 Petra Lamborn
|
e30e28a747
|
Save an rds file of all models
|
5 years ago |
|
Petra Lamborn
|
d4e459128d
|
Bunch of proper python scripts
|
5 years ago |
|
Petra Lamborn
|
200dfde9ab
|
Fix a very important error; continue aggregation
|
5 years ago |
|
Petra Lamborn
|
89578e7989
|
Start agg py
|
5 years ago |
|
Petra Lamborn
|
bcbb6fe5b9
|
Clustering file
|
5 years ago |
|
Petra Lamborn
|
060efd02d4
|
Make separate dcorr script
|
5 years ago |
|
Petra Lamborn
|
b64ff94e94
|
Update requirements
|
5 years ago |
|
Petra Lamborn
|
07ff8e137b
|
Need to actually save tcorr
|
5 years ago |
|
Petra Lamborn
|
e9a112ad75
|
Algorithm for calculating correlation matrix for 5k
|
5 years ago |
|
Petra Lamborn
|
83039c973a
|
Collate 5k
|
5 years ago |
|
Petra Lamborn
|
91e0ad4943
|
Download only 5k sample
|
5 years ago |
|
Petra Lamborn
|
4c3c2ff9d0
|
Ok, so this should make a distance-correlation matrix of everything
Unfortunately, it would take approximately a week on an r5.large, so
I'm going to work with a large sample instead
|
5 years ago |
|
Petra Lamborn
|
aa49458bfa
|
Save correlation matrix
|
5 years ago |
|
Petra Lamborn
|
18aaf160f1
|
Diy corr?
|
5 years ago |
|
Petra Lamborn
|
6231929972
|
Trying to make my own corr
|
5 years ago |
|
Petra Lamborn
|
396cad8454
|
Download and collate in wide form, month-by-month
|
5 years ago |
|
Petra Lamborn
|
19f851fb09
|
Also print the time
|
5 years ago |
|
Petra Lamborn
|
3148bc89da
|
Further optimisations
|
5 years ago |
|
Petra Lamborn
|
ce7b167707
|
Grab a month at a time (Still need to merge!)
|
5 years ago |
|
Petra Lamborn
|
e2691dcf8a
|
Revert "Go back to downloading all"
This reverts commit 88a5daed65.
|
5 years ago |
|
Petra Lamborn
|
491ebc8e73
|
Go back to downloading all
|
5 years ago |
|
Petra Lamborn
|
d2e524cc7c
|
1 month of all data - test
|
5 years ago |
|
Petra Lamborn
|
ec351c80ed
|
Add images
|
5 years ago |
|
Petra Lamborn
|
bce871d1f8
|
Save figure instead of showing; download all data
|
5 years ago |
|
Petra Lamborn
|
427fca640c
|
Fix matplotlib errors; download the sample data
|
5 years ago |
|
Petra Lamborn
|
1f8ec45d0a
|
Further readme
|
5 years ago |
|
Petra Lamborn
|
100e043c57
|
Start writing readme
Fix readme
|
5 years ago |
|
Petra Lamborn
|
0ecaac5e52
|
Make a table of 5k sample
|
5 years ago |
|
Petra Lamborn
|
9ed4f3828e
|
Assign a new sample to clusters; look at how well it is modeled
|
5 years ago |
|
Petra Lamborn
|
449123773a
|
Fill-forward algorithm
|
5 years ago |
|
Petra Lamborn
|
9defcc38aa
|
A mostly-done cluster assigning algorithm
|
5 years ago |
|
Petra Lamborn
|
a7768eac28
|
Data frame reshaping for cluster assignment
|
5 years ago |
|
Petra Lamborn
|
5e15298c2a
|
Weather including rolling maximum
|
5 years ago |
|
Petra Lamborn
|
2ea6c125b2
|
Refactor out filenames for data pickles
|
5 years ago |
|
Petra Lamborn
|
e0bf20d2bd
|
Graph predictions
|
5 years ago |
|
Petra Lamborn
|
c82da5a058
|
Workable harmonic/weather model
|
5 years ago |
|
Petra Lamborn
|
66d259f7e6
|
Early attempt at combining weather and kwh models
|
5 years ago |
|
Petra Lamborn
|
654fe4c15d
|
A harmonic model for annual patterns in rolling 24-hour minimum temperature
|
5 years ago |
|
Petra Lamborn
|
8009428408
|
Working with old 1-year dataset
|
5 years ago |
|
Petra Lamborn
|
3dae6b02ac
|
Harmonic stuff
|
5 years ago |
|
Petra Lamborn
|
2355086ff5
|
Moving from getting a longer dataset to some more decomposition stuff
|
5 years ago |
|
Petra Lamborn
|
a5715417bc
|
Data back to April 2016
|
5 years ago |
|
Petra Lamborn
|
a95ac06382
|
Fourier stuff
|
5 years ago |
|
Petra Lamborn
|
3e423aea81
|
Working on seasonality; very much a WIP
|
5 years ago |
|
Petra Lamborn
|
e836325c1c
|
Create virtualenv requirements file
|
5 years ago |
|
Petra Lamborn
|
1705c988c3
|
Cluster visualisation images
|
5 years ago |
|
Petra Lamborn
|
709d2f8892
|
Some R plotting
|
5 years ago |
|
Petra Lamborn
|
07ca9e5130
|
Saving clustered datasets
|
5 years ago |
|
Petra Lamborn
|
deb90b7551
|
Separated graphs
|
5 years ago |
|
Petra Lamborn
|
0af67c2b17
|
parameter change
|
5 years ago |
