Multipod processing code and fully processed data
Copyright (c) Dan Zhao 2018-2021 & Shai Revzen 2008-2021

The code in this folder uses the Multipod RAW dataset found at https://doi.org/10.7302/m05a-0d90
to re-create Multipod_data_processed*_Leg.csv.gz with * = 6,8,10,12.
The raw dataset also contain photos showing the locations of the motion tracking markers.

However, for most applications there is no need to re-create these data files. 
You can use the .csv.gz file as is. There is one header row. To read all the data, use

from numpy import loadtxt
dat = loadtxt( FILENAME, delimiter=',', skiprows=1)
This can take a hot minute to load; at the end dat.shape will equal
for 6 legs (103044, 59)
for 8 legs (102849, 73)
for 10 legs (103871, 87)
for 12 legs (104274, 101)

If you want the column headings, you can use:
import gzip
with gzip.open( FILENAME,"rb") as f:
    hdr = f.readline().strip().split(b",")

The processing code and the associated libraries are provided subject to the GPL 3.0 license.
This is the "greater" GPL, and it requires (among other things) that any code using our code
be itself free and open-source. 

To run the processing code, first download the Multipod-raw.tar file using the DOI and put in 
the current folder. Then give the command:

python3 processData.py

if you want to speed things up, you can also indicate which leg number to process on the command-line, e.g.

python3 processData.py 6_Leg

On a strong machine this allows you to process the dataset about 4 times faster by launching multiple processing tasks in parallel.

To Cite this data:
BIRDS Lab, U. BIRDS Lab Multipod robot motion tracking data - processed data and code [Data set], University of Michigan - Deep Blue Data. https://doi.org/10.7302/0fpj-dz57