Pranav Nerurkar comments

Results 5 comments of


                                            Pranav Nerurkar

'Explanation' object has no attribute 'conjugate'

Try scatterplot https://shap.readthedocs.io/en/latest/example_notebooks/api_examples/plots/scatter.html

Data with only numeric features - alternative to get_he_preprocessor() to avoid argument category_map

Thanks. I used the solution suggested and it is working without errors.

Data with only numeric features - alternative to get_he_preprocessor() to avoid argument category_map

So if no immutable/categorical data nor range is given- we can write as below? ```py # Define constants COEFF_SPARSITY = 0.5 # sparisty coefficient COEFF_CONSISTENCY = 0.5 # consisteny coefficient...

Data with only numeric features - alternative to get_he_preprocessor() to avoid argument category_map

yes it is working thanks. From my understanding the example given https://docs.seldon.io/projects/alibi/en/stable/examples/cfrl_adult.html. 1. The quality of counterfactuals generated depends on the black box model - in the above link randomforest...

IndexError: tuple index out of range

Thanks I will try the new link. The error goes if i one-hot encode the labels. I did not do this as the data was binary. My bad.