Andrewwango

Andrewwango's repositories

open-set-resnet

Open set classification of car models. This 3-step classifier solves the problem where dogs are classified as cars, by first filtering these images out using ResNet CNNs transfer-trained on different datasets.

Language:Jupyter Notebook8 40

dynamic_vaes

Language:Jupyter Notebook2 40

gpx-overpass-googlemaps-tutorial

Repo hosting blog "How to find places along a GPX route in Python using Overpass API and Google Maps Places API"

Language:Jupyter NotebookApache-2.02 20

perspective-equivariant-imaging

Repository for Perspective-Equivariant Imaging: Unsupervised Framework for Multispectral Pansharpening

Language:Jupyter NotebookBSD-3-Clause200

andrewwango.github.io

Language:HTMLGPL-3.01 30

aws-custom-calls

An AWS serverless AI services pipeline for analysing call center calls, with custom analytics built around AWS Transcribe and AWS Comprehend and insights published on AWS Quicksight.

Language:Python1 30

deepinv-experiments

Template for deep learning experiments using the DeepInverse library

Language:PythonBSD-3-Clause100

femda

FEMDA: Robust classification with Flexible Discriminant Analysis in heterogeneous data. Flexible EM-Inspired Discriminant Analysis is a robust supervised classification algorithm that performs well in noisy and contaminated datasets.

Language:PythonMIT1 30

govuk-accessible-search

An intelligent search bar to improve accessibility on GOV.UK/NHS.UK pages using OpenAI

Language:CSS1 5 54

gpr-mcmc

Gaussian Process Regression (GPR) with non-Gaussian likelihoods using robust infinite-dimension Monte Carlo Markov Chain (MCMC) sampling for spatial inference problems

Language:Jupyter NotebookApache-2.01 30

neo4j-demos

Demonstrating and visualising graph data science and ML using neo4j and GDS with the Northwind and Cora datasets

Language:Jupyter Notebook1 20

143-hackcambridge

IoT Waste Management project for 2020 Hack Cambridge

Language:Python040

4M24

Language:Jupyter Notebook020

AMT

Language:MATLAB030

prow-map

Comparing public rights of way (PRoW) with public recorded GPS activity data to support campaigns reclaiming public rights of way using Python geospatial libraries.

Language:HTMLGPL-3.0000