模型提交异常 — 类型不匹配（expect float, get: double）

[treeforest]

Issue Type

Running

Have you searched for existing documents and issues?

No

OS Platform and Distribution

CentOS 7

All_in_one Version

v0.11.0b1

Kuscia Version

0.13.0b0

What happend and What you expected to happen.

## all-in-one 使用的镜像版本列表

{
    "images": {
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/capsule-manager-sim-ubuntu20.04": "v0.1.0b0",
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/dataproxy": "0.3.0b0",
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/kuscia": "0.13.0b0",
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/scql": "0.9.2b1",
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/secretflow-lite-anolis8": "1.11.0b1",
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/secretpad": "0.12.0b0",
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/serving-anolis8": "0.8.0b0",
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/sf-tee-dm-sim": "0.1.0b0",
        "secretflow-registry.cn-hangzhou.cr.aliyuncs.com/secretflow/teeapps-sim-ubuntu20.04": "0.1.2b0"
    }
}


## 场景复现
在secretpad上定义好训练流，并成功执行所有节点后，在`提交模型`阶段出错。

Log output.

## secretpad web端返回的错误信息

The remaining no-failed party task counts 1 are less than the task success threshold 2. pending party[], running party[tckhsamm], successful party[], failed party[vrwpdnqj]


## Lite `tckhsamm` 的 secretpad 错误日志

2025-07-30 13:51:27 [] [grpc-default-executor-17] INFO  o.s.s.m.integration.job.JobManager - starter jobEvent ... type: MODIFIED
object {
  job_id: "ysuo"
  status {
    state: "Failed"
    create_time: "2025-07-30T05:51:18Z"
    start_time: "2025-07-30T05:51:18Z"
    end_time: "2025-07-30T05:51:27Z"
    tasks {
      task_id: "wbsx-model-export"
      state: "Failed"
      err_msg: "The remaining no-failed party task counts 1 are less than the task success threshold 2. pending party[], running party[tckhsamm], successful party[], failed party[vrwpdnqj]"
      create_time: "2025-07-30T05:51:18Z"
      start_time: "2025-07-30T05:51:18Z"
      end_time: "2025-07-30T05:51:27Z"
      parties {
        domain_id: "tckhsamm"
        state: "Failed"
        endpoints {
          port_name: "inference"
          scope: "Cluster"
          endpoint: "wbsx-model-export-0-inference.tckhsamm.svc"
        }
        endpoints {
          port_name: "spu"
          scope: "Cluster"
          endpoint: "wbsx-model-export-0-spu.tckhsamm.svc"
        }
        endpoints {
          port_name: "fed"
          scope: "Cluster"
          endpoint: "wbsx-model-export-0-fed.tckhsamm.svc"
        }
        endpoints {
          port_name: "global"
          scope: "Domain"
          endpoint: "wbsx-model-export-0-global.tckhsamm.svc:29599"
        }
      }
      parties {
        domain_id: "vrwpdnqj"
        state: "Failed"
        err_msg: "container[secretflow] terminated state reason \"Error\", message: \".py\\\", line 458, in result\\n    return self.__get_result()\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 403, in __get_result\\n    raise self._exception\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/thread.py\\\", line 58, in run\\n    result = self.fn(*self.args, **self.kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/call_holder.py\\\", line 91, in _task\\n    ret = self._task_impl(*resolved)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/actor.py\\\", line 147, in _execute_impl\\n    ret = func(*args, **kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/device/proxy.py\\\", line 65, in wrapper\\n    return method(*args, **kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow_serving_lib/graph_builder.py\\\", line 251, in build_proto\\n    libserving.graph_validator_impl(graph_def_str)\\nRuntimeError: what: \\n\\t[Enforce fail at secretflow_serving/util/arrow_helper.cc:384] src_f->type()->id() == dst_f->type()->id(). edge schema check failed, src: onehot_encode_1, dst: feature_calculate_2. field: bmi type not match, expect: float, get: double\\n\\n2025-07-30 05:51:26.184 ERROR global_context.py:183 [vrwpdnqj] -- FedLocalError on seq id 125\\nTraceback (most recent call last):\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/global_context.py\\\", line 177, in _send_check\\n    send_future.result()\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 458, in result\\n    return self.__get_result()\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 403, in __get_result\\n    raise self._exception\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/thread.py\\\", line 58, in run\\n    result = self.fn(*self.args, **self.kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/global_context.py\\\", line 170, in _send\\n    raise local_err\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/global_context.py\\\", line 158, in _send\\n    obj = fed_obj.get_object()\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/object.py\\\", line 117, in get_object\\n    self._object = self._object.get()\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/object.py\\\", line 31, in get\\n    future_result = self.future.result(timeout=2)\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 458, in result\\n    return self.__get_result()\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 403, in __get_result\\n    raise self._exception\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/thread.py\\\", line 58, in run\\n    result = self.fn(*self.args, **self.kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/call_holder.py\\\", line 88, in _task\\n    resolved = self.resolve_dependencies(*args, **kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/call_holder.py\\\", line 68, in resolve_dependencies\\n    resolved.append(arg.get_object())\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/object.py\\\", line 117, in get_object\\n    self._object = self._object.get()\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/object.py\\\", line 31, in get\\n    future_result = self.future.result(timeout=2)\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 458, in result\\n    return self.