Category Archives: CNTK

Posts about the Microsoft CNTK deep learning code library

The Multinomial Distribution

The multinomial probability distribution occurs surprisingly (to me) often in machine learning scenarios. Suppose you have an American roulette wheel. There are a total of 38 slots for the ball to land in: 18 red, 18 black, 2 green (the … Continue reading

Posted in CNTK, Machine Learning | Leave a comment

Exploring Neural Network Input-Output using CNTK

The CNTK machine learning code library has become one of my most-used tools. I set out to explore the neural network input-output mechanism using CNTK. My goal was to completely understand how to create a network, assign values to the … Continue reading

Posted in CNTK, Machine Learning | Leave a comment

When to Apply Softmax on a Neural Network

Suppose you want to predict the political party affiliation (democrat, republican other) of a person based on age, income, and education. A training data set for this problem might look like: 32 48 14 0 1 0 24 28 12 … Continue reading

Posted in CNTK, Machine Learning | Leave a comment

Recap of the 2017 DevIntersection Conference

I recently spoke at (and attended talks at) the 2017 DevIntersection Conference. The event was held October 29 to November 3, 2017, at the MGM Grand hotel in Las Vegas. There are actually several DevIntersection conferences every year, but the … Continue reading

Posted in CNTK, Conferences, Machine Learning | Leave a comment

Protected: My Presentation at the 2017 DevIntersection Conference

There is no excerpt because this is a protected post.

Posted in CNTK, Conferences, Machine Learning

Making a Prediction from an MNIST Image Classification Model

I’ve been taking a deep look at the MNIST image classification problem. So far I wrote a utility function to convert the raw MNIST dataset images from their native binary format into text files. Then I used the CNTK code … Continue reading

Posted in CNTK, Machine Learning

The MNIST Image Classification Problem using CNTK

One of the standard deep learning problems is to classify the MNIST handwritten digits dataset. I took a deep dive into this problem, using the Microsoft CNTK deep learning code library. The MNIST dataset consists of 60,000 images used to … Continue reading

Posted in CNTK, Machine Learning | 1 Comment