-
The current CCAO feature evaluation process for new model features is very ad-hoc. We typically look at the change in model performance metrics before and after the addition of a new feature, as well …
-
Working with chemists and engineers, they often mention that they suspect that factor interactions are important in their chemical processes. They would be very interested in seeing whether this is t…
-
https://github.com/slundberg/shap
https://towardsdatascience.com/explainable-ai-interpretability-of-machine-learning-models-412840d58f40
-
### Issue Description
`DeepExplainer` currently seems unable to handle a Pytorch model loaded from TorchScript, and will throw `RuntimeError: register_forward_hook is not supported on ScriptModules`.…
-
https://github.com/slundberg/shap
-
### Issue Description
When finding shap values for the deep learning model using Tensorflow, there is some operation lookup error for following operations:
1. TensorListStack
2. BatchMatMulV2 - Tra…
-
Hey there I'm totally new into programming with python. Doing it for my thesis. I managed to learn a lot by reading all your issues but right now I'm a little desperate.
The following:
x_train…
-
### Issue Description
The background dataset always gets subsampled to 100 samples
### Minimal Reproducible Example
```python
from sklearn.datasets import fetch_openml
import shap
import n…
-
Sklearn, random forest for regression. I am using `shap.explainers.GPUTree` and manage to generate an instance of `shap._explanation.Explanation`.
The problem is, global visualizations do work (`sh…
-
### Problem Description
shap plots should consistently include axes and figsize parameters.
For example,
`shap.plots.force(... , figsize=(20, 3), ...)` includes `figsize` but not `ax`.
`shap.…