August 2017

Word2Vec Keras - negative sampling architecture

A Word2Vec Keras tutorial

Understanding Word2Vec word embedding is a critical component in your machine learning journey.  Word embedding is a necessary step in performing efficient natural language processing in your machine learning models.  This tutorial will show you how to perform Word2Vec word embeddings in the Keras deep learning framework – to get an introduction to Keras, check out… Read More »A Word2Vec Keras tutorial

TensorFlow queuing - example queue

An introduction to TensorFlow queuing and threading

One of the great things about TensorFlow is its ability to handle multiple threads and therefore allow asynchronous operations.  If we have large datasets this can significantly speed up the training process of our models.  This functionality is especially handy when reading, pre-processing and extracting in mini-batches our training data.  The secret to being able to… Read More »An introduction to TensorFlow queuing and threading

PyTorch tutorial - fully connected neural network example architecture

A Microsoft CNTK tutorial in Python – build a neural network

In previous tutorials (Python TensorFlow tutorial, CNN tutorial, and the Word2Vec tutorial) on deep learning, I have taught how to build networks in the TensorFlow deep learning framework.  There is no doubt that TensorFlow is an immensely popular deep learning framework at present, with a large community supporting it.  However, there is another contending framework which… Read More »A Microsoft CNTK tutorial in Python – build a neural network