microsoft / augmented-interpretable-models Goto Github PK

Hi, thank you for providing this repo!

I want to use emb-gam for multiclassification (4 classes, embedding size is 768), but I have the following error when the linear coefficients are calculated : "matmul: Input operand 1 has a mismatch in its core dimension 0, with gufunc signature (n?,k),(k,m?)->(n?,m?) (size 4 is different from 768)"

(line 199 in cache_linear_coefs)

The error is corrected with linear_coef = embs @ coef_embs.T instead of linear_coef = embs @ coef_embs

Now I'm looking for a way to correctly predict the classes (I will probably modify _predict_cached function).

There are important files that Microsoft projects should all have that are not present in this repository. A pull request has been opened to add the missing file(s). When the pr is merged this issue will be closed automatically.

Microsoft teams can learn more about this effort and share feedback within the open source guidance available internally.

Merge this pull request