Text correction

We do not use a leaky ReLU in Generator.

Author: Abdullah-Jannadi

dcgan-svhn/DCGAN_Exercise.ipynb

@@ -251,7 +251,7 @@
     "\n",
     "What's new here is we'll use transpose convolutional layers to create our new images. \n",
     "* The first layer is a fully connected layer which is reshaped into a deep and narrow layer, something like 4x4x512. \n",
-    "* Then, we use batch normalization and a leaky ReLU activation. \n",
+    "* Then, we use batch normalization and ReLU activation. \n",
     "* Next is a series of [transpose convolutional layers](https://pytorch.org/docs/stable/nn.html#convtranspose2d), where you typically halve the depth and double the width and height of the previous layer. \n",
     "* And, we'll apply batch normalization and ReLU to all but the last of these hidden layers. Where we will just apply a `tanh` activation.\n",
     "\n",