All Posts About

Energize Andover Summer 2017 Blog

Matt Rossman

Search results for

  • Jun 6
    Day 1

    Introduction

    . . . (read more)

  • Jun 7
    Day 2

    Initial distribution analysis

    . . . (read more)

  • Jun 8
    Day 3

    Night distribution continued

    . . . (read more)

  • Jun 9
    Day 4

    Median absolute deviations

    . . . (read more)

  • Jun 12
    Day 5

    Daytime usage first look

    . . . (read more)

  • Jun 13
    Day 6

    Time filtering

    . . . (read more)

  • Jun 14
    Day 7

    Generating complex time ranges with ical files, and improvements to time filtering

    . . . (read more)

  • Jun 15
    Day 8

    Debugging the calendar parser

    . . . (read more)

  • Jun 16
    Day 9

    Rolling windows intro

    . . . (read more)

  • Jun 19
    Day 10

    Rolling MAD, switching to time based windows

    . . . (read more)

  • Jun 20
    Day 11

    Centering time based windows, temperature data intro

    . . . (read more)

  • Jun 21
    Day 12

    Night data plotting, temperature data update

    . . . (read more)

  • Jun 22
    Day 13

    Percentage bounds and density plots

    . . . (read more)

  • Jun 23
    Day 14 - Summary

    Overview of my first couple of weeks, and a guide on how to make a blog

    . . . (read more)

  • Jun 26
    Day 15

    Team meeting, researching distribution functions and tests

    . . . (read more)

  • Jun 27
    Day 16

    Settling the null values and time gaps, looking at more filtered sample density estimations

    . . . (read more)

  • Jun 28
    Day 17

    Daily power trends and couple of stats test resources

    . . . (read more)

  • Jun 29
    Day 18

    Scipy, calculating the best fit distribution

    . . . (read more)

  • Jun 30
    Day 19

    Trapezoidal approximation of energy usage from power data

    . . . (read more)

  • Jul 3
    Day 20

    Team meeting with guest Viraj

    . . . (read more)

  • Jul 5
    Day 21

    Going over examples with the team

    . . . (read more)

  • Jul 6
    Day 22

    Working with the team (cont.)

    . . . (read more)

  • Jul 7
    Day 23

    Fitting the lognorm distribution

    . . . (read more)

  • Jul 10
    Day 24

    Abstracting the fit implementation, trouble with likelihood, and a way to compare quantiles

    . . . (read more)

  • Jul 11
    Day 25

    Plotting and comparing sample/model quartiles

    . . . (read more)

  • Jul 12
    Day 26

    Quantile translation for an adjusted data set

    . . . (read more)

  • Jul 13
    Day 27

    Tour of AHS, sample adjustment function and testing

    . . . (read more)

  • Jul 14
    Day 28

    More data to work with

    . . . (read more)

  • Jul 17
    Day 29

    Some Bancroft plots, time filter KeyError fix, and a promising lead on nonlinear regression

    . . . (read more)

  • Jul 18
    Day 30

    The holy grail of smart energy resources

    . . . (read more)

  • Jul 19
    Day 31

    Nonlinear regression lecture (polynomials, RBFs, kernels), testing out polynomial modeling on temperature data

    . . . (read more)

  • Jul 20
    Day 32

    Some remarks on least squares polynomial regression and an amazing machine learning library

    . . . (read more)

  • Jul 21
    Day 33

    Regression on higher dimensional input, 3D plotting with mplot3d

    . . . (read more)

  • Jul 24
    Day 34

    Regularized estimators

    . . . (read more)

  • Jul 25
    Day 35

    Distribution of the residuals, robust estimators, comprehending and implementing normalization

    . . . (read more)

  • Jul 26
    Day 36

    Fixing yesterday’s normalization misuse

    . . . (read more)

  • Jul 27
    Day 37

    Cross validation intro

    . . . (read more)

  • Jul 28
    Day 38

    Quick look at the cross validated residual plot

    . . . (read more)

  • Jul 31
    Day 39

    Research papers on power usage forecasting, residual plots of more data sets

    . . . (read more)

  • Aug 1
    Day 40

    Overview of forecasting methods, very promising results from multiple regression with lag features

    . . . (read more)

  • Aug 2
    Day 41

    Multi-step forecasting with multiple output regression, overview of the realtime anomaly detection research paper

    . . . (read more)

  • Aug 3
    Day 42

    Random forest vs ANN, forecast flow, more efficient window calculation with NumPy strides

    . . . (read more)

  • Aug 4
    Day 43

    Preparing the inputs/output matricies, testing the RF fit with median APE using different input/output/gap sizes, starting improvements to train speed

    . . . (read more)

  • Aug 7
    Day 44

    Index windows, more input features (holidays, descriptive stats, time data), trying non-default hyperparameters, flagging anomalies

    . . . (read more)

  • Aug 8
    Day 45

    Presentation, generalizing data frequency, quantile-based prediction intervals as threshold

    . . . (read more)

  • Aug 9
    Day 46

    Changing the prediction interval approach, starting abstraction of the training process

    . . . (read more)

  • Aug 10
    Day 47

    Starting on a model class

    . . . (read more)

  • Aug 14
    Day 48

    Adding extra features to the model, streamlining the training and prediction process

    . . . (read more)

  • Aug 15
    Day 49

    Getting prediction variances, auto-upsampling extra features, weighting samples by time, learning how to schedule commands

    . . . (read more)

  • Aug 16
    Day 50

    Restructuring the classes, predictions on multiple columns, outputting a CSV file, trial run

    . . . (read more)

  • Aug 17
    Day 51

    Slice based windows, more tests, bugfixing, writing up documentation

    . . . (read more)

  • Aug 18
    Day 52

    Bugfixing and class organization improvements

    . . . (read more)

  • Aug 21
    Day 53

    Multi-core processing for a performance boost

    . . . (read more)

  • Aug 22
    Day 54

    Logging the model properties, considering the implications of various energy savings quantification methods

    . . . (read more)

  • Aug 23
    Day 55

    Meeting, server setup intro

    . . . (read more)

  • Aug 24
    Day 56

    Slow progress with the setup

    . . . (read more)

  • Aug 25
    Day 57

    Getting the server working, starting to understand the mechanics

    . . . (read more)

  • Aug 26
    Day 58

    Future-proofing the energize module with a user guide and demo

    . . . (read more)

  • Aug 28
    Day 59

    Drafting a new data logging script, testing out APScheduler

    . . . (read more)

  • Aug 29
    Day 60

    Getting the team caught up, figuring out how to run the BACnet logging script

    . . . (read more)

  • Sep 1
    Day 61

    Successfully logging data, verifying compatibility with prediction process

    . . . (read more)

Page Source

© Copyright 2019