From 71e56a56bf72bf5a105a447167fedc801628c9da Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Luis=20Coss=C3=ADo?= Date: Thu, 7 Dec 2023 18:05:05 -0300 Subject: [PATCH] min -> \min --- qdrant-landing/content/documentation/concepts/explore.md | 2 +- qdrant-landing/content/documentation/concepts/search.md | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/qdrant-landing/content/documentation/concepts/explore.md b/qdrant-landing/content/documentation/concepts/explore.md index 6b599ae0c..1de7b0baf 100644 --- a/qdrant-landing/content/documentation/concepts/explore.md +++ b/qdrant-landing/content/documentation/concepts/explore.md @@ -569,7 +569,7 @@ Conversely, in the absence of a target, a rigid integer-by-integer function does We can directly associate the score function to a loss function, where 0.0 is the maximum score a point can have, which means it is only in positive areas. As soon as a point exists closer to a negative example, its loss will simply be the difference of the positive and negative similarities. $$ -\text{context score} = \sum min(s(v^+_i) - s(v^-_i), 0.0) +\text{context score} = \sum \min(s(v^+_i) - s(v^-_i), 0.0) $$ Where $v^+_i$ and $v^-_i$ are the positive and negative examples of each pair, and $s(v)$ is the similarity function. diff --git a/qdrant-landing/content/documentation/concepts/search.md b/qdrant-landing/content/documentation/concepts/search.md index 1da93803e..a54693f21 100644 --- a/qdrant-landing/content/documentation/concepts/search.md +++ b/qdrant-landing/content/documentation/concepts/search.md @@ -1,4 +1,4 @@ - +--- title: Search weight: 50 aliases: