mgqa34

If you want to use binning as a filter in selection component ,make sure that all selecting feature columns should be included in binning component's iv calculating process.

这个报错应该是三方（有协调方的）LR组件的报错？ 可以加一个scale组件或者binning+onehot去解决这个问题，对数据进行归一化后再进入LR组件，报错是模型权重太大(超过1e8)，发散了

fate-client can be find in this [link](https://pypi.org/project/fate-client/#history)

Please upload example data first.

Conf [uploading guide](./https://github.com/FederatedAI/FATE/blob/master/doc/tutorial/dsl_conf/upload_data_guide.md) Pipeline [uploading guide](./https://github.com/FederatedAI/FATE/blob/master/doc/tutorial/pipeline/pipeline_tutorial_upload.ipynb)

Have a look at this [link](https://github.com/FederatedAI/FATE/tree/master/examples/data#data-set-naming-rule)，and find out which datasets are needed. Or uploading all examples's data by [fate_test](https://github.com/FederatedAI/FATE/blob/master/doc/tutorial/fate_test_tutorial.md#command-types) : fate_test data upload -t all_examples. If you want to use...

Evaluation组件是支持multi的，文档的表达可能旧了，evaluation的参数eval_type和上游组件关系不大，是一个独立解耦的，只是说要用什么类型的指标去评估输入数据。可以在fateboard上查看这个参数配置是否真的生效，然后log的话也麻烦给出对应哪一行代码打出来的，这样方便查案

Which version is your fate-client?

> Hi @mgqa34 > > I installed FATE=1.9.0 and as it shown after pip install fate_client :it shows: > > (from requests-oauthlib>=0.7.0->google-auth-oauthlib=0.4.1->tensorboard=2.10->tensorflow>=2.4.1->fate_client) (3.2.0) Execute "pip list | grep fate-client" to...

Oh, the version of fate-client is a little old. If you are using fate-v1.9.0, please upgrade fate-client to 1.9.0 too, the version of fate-client and fate should be matched. If...