From 52fbfc6b9bf7aae292466e2e7b67d3d419da59dd Mon Sep 17 00:00:00 2001 From: Yuchen Pei Date: Sun, 2 Dec 2018 22:17:29 +0100 Subject: minor change --- posts/2018-12-02-lime-shapley.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/posts/2018-12-02-lime-shapley.md b/posts/2018-12-02-lime-shapley.md index 1f1fe34..5ccf701 100644 --- a/posts/2018-12-02-lime-shapley.md +++ b/posts/2018-12-02-lime-shapley.md @@ -76,8 +76,8 @@ feature contributions of supervised learning models locally. Let $f: X_1 \times X_2 \times ... \times X_n \to \mathbb R$ be a function. We can think of $f$ as a model, where $X_j$ is the space of -$j$th feature. For example, in a language model, $X_j$ may be the count -of the $j$th word in the vocabulary. +$j$th feature. For example, in a language model, $X_j$ may correspond to +the count of the $j$th word in the vocabulary. The output may be something like housing price, or log-probability of something. -- cgit v1.2.3