{
  "fetched_at": "2026-10-19T16:37:09Z",
  "source_url": "https://www.kaggle.com/competitions/dlai-8-porosity-challenge/data",
  "pages": [
    {
      "name": "data-description",
      "content": "The dataset consists of the following files and folders:\n\ntrain.csv: This file contains the training data, including features and the target variable (log permeability).\ntest.csv: This file contains the test data, which includes the features for which you need to predict the log permeability.\nsample_submission.csv: This file provides the format required for submitting your predictions for the test set.\nimages/: This folder contains the images associated with each sample in the dataset.\n\n<h3>train.csv</h3>\nThe train.csv file includes the following columns:\n\nID: Unique identifier for each sample.\nfeature_1, feature_2, ..., feature_n: Various features that describe the geological properties of the sample.\nLog Permeability: The target variable representing the logarithm of the permeability of the sample.\n\n<h3>test.csv</h3>\nThe test.csv file includes the following columns:\n\nID: Unique identifier for each sample.\nfeature_1, feature_2, ..., feature_n: Various features that describe the geological properties of the sample. \n\n<h3>sample_submission.csv</h3>\nThe sample_submission.csv file includes the following columns:\n\nID: Unique identifier for each sample in the test set.\nLog Permeability: The predicted logarithm of the permeability for each sample.\n\n<h3>images/</h3>\nThe images/ folder contains the images associated with each sample. The naming convention for the images corresponds to the ID of each sample, ensuring that each image can be uniquely linked to its respective sample in the train.csv and test.csv files.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Porous media, found in both natural forms (e.g., sand, oil-bearing rocks, honeycombs) and synthetic forms (e.g., lattice materials, metallic fibers, metal foams), play crucial roles in numerous natural and industrial processes. In nature, porous media are vital for soil formation, groundwater flow, and the movement of nutrients and pollutants. Industrial applications include filtration, fuel cells, energy storage, and biomedical devices.\n\nThe characteristics of porous media, such as pore size, grain size, and connectivity, significantly influence their physical and chemical properties like permeability, porosity, and surface area. Determining permeability typically involves resource-intensive experiments, as it varies widely among different porous media structures. Computational Fluid Dynamics (CFD) has become essential in studying fluid flow in complex geometries, providing valuable insights into behaviors such as multiphase flow and chemical reactions. However, CFD simulations can be computationally demanding, often requiring high computing power and extensive mesh elements.\n\nTo address these challenges, Artificial Intelligence (AI) techniques, including machine learning and deep learning, offer efficient and accurate methods to predict fluid flow behavior in porous media. This competition aims to leverage AI to enhance the understanding and prediction of fluid dynamics within porous structures, providing a cost-effective and scalable solution to traditional methods.\n\nThis competition is a part of DLAI8.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Submissions are evaluated on Mean Squared Error between the predicted permeability and the observed target.\n\n\n## Submission File\nFor each ```ID``` in the test set, you must predict a value for the ```log permeability``` variable. The file should contain a header and have the following format:\n\n    ID,Log Permeability\n    vv2,-9\n    vv5,-9\n    GG6,-9\n    etc.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": " - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "The goal of this competition is to predict permeability values of porous materials. You will develop a model trained on image and tabular features generated from computational fluid dynamics (CFD) simulation of 2D porous domains.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}