statsgary Goto Github PK

followers: 121.0 following: 115.0 repos: 267.0 gists: 148.0

Name: Gary Hutson

Type: User

Company: Crisp Thinking / Google Certified Tensorflow Developer / NHS-R Community Senior Fellow / Kaggler

Bio: Head of Machine Learning | Former Head of Data Science / Senior DS. Loves: Python, R, ML/DL, modelling , Stats, Kaggle, anything data and my little dude.

Location: Watnall, Nottinghamshire

Blog: http://hutsons-hacks.info/

A little about me

📕 Co-author of Graph Data Modeling in Python.
📈 I'm currently working on a number of Machine Learning projects utilising Transformers for natural language text modelling of some of my organisation's key aims around protecting people online against bad actors and bad actor groups
🎓 I have a degree in applied business statistics and an MBA in Business & Accountancy (with modules in OR and Statistics). A fan of statistics, machine learning and anything that can be used to solve real world problems.
🌱 I’m currently learning how to use Google Cloud Platform to its greatest potential, as well as pressing on with some Python package and class development, as well as Named Entity Recognition token classification and generative AI.
👯 I’m looking to collaborate on R or Python package development, as I have 8 active packages on CRAN and have two packages on PyPi. I would be happy to take you through the steps needed to get there.
Creator and author of several R packages namely OddsPlotty, ConfusionTableR, NHSDataDictionaRy, FeatureTerminatoR and MLDataR. I have collaborated on a couple more, namely NHSRdatasets and SangerTools. Authored the packages on PyPI Torchtilities (utilities for working with PyTorch) and I am working on a general sci-kit learn utilities package.
💻 - specialised in R and Python for data science, machine learning and statistics. Object orientated knowledge in C# for Windows Forms design and databases. Former web developer in technologies HTML, Javascript, PHP and CSS (also know a little Node.JS, enough to be dangerous). I find learning new programming languages interesting, the same way polyglots do learning new spoken languages. ML and Deep Learning skillsets in Computer Vision Classification, Object Detection, Semantic Segmentation and OCR, supervised and unsupervised methods and natural language processing, currently working on pretraining and finetuning Transformer models.
📑 - blogger and website owner of hutsons-hacks.
📧 how to reach me - @StatsGary - you can message me, follow me, or hit me up on my Twitter channel.

Gitty statistics

Here is a summary of the key Git statistics:

Gary Hutson's Projects

controlnet-playground

A repository to show the power of the ControlNet framework

correlationplot

A simple function to build correlation plots

ctgan

Conditional GAN for generating synthetic tabular data.

curated-transformers

:robot: A PyTorch library of curated Transformer models and their composable components

Designed a CycleGAN model for image-to-image translation from unpaired training dataset using an unsupervised approach. The objective was to train generators that learn to transform an image from domain X into an image that looks like it came from domain Y (and vice versa).

data

Samples of data used in tutorials

data-engineer-learning-path

data-engineering

Repository is to maintain the artifacts and codes for demo purposes

data-engineering-with-aws

Data Engineering with AWS, Published by Packt

data.tabletutorial

A tutorial on the data.table package

deep-learning-drizzle

Drench yourself in Deep Learning, Reinforcement Learning, Machine Learning, Computer Vision, and NLP by learning from these exciting lectures!!

deep-reinforcement-learning-with-python

Master classic RL, deep RL, distributional RL, inverse RL, and more using OpenAI Gym and TensorFlow with extensive Math

detoxify

Trained models & code to predict toxic comments on all 3 Jigsaw Toxic Comment Challenges. Built using ⚡ Pytorch Lightning and 🤗 Transformers. For access to our API, please email us at [email protected].