FATE
FATE copied to clipboard
An Industrial Grade Federated Learning Framework
我使用数据集进行横向分箱,设置如下,分别设置了连续列,离散列和转换列:  但是会报以下异常: ``` ==== detail start, at 20231219.235952.337 ==== Traceback (most recent call last): File "/data/projects/fate/eggroll/python/eggroll/core/utils.py", line 187, in wrapper return func(*args, **kw) File "/data/projects/fate/eggroll/python/eggroll/roll_pair/egg_pair.py", line 265, in run_task...
Signed-off-by: circlekkk Fixes ISSUE #xxx Changes: 1.add file: - python/federatedml/secureprotol/ecc/sm2.py 2.add SM2 parameters to the following file: - python/fate_client/pipeline/param/consts.py - python/federatedml/param/intersect_param.py - python/federatedml/util/consts.py 3.The following file code has been slightly...
fate-flow注册到zookeeper时服务地址乱码,  例如 172.16.111.11:9360 再zookeeper中变成了172.16.111.11%3A9360
  fateflow 注册失效

在项目中遇到一个问题,需要设置某些特征为单调约束,在XGBoost中可以通过设置参数“monotone_constraint”来实现(如下图),但在SecureBoost中始终未找到相关的设置。 
**Describe the bug** Hetero-nn does not calculate the validation loss. It only runs the validation set during eval. **To Reproduce** Run Hetero-nn with a validation set. **Expected behavior** A graph...
我使用pipeline构建神经网络,横向联邦学习任务。 自定义模型和数据问题解决,使用pipeline构建任务,可以运行pipeline.fit开始训练,可以看见日志显示任务训练了设定的epoch,但是训练完毕后报错socket.timeout,我猜测是run_ip显示为'run_ip': 'xxx.xxx.xxx.xxx'导致。 日志显示的参数如下 {'job_id': '202311290103155381610', 'component_name': 'reader_0', 'task_id': '202311290103155381610_reader_0', 'task_version': '0', 'role': 'guest', 'party_id': '9999', 'run_ip': 'xxx.xxx.xxx.xxx', 'run_pid': 467154, 'run_port': '9380', 'party_status': 'running'}
**Is your feature request related to a problem? Please describe.** A clear and concise description of what the problem is. Ex. I'm always frustrated when [...] **Describe the solution you'd...
Signed-off-by: elvis-xiao [email protected] Changes: 1. federatedml , data_statistics