From 2b0b8a1c1a793f5f88a5cacd60b0a637a8c54cf6 Mon Sep 17 00:00:00 2001
From: Pietro Monticone <38562595+pitmonticone@users.noreply.github.com>
Date: Fri, 13 Oct 2023 15:19:04 +0200
Subject: [PATCH] Update README.md

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 98cefe2..a56f954 100644
--- a/README.md
+++ b/README.md
@@ -27,7 +27,7 @@ There is, however, only partial support for multidimensional layers (convolution
 
 ## How it works
 
-There is a distinction between netwoks starting with a layer with fixed-size input (Dense and Recurrent) and networks starting with a layer with variable-size input (Convolutional, Pooling, and functional).
+There is a distinction between networks starting with a layer with fixed-size input (Dense and Recurrent) and networks starting with a layer with variable-size input (Convolutional, Pooling, and functional).
 
 In the former case, just passing a network `m = Chain(...)` to plot works, e.g. `plot(m)`. In the latter case, one needs to pass along an initial input `inp`, or input size `inpsz = size(inp)`, as the second argument, like `plot(m, inp)` or `plot(m, inpsz)`, so that the plot recipe can properly figure out the size of each layer.