__get_result()\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 403, in __get_result\\n    raise self._exception\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/thread.py\\\", line 58, in run\\n    result = self.fn(*self.args, **self.kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/call_holder.py\\\", line 88, in _task\\n    resolved = self.resolve_dependencies(*args, **kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/call_holder.py\\\", line 68, in resolve_dependencies\\n    resolved.append(arg.get_object())\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/object.py\\\", line 117, in get_object\\n    self._object = self._object.get()\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/object.py\\\", line 31, in get\\n    future_result = self.future.result(timeout=2)\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 458, in result\\n    return self.__get_result()\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 403, in __get_result\\n    raise self._exception\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/thread.py\\\", line 58, in run\\n    result = self.fn(*self.args, **self.kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/call_holder.py\\\", line 88, in _task\\n    resolved = self.resolve_dependencies(*args, **kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/call_holder.py\\\", line 68, in resolve_dependencies\\n    resolved.append(arg.get_object())\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/object.py\\\", line 117, in get_object\\n    self._object = self._object.get()\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/object.py\\\", line 31, in get\\n    future_result = self.future.result(timeout=2)\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 458, in result\\n    return self.__get_result()\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/_base.py\\\", line 403, in __get_result\\n    raise self._exception\\n  File \\\"/usr/local/lib/python3.10/concurrent/futures/thread.py\\\", line 58, in run\\n    result = self.fn(*self.args, **self.kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/call_holder.py\\\", line 91, in _task\\n    ret = self._task_impl(*resolved)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/distributed/fed/actor.py\\\", line 147, in _execute_impl\\n    ret = func(*args, **kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow/device/proxy.py\\\", line 65, in wrapper\\n    return method(*args, **kwargs)\\n  File \\\"/usr/local/lib/python3.10/site-packages/secretflow_serving_lib/graph_builder.py\\\", line 251, in build_proto\\n    libserving.graph_validator_impl(graph_def_str)\\nsecretflow.distributed.fed.exception.FedLocalError: what: \\n\\t[Enforce fail at secretflow_serving/util/arrow_helper.cc:384] src_f->type()->id() == dst_f->type()->id(). edge schema check failed, src: onehot_encode_1, dst: feature_calculate_2. field: bmi type not match, expect: float, get: double\\n\\n2025-07-30 05:51:26.185 WARNING global_context.py:133 [vrwpdnqj] -- Signal SIGINT to exit.\\n2025-07-30 05:51:26.185 WARNING api.py:116 [vrwpdnqj] -- Stop signal received (e.g. via SIGINT/Ctrl+C), try to shutdown fed. Press CTRL+C (or send SIGINT/SIGKILL/SIGTERM) to skip.\\n2025-07-30 05:51:26.185 INFO api.py:214 [vrwpdnqj] -- Shutdowning unintendedly, wait_for_sending True\\n2025-07-30 05:51:26.185 INFO global_context.py:214 [vrwpdnqj] -- Try stop context, wait_for_sending True, on_error True\\n2025-07-30 05:51:26.185 INFO global_context.py:218 [vrwpdnqj] -- task_executor stopped\\n2025-07-30 05:51:26.185 INFO global_context.py:220 [vrwpdnqj] -- recv_executor stopped\\n2025-07-30 05:51:26.186 INFO global_context.py:224 [vrwpdnqj] -- send_executor stopped\\n2025-07-30 05:51:26.186 INFO global_context.py:237 [vrwpdnqj] -- Context stopped\\n2025-07-30 05:51:26.283 [warning] [channel.h:~Channel:163] Channel destructor is called before WaitLinkTaskFinish, try stop send thread\\n2025-07-30 05:51:26.284 ERROR brpc_link.py:150 [vrwpdnqj] -- Receiving exception: <class \'secretflow.distributed.fed.exception.FedRemoteError\'>, FedRemoteError occurred at tckhsamm caused by what: \\n\\t[Enforce fail at secretflow_serving/util/arrow_helper.cc:384] src_f->type()->id() == dst_f->type()->id(). edge schema check failed, src: onehot_encode_1, dst: feature_calculate_2. field: age type not match, expect: float, get: double\\n from tckhsamm, seq id 124. Re-raise it.\\n2025-07-30 05:51:26.284 INFO api.py:222 [vrwpdnqj] -- Shutdowned\\n2025-07-30 05:51:26.285 CRITICAL api.py:225 [vrwpdnqj] -- Exit now due to the previous error.\\n\""
        endpoints {
          port_name: "fed"
          scope: "Cluster"
          endpoint: "wbsx-model-export-0-fed.vrwpdnqj.svc"
        }
        endpoints {
          port_name: "global"
          scope: "Domain"
          endpoint: "wbsx-model-export-0-global.vrwpdnqj.svc:32277"
        }
        endpoints {
          port_name: "inference"
          scope: "Cluster"
          endpoint: "wbsx-model-export-0-inference.vrwpdnqj.svc"
        }
        endpoints {
          port_name: "spu"
          scope: "Cluster"
          endpoint: "wbsx-model-export-0-spu.vrwpdnqj.svc"
        }
      }
      alias: "wbsx-model-export"
    }
    stage_status_list {
      domain_id: "tckhsamm"
      state: "JobCreateStageSucceeded"
    }
    stage_status_list {
      domain_id: "vrwpdnqj"
      state: "JobCreateStageSucceeded"
    }
    approve_status_list {
      domain_id: "tckhsamm"
      state: "JobAccepted"
    }
    approve_status_list {
      domain_id: "vrwpdnqj"
      state: "JobAccepted"
    }
  }
}
,nodeId=vrwpdnqj
2025-07-30 13:51:27 [] [grpc-default-executor-17] INFO  o.s.s.m.integration.job.JobManager - watched jobEvent: jobId=ysuo, jobState=Failed, task=[taskId=wbsx-model-export,alias=wbsx-model-export,state=Failed], endTime=2025-07-30T05:51:27Z
2025-07-30 13:51:27 [] [grpc-default-executor-17] INFO  o.s.s.m.integration.job.JobManager - watched jobEvent: jobId=ysuo, but project job not exist, skip

