FATE实践（参考联邦学习实战）

1. 以管理员身份进入FATE容器

[root@localhost cen]# docker exec -it fate_python bash

2.编辑配置upload_data_host.json(upload_data_guest.json)文件，主要修改file,  table_name,  namespace字段

vim /fate/examples/dsl/v1/upload_data_host.json

修改内容

{
    "file": "examples/data/breast_homo_host.csv",
    "head": 1,
    "partition": 10,
    "work_mode": 0,
    "table_name":"breast_homo_host",                  //指定DTable表名
    "namespace": "experiment"                         //指定DTable表名的命名空间
}

3.执行upload命令，将原始的本地文件breast_homo_host 转换为DTable格式

python /fate/python/fate_flow/fate_flow_client.py -f upload -c /fate/examples/dsl/v1/upload_data_host.json

4. 接下来是模型训练，主要包括对dsl文件和conf文件的编辑

vim /fate/examples/dsl/v1/homo_logistic_regression/test_homolr_train_job_dsl.json

dsl文件来描述任务模块，将任务模块以有向无环土的形式组合在一起

{
    "components" : {
        "dataio_0": {                             //数据I/O组件，将本地数据转化为DTable形式
            "module": "DataIO",
            "input": {
                "data": {
                    "data": [
                        "args.train_data"
                    ]
                }
            },
            "output": {
                "data": ["train"],
                "model": ["dataio"]
            }
         },
         "dataio_1": {
              "module": "DataIO",
              "input": {
                   "data": {
                       "data": [
                            "args.eval_data"
                            ]
                           },
                        "model":["dataio_0.dataio"]
                   },
                   "output":{
                           "data":["eval_data"]
                   }
                  },

"homo_lr_0": {                           //横向逻辑回归组件
            "module": "HomoLR",   
            "input": {
                "data": {
                    "train_data": [
                        "dataio_0.train"
                    ]
                }
            },
            "output": {
                "data": ["train"],
                "model": ["homolr"]
            }
        },
        "homo_lr_1":{
                "module":"HomoLR",
                "input":{
                        "data":{
                                "eval_data":[
                                        "dataio_1.eval_data"
                                ]
                        },

 //模型评估组件，若无测试数据集，则自动使用训练数据集进行模型评估
"evaluation_0": {                              
            "module": "Evaluation",
            "input": {
                "data": {
                    "data": [
                        "homo_lr_0.train"

5  conf文件用来描述所有参数信息需要修改role_parameter字段下的train_data,l  label_name表示的是标签列对应的属性名。

"role_parameters": {
        "guest": {
            "args": {
                "data": {
                    "train_data": [
                        {
                            "name": "CM1_guest",
                            "namespace": "experiment"
                        }
                    ],
                    "eval_data":[{
                                "name":"homo_breast_test",
                                "namespace":"experiment"
                    }]
                }
            },
                    "dataio_0":{
                            "with_label":[true],
                            "label_name":["y"],
                            "label_type":["int"],
                            "output_format":["dense"]
                }

        },
 "host": {
            "args": {
                "data": {
                    "train_data": [
                        {
                            "name": "KC1_host",
                            "namespace": "KC1_host"
                        }
                    ],
                    "eval_data":[{
                        "name":"homo_breast_test",
                        "namespace":"KC1_host"
                    }]
                }
            },
            "dataio_0":
            {
              "with_label":[true],
              "label_name":["y"],
              "label_type":["int"],
              "output_format":["dense"]
            },

6.执行submit_job命令，评估函数

python /fate/python/fate_flow/fate_flow_client.py -f submit_job -c /fate/examples/dsl/v1/homo_logistic_regression/test_homolr_evaluate_job_conf.json -d/fate/examples/dsl/v1/homo_logistic_regression/test_homolr_evaluate_job_dsl.json

7. 输入虚拟机IP:8080