From bdd246bf07bfef58542e93e361384b7814a0f862 Mon Sep 17 00:00:00 2001
From: GarrettJenkinson <GarrettJenkinson@users.noreply.github.com>
Date: Mon, 27 Jun 2022 09:27:51 -0500
Subject: [PATCH] Fixing wide-and-deep to use normalized inputs

Thanks for this great book/repo. I noticed the wide and deep example is using non-normalized inputs which leads to performance degradation.
---
 10_neural_nets_with_keras.ipynb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/10_neural_nets_with_keras.ipynb b/10_neural_nets_with_keras.ipynb
index c48f989..fea18ab 100644
--- a/10_neural_nets_with_keras.ipynb
+++ b/10_neural_nets_with_keras.ipynb
@@ -1568,7 +1568,7 @@
     "normalized = normalization_layer(input_)\n",
     "hidden1 = hidden_layer1(normalized)\n",
     "hidden2 = hidden_layer2(hidden1)\n",
-    "concat = concat_layer([input_, hidden2])\n",
+    "concat = concat_layer([normalized, hidden2])\n",
     "output = output_layer(concat)\n",
     "\n",
     "model = tf.keras.Model(inputs=[input_], outputs=[output])"