Amit Sharma

@finnschwall From the error description, looks like the error might be related to your dataset. Have you tried the dataset with a standard sklearn model (e.g., `LinearRegression`) and check if...

@finnschwall you are almost there! If I understand correctly, you just need to replace the below line, ``` m = dice_ml.Model(model=model, backend="sklearn") ``` with a custom `Model` that extends `BaseModel`....

@Labaien96 DiCE works for continuous features. So it will interpret the timeseries also as a (unordered) set of continuous features. There are ways to preprocess the data to make it...

The running time should scale linearly with the number of input points. Is that what you are observing? Or there is a marked increase?

@Saladino93 the question of obeying (causal) relationships is an important one. Thanks for raising this here. Current research literature has taken a graph-based approach to specifying relationships. However in practice,...

Re causal constraints paper: Our paper on using causal constraints (the one you refer) does need Pytorch to work, because it uses the gradient of the model. So it is...

yes, that's a good idea. We just need to call the classifier again on the returned CF data to obtain the predicted probabilities. This would be a general solution that...

We will have to see how to prioritize this change, given the other enhancements we are working on. Hopefully this will be included in the next release, near Dec.

thanks for reporting this, @hadjipantelis Let me have a look and try to reproduce this--the correct behavior is to return no CFs in case `features_to_vary` cannot lead to a CF.

This scenario is not supported as of now. Can look to implement it for dice_random and dice_genetic. Dice_kd does require access to the dataset so it will not work. Perhaps...