Kevin P Murphy

yes, sounds great. You could even have a unit test like reproducing some of the demos below https://github.com/probml/pyprobml/blob/master/notebooks/book1/04/beta_binom_post_plot.ipynb https://github.com/probml/pyprobml/blob/master/notebooks/book1/04/beta_binom_post_pred_plot.ipynb https://github.com/probml/pyprobml/blob/master/notebooks/book2/03/gauss_seq_update_sigma_1d.ipynb https://github.com/probml/pyprobml/blob/master/notebooks/book1/03/gauss_infer_2d.ipynb [image: Screenshot 2024-11-18 at 10.25.44 AM.png] On Mon, Nov...

Thank you for you kind words. @patel-zeel helped me a lot with it.

that's weird. I cannot see it on my pdf version, either locally or on the web.

"Should the kmeans implementation be marked as reserved for internal use only?" No! K-means in jax could be of independent interest, so it is worth making it a first class...

Also on p56-p57 there is an inconsistency between y-hat meaning logits (colored equation for loss(sm(yhat), y)) and y-hat meaning probabilities (code snippet, y_hat = F.softmax(logits)). Maybe call the latter p_hat?