Links with relative paths

Author: havk64

08_Convolutional_Neural_Networks/01_Intro_to_CNN/readme.md

@@ -4,10 +4,10 @@ Convolutional Neural Networks (CNNs) are responsible for the latest major breakt
 
 In mathematics, a convolution is a function which is applied over the output of another function.  In our case, we will consider applying a matrix mutliplication (filter) across an image. See the below diagram for an example of how this may work.
 
-![Convolutional Filter](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/01_intro_cnn.png)
+![Convolutional Filter](../images/01_intro_cnn.png)
 
 CNNs generally follow a structure.  The main convolutional setup is (input array) -> (convolutional filter layer) -> (Pooling) -> (Activation layer).  The above diagram depicts how a convolutional layer may create one feature.  Generally, filters are multidimensional and end up creating many features.  It is also common to have a completely separate filter-feature creator of different sizes acting on the same layer.  After this convolutional filter, it is common to apply a pooling layer.  This pooling may be a max-pooling or an average pooling or another aggregation.  One of the key concepts here is that the pooling layer has no parameters- while decreasing the layer size.  See the below diagram for an example of max-pooling.
 
-![Convolutional Filter](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/01_intro_cnn2.png)
+![Convolutional Filter](../images/01_intro_cnn2.png)
 
-After the max pooling, there is generally an activation layer.  One of the more common activation layers is the ReLU (Rectified Linear Unit).  See [Chapter 1, Section 6](https://github.com/nfmcclure/tensorflow_cookbook/tree/master/01_Introduction/06_Implementing_Activation_Functions) for examples.
+After the max pooling, there is generally an activation layer.  One of the more common activation layers is the ReLU (Rectified Linear Unit).  See [Chapter 1, Section 6](../../01_Introduction/06_Implementing_Activation_Functions) for examples.

08_Convolutional_Neural_Networks/02_Intro_to_CNN_MNIST/readme.md

@@ -5,11 +5,11 @@ Here we illustrate how to use a simple CNN with three convolutional units to pre
 
 When the script is done training the model, you should see similar output to the following graphs.
 
-![Loss and Accuracy](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/02_cnn1_loss_acc.png "Loss and Accuracy")
+![Loss and Accuracy](../images/02_cnn1_loss_acc.png "Loss and Accuracy")
 
 Training and test loss (left) and test batch accuracy (right).
 
 
-![Sample Test Batch](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/02_cnn1_mnist_output.png "Sample of 6 Images")
+![Sample Test Batch](../images/02_cnn1_mnist_output.png "Sample of 6 Images")
 
 A random set of 6 digits with actuals and predicted labels. You can see a prediction failure in the lower right box.

08_Convolutional_Neural_Networks/03_CNN_CIFAR10/readme.md

@@ -5,6 +5,6 @@ Here we will build an convolutional neural network to predict the CIFAR-10 data.
 
 The script provided will download and unzip the CIFAR-10 data.  Then it will start training a CNN from scratch.  You should see similar output at the end to the following two graphs.
 
-![Loss and Accuracy](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/03_cnn2_loss_acc.png)
+![Loss and Accuracy](../images/03_cnn2_loss_acc.png)
 
 Here we see the training loss (left) and the test batch accuracy (right).

08_Convolutional_Neural_Networks/05_Stylenet_NeuralStyle/readme.md

@@ -5,10 +5,8 @@ The purpose of this script is to illustrate how to do stylenet in TensorFlow.  W
 
 ## Prerequisites
  * Download the VGG-verydeep-19.mat file [here](http://www.vlfeat.org/matconvnet/models/beta16/imagenet-vgg-verydeep-19.mat).
- * You must download two images, a [style image](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/starry_night.jpg) and a [content image](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/book_cover.jpg) for the algorithm to blend. (Image links are to the images used in the book.)
+ * You must download two images, a [style image](../images/starry_night.jpg) and a [content image](../images/book_cover.jpg) for the algorithm to blend. (Image links are to the images used in the book.)
 
 The algorithm will output temporary images during training.
 
-![Stylenet Example](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/05_stylenet_ex.png)
-
-
+![Stylenet Example](../images/05_stylenet_ex.png)

08_Convolutional_Neural_Networks/06_Deepdream/readme.md

@@ -5,7 +5,7 @@ Note: There is no new code in this script.  It originates from the TensorFlow tu
 
 Here are some potential outputs.
 
-![deepdream outputs](https://github.com/nfmcclure/tensorflow_cookbook/blob/master/08_Convolutional_Neural_Networks/images/06_deepdream_ex.png)
+![deepdream outputs](../images/06_deepdream_ex.png)
 > Deepdream results on four features with the book's cover image.
 
 Our interpretation so far: