Over the past few weeks, I’ve been looking at time series regression where you have data associated with times and you want to predict the next value. My standard example is airline passenger count data each month, from January 1949 through December 1960. Using that data, the goal is to create a model that predicts […]

# Deep Learning in Real Time – Inference Acceleration and Continuous Training

Introduction Deep learning is revolutionizing many areas of computer vision and natural language processing (NLP), infusing into increasingly more consumer and industrial products intelligence capabilities with the potential to impact the everyday experience of people and the standard processes of industry practices. On a high level, deep learning, similar to any automated system based on […]

# The Maclaurin Series and Machine Learning

In the very early days of computers (say the 1950s and 1960s), most guys who entered the new field of “computer science” came from a background in either mathematics or electrical engineering. There’s always been a strong connection between mathematics and computer science. More specifically, with regards to machine learning, every now and then I’ll […]

# Neural Network Momentum using Python

I wrote an article titled “Neural Network Momentum using Python” in the August 2017 issue of Visual Studio Magazine. See https://visualstudiomagazine.com/articles/2017/08/01/neural-network-momentum.aspx Momentum is a technique intended to speed up neural network training. Training a neural network is the process of determining the values of the weights and biases that essentially define the behavior of the […]

# Using Random Forests and Wordclouds to Visualize Feature Importance in Document Classification

What characteristics do the works of famous authors have that make them unique? This post uses ensemble methods and wordclouds to explore just that. Project Gutenberg offers a large number of freely available works from many famous authors. The dataset for this post consists of books, taken from Project Gutenberg, written by each of the […]

# Replicator Neural Networks

A standard neural network classifier builds a model that predicts output values from input values. For example, the famous Iris Data has 150 items. Each item has four predictor variables (sepal length, sepal width, petal length, petal width) followed by one of three species to predict: setosa encoded as (1,0,0), versicolor encoded as (0,1,0), and […]

# Humans in the Loop

This guest blog post from Paco Nathan dives into how people and machines collaborating together to perform work is real and not science fiction. Paco Nathan is the Director, Learning Group at O’Reilly Media and an advisor for Amplify Partners. His expertise includes machine learning, distributed systems, and cloud computing. He was cited in 2015 as […]

# Time Series Regression using a Raw Python Neural Network

I’ve been looking at time series regression recently. Just for fun I coded up an example using a raw Python (with the NumPy library for numerical functions) neural network. For my example I used a standard benchmark data set that has the total number of airline passengers for the 144 months from January 1949 through […]

# Calculating Standard Deviation using custom UDF and group by in H2O

Here is the full code to calculate standard deviation using H2O group by method as well as using customer UDF: library(h2o) h2o.init() irisPath <- system.file(“extdata”, “iris_wheader.csv”, package = “h2o”) iris.hex <- h2o.uploadFile(path = irisPath, destination_frame = “iris.hex”) # Calculating Standard Deviation using h2o group by SdValue <- h2o.group_by(data = iris.hex, by = “class”, sd(“sepal_len”)) # […]

# Calculate mean using UDF in H2O

Here is the full code to write a UDF to calculate mean for a given data frame using H2O machine learning platform: library(h2o) h2o.init() ausPath <- system.file(“extdata”, “australia.csv”, package=”h2o”) australia.hex <- h2o.uploadFile(path = ausPath) # Writing the UDF myMeanUDF = function(Fr) { mean(Fr[, 1]) } # Applying UDF using ddply MeanValue = h2o.ddply(australia.hex[, c(“premax”, […]