{
  "fetched_at": "2026-10-19T08:57:38Z",
  "source_url": "https://www.kaggle.com/competitions/cse151b-spring2022/data",
  "pages": [
    {
      "name": "data-description",
      "content": "This page describes the details of the dataset. We provide the training set and the input of the test set. You are requested to submit the output (predictions) for the test set. The test set is further split into 80% for the public leaderboard and 20% for the private leaderboard.\n\nWe also provide a sample submission file with zero predictions. You are expected to replace the zeros with the predictions generated by your deep learning model.\n\n\n\n## Files\n\n*   **train** - the training set folder\n*   **test** - the test set folder\n*  **Load_Argo2_Publlic.ipynb** - a jupyter notebook to load the data \n*   **sample_submission.csv** - a sample submission file in the correct format\n\n## Format\n\n*   `austin_input` - input trajectories in Austin. A numpy array with the shape ( N, 50, 2). N is the number of scenes. 50 is the input 50 time steps. 2 is the x and y dimension.\n*   `austin_output` - output trajectories in Austin. A numpy array with the shape ( N, 60, 2). N is the number of scenes. 60 is the output 60 time steps. 2 is the x and y dimension.\n*   same for all other cities",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "This is the home page of the deep learning class competition (CSE 151b Spring 2022). \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).\n\n\n## Background\nAutonomous vehicles (AV) are expected to dramatically redefine the future of transportation. However, there are still significant engineering challenges to be solved before one can fully realize the benefits of self-driving cars. One such challenge is building models that reliably predict the movement of traffic agents around the AV, such as cars, cyclists, and pedestrians.\n\nIn this competition, you’ll apply your deep learning skills to build motion forecasting models for self-driving vehicles. You'll have access to a large AV motion dataset to train and test your models. The goal of the prediction task is to predict the positions of a tracked AV 6 seconds into the future, given an initial 5-second observation. \n\nThe training, validation, and test sequences are taken from different areas of 6 cities so that there is no geographical overlap. Each sequence follows the trajectory of the main agent for 11 seconds, while keeping track of all other traffic agents including cars and pedestrians. Agents follow very complex trajectories, such as changing lanes, navigating intersections, and turning. \n\n## Team Guideline\nThis is a group project for the class. You can request the computational resources from [UCSD Data Science/Machine Learning platform](https://datahub.ucsd.edu/hub/login) allocated for this course. See more details on the [Course Website](https://sites.google.com/view/cse151b/competition).\n\nMaximum submission per day: 10\nSubmission deadline: 11:59 pm UTC May 29, 2021. \n\nPlease note UTC is 7 hours ahead of PST,\n\n## Evaluation Guideline\nSee evaluation page for submission format and evaluation metric. The final rank counts towards 10% of the final score. \n\n## Acknowledgements\n\nWe thank [Argo](https://www.argo.ai/) for sharing this dataset.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The evaluation page describes how submissions will be scored and how you should format your submissions. \n\n---\n\nThe evaluation metric for this competition is [Mean Squared Error](https://en.wikipedia.org/wiki/Mean_squared_error). The MSE score measures the average squared difference between the estimated values and the actual value. \n\n\n## Submission Format\n\nYou will make predictions for each scene in the test set, where only the input is provided. You will generate the prediction and submit your solution in a single csv file. In the solution file, each row starts with a ID of the scene, followed by 120 columns.  These 120 columns are the x and y coordinates for the future 60 time steps (6 seconds). The submission file has the following format.\n```\n# ID      v1   v2   v3  v4 ...\n0_austin  x0   y0   x1  y1 ..\n1_austin\n```\nThe ID contains two parts. The prefix is the index in the input file, followed by the name of the city. You can also refer to the sample submission file as an example. ",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "You must accept the following before joining.\n\n - Don't cheat!\n - Work in teams!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}