From 36136cdd13bacd3413726335f43c2de98ef8018c Mon Sep 17 00:00:00 2001
From: Tijs van der Velden <47599600+safersephy@users.noreply.github.com>
Date: Sun, 8 Sep 2024 15:44:21 +0200
Subject: [PATCH] corrected multiclass to multilabel

---
 notebooks/2_convolutions/01_losses.ipynb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/notebooks/2_convolutions/01_losses.ipynb b/notebooks/2_convolutions/01_losses.ipynb
index 282b60af..04e4e28a 100644
--- a/notebooks/2_convolutions/01_losses.ipynb
+++ b/notebooks/2_convolutions/01_losses.ipynb
@@ -457,7 +457,7 @@
     "\n",
     "- For regression models, typically use a MSE\n",
     "- For classification, use CrossEntropyLoss (note: this might be implemented different in other libraries like Tensorflow!)\n",
-    "- For multiclass, use BinaryCrossEntropy\n",
+    "- For multilabel, use BinaryCrossEntropy\n",
     "\n",
     "There are other, more complex losses for more complex usecases but these three will cover 80% of your needs."
    ]