►
From YouTube: SimPEG Meeting June 23rd
Description
Weekly SimPEG Meeting from June 23rd, 2021
A
A
A
As
far
as
agenda
items,
thanks
things
that
other
things
we
should
hopefully
start
looking
up
to
have
a
vancouver
meet
up
for
the
people
that
are
around
at
some
point
in
the
future
soon,
so
keep
it
on
your
on
your
mind
that
we're
going
to
do
an
outside
meet
up
somewhere,
that'd
be
great
for
everyone,
at
least
for
the
people
here,
it'd
be
nice
to
see
in
person.
Again
it's
been
a
long
time
since
we've
had
one
of
those.
A
A
B
E
A
F
C
A
C
That
sounds
good.
That's
that
sounds
good.
Actually,
we
can
yeah
I'm
away
till
the
6th,
but
after
the
7th
I'm
I'm
free,
so
we
could
yeah.
We
could
just
get
together.
You
know
whoever
could
come
partial
team
whatever,
but
I
think
getting
together
early
in
july
would
be
would
be
a
good
idea.
It's
time
we
made
use
of
this
sunshine.
C
A
So
then
we
can
move
on
to
some
quick
reports
here.
There's
some
updates.
On
my
end,
I've
I've
put
a
pr
on
there
to
merge
in
my
edits
for
the
en1d
branch,
it's
currently
reliant
on
a
few
things
from
like
a
few
analytic
things
I
added
to
geoanna,
I
just
kind
of
added
simple
functions
that
reproduce
some
of
the
specific
equations
in
like
warden
coleman
to
geoanna.
A
A
A
A
B
B
A
A
Some
of
those
I
o
related
functions
inside
the
emoji
like
kind
of
home
for
them
need
to
get
the
stitched
regularization
into
like
the
regularization
module
inside
simply
make
sure
it
works
there
and
then
update
some
of
the
stitch
codes
like
the
stitch
simulations
to
use
like
the
updated
ones
that
I
made.
I'm
like
sorry,
I
know
you
and
I
I
showed
you
where
like
where
it
could
be
changed
and
kind
of
like
assume
they're
all
the
same
model.
A
A
F
So
if
there
has
been
a
release,
that's
great,
if
not
that's,
also
fine,
we
can
send
out
kind
of
some
of
the
highlights
and
then
also
any
other
things
that
are
like
shout
outs
to
just
things
that
have
happened
in
the
community.
Somebody
publishes
a
paper
or
is
given
a
talk.
Elsewhere,
it'd
be
nice
to
kind
of
use
this
as
a
bit
of
a
forum
to
share
some
of
those
things.
F
F
I
think
the
next
step
will
be
figuring
out
who's
willing
to
perhaps
champion
this
and
sort
of
be
in
charge
of
each
month,
just
collecting
sort
of
any
items
from
people
and
and
getting
that
out
there
I'm
working
on
setting
up.
I
was
talking
to
rowan
about
different
mail
out
tools
or
apps
or
or
whatever
we
previously
used
mailchimp.
F
When
I
mentioned
that
to
him,
he
went
on
a
very
long
rant
about
how
we
should
not
use
mailchimp,
so
we're
not
gonna
use
mailchimp
and
he's
pointed
me
to
another
one
called
send
grid.
So
I'm
just
working
on
learning
that
right
now
and
and
getting
that
set
up
I'll
share
a
bit
of
that
once
I
actually
know
what
I'm
doing
but
right
now,
I
do
not
so
I'll
share
a
bit
about
that.
Maybe
next
week.
F
A
E
Yeah
just
moving
through
the
dock
strings
and
made
a
lot
of
progress
so
yeah
your
your
brief
review
was
helpful
and
I
tried
to
make
those
changes
as
soon
as
possible
just
to
get
in
the
right
habits.
I
guess
I
still
have
a
couple
questions
about
how
much
detail
we
want
to
go
into
the
doc
strings
when
it
comes
to
discretized
operators,
so
I
mean
we
obviously
have
to
define
what
the
quantity
is.
E
That
would
be
useful,
but
I
don't
know
if
I
want
to
really
make
something
that
lengthy,
because
if
you're,
if
you're
working
around,
I
don't
know
if
you
need
too
much
detail
to
be
able
to
implement
it
but
yeah,
I
guess
I
I'd
like
to
come
to
consensus
on
what
are
the
most
important
things
to
that.
We,
the
must-haves
for
the
doc
strings
on
those
operators
before
I
go
through
the
trouble
of
making
them
so.
C
Devin
could
one
imagine
where
okay
you've
got
a
doc
string,
it's
got
most
every
well,
it's
got
everything
that
you
need
to
know
on
how
to
implement
it,
but
is
it
reasonable
to
think
about
a
link
that
you
could
have
to
some
of
maybe
other
documents
that
we
have
that
would
you
know
delve
in
more
detail
as
to
you
know,
this
is
how
this
is.
This
is
derived,
or
this
is
more
what
it
is
yeah.
E
I
mean
that's,
that's
basically
the
idea
yeah
so
how
to
implement
it,
what
it
does
and
then
I
guess
I
have
to
I
have
to
have
a
page
for
it
to
link
to
before.
I
actually
make
the
link,
so
that
might
be
kind
of
like
a
something
where,
when
we,
when
we
go
back
to
the
tutorials
that
I
had
been
making
for
discretize,
then
I
can
go
and
add
the
links
to
the
doc
strings
to
the
finalized
tutorials.
E
E
E
A
For,
like
the,
I
don't
think
it
needs
to
be
that
explicit
in
the
doc
strings
like
that,
like
kind
of
those
like
it'd
be
nice
to
have
them
somewhere,
but
like
just
within
the
doctrine
of
that
function,
I
think
all
you
need
is
just
like
a
a
math
thing.
