TensorFlow 1.x Deep Learning Cookbook

How to do it...

Here is how is how we proceed with the recipe:

1. Clone the code from github under an Apache Licence:

git clone https://github.com/naturomics/CapsNet-Tensorflow.git $ cd CapsNet-Tensorflow

2. Download MNIST and create the appropriate structure:

mkdir -p data/mnistwget -c -P data/mnist \\http://yann.lecun.com/exdb/mnist/{train-images-idx3-ubyte.gz,train-labels-idx1-ubyte.gz,t10k-images-idx3-ubyte.gz,t10k-labels-idx1-ubyte.gz}gunzip data/mnist/*.gz

3. Start the training process:

python main.py

4. Let's see the code used for defining a capsule. Each capsule takes a 4D tensor as input and returns a 4D tensor. It is possible to define a capsule as either a fully connected network (the DigiCaps) or a convolutional one (the Primary Capsules). ...