{ "fetched_at": "2026-10-19T08:44:04Z", "source_url": "https://www.kaggle.com/competitions/2021-linear-model-final-project/data", "pages": [ { "name": "data-description", "content": "This dataset consists of songs collected by year, with 515345 instances (samples) and 90 attributes (variables). Songs are mostly western, commercial tracks ranging from 1922 to 2011, with a peak in the year 2000s. The task is to predict the albums release year (i.e., V1) by the audio features transformed under MFCC. In this competition, you have to train a model based on the train.csv file, and make a prediction about the release year of the songs for those given in test.csv.\n\n## Files\n\n* **train.csv** - the training set\n* **test.csv** - the test set\n* **sample_submission.csv** - a sample submission file in the correct format\n\n## Columns\n\n* `V1` - The year of the albums were published\n* `V2` to `V14` - Timbre average\n* `V15` to `V91` - Timbre covariance\n\nFeatures extracted from the 'timbre' features from The Echo Nest API.\nWe take the average and covariance over all 'segments', each segment being described by a 12-dimensional timbre vector. If you are interested in the feature extraction process, you may search using keywords like MFCC and DCT. ", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Description", "content": "After a semester of learning linear models, we setup this competition to encourage participates to apply their learned methods to real data. In class, we have learn the theory and mathematics. Now it is the time to get familiar with the real application!! The competition helps you to get started fast.\n\nHere are some suggestions:\n1. In the real world, it is not sufficient to fit your model directly to the data. Try to do some exploratory data analysis (EDA) before you fitting the model, which may help you pass through the baseline (See Score). \n2. More often than not, \"pure linear\" models may not be useful. But a too complicated model also easily performs awfully for prediction in the test set. \n3. Usually, we have to try several candidate models, and select a final one. This competition allows you to upload two final models, and we count your best one.\n\nEnjoy and have fun!!", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "The evaluation metric for this competition is [RMSE (Root Mean Squared Error)](https://en.wikipedia.org/wiki/Root-mean-square_deviation). The RMSE score, commonly used in regression tasks, measures the goodness-of-fit using the predicted values and the true values of the outcome varaible y\n\n$$\n\\mbox{RMSE} = \\sqrt{\\sum\\_{i=1}^n \\frac{(y_i - \\hat y\\_i)^2}{n}}\n$$\n\n## Submission Format\n\n**For every prediction in the dataset**, submission files should contain two columns: `Id` and `Predicted`. `Predicted` should be float numbers. Every `Id` represent the song Id, and prediction should be listed with respect to its ids. For example, if you predict a release song with Id `511162`, you should assign your prediction at the row which Id is `511162`. The size of the submission file should be around 774.47KB.\n\n\nThe file should contain headers for columns in the following format:\n```\nId,Predited\n511162,2000\n```", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "rules", "content": "The following is the competition rules, if you violate one of them, you may have a zero score on the final project.\n\n## Team Limits\n\nThe maximum number of team size is 1, and you must join one and only one team.\n\nYour team name should be your student ID + \\_(anything), for example, M1020400xx\\_樓下請雞排\n\n## Submission Limits\n\nYou may submit a maximum of 5 entries per day.\n\nYou may select up to 2 final submissions for judging.\n\nYou have to submit your codes to Cyber University by deadline January 17, 2022.\n\n## Data and Model\n\nYou are not allowed to use any deep learning models or other non-open source models\n\nYou are not allowed to use extra data or marking on the data by yourself\n\n## No private sharing outside teams\n\nIt is alright to share codes if made available to all participants on the forums. Privately sharing codes or data outside of teams is not permitted. \nMoreover, if you have code issues or data issues, post them on the Discussion board.\n\n## Remarks\n\nThis competition doesn't award ranking points\n\nThis competition doesn't count towards tiers\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Timeline", "content": "- December 21, 2021 - Start Date.\n- January 10, 2022 - Entry Deadline. You must accept the competition rules before this date in order to compete.\n- January 16, 2022 - Final Submission Deadline.\n\nAll deadlines are at 11:59 PM UTC on the corresponding day unless otherwise noted. ", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Score", "content": "## Your final scoring\n\nYou need to make your code available to all the participants after the competition.\nWe set a public simple baseline. If your pass the baseline, you will get 60pts. \nYour final score will depend on the rank on the private leaderboard： \nScore = (25 - position rank) x 1.45 + 60 \nif you pass the baseline. \n\n## The Organizer reserves the right to change the rules as necessary.\nIf most of the participates can't achieve the baseline, we may decrease the baseline standard", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }