FATE icon indicating copy to clipboard operation
FATE copied to clipboard

An Industrial Grade Federated Learning Framework

Results 385 FATE issues
Sort by recently updated
recently updated
newest added

**Describe the bug** There is error when trying to define pipeline. **To Reproduce** Steps to reproduce the behavior: when trying to define pipeline after successful data uploaded the following error...

在conf文件中,设置了eval_type:'multi',但是在log打印时候仍然显示 eval_type is binary。然后在EvaluateParam类中看到注释 Parameters ---------- eval_type : {'binary', 'regression', 'multi'} support 'binary' for HomoLR, HeteroLR and Secureboosting, support 'regression' for Secureboosting, 'multi' is not support these version 请问,homoNN的验证集是否支持‘multi'呢? Fate版本1.8

**Is your feature request related to a problem? Please describe.** Recoding Best Metrics When Running Hetero Feature Binning Using Optimal Binning Method

enhancement
federatedml

能增加一个联邦随机森林模块么?

**双边测试通过** **单边测试阻塞**: - kubefate 部署spark的模式是standalone - spark-submit 提交时传递的参数--num-executors(这个参数在yarn模式下生效)失效,导致spark中application抢占所有资源

Signed-off-by: Xinye Fan [[email protected]](mailto:[email protected]) Fixes ISSUE #3930 Changes: 1. Adding semi-supervised learning component

**Describe the bug** 上传数据,运行examples/pipeline/homo_nn/pipeline_homo_nn_single_layer.py样例代码时,模型训练成功。 ![image](https://user-images.githubusercontent.com/73977132/192463935-2e0e052d-0115-45ea-b2b8-4736e74039ce.png) 在运行后续的predict代码。 ``` predict_pipeline = PipeLine() predict_pipeline.add_component(reader_0) predict_pipeline.add_component( pipeline, data=Data( predict_input={pipeline.data_transform_0.input.data: reader_0.output.data} ), ) predict_pipeline.predict() ``` 发布预测任务后,运行时报错。查看Fate Board中,发现如下问题: ![image](https://user-images.githubusercontent.com/73977132/192463712-c1c8e459-d45d-42e3-9bde-0f3a6b427346.png) [ERROR] [2022-09-27 03:41:42,519] [202209270339544274350] [2134:140356215727936] - [task_executor._run_]...

请问FATE框架中有实现异步联邦学习的功能吗?如果没有,那是否可以用FATE来实现一个异步联邦学习系统?

在dsl 和config中添加了验证集导入,然后在HomoNNClient类的fit函数中添加了validate_data参数,后面的代码也针对验证集做了对应修改,但是fateflow报错显示:TypeError: fit() takes 2 positional arguments but 3 were given

**Describe the bug** 有两个job, job-A:包含DataTransform、Intersection、HeteroFeatureSelection三个组件,job完成后进行deploy job-B: 包含HeteroSecureBoost,input_data是job-A的output_data, job-B成功,自定义dsl(包含job-A和job-B的组件)进行deploy 基于job-B deploy的mode进行预测任务时,model_loader 异常,目前发起的是所有的model_loader 加载其他模型组件的model都会异常,在1.8版本是正常的,1.9版本异常 异常信息如下 ``` [ERROR] [2022-09-22 14:53:22,058] [202209221452420715180] [12081:139734406686528] - [task_executor._run_] [line:265]: 'NoneType' object is not iterable Traceback (most recent...