That's
like
okay,
some
of
the
values
like
some
of
the
vector
values
of
each
phase,
times
area
that
divided
by
the
cell
volume,
or
something
like
that,
something
pretty
simple,
simple,.
E
E
So
I
I
like
keeping
that
equation
there
but
yeah.
I
guess
this
is
why
we're
having
the
conversation
so
coming
to
consensus
on
like
an
itemized
list.
This
is
what
each
one
should
have
in
it
before
we
go
and
crazy
and
start
writing
a
bunch
of
doc's
drinks.
So
I'd
like
to
know
what
everyone
thinks
is
the
most
important
information
and
there
is
a
pull
request
for
the
doc
strings,
and
this
is
kind
of
this
is
mentioned
in
there.
So
any
feedback
would
be
useful.
A
That
that's
good,
I
mean
I'll,
have
a
few
thoughts
and
I'll
add
in
there
in
general.
I
I
think
this
like
talking
about
like
the
individual
like
ways
that
they're
implemented
within
each
side
of
within
each
mesh
is
a
little
bit,
maybe
too
much,
whereas
just
kind
of
like
it's
like
start
even
talking
about
the
stencil
and
everything
else
is
like
still
just
dependent
on
that.
It's
a
tensor
mesh
doing
this
and
it's
not
like,
I
guess
to
me-
that's
a
little
bit
still
too
much
detail.
E
Yeah
one
for
like
for
the
tree
mesh
as
well,
you
can
kind
of
go
down
the
rabbit
hole
of
does.
How
does
it
account
for
hanging
and
non-hanging
faces
and,
like
I
want
to
avoid
that
right?
The
whole
point
is
somebody
reading
it
understanding
what
it
does
and
being
able
to
implement
it
they're
never
going
to
have
to
code
it
that's.
The
whole
point
is
that
we've
we've
done
that
work.
They
just
need
to
know
what
it
is
and
how
to
implement
it
properly.
E
F
It's
like
one
sentence
about
what
does
this
thing
do
but
then
later
on,
there's
a
notes
section,
and
that
is
where
you
know
there
can
be
formulas
or
some
of
these
details
or
perhaps
gotchas
and
things
like
that
and
that's
something
we
can
if
you've
already
got
some
content,
that
kind
of
fits
into
that
category
be
easy
to
move
it
to
that
kind
of
section
and
if
not
that's
something,
we
can
also
kind
of
fill
out
later
on
too.
If
there
are
things
that
we
find
we're
consistently
getting.
F
E
Yeah
I
mean
it's
yeah,
it's
a
good
way
of
doing
it.
If
you,
if
you
want
to
put
everything
there,
it's
quite
comprehensive,
but
without
being
very
wordy
up
front,
so
that
is
kind
of
nice.
I
did.
I
did
like
having
a
separate
tutorial
section
to
kind
of
show
an
application
of
it
instead
of
spending
a
lot
of
time
talking
about
the
operator
itself,
but
that
that
would
be
one
useful
way
of
putting
additional
information.
F
Yeah-
and
I
don't
think
this
negates
having
the
tutorials
I
totally
agree
having
those
is
is
great,
but
even
actually
in
the
notes
section,
it
could
say
like
go
see
this
tutorial
if
that's
sort
of
the
right
place
and
then,
if
there's
other
simple
notes
that
you
know,
if
you
do
want
to
talk
about
how
hanging
faces
are
handled
in
a
given
operator,
then
that
would
be
a
good
place
to
do
it.
But
I
just
thought
that
parsing
was
good.
Okay,.
G
G
D
Yeah,
I
didn't
get
too
much
time
this
past
week
to
work
on
things,
but
I've
got
it
narrowed
down.
It's
it's
just
in
the
package.
It's
like
yeah
one
of
the
testing
examples
where
it
just
tests
a
group
of
examples,
and
I
think
it's
the
lugan,
oh
yeah,
the
laguna
example
I
haven't
dug
into
it.
Quite
yet
the
example
I
ran
it
locally
and
it
seemed
to
work.
D
Okay,
well
I'll,
take
a
look
at
it.
It's
it's
narrowed
down
to
about
there,
so
that
should
be
about
it
once
I
figure
that
out,
that'll
be
all
wrapped
up
and
yeah
still
playing
with
the
mt
version.
It's
actually
working
really
well
we're
running
lots
of
inversions
with
it.
It
runs
pretty
quick
and
we're
getting
comparable
results
results
to
our
dc
survey
that
we
did
in
the
same
spot,
so
yeah
looking
good.
F
D
A
D
B
B
It's
actually
like
a
it's,
not
differentiable,
so
it
took
me
a
while
to
actually
understand,
and
actually
I
tried
many
times
and
it
kept
failing,
and
actually
yesterday
I
figured
out
how
to
and
how
simple
it
was
anyway.
So
actually
I
am
now
I
tested
for
the
order
test
and
I
actually
was
passing
so
I
actually
am
now
ready
to
invert
for
amplitune
phase,
starting
from
the
complex
value
anyway,
so
this
was
kind
of
personally
pleasing.
A
B
C
B
That
was,
I
think
that
was
the
main
problem
that
I
had,
but
actually
the
derivative
that
I
need
was
not.
It
was
in
the
other
direction,
so
I
need
the
perturbation
of
a
complex
value
with
regard
to
perturbation
and
amplitude
in
in
such
a
case,
it's
actually
a
continuous
function.
The
other
way
is
discontinuous,
so
actually
there's
no
derivative.