shapash icon indicating copy to clipboard operation
shapash copied to clipboard

πŸ”… Shapash: User-friendly Explainability and Interpretability to Develop Reliable and Transparent Machine Learning Models

Results 80 shapash issues
Sort by recently updated
recently updated
newest added

**Description of Problem**: Currently only Python 3.6 to 3.9 seem to be officially supported. **Overview of the Solution**: Drop support for 3.6 which is EOL, add support of 3.10. Check...

# Description Python 3.6, is rarely used for Shapash and Python 3.10 begins to be used more. So we replace the python 3.6 support with python 3.10. ![image](https://user-images.githubusercontent.com/53937041/153877787-0ce18ea3-5390-4553-832b-f5d71d4819c7.png) Fixes #293...

# Description This PR creates a new plot that allows users to compare contributions of 2 methods across each feature Fixes #362 # Examples ![Capture d’écran 2022-06-14 aΜ€ 01 13...

**Description of Problem**: As different explainability methods often provide different explanations, we provide a new graph that helps investigating where those differences are located in the data **Overview of the...

**Description of Problem**: Can Shapash be used for image classification and object detection problems...? like Lime and Shap libraries?

I was going through the documentation and came across the below plot and its header ![image](https://user-images.githubusercontent.com/30723825/160334335-bb88bb5e-a821-4243-b3bc-21270e679ea2.png) Here, for the local explanation id = 3, we see that SHAPASH graph plots...

I went through the SHAPASH documentation [https://shapash.readthedocs.io/en/latest/tutorials/tutorial02-Shapash-overview-in-Jupyter.html](here) and came across the below plot While I understand that the taller/lengthier bar indicates better predictive power. But how do we know whether...

I am using a very large dataset in which some float columns have low cardinality (but it is not an integer column, think of float values with 90% missing data)....

Hi, I've just checked out Shapash. I've seen a lot of this line in the document: `preprocessing=encoder, # Optional: compile step can use inverse_transform method` However, I'm not sure how...

Currently [setup.py](https://github.com/MAIF/shapash/blob/v1.6.1/setup.py#L12-L26) is very strict on some requirements (and quite loose on others). This will make installing shapash in combination with other packages difficult as it can lead to incompatible...