DiCE

DiCE copied to clipboard

Generating CF examples for the multi-class classification case using the GD-based method

4

Hi, I was wondering if the current implementation of the GD-based method (I mean the loss-based method described in your paper, [Mothilal et al. (2020)](https://arxiv.org/abs/1905.07697)) can handle the multi-class classification...

ah-ansari

Counterfactual for subpopulation

2

Can DiCE be extended for calculating the counterfactuals for a subpopulation based on certain conditions?

tonyabracadabra

Hi everyone, How could I get distance from counterfactuals with KDtree method? I observed the line `cfs = cfs.drop(["distance", "sparsity", "distancesparsity"], axis=1)` in the method ` do_sparsity_check(self, cfs, query_instance, sparsity_weight)`...

immaF

How to early stop? The regression task runs without results when the expected output is too high

3

The output range of my regression prediction is [1, 30], while most targets are lower than 2. When I generate counterfactuals with an expected output of 10 or more for...

xueyagaga

Wachter implementation

1

Hey everyone, in the paper is mentioned that DiCE is compared to SimpleCF proposed by Wachter, is there the code of generating those Counterfactuals in this repo? I am afraid...

ninaspreitzer

DiCE doesn't change my outcome

3

Hello, when I try to run DiCE, the proposed explanations do not change my outcome from 0 to . I am only trying the basic settings with the adult dataset...

ninaspreitzer

DataFrame.dtypes for data must be int, float or bool. Error even if the data is float

8

the code was running fine last week but today as i try to run the same code i'm bumping into this issue/ Error:  even if the data is in...

ShadiKhoury

error "unexpected keyword argument 'desired_range'" for TF regression models

4

Hello, I have a DNN model built in Keras and TensorFlow with a regression output. When I call the `generate_counterfactuals` function, I receive the error: `TypeError: generate_counterfactuals() got an unexpected...

ddefranza

Implementation of causality in DICE?

5

I am not sure if I can ask here, or email the authors of the paper.... I really do not know with whom to discuss this, but I will try...

Saladino93

Feature ranges

5

I am using the drug consumption dataset for a multiclass classification and trying to enforce some permitted ranges like: ``` d = dice_ml.Data(dataframe = eval_data, continuous_features = continuous_features, outcome_name =...

PaoloFantine