Chen Zihao

Is Anchor in this implementation suitable for one-hot encoding tabular dataset ?

3

Just shifted from that original Anchor implementation (implemented by paper's author), encounted a problem so wonder if it is exist in this implementation too. My features had been encoded into...

Why some fundament algorithms like LR DT RF is comparable with DES methods on my dataset.

3

I mean, the des method does not improve or even worse in the indicators run by my data set.

克隆到本地安装后，运行HereWeGo.py程序报错。

请问包中具体用的分箱算法是哪一种

14

您好，我了解到目前的监督分箱算法包括最优分箱 卡方分箱 决策树分箱等，请问这个包中的分箱算法是用的哪一种

I want use LIME to interprete classifiers trained by nimbusml, this package has also a `predict_proba` function which LIME needs( the usage basically the same from my simple experiment) but...

My dataset has already binarization into one-hot format, like features:['loan_amount=500'] with values [[0,1],[1,0],[0,1]]. So the anchor we expected should be like: "loan_amount=500=**1**" However, when i ran the algorithm, it showed...

 I clone the repository and run "Anchor on tabular data.ipynb" and occure this error comment this two line like:  then everything goes well.

Text、tabular data、image, I have no idea how to perturb different data format specifically, and seems there is no explicit mention in paper.