{
  "id": 177405,
  "title": "Does evaluation on validation dataset take too long?",
  "url": "/competitions/lyft-motion-prediction-autonomous-vehicles/discussion/177405",
  "author_name": "",
  "post_date": "2020-08-25T19:16:43.459074400Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>When I try to run the evaluation loop according to <a href=\"https://github.com/lyft/l5kit/blob/master/examples/agent_motion_prediction/agent_motion_prediction.ipynb\" target=\"_blank\">l5kit example</a>, it is taking way too long. Is this normal? Moreover, how can we test on test dataset since I cannot load the test dataset. </p>",
  "messages": [
    {
      "id": "985517",
      "postDate": "08/25/2020 19:16:43",
      "content": "<p>When I try to run the evaluation loop according to <a href=\"https://github.com/lyft/l5kit/blob/master/examples/agent_motion_prediction/agent_motion_prediction.ipynb\" target=\"_blank\">l5kit example</a>, it is taking way too long. Is this normal? Moreover, how can we test on test dataset since I cannot load the test dataset. </p>",
      "rawMarkdown": "When I try to run the evaluation loop according to [l5kit example](https://github.com/lyft/l5kit/blob/master/examples/agent_motion_prediction/agent_motion_prediction.ipynb), it is taking way too long. Is this normal? Moreover, how can we test on test dataset since I cannot load the test dataset.",
      "votes": null
    },
    {
      "id": "985584",
      "postDate": "08/25/2020 20:52:54",
      "content": "<p>Yes it does if you run on the CPU. There's a problem with torch in the current environment. <a href=\"https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/177125\" target=\"_blank\">https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/177125</a></p>",
      "rawMarkdown": "Yes it does if you run on the CPU. There's a problem with torch in the current environment. https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/177125",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 985584,
      "author_name": "jpmiller",
      "author_url": "",
      "post_date": "08/25/2020 20:52:54",
      "content": "<p>Yes it does if you run on the CPU. There's a problem with torch in the current environment. <a href=\"https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/177125\" target=\"_blank\">https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/177125</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "985517": "When I try to run the evaluation loop according to [l5kit example](https://github.com/lyft/l5kit/blob/master/examples/agent_motion_prediction/agent_motion_prediction.ipynb), it is taking way too long. Is this normal? Moreover, how can we test on test dataset since I cannot load the test dataset.",
    "985584": "Yes it does if you run on the CPU. There's a problem with torch in the current environment. https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/177125"
  },
  "source": "meta"
}