Usage: ./network_up.sh { start | stop | restart }
- -b $blockchainType,网络类型,取值xchain或fabric,默认xchain网络
- -s $storageModeType,底层存储引擎,本地存储为 local,IPFS 为 ipfs,默认 local
./network_up.sh start -b $blockchainType -s $storageModeType
- -p 如果需要使用 PaddleFL 的能力, 需要启动 PaddleFL 的容器, 默认为不启动
./network_up.sh start -p true
- -h 如果需要使用PaddleDTX可视化能力,需启动可视化服务paddledtx-visual,启动时指定可被浏览器访问的计算节点IP地址,默认为不启动。在host为“106.13.169.234”的机器上启动PaddleDTX及其可视化服务,命令如下:
./network_up.sh start -h 106.13.169.234
paddledtx-visual启动之后,浏览器输入 http://106.13.169.234:8233/ 即可访问,使用之前需要输入区块链节点、数据持有节点相关信息,可直接导入如下配置paddledtx_setting.json:
{
"users": [
{
"publicKey": "e790393685a359e37a73457b3eef55c87264a61c968e5c136b70b8b5e6941f3605a67561af41633035239f6393b949584470da7a67b5b8fe284bd69cfb0d3d59",
"privateKey": "f0f6ad5422b37bdf18f3ef6464ce682d7412f25b5f5f5e800454f195055bffb1",
"mnemonic": "提 现 详 责 腐 贪 沉 回 涨 谓 献 即",
"address": "eFHH6ovPcG6eMszLB4DxFWeY3EBPZ9Hrb",
"default": true
}
],
"contractName": "paddlempc",
"node": "106.13.169.234:8908",
"dataOwners": [
{
"address": "106.13.169.234:8441",
"default": false
},
{
"address": "106.13.169.234:8442",
"default": true
},
{
"address": "106.13.169.234:8443",
"default": false
}
]
}
./network_up.sh stop -b $blockchainType
./network_up.sh restart -b $blockchainType -s $storageModeType
Usage: ./paddledtx_test.sh {upload_sample_files | start_vl_linear_train | start_vl_linear_predict | start_vl_logistic_train | start_vl_logistic_predict | tasklist | gettaskbyid}
./paddledtx_test.sh upload_sample_files
- vlLinTrainfiles 取值为步骤2.1获取到的 vertical linear train sample files
./paddledtx_test.sh start_vl_linear_train -f $vlLinTrainfiles
./paddledtx_test.sh start_vl_linear_train -f $vlLinTrainfiles -e true
./paddledtx_test.sh start_vl_linear_train -f $vlLinTrainfiles -l true
- vlLinPredictfiles 取值为步骤2.1获取到的 vertical linear predict sample files
- linearModelTaskId 取值为步骤2.2的模型训练任务ID
- 请确保2.2训练任务已经完成
./paddledtx_test.sh start_vl_linear_predict -f $vlLinPredictfiles -m $linearModelTaskId
- vlLogTrainfiles 取值为步骤2.1获取到的 vertical logistic train sample files
./paddledtx_test.sh start_vl_logistic_train -f $vlLogTrainfiles
./paddledtx_test.sh start_vl_logistic_train -f $vlLogTrainfiles -e true
./paddledtx_test.sh start_vl_logistic_train -f $vlLogTrainfiles -l true
- vlLogPredictfiles 取值为步骤2.1获取到的 vertical logistic predict sample files
- logisticModelTaskId 取值为步骤2.4的模型任务ID
- 请确保2.4训练任务已经完成
./paddledtx_test.sh start_vl_logistic_predict -f $vlLogPredictfiles -m $logisticModelTaskId
./paddledtx_test.sh tasklist
docker exec -it executor1.node.com sh -c "
./executor-cli task list --host 127.0.0.1:80 -p 6cb69efc0439032b0d0f52bae1c9aada3f8fb46a5f24fa99065910055b77a1174d4afbac3c0529c8927587bb0e2ad90a85eaa600cfddd6b99f1212112135ef2b
"
- taskID 为目标任务ID
./paddledtx_test.sh gettaskbyid -i $taskID
- taskID 为目标任务ID
docker exec -it executor1.node.com sh -c "./executor-cli task getbyid --host 127.0.0.1:80 -i $taskID"