Yang Wenzhuo

Hi, this is an issue coming from installing the hnswlib. It looks that there is no C++11 compiler. Could you please install it and try to install hnswlib again?

No hesitate to contact us if there are still some problems in installing hnswlib. And could you list your OS version and python version?

Thank you so much for adding the conda installation. This is very helpful for users. When the PR is ready, I will merge it to the main branch.

Hi, could you provide more background information about your application? For the current version, OmniXAI only supports image inputs with shape (batch_size, h, w, c). For a 3D images with...

Thanks for trying OmniXAI. If you have other issues, feel free to contact us.

Hi, Max, if there is still some problem with hnswlib, please try to install it separately as mentioned in https://github.com/nmslib/hnswlib. The problem is related to CPU versions. On Mon, Feb...

The problem comes from hnswlib. Could you please try to install it by following: https://github.com/nmslib/hnswlib? If there is still an installation error, try older version, e.g., 0.6.0.

Hi, the current version requires scikit-learn < 1.2 because some dependencies have issues with the latest scikit-learn. Please reinstall scikit-learn to proceed.

You can simply implment a prediction function using the trained pyspark model, and pass it to the "model" parameter when initializing a "TabularExplainer". Some examples: How to use TabularExplainer: https://github.com/salesforce/OmniXAI/blob/main/tutorials/tabular_classification.ipynb...

If you want to generate explanations for 1M examples, please also use pyspark to run the explainer, e.g. distribute workloads into multiple workers.