►
From YouTube: SimPEG meeting January 29, 2019
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
B
B
B
We
added
to
the
data
over
so
one
time
to
read
notebook,
just
an
invert,
a
single
sounding
and
just
like,
walked
through
that
was
with
everyone
and
so
yeah
it
was.
It
was
really
good.
It
was
nice
to
get
like
questions
and
I
think
this
is
a
good
format.
That's
something
that
we
should
think
about
like
trying
to
do
every
couple
of
months
together,
you
know
regularization
or
something
like
that,
and
you
couldn't
do
one
on
your
like
interesting
use
cases
there.
The
regularization
tube,
oh
yeah,.
B
A
B
So
there
was,
there
were
three
pcs
or
master
students
who
are
in
Italy
and
in
later,
and
so
they
were
working
on
time
domain,
iam
inversions
as
well.
This
actually
magnetics.
So
we
didn't
touch
magnetics
at
all,
but
like
a
lot
of
the
same
machinery,
of
course
is
used
place.
A
B
So
and
then
we've
asked
that,
like
they
jump
in
and
actually
annotate
the
notebook
and
put
more
notes
in
it
and
can't
you
read
that
back
yeah,
so
I
think
that,
like
they
seem
quite
keen
to
do
that
and
I
think
that's
a
really
cool
way
to
start,
like
you
mean
tutorials
like
this
is
like
we,
you
know
we
put
together
as
much
as
feasible
for
us,
but
then
actually
ask
that
other
people
like
based
on
what
they
learned
then
fill
it
in
or
somebody
else.
Okay,.
A
B
A
B
B
C
B
Yeah
yeah,
okay,
so
I,
don't
actually
have
like
too
many
items
and
feel
free
to
jump
in.
If
there's
other
things
that
you
both
would
like
to
go
over
yeah.
So
regrouping
on
the
e/m
tutorial.
If
you
have
a
chance
to
watch
it
or
you're
curious
to
watch
it
and
have
any
feedback
based
on
what
we
did
like
I
would
love
to
look
to
hear
that
in
capturing
I
might.
B
A
B
I'll
generate
a
pull
request
this
afternoon
and
we
can
capture
it
and
like
iterate
a
little
bit
in
the
pull
request
before
we
merge
it.
But
we
talked
a
bit
about
this
last
time
so
having
it's
broken
up
into
like
getting
so
headings
is
like
getting
started
than
a
bit
of
a
user
guide
and
then
the
API
documentation.
B
A
B
B
So
this
is
actually
where
I
see
the
biggest
need
for
development,
both
with
respect
to
like
setting
up
penser
matches
and
then
Deiters
been
asking
a
lot
of
questions
just
unlike
finite
volume
and
how
do
the
differential
operators
work?
Inner
products
work,
so
I
think
we
need
to
actually
like
beef
up
our
user
documentation
on
a
lot
of
that
and
I
think
you
mean
even
for
us
like
going
through.
B
The
API
reference
took
a
bit
to
like
get
into
better
shape
and
there's
still
a
few
bugs
like
we
need
to
add
docs
to
these
guys,
for
example.
But
if
we
look
now
at
the
SIL
mesh
the
way
I've
tried
to
break
this
up
is
so
we
go
by
class.
It'll.
Tell
you
everything
that
it's
an
from.
This
is
the
documentation
on
the
class.
So
let
me
just
pull
up.
B
This
so
that
you
can
see
it
where
it's
pulling
from
so
like
this
text
here
is
just
on
the
like
the
class
documentation
that
makes
sense.
You
see
what
it
okay
cool
and
then
added
some
plots.
I
think
what
we
should
try
to
do
is
with
every
single
class.
So
we
should
make
sure
that
we've
got
a
minimal
example
and
a
plot
of
something
just
so
that
it's
a
little
more
clear
what
it
is
and
why
you
would
use
it.
B
B
So
these
are
now
all
using
the
numpy
style
documentation.
