.edu/Image_Datasets/Caltech256/\n\nWhen using this dataset, please remember to cite:\n\nGriffin, G. Holub, AD. Perona, P.\nThe Caltech 256.\nCaltech Technical Report.\n\n## Inspiration\nCan you build a model that classifies images in the Caltech-256 dataset into their respective categories?\nFor example, by taking input as an image, can your deep learning model (a Convolutional Neural Network ) classify whether the object is a ostrich, fried egg, leopard, skyscraper, penguin, or one of the other 251 object categories?\n\n## Files\n\n*   **Caltech 256_Test** - the test set\n*   **Caltech 256_Train** - the training set\n*   **sample_submission.csv** - a sample submission file in the correct format\n\n\n## Columns of the sample submission file\n\n*   `img_path` - path of the image\n*   `label` - category/class of the object/image",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "## Objective: \n\nTo classify images in the Caltech-256 dataset, which is an improvement over Caltech-101 dataset using a Convolutional Neural Network.\n\n## Problem Statement\n\nTo build and implement a Convolutional Neural Network model to classify images in the Caltech-256 dataset.\n\nAt the end of this competition, you will be able to:\n\n1. Load and extract features of images available in the Caltech-256 dataset using ImageDataGenerator\n2. Build convolutional neural networks using either Keras or PyTorch deep learning libraries\n3. Use the pre-trained models using either Keras or PyTorch deep learning libraries\n\n## Description: \n\nCaltech-256 is an object recognition dataset containing approximately 30,000 real-world images, of different sizes, spanning 256 classes (256 object classes and an additional clutter class). Each class is represented by at least 80 images. The dataset is a superset of the Caltech-101 dataset.\n\nHere is a handy link to Kaggle's competition documentation (https://www.kaggle.com/docs/competitions), which includes, among other things, instructions on submitting predictions (https://www.kaggle.com/docs/competitions#making-a-submission).",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The evaluation pages describes how submissions will be scored and how students should format their submissions.\n\nThe evaluation metric for this competition is Categorization Accuracy.\n\nThe file should contain a header and have the following format:\n\nimg_path, label\ntest/1.jpg, Faces-Easy-101",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "One account per participant You cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.\n\nNo private sharing outside teams Privately sharing code or data outside of teams is not permitted. It's okay to share code if made available to all participants on the forums.\n\nTeam Mergers Team mergers are allowed and can be performed by the team leader. In order to merge, the combined team must have a total submission count less than or equal to the maximum allowed as of the Team Merger Deadline. The maximum allowed is the number of submissions per day multiplied by the number of days the competition has been running.\n\nTeam Limits The maximum team size is 4.\n\nSubmission Limits You may submit a maximum of 20 entries per day.\n\nCompetition Timeline Competition Timeline dates (including Entry Deadline, Final Submission Deadline, Start Date, and Team Merger Deadline, as applicable) are reflected on the competition’s Overview > Timeline page.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}