From db82ce551bc4a0a3019f3769068e1fbdfad4bc5c Mon Sep 17 00:00:00 2001
From: kismat02 <31696003+kismat02@users.noreply.github.com>
Date: Sun, 6 Nov 2022 00:33:21 +0400
Subject: [PATCH] Update warp_loss.rst

Fix typo
---
 doc/examples/warp_loss.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/examples/warp_loss.rst b/doc/examples/warp_loss.rst
index 5985623..6da8e3b 100644
--- a/doc/examples/warp_loss.rst
+++ b/doc/examples/warp_loss.rst
@@ -17,7 +17,7 @@ the form of the k-OS WARP loss, also implemented in LightFM.
 
 Like the BPR model, WARP deals with (user, positive item, negative item)
 triplets. Unlike BPR, the negative items in the triplet are not chosen
-by random sampling: they are chosen from among those negatie items which
+by random sampling: they are chosen from among those negative items which
 would violate the desired item ranking given the state of the model.
 This approximates a form of active learning where the model selects
 those triplets that it cannot currently rank correctly.