From 4cc85b7cc47a6b4c39f2fd2de918b3a30167c380 Mon Sep 17 00:00:00 2001
From: WANG-Yuchen-Alice <45040630+WANG-Yuchen-Alice@users.noreply.github.com>
Date: Tue, 22 Feb 2022 16:52:09 +0800
Subject: [PATCH] Correct a typo :)

---
 workspace/uncond/Experiments.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/workspace/uncond/Experiments.md b/workspace/uncond/Experiments.md
index 895c05f..6be3d06 100644
--- a/workspace/uncond/Experiments.md
+++ b/workspace/uncond/Experiments.md
@@ -102,4 +102,4 @@ For the CP+Linear setting, it only takes less than half minutes to generate a so
 We find that the generated pieces of REMI-XL tend to stick to some patterns and occasionally fall to loop them for a quite long time, or even the entire song. The quality within the loop is suprisingly organized (clearly arpeggio in left hand, melody line in right hand), but also a bit of tedious. The samples from CP-linear have a rather different texture, the "structure" diversity is richer but it's also more aggressive in selecting pitches.
 
 ### About EOS
-It turns out that REMI-XL failed in generating EOS sequence, which implies the sequence length might exceed the "effective length" becuase of it's rnn nature. In fact, we also tried remi+linear and cp+linear, and both of them success in this criterion.
+It turns out that REMI-XL failed in generating EOS sequence, which implies the sequence length might exceed the "effective length" becuase of it's rnn nature. In fact, we also tried remi+linear and cp+linear, and both of them succeed in this criterion.