Jesse Horne
Jesse Horne
Is there an easy solution for local installations facing this issue? Is it just upgrading jaxlib with pip? Same as #162
@scb2310 Did you ever look into this or find any additional sources that evaluated this? I’m wondering about this as well for my own work. Thanks.
Was anyone able to get SHAP values for individual predictions with an MLJAR model? Or does it depend on the model type selected in the end?
Thanks for the quick reply @pplonski. I can actually handle for it to be slow in this case, but using the `automl.predict` in SHAP was not working. I'll see if...
Would it be possible to extract only the feature processing part before passing back to SHAP? FWIW, I’m not currently using golden features. Thanks again for the input. I’ll see...
Any updates for macOS Sonoma / 14.2.1?
Did you ever implement this with edge features (i.e. an edge weight)? Thanks!
How large of a lift would it be to support partitioned tables? Is that something I could contribute to? I'm guessing some involved logic in the create stage, given you...
Would it simplify things to tie back to snapshot intervals? So forcing partitions to align (but contain possibly multiple of) with the interval unit? Given the way SQLMesh also creates...
Along with this, options to skip the existing image features and umap process if we're bringing our own embeddings.