{
  "fetched_at": "2026-10-19T19:04:08Z",
  "source_url": "https://www.kaggle.com/competitions/cmf-dataanalysis-training/data",
  "pages": [
    {
      "name": "data-description",
      "content": "<p>Data Fields</p>\r\n<p>Elevation - Elevation in meters<br />Aspect - Aspect in degrees azimuth<br />Slope - Slope in degrees<br />Horizontal_Distance_To_Hydrology - Horz Dist to nearest surface water features<br />Vertical_Distance_To_Hydrology - Vert Dist to nearest surface water features<br />Horizontal_Distance_To_Roadways - Horz Dist to nearest roadway<br />Hillshade_9am (0 to 255 index) - Hillshade index at 9am, summer solstice<br />Hillshade_Noon (0 to 255 index) - Hillshade index at noon, summer solstice<br />Hillshade_3pm (0 to 255 index) - Hillshade index at 3pm, summer solstice<br />Horizontal_Distance_To_Fire_Points - Horz Dist to nearest wildfire ignition points<br />Wilderness_Area (4 binary columns, 0 = absence or 1 = presence) - Wilderness area designation<br />Soil_Type (40 binary columns, 0 = absence or 1 = presence) - Soil Type designation<br />Cover_Type (7 types, integers 1 to 7) - Forest Cover Type designation</p>\r\n<p>The wilderness areas are:</p>\r\n<p>1 - Rawah Wilderness Area<br />2 - Neota Wilderness Area<br />3 - Comanche Peak Wilderness Area<br />4 - Cache la Poudre Wilderness Area</p>\r\n<p>The soil types are:</p>\r\n<p>1 Cathedral family - Rock outcrop complex, extremely stony.<br />2 Vanet - Ratake families complex, very stony.<br />3 Haploborolis - Rock outcrop complex, rubbly.<br />4 Ratake family - Rock outcrop complex, rubbly.<br />5 Vanet family - Rock outcrop complex complex, rubbly.<br />6 Vanet - Wetmore families - Rock outcrop complex, stony.<br />7 Gothic family.<br />8 Supervisor - Limber families complex.<br />9 Troutville family, very stony.<br />10 Bullwark - Catamount families - Rock outcrop complex, rubbly.<br />11 Bullwark - Catamount families - Rock land complex, rubbly.<br />12 Legault family - Rock land complex, stony.<br />13 Catamount family - Rock land - Bullwark family complex, rubbly.<br />14 Pachic Argiborolis - Aquolis complex.<br />15 unspecified in the USFS Soil and ELU Survey.<br />16 Cryaquolis - Cryoborolis complex.<br />17 Gateview family - Cryaquolis complex.<br />18 Rogert family, very stony.<br />19 Typic Cryaquolis - Borohemists complex.<br />20 Typic Cryaquepts - Typic Cryaquolls complex.<br />21 Typic Cryaquolls - Leighcan family, till substratum complex.<br />22 Leighcan family, till substratum, extremely bouldery.<br />23 Leighcan family, till substratum - Typic Cryaquolls complex.<br />24 Leighcan family, extremely stony.<br />25 Leighcan family, warm, extremely stony.<br />26 Granile - Catamount families complex, very stony.<br />27 Leighcan family, warm - Rock outcrop complex, extremely stony.<br />28 Leighcan family - Rock outcrop complex, extremely stony.<br />29 Como - Legault families complex, extremely stony.<br />30 Como family - Rock land - Legault family complex, extremely stony.<br />31 Leighcan - Catamount families complex, extremely stony.<br />32 Catamount family - Rock outcrop - Leighcan family complex, extremely stony.<br />33 Leighcan - Catamount families - Rock outcrop complex, extremely stony.<br />34 Cryorthents - Rock land complex, extremely stony.<br />35 Cryumbrepts - Rock outcrop - Cryaquepts complex.<br />36 Bross family - Rock land - Cryumbrepts complex, extremely stony.<br />37 Rock outcrop - Cryumbrepts - Cryorthents complex, extremely stony.<br />38 Leighcan - Moran families - Cryaquolls complex, extremely stony.<br />39 Moran family - Cryorthents - Leighcan family complex, extremely stony.<br />40 Moran family - Cryorthents - Rock land complex, extremely stony.