So
if
we
look
here,
it
takes
up
a
little
bit
more
vertical
space,
but
it
gets
rid
of
that
like
this
strange
Sphinx,
documentation
style,
where,
like
you,
have
the
colon
program,
colon
whatever,
which
is
just
like
hard
to
read,
especially
in
the
notebook
so
yeah.
This
is
like
much
friendlier
in
plain
text
and
when
you
pull
it
up
in
the
notebook
I
think
it's
much
better.
So
yeah
do
you
guys
have
any
questions
good
fat
or
like
a.
A
B
Because
it
actually
like
gives
me
the
text
from
the
commit
which
is
kind
of
cool,
and
it's
also
nice
to
see
like
when
something
has
changed
more
recently
versus
like
this
was
four
years
ago.
B
A
B
And
the
cylindrical
mesh
is
just
a
placeholder
right
now
like
what
I
was
sort
of
thinking
is.
We
should
have
one
for
setting
up
each
of
the
different
mesh
types,
so
this
like
overview,
gives
a
bit
of
an
overview
of
just
what
each
of
the
things
are.
Where
things
live.
All
of
that
first
step,
then
we
should
have
something
that
constructs
and
walks
through
each
of
the
mesh
types.
B
We
should
probably
have
something
on
like
IO,
so
reading
in
UBC
meshes
yeah.
Somebody
maybe
write
these
down
son.
A
My
you
see
the
link.
The
slack
was
a.
A
A
A
B
B
A
B
C
B
I
think
it's
like
what
I
could
actually
see
happening
is
we've
got
a
finite
volume,
one
that
is
perhaps
just
an
overview
of
where
things
live
and
then
I
think.
Potentially
afterwards,
we've
got
one
on
the
edge
Perl.
We've
got
one
on
the
cell
gradient
one
on
the
nodal
gradient
like
those
kinds
of
things
it.
B
A
B
A
Have
to
admit
that
some
I'm
a
bit
confusing
about
something
to
like
exact
feels
like
in
the
user
guide
in
the
API
is
there
might
be
a
lot
of
overlap
because
at
some
I
do
admit
at
some
point,
I
thought
we
wearing
the
user
guide
for
the
same
recall
mesh.
But
then
that
was
what
you
vote
in
the
API
yeah.
B
So
I
think
we
should.
We
should
maybe
write
a
small
documentation
guide.
Cuz
I
think
we're
like
when
we
here
so
for
the
tensor
mesh
like
this
page,
is
much
more
in-depth
than
what
the
class
documentation
needs.
I
think
the
class
documentation.
We
should
still
strive
to
have
like
a
minimal
example
and
a
plot
wherever
that
makes
sense.
So
the
minimal
example
being
here
are
four
lines
of
code
that
build
a
mesh
like
nothing,
fancy,
no
bells
and
whistles.
B
B
B
B
B
B
Yeah,
so
now
what
we
have
there's
we're
adding
like
the
subject
units,
so
that
was
previously
what
bomb
was
calling
I
think
like
a
homogeneous
Matt,
but
the
idea
being.
If
you
know
the
indices
of
giving
units,
then
we
can
define
like
the
value
for
each
of
them
and
then
just
like
Sir
Jack,
those
onto
the
mesh.
So
if
you've
got
a
geologic
sketch
or
whatever,
and
there
are
two
zones,
then
we
can
just
invert
for
the
values
of
those
mm-hmm.
B
Then
I
also
went
through
and
there's
examples
well,
there's
code
now
for
inverting
for
electric
fields
and
current
density
from
the
time
domain.
Just
the
HJ.
Sorry,
there's
two
things
that
I'm
talking
about
right
now.
Sorry
I
went
the
DC
code
and
I
added
the
ability
to
invert
for
electric
field
data
and
then
on
the
time
domain
side
of
things,
there's
the
addition
of
a
grounded
source
where
we
also
solved
the
MMR
problem.
So
if
we
have,
if
we've
got
a
permeable
object
in
a
time
domain
problem,
we
can
handle
that.
B
So
it's
like
not
the
best
implementation,
there's
some
things
like
in
terms
of
storing
matrices
and
