From 7161590fe306578ad02d851e74e1141bd39004af Mon Sep 17 00:00:00 2001
From: Damon Gwinn <15755242+gwinndr@users.noreply.github.com>
Date: Wed, 25 May 2022 20:01:12 -0400
Subject: [PATCH] Fix setting tgt when using a dataset with a performance less
 than the max sequence length

---
 dataset/e_piano.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dataset/e_piano.py b/dataset/e_piano.py
index a43e67d..024e564 100644
--- a/dataset/e_piano.py
+++ b/dataset/e_piano.py
@@ -89,7 +89,7 @@ def process_midi(raw_mid, max_seq, random_seq):
     if(raw_len < full_seq):
         x[:raw_len]         = raw_mid
         tgt[:raw_len-1]     = raw_mid[1:]
-        tgt[raw_len]        = TOKEN_END
+        tgt[raw_len-1]      = TOKEN_END
     else:
         # Randomly selecting a range
         if(random_seq):