From 08e3f94357723a7ec5d5459b5cc10200b68e656b Mon Sep 17 00:00:00 2001 From: venture_kwok Date: Wed, 11 Sep 2024 20:14:02 -0700 Subject: [PATCH] Correct dropout p definition based on test cases --- hw2.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw2.ipynb b/hw2.ipynb index 5616e96..a1a5acf 100755 --- a/hw2.ipynb +++ b/hw2.ipynb @@ -491,7 +491,7 @@ "**Important**: If the Dropout module the flag `training=False`, you shouldn't \"dropout\" any weights. That is, dropout applies during training only, not during evaluation. Note that `training` is a flag in `nn.Module`.\n", "\n", "##### Parameters\n", - "- `p` - the probability of an element to be zeroed.\n", + "- `p` - the probability of an element to be kept.\n", "\n", "___" ]