{
  "id": 23375,
  "title": "Python : Scikit Random Forest on Decision trees",
  "url": "/competitions/bosch-production-line-performance/discussion/23375",
  "author_name": "",
  "post_date": "2016-09-05T11:01:21.810Z",
  "votes": null,
  "comment_count": 2,
  "views": 807,
  "content": "<p>Hi All,\nI am using Random Forest Classifier to predict the values. The problem that I am facing is that the  categorical values of Training_categorical.csv and the Testing_Categorical.csv differ because of which  the total number of features differ at the end.\nAny suggestion on how I can handle it </p>",
  "messages": [
    {
      "id": "134137",
      "postDate": "09/05/2016 11:01:21",
      "content": "<p>Hi All,\nI am using Random Forest Classifier to predict the values. The problem that I am facing is that the  categorical values of Training_categorical.csv and the Testing_Categorical.csv differ because of which  the total number of features differ at the end.\nAny suggestion on how I can handle it </p>",
      "rawMarkdown": "Hi All,\r\nI am using Random Forest Classifier to predict the values. The problem that I am facing is that the  categorical values of Training_categorical.csv and the Testing_Categorical.csv differ because of which  the total number of features differ at the end.\r\nAny suggestion on how I can handle it",
      "votes": null
    },
    {
      "id": "134428",
      "postDate": "09/06/2016 15:30:07",
      "content": "<p>Remove the features (i.e., defect codes) from the test data that aren't found in the train data.</p>",
      "rawMarkdown": "Remove the features (i.e., defect codes) from the test data that aren't found in the train data.",
      "votes": null
    },
    {
      "id": "134584",
      "postDate": "09/07/2016 11:28:10",
      "content": "<p>how is import in data set find any answer........</p>",
      "rawMarkdown": "how is import in data set find any answer........",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 134428,
      "author_name": "inversion",
      "author_url": "",
      "post_date": "09/06/2016 15:30:07",
      "content": "<p>Remove the features (i.e., defect codes) from the test data that aren't found in the train data.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 134584,
      "author_name": "sathish123",
      "author_url": "",
      "post_date": "09/07/2016 11:28:10",
      "content": "<p>how is import in data set find any answer........</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "134137": "Hi All,\r\nI am using Random Forest Classifier to predict the values. The problem that I am facing is that the  categorical values of Training_categorical.csv and the Testing_Categorical.csv differ because of which  the total number of features differ at the end.\r\nAny suggestion on how I can handle it",
    "134428": "Remove the features (i.e., defect codes) from the test data that aren't found in the train data.",
    "134584": "how is import in data set find any answer........"
  },
  "source": "meta"
}