{
  "id": 241787,
  "title": "Doubts in the given question statement and its information",
  "url": "/competitions/pku-autonomous-driving/discussion/241787",
  "author_name": "puranjay mishra",
  "post_date": "2021-05-26T06:43:54.879000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hello.<br>\nI am an absolute begginer and would really appreciate any help.<br>\nThe doubts might look pretty silly but i couldn't come up with the answers….<br>\nMy doubts are -<br>\n[First] Why has the predicted string of the training data have multiple car models in the output , when we have to use the closest object to determine the yaw , pitch and roll?<br>\n[Second] What are the train and test masks exactly used for ? Is it used for blackening out the irrelevant parts of the image for training and testing? If so , then why is most of the mask black with only a little bit white on the wheel areas of the vehicles?<br>\n[Third] Do we have to identify the car models too? becuase the prediction string in training data file has multipe car models and their yaw,roll and pitch in it.  If so , then how is it possible to do it when the masks is making the model only focus on the wheel area?</p>",
  "messages": [
    {
      "id": 1323317,
      "postDate": "2021-05-26T06:43:54.880Z",
      "content": "<p>Hello.<br>\nI am an absolute begginer and would really appreciate any help.<br>\nThe doubts might look pretty silly but i couldn't come up with the answers….<br>\nMy doubts are -<br>\n[First] Why has the predicted string of the training data have multiple car models in the output , when we have to use the closest object to determine the yaw , pitch and roll?<br>\n[Second] What are the train and test masks exactly used for ? Is it used for blackening out the irrelevant parts of the image for training and testing? If so , then why is most of the mask black with only a little bit white on the wheel areas of the vehicles?<br>\n[Third] Do we have to identify the car models too? becuase the prediction string in training data file has multipe car models and their yaw,roll and pitch in it.  If so , then how is it possible to do it when the masks is making the model only focus on the wheel area?</p>",
      "rawMarkdown": "Hello.\nI am an absolute begginer and would really appreciate any help.\nThe doubts might look pretty silly but i couldn't come up with the answers....\nMy doubts are -\n[First] Why has the predicted string of the training data have multiple car models in the output , when we have to use the closest object to determine the yaw , pitch and roll?\n[Second] What are the train and test masks exactly used for ? Is it used for blackening out the irrelevant parts of the image for training and testing? If so , then why is most of the mask black with only a little bit white on the wheel areas of the vehicles?\n[Third] Do we have to identify the car models too? becuase the prediction string in training data file has multipe car models and their yaw,roll and pitch in it.  If so , then how is it possible to do it when the masks is making the model only focus on the wheel area?\n"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1323317": "Hello.\nI am an absolute begginer and would really appreciate any help.\nThe doubts might look pretty silly but i couldn't come up with the answers....\nMy doubts are -\n[First] Why has the predicted string of the training data have multiple car models in the output , when we have to use the closest object to determine the yaw , pitch and roll?\n[Second] What are the train and test masks exactly used for ? Is it used for blackening out the irrelevant parts of the image for training and testing? If so , then why is most of the mask black with only a little bit white on the wheel areas of the vehicles?\n[Third] Do we have to identify the car models too? becuase the prediction string in training data file has multipe car models and their yaw,roll and pitch in it.  If so , then how is it possible to do it when the masks is making the model only focus on the wheel area?\n"
  }
}