ProtoDash: local variable 'newinnerProduct' referenced before assignment

[laramdemajo]

I am using the HELOC Dataset and trying to explain a single test instance using prototypes from my training subset using below code:

explainer = ProtodashExplainer()
(W, S, _) = explainer.explain(dfTrain.to_numpy(), dfTest.iloc[0:1,:].to_numpy(), m=2)

However, I am getting below error:

Is this intentional? Please help.

Thank you

[laramdemajo]

Hi I would really appreciate some feedback please. I am using this library for my disseration and currently stuck due to this issue. Your help will be very much appreciated.

[ArpitSisodia]

I doubt you can use prototype for single instance. It is a way of sampling the data. Does it make sense to sample single instance ?

[laramdemajo]

It should be possible, because IBM's Arya et al. state in their paper that ProtoDash can be used to explain a single test sample using samples from the train set (https://arxiv.org/pdf/1909.03012.pdf).

ProtoDash [26]: Selects diverse and representative samples that summarize a dataset or explain a test instance. Non-negative importance weights are also learned for each of the selected samples.

[sadhamanus]

So you can explain a single instance but the dataset you use to explain it has to have multiple instances. You cannot have both X and Y have cardinality 1.

[laramdemajo]

Of course, in fact in my code dfTrain contains 7403 instances as shown below:

[sadhamanus]

So then try swapping the arguments. The thing to be explained is to come first. Also check they have the same dimensionality. If this doesnt work try looking at usage in the example tutorial notebooks.

[laramdemajo]

Many thanks for your help. Swapping the arguments did the trick.

I followed the below comments in the PDASH.py file of this library so maybe you need to change the explanation there to avoid anyone doing the same mistake.

[sadhamanus]

Awesome! Thanks.