</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<h3>It is slightly changed Kaggle Forest Cover Type Prediction competition.</h3>\r\n<p>In this competition you are asked to predict the forest cover type (the predominant kind of tree cover) from strictly cartographic variables (as opposed to remotely sensed data). The actual forest cover type for a given 30 x 30 meter cell was determined from US Forest Service (USFS) Region 2 Resource Information System data. Independent variables were then derived from data obtained from the US Geological Survey and USFS. The data is in raw form (not scaled) and contains binary columns of data for qualitative independent variables such as wilderness areas and soil type.</p>\r\n<p>This study area includes four wilderness areas located in the Roosevelt National Forest of northern Colorado. These areas represent forests with minimal human-caused disturbances, so that existing forest cover types are more a result of ecological processes rather than forest management practices.</p>\r\n<h2>Acknowledgements</h2>\r\n<p>Kaggle is hosting this competition for the machine learning community to use for fun and practice. This dataset was provided by Jock A. Blackard and Colorado State University. We also thank the UCI machine learning repository for hosting the dataset. If you use the problem in publication, please cite:</p>\r\n<p>Bache, K. &amp; Lichman, M. (2013). UCI Machine Learning Repository. Irvine, CA: University of California, School of Information and Computer Science</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<h2>Evaluation</h2>\r\n<p>Submissions are evaluated using the multi-class logarithmic loss. Each incident has been labeled with one true class. For each animal, you must submit a set of predicted probabilities (one for every class). The formula is then,</p>\r\n<p>\\[logloss=-\\frac{1}{N}\\sum_{i=1}^N\\sum_{j=1}^M y_{ij}\\log(p_{ij})\\]<br /><br />where N is the number of animals in the test set, M is the number of outcomes, \\\\(log\\\\) is the natural logarithm, \\\\(y_{ij}\\\\) is 1 if observation \\\\(i\\\\) is in outcome \\\\(j\\\\) and 0 otherwise, and \\\\(p_{ij}\\\\) is the predicted probability that observation \\\\(i\\\\) belongs to outcome \\\\(j\\\\).</p>\r\n<p>The submitted probabilities for a given animal are not required to sum to one because they are rescaled prior to being scored (each row is divided by the row sum). In order to avoid the extremes of the log function, predicted probabilities are replaced with \\\\(max(min(p,1-10^{-15}),10^{-15})\\\\).</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "\r\n\t<!-- Begin Base Rules -->\r\n    <h3>One account per participant</h3>\r\n    <p>You cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.</p>\r\n    <h3>No private sharing outside teams</h3>\r\n    <p>\r\n        Privately sharing code or data outside of teams is not permitted.\r\n        It's okay to share code if made available to all participants on the forums.\r\n    </p>\r\n    <h3>Team Mergers</h3>\r\n    <p>Team mergers are not allowed in this competition.</p>\r\n\r\n    <h3>Team Limits</h3>\r\n    <p>The maximum size of a team is 4 participants.</p>\r\n    <h3>Submission Limits</h3>\r\n    <p>You may submit a maximum of 10 entries per day.</p>\r\n    <p>You may select up to 2 final submissions for judging.</p>\r\n    <h3>Competition Timeline</h3>\r\n    <p>Start Date: <strong>4/5/2016 6:21 PM UTC</strong></p>\r\n    <p>Merger Deadline: <strong>None</strong></p>\r\n    <p>Entry Deadline: <strong>None</strong></p>\r\n    <p>End Date: <strong>6/1/2016 11:59 PM UTC</strong></p>\r\n\t<!-- End Base Rules -->\r\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}