Notes, essays & the occasional brew day

Beer

Copper Mosaic IPA: My Latest Homebrewing Adventure

A five-gallon Mosaic IPA experiment, an enthusiastic fermentation, and a beer that tasted suspiciously like a stout.

Data Science · Development

Fine Tuning Justice: The Role of Pre-Trained LLMs in Enhancing Federal Investigations and Legal Procedures

Large language models (LLMs) represent the next generation of artificial intelligence applications, attaining widespread attention and adoption. These models demand substantial energy and resources for training. Consequently, there has been a shift towards developing pre-trained models like bidirectional encoder representations from transformers (BERT), utilizing millions of parameters from training texts to create general use models. An evolution of this approach involves fine-tuning models with domain-specific data, enhancing their utility in fields such as medicine, law, and science.

Data Science

Optimizing Gradient Boosting Models

Gradient boosting classifier models are a powerful type of machine learning algorithm that outperform many other types of classifiers. In simplest terms, gradient boosting algorithms learn from the mistakes they make by optmizing on gradient descent. A gradient boosting model values the gradient descent, or the direction of the steepest increase of a function, to make adjustments so that the function can increase rapidly over each iteration. Gradient boosting models can be used for classfication or regression. This post will concern itself with gradient boosting classification. In particular, we will study and optimize three gradient boosting models, AdaBoost (sklearn), XGBoost, and Gradient Boosting Classifier (sklearn). We will study these classifiers with the Kaggle data set, Kaggle: Diabetes Health Indicators Dataset . In particular, we will use the binary data set. I will not be doing any cleansing or engineering of the data. I am strictly experimenting with the effects of the learning rate and number of estimators hyperparameters on gradient boosting model performance.

Data Science

Traditional Methods vs AI Methods in Forecasting

Forecasting remains an indispensable tool for businesses to make informed decisions about future trends, demands, and opportunities. Accurate forecasts can lead to better resource allocation, inventory management, and strategic planning. When it comes to forecasting, data scientists have many tools at their disposal, ranging from traditional and time-tested methods to advanced artificial intelligence (AI) techniques. While AI methods have gained significant attention in recent years, traditional forecasting methods like Moving Average, Exponential Moving Average, and Linear Regression Forecasting still hold their ground in many real-world scenarios. As a data scientist with substantial experience in machine learning and big data analytics, I will shed light on why, in most cases, traditional forecasting methods remain preferable to their more advanced counterparts.

Lean Management Style

As a data science team leader, I understand the importance of effective time management. Data-driven projects require careful planning, focused execution, and constant collaboration. However, one common obstacle that can hinder productivity is the abundance of time-wasting meetings. To address this challenge, I have adopted a management style that minimizes unnecessary meetings, allowing my team to maximize their productivity and focus on delivering results. In this blog post, I will share my approach, emphasizing the significance of streamlined communication and outlining the key standing meetings that have proven invaluable in managing my small data science team.

3d printing

3D Printing

3D printing has revolutionized the manufacturing industry in the last few years. This technology enables individuals and businesses to create objects from a digital model using a 3D printer. The process of fused deposition modeling, or FDM for short, 3D printing involves adding layers of material until the desired object is created. This technology has become increasingly popular and is being used in a variety of fields including medicine, architecture, and engineering. Most hobbyist 3D printers, like myself, use a variety of plastics. Most often, PLA as it is economical and easy to use. Some other plastics like ABS and TPU are used as well.

Social Media

Twitch Streaming

Twitch, parent company Amazon, is the leading streaming platform on the web reaching over 7.03 million users in 2022. The vast majority of streamers using the Twitch service are streaming video game gameplay or “Just Chatting”. Among the video games categories you can find almost any video game imaginable. Amongst the “Just Chatting” category you will find streamers that primarily interact with chat with any combination of engaging gimmicks. As a long time Twitch user, I have recently begun experimenting with what I hope will be a unique interactive and engaging concept that will allow Twitch users that are interested in STEM concepts like engineering, mathematics, and technology to participate in my interest by watching and participating in my stream. My first few streams have taken place this week as I set out on a mission to become a Twitch Affiliate. A Twitch Affiliate is a streamer that has reached several goals, such as follwer count, regular streaming, and an average user count above a defined minimum threshold.

Social Media

Migrating to Mastodon from Twitter

Like many people, I’ve made the decision to migrate my sparce social media presence from Twitter to Mastodon. I am a big advocate of open source. I think the fediverse, as the open-source Mastodon ecosystem has been referred to, is a flexible and very useful system for a well-curated social media presence. It is a little more involved to get started, so here are some tips.

Beer

My First Brew Day

Today was a big day for me. A day I’ve been thinking about and planning for a long time. My first beer brewing day! And I want to tell anyone that might be interested in learning the ancient art of brewing that it was not hard at all. There is a time commitment of several hours but the project I have been intimidated by for years was actually not difficult.

Productivity

My Morning Ritual

I have a productivity ritual and set of habits I follow most mornings to get my ready to be productive and I wanted to share them. I live in the Eastern Time Zone while most of my coworkers work from the Central Time Zone. Because we work remotely, I have a set of rituals that I follow that I believe replace my commute and office arrival with more productive activities. In the before days (before COVID), I would drive about 20 minutes, sometimes grabbing a breakfast sandwich on the way. When I got to the office, I would get the lights turned on, set up my workspace, and make some coffee. This was a very natural set of events that I followed 4-5 days a week without much deviation.

Linux

Distro Hopping

I’ve been primarily using different flavors of Linux personally and professionally since around 2014. For most of those years, I’ve been using Ubuntu. Recently, I got the chance to upgrade my company laptop after 5 years on a Macbook Pro. I opted for a Dell XPS 13 Developer Edition shipped with Ubuntu. For the first few weeks, I kept Ubuntu but noticed a lot of issues with my docking and random video freezes that requiered a hard reboot. I finally reached my limit and decided to switch Linux distributions.

Introduction

My First Post

If you are interested in programming, data analytics, 3d printing, beer, and other endeavors, you have just arrived at the right place. I will be using my blog to discuss my interests and experience. I am the Data and Analytics Manager at a startup specializing in big data analytics and machine learning approaches to solving fraud problems. Before I worked in big data and analytics, I worked in the emergency services. Specifically, I was an Emergency 911 dispatcher for a county in Tennessee for 5 years. I volunteered at my local fire department as a firefighter and first responder/EMT. After five years, I made the decision to go to college for nursing while also getting my paramedic license so I could be a Vanderbilt LifeFlight nurse. While attending my night classes at Nashville State Community College, I discovered a love for math and physics. Four years later, I graduated from Austin Peay State University in Clarksville, TN with my Bachelor of Science in Applied Mathematics. Shortly after, I landed my first startup job as a Data Analyst. In 2021, I graduated from Northern Illinois University with my Master of Science in Data Analytics.