Open AlexsandarB opened 2 years ago
Hi
You could try using the explainer for linear models, which might be more efficient given the narrower usage compared to the general Explainer.
/Magnus
Alternatively, you might benefit from playing around with the min_df
parameter in the TfidfVectorizer
, which can drastically reduce the amount of features it returns.
/Magnus
Hi I am having a problem trying to create SHAP values with the code below, the program stops running because its unable to locate enough memory with the memory needed to be located is over 35gb.