Count tokens per class with Laplace smoothing. Work in log space to avoid multiplying tiny probabilities; repeated tokens contribute repeatedly.

import math
print(math.log(0.5) + math.log(3/4))

Count tokens per class with Laplace smoothing. Work in log space to avoid multiplying tiny probabilities; repeated tokens contribute repeatedly. The trace reads the local variables of your function as it executes. Change one input or parameter, predict the result, then run the checks again. A small deterministic example makes arithmetic mistakes easier to locate than a long training run. State the shape and meaning of the returned value before coding. Keep preprocessing, evaluation and model updates in their proper order, and examine the boundary case in each check. The companion lab lets you replay intermediate values while the same algorithm runs.

log_scores(documents, labels, query, vocabulary) returns log posterior scores (without normalization) for classes 0 and 1. Both classes have documents; vocabulary is nonempty and unique. Ignore query tokens outside vocabulary, use alpha=1 and empirical class priors.

The explanation opens here once every check passes.