找到正确的方法来产生罚款

Question

我尝试在本地配置this存储库

当我尝试执行这个命令python feature_extraction.py -t 1 [-nlon] -i mairesse_Apache.arff 时，我明白需要一个更精细的命名结果

这是我用于测试的 mailcorpus.json 文件：

[
   {
      "id":12,
      "mailing_list_url":"12",
      "type_of_recipient":"From",
      "email_address":"test@test.org",
      "message_body":"Here is one text to test sentiment and feel happy",
      "is_response_of":"before"
   },
   {
      "id":21,
      "mailing_list_url":"21",
      "type_of_recipient":"From",
      "email_address":"test@gmail.com",
      "message_body":"Here is one text to test sentiment and feel happy and feel fine",
      "is_response_of":"before"
   },
   {
      "id":21,
      "mailing_list_url":"23",
      "type_of_recipient":"From",
      "email_address":"23@gmail.com",
      "message_body":"Not feel so good for this code",
      "is_response_of":"after"
   }
]

存储库中有一个名为骨架的示例，具有此结构

[
    {
        "id_test": "uid",
        "time": "mm:ss",
        "email": "email@ddress\t",
        "extraversion": 1.0,
        "coscientiousness": 2.0,
        "agreeableness": 3.0,
        "openness": 4.0,
        "neuroticism": 5.0
    },
    ...
]

我知道如何接收我的 mailcorpus.json 文件的这个结果文件吗？

Answer 1

您链接中提供的存储库似乎不是供公众使用的。您列出的文件示例实际上不是输出，而是输入（存储库中没有用于转储 json 的代码）。我不希望你能够运行代码。

In the data/ folder, you will also find:

the training_data.rda model from the NLoN R package;
the skeleton of the json that will contain the responses from the mini-IPIP test taken by your subjects.

在这种情况下，您可能需要自己管理 mini-IPIP 测试。