TODO

ALL TO DO ITEMS HAVE BEEN MOVED HERE:

/opt/projects/neurome/TODO

Refer to this file from now on.



9/10/2021:

[1) Establish a convention for saving the dataset's creation time
(i.e., when recordings start and end).  Right now, things are
expressed in terms of sample numbers offset from the start of
recording.  This is not enough if we want a concordance between H10
recordings and Libre Freestyle datapoints.

A high-precision "datestamp" should be preserved at the start of a
recording session, and then propagated through the nbf and .db
formats, perhaps through the 'properties' table.  This datestamp will
fix the start sample number to a specific date and time.]


OLD:

2) DONE: Using the convention defined in (1), implement an
"auto-datestamp" utility that will automatically datestamp a .db file
(and its corresponding data store) according to the filename.  This is
a cheap way of anchoring early glucose datasets to a specific date and
time.  Once (3) is implemented, we won't need this utility since
recordings will automatically datestamp everything.


- Legends & labels for clustered entities:

vrasters
vwave
vscatter


- Alternative visualizations of clusters.