Jack Kelly

neuralnilm. Deep Neural Networks Applied to Energy Disaggregation

climate-change-mitigation. Using computer science to mitigate climate change: lists of topics, conferences, etc.

light-speed-io. Read & decompress many chunks of files at high speed

neuralnilm_prototype. Jupyter Notebook

hypergrib. Lazily open petabyte-scale GRIB datasets consisting of trillions of GRIB messages from xarray

bayesianchangepoint. An implementation of Adams & MacKay 2007 "Bayesian Online Changepoint Detection"

rfm_edf_ecomanager. Communicate with EDF EcoManager Transmitter Plugs; Whole House TX and CurrentCost Whole House TX using a 433MHz RFM12b and a ATmega328 micro controller

lets_make_it_super_easy_to_use_weather_forecast_data. Let's make it super easy to use weather forecast data

UK-DALE_metadata. Metadata for my UK Domestic Appliance-Level Electricity (UK-DALE) dataset

rfm_ecomanager_logger. Python front-end for the rfm_edf_ecomanager RF base unit

ukdale_plots. Plots (using NILMTK) for my UK-DALE dataset paper

snd_card_power_meter. An electricity meter using a sound card to digitise signals from current transformer probes and AC-AC adapters.

linux_home. Config files in my Linux home directory

BuildSys_2015_NeuralNILM. Presentation of Neural NILM for BuildSys 2015 conference in November 2015

reference-library. My academic reference library in BibLaTeX format

rfm_current_cost. An attempt to build a DIY receiver for Current Cost transmitters

currentCostCosmTX. Simple script to send data from a Current Cost to Cosm

cosmSender. Python class to send data to Cosm. Caches data to minimise API requests and to protect against network errors.

PGM. Helper functions for free online Stanford Probabilistic Graphical Modelling course

domesticPowerData. Electricity power consumption data recorded from two domestic houses

slicedpy. SlicedPy: Estimate the energy consumed by individual appliances from whole-house power meter readings.

iam_logger. Current Cost Individual Appliance Monitor (IAM) logger. Handles multiple current cost monitors using multiple threads.

ukdale_utils. Utilities for UK-DALE dataset

LaTeX. LaTeX bits and bobs

pda. Power Data Analysis: finding correlations and patterns in domestic electricity data

StressAndLog. Stress a Linux machine whilst logging utilisation and power consumption using a WattsUp meter

JackKelly.github.io. My new blog; using Jekyll.

awesome-public-datasets. An awesome list of high-quality open datasets in public domains (on-going).

matplotlib. Python

babysitter. Monitor disk space, multiple files and multiple processes. If errors are found then an email is sent and corrective action may be taken.

CurrentCostLogger. A simple Perl script for logging data from the Current Cost EnviR

neuralnilm_plots. Generating figures for illustrating Neural NILM

EFL_talk. To view this presentation in your browser, go to:

migration-from-drupal-to-jekyll. Move blog from Drupal 7 to Jekyll

clean_uk_pv. Cleaning the solar power dataset "uk_pv"

NILM_Workshop_talk_2016. Slides for my talk on "Does disaggregated electricity feedback reduce domestic electricity consumption? A systematic review of the literature"

mechanic. My Eclipse (with PyDev & Aptana) settings (as saved by Workspace Mechanic)

Sustainable-OSM-maps. Tools to create pretty maps of sustainable living features from OpenStreetMap

UCL_effectiveness_of_disag_talk. Slides for my talk on "Do disaggregated electricity bills really help people to save energy?"

powerstats. Produce simple stats for REDD-formatted electricity power consumption log files

owfsCosmTX. Transmit readings from a 1-wire network to Pachube

Electricity-load-disaggregation. Smart meter load disaggregation with Hidden Markov Models