[treeforest]

训练流

secretpad错误截图

错误日志

secretpad.log

[treeforest]

不同组件的输出表展示

• 隐私求交

• 常规分箱

• 独热编码

• 特征计算

[treeforest]

补充：kuscia容器中pod的日志：

0.log

[peakshen]

官网的组件没看到 独热编码 你是在二开自定义开发的组件吗
考虑吧这个组件去掉后 再尝试下提交模型？

[treeforest]

您好，这里的 独热编码 就是 onehot_encode，仅做了统一的中文化标签处理：

[treeforest]

经过对比测试发现如下问题：

[peakshen]

经过对比测试发现如下问题：

方便发下你的特征计算的配置吗
选择的哪个算子和配置。
从报错上看，可能是float计算的数据结果是double类型了，跟期望不匹配。我本地复现下你的问题。

[treeforest]

您好！

训练流复现步骤

集群采用的是 master-lite 部署，包括两个参与方 alice 和 bob

0.数据准备（数据集）

alice: charges_alice.csv
bob:charges_bob.csv

1.隐私求交

alice: id
bob: uid
求交类型: inner joind

2.常规分箱

特征列: age、bmi、children
分箱方式: eq_num
bin_num: 10

3.onehot_encode

特征列: sex、smoker、region
丢弃: 不丢弃

4.特征计算

特征列: bmi、children、age、sex_0、sex_1、smoker_0、smoker_1、region_0、region_1、region_2、region_3
算子选择: Standardize(标准化)

5.逻辑回归训练

特征列: bmi、children、age、sex_0、sex_1、smoker_0、smoker_1、region_0、region_1、region_2、region_3
标签: charges

备注：训练流可以正常执行，只是模型提交时报错。

[peakshen]

您好！

训练流复现步骤

集群采用的是 master-lite 部署，包括两个参与方 alice 和 bob

0.数据准备（数据集）

alice: charges_alice.csv bob:charges_bob.csv

1.隐私求交

alice: id bob: uid 求交类型: inner joind

2.常规分箱

特征列: age、bmi、children 分箱方式: eq_num bin_num: 10

3.onehot_encode

特征列: sex、smoker、region 丢弃: 不丢弃

4.特征计算

特征列: bmi、children、age、sex_0、sex_1、smoker_0、smoker_1、region_0、region_1、region_2、region_3 算子选择: Standardize(标准化)

5.逻辑回归训练

特征列: bmi、children、age、sex_0、sex_1、smoker_0、smoker_1、region_0、region_1、region_2、region_3 标签: charges

备注：训练流可以正常执行，只是模型提交时报错。

问题已经稳定复现确定个代码bug。代码会内部提交测试后，随着下个版本发布带进去，可以关注下代码更新。

[treeforest]

好的，谢谢

[peakshen]

问题已经稳定复现确定个代码bug。代码会内部提交测试后，随着下个版本发布带进去，可以关注下代码更新。

这边先关闭issue，后续如果有其他问题，欢迎随时再沟通。