{
  "fetched_at": "2026-10-19T07:25:14Z",
  "source_url": "https://www.kaggle.com/competitions/face-classification-deep-learning/data",
  "pages": [
    {
      "name": "data-description",
      "content": "## Directories:\n\n- **train** - the training set; each identity has its own folder with 20 images\n- **test** - the test set; it contains a total of 35k images (5 for each person)\n- **sample_submission.csv** - a sample submission file in the correct format showing you what needs to be submitted\n\nYou are free to create a validation set from the training set. Once your model has been completely trained, run it on the test set. Your final submission will be in the form of a CSV file that contains test set image names in the first column and their predicted labels in the second column. Make sure that the CSV file you submit is exactly in the same format as sample_submission.csv\n \n**Note:** You still have to submit your code / notebook. The CSV file is just for automatic ranking on the test set for Kaggle leaderboard. For submission on Kaggle, create a zip file that contains both the CSV as well as the code file.\n\nAlso, make sure that your team name is set to your full name as it is present on Qalam.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "In the Face Classification (Deep Learning) competition, we delve into the fascinating domain of computer vision, where the task at hand is to classify faces with the power of deep learning models. This competition challenges participants to create models that can associate various facial images with their respective identities.\n\n## The Challenge:\n\nAs the foundation of numerous applications, including facial recognition, security systems, and human-computer interaction, the accurate classification of faces plays a pivotal role in technology and society. Deep learning has revolutionized this field, offering immense potential for improving accuracy and efficiency.\n\nThe competition dataset is a diverse collection of facial images, spanning various demographics, expressions, and conditions. Your goal is to train DL models that can not only identify the presence of faces but also correctly label them with the corresponding identity. Participants are encouraged to explore different architectures, optimization techniques, and data preprocessing methods to achieve superior results.\n\nYour innovative solutions may also gain recognition within the machine learning and data science communities. This competition welcomes participants of all skill levels. You can also find valuable resources, forums, and tutorials to assist you on your journey.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Your models will be evaluated based on classification accuracy. The competition allows you to make multiple submissions, enabling you to fine-tune your models and improve your performance throughout the competition. However, only the final submission will be marked at the end.\n\n## Metric:\n\nThe primary evaluation metric for this competition is **classification accuracy**.\n\n## Submission Format:\n\nParticipants are required to submit a CSV file that contains two columns:\n\n**\"image_name\":** A unique identifier for each test image.\n**\"label\":** The predicted label or identity associated with each test image.\nThe \"label\" should be a representation corresponding to the identity you predict for each image in the test dataset.\n\n## Scoring:\n\nThe score on the leaderboard is calculated based on the accuracy of your model's predictions on the test dataset. The higher the accuracy of your model, the better your leaderboard score.\n\n## Final Ranking:\n\nThe final competition rankings will be based on the scores from the leaderboard as well as your code / notebook submissions. These will be revealed at the end of the competition. You are encouraged to fine-tune your models, experiment with different architectures, and optimize your approaches to achieve the highest classification accuracy possible. Make sure your submissions adhere to the specified format to ensure accurate evaluation and ranking.\n\nBest of Luck!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "To ensure a fair and enjoyable experience for all participants, we've outlined a set of rules that everyone must adhere to. Please read and understand these rules before joining the competition:\n\n**Fair Play:** Participants must compete fairly and follow ethical guidelines. Any form of cheating, plagiarism, or unfair practices is strictly prohibited. This includes, but is not limited to, using external data sources not provided by the competition, sharing solutions, or submitting misleading information.\n\n**Respect Others:** Be respectful and considerate when interacting with fellow participants, staff, and the community at large. Inappropriate behavior, harassment, or offensive content will not be tolerated.\n\n**Model Training:** Participants are encouraged to build their models using the competition data provided. The use of external data or pre-trained models is allowed, but it must be clearly disclosed in your code and any discussions related to your solution.\n\n**Code of Conduct:** All code submitted must adhere to Kaggle's code of conduct and terms of service. Respect copyright and intellectual property rights.\n\n**Multiple Submissions:** While you are allowed to make multiple submissions, please avoid excessive submissions, as it may result in penalties. Focus on improving your models and avoid submitting similar solutions repeatedly.\n\n**Data Usage:** The competition data is provided for the sole purpose of this competition. Participants are not permitted to use the data for any other purposes.\n\n**Late Submissions:** Late submissions will not be accepted after the competition deadline. Please ensure your submissions are made before the specified closing date and time.\n\n**Final Rankings:** The final rankings in this competition will be determined based on the private leaderboard, which will remain hidden until the competition concludes.\n\nBy participating in the Face Classification (Deep Learning) competition, you agree to abide by these rules. Failure to comply with these may result in disqualification and the removal of your submission. We believe in the spirit of fair competition and cooperation. Excited to see the innovative solutions you create in this competition. Good luck and enjoy the challenge!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "Welcome to the Face Classification (Deep Learning) competition! In this exciting challenge, you are to harness the power of cutting-edge neural networks for the task of associating different identities through facial images.\n\n## Goal:\n\nThe primary objective of this competition is to develop deep learning models that can accurately classify faces, connecting each image with its corresponding identity. By leveraging the potential of deep neural networks, you'll have the opportunity to advance the field of computer vision and contribute to the development of facial recognition technologies.\n\nPrepare to showcase your skills, compete with a community of data scientists, and make significant strides in the world of face classification through DL models. Get ready to dive in, and may the best models prevail!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}