JSON 训练数据配置文件示例 - Amazon Neptune
Amazon Web Services 文档中描述的 Amazon Web Services 服务或功能可能因区域而异。要查看适用于中国区域的差异，请参阅 中国的 Amazon Web Services 服务入门 (PDF)。

本文属于机器翻译版本。若本译文内容与英语原文存在差异，则一律以英文原文为准。

JSON 训练数据配置文件示例

以下是描述节点分类任务图形的示例训练数据配置文件：

{
  "version" : "v2.0",
  "query_engine" : "gremlin",
  "graph" : [
    {
      "edges" : [
        {
          "file_name" : "edges/(movie)-included_in-(genre).csv",
          "separator" : ",",
          "source" : ["~from", "movie"],
          "relation" : ["", "included_in"],
          "dest" : [ "~to", "genre" ]
        },
        {
          "file_name" : "edges/(user)-rated-(movie).csv",
          "separator" : ",",
          "source" : ["~from", "movie"],
          "relation" : ["rating", "prefixname"], # [prefixname#value]
          "dest" : ["~to", "genre"],
          "features" : [
            {
              "feature" : ["rating", "rating", "numerical"],
              "norm" : "min-max"
            }
          ]
        }
      ],
      "nodes" : [
        {
          "file_name" : "nodes/genre.csv",
          "separator" : ",",
          "node" : ["~id", "genre"],
          "features" : [
            {
              "feature": ["name", "genre", "category"],
              "separator": ";"
            }
          ]
        },
        {
          "file_name" : "nodes/movie.csv",
          "separator" : ",",
          "node" : ["~id", "movie"],
          "features" : [
            {
              "feature": ["title", "title", "word2vec"],
              "language": ["en_core_web_lg"]
            }
          ]
        },
        {
          "file_name" : "nodes/user.csv",
          "separator" : ",",
          "node" : ["~id", "user"],
          "features" : [
            {
              "feature": ["age", "age", "numerical"],
              "norm" : "min-max",
              "imputation": "median",
            },
            {
              "feature": ["occupation", "occupation", "category"],
            }
          ],
          "labels" : [
            {
              "label": ["gender", "classification"],
              "split_rate" : [0.8, 0.2, 0.0]
            }
          ]
        }
      ]
    },
    "warnings" : [ ]
  ]
}