-
I used `deepExplainer` for explaining a pytorch-implemented classification model(basically multiple layers of transformer encoders succeeded by a classifier). My data is tabular timeseries in the shap…
-
 and have downsampled the negative class to give me a 50/50 balance in the two classes. Ignoring the challenges that comes with undersampling (l…
-
Hello,
I have a model that's built with h2o version 3.28.1.2 that i'm running predict_contributions() for, and the values are not making sense. There is one feature that produces a massive contribu…
-
Hello, @slundberg .
Thank you for your papers about SHAP values.
Could you, please, provide an explanation what is going on exactly when in the TreeExplainer we set `feature_dependence = “indepe…
-
### Issue Description
While removing the Boston dataset in #3200, it was observed that it is difficult to specify the layout of the resulting waterfall plot correctly. Please take a look at the min…
-
Hi, all,
Thank you for reading my issue.
I am trying to install ```shap``` and use GPU to accelerate. I follow this:
Check to makes sure you have the NVIDA CUDA Toolkit installed by running t…
-
Display the clicked shap during the drag action.
-
How to use the SGDClassifier based on Output from a TFIDFVectorizer and chi-sq transform (vectorized text)
X_test_matrix = pd.DataFrame(X_test.toarray())
sample = shap.sample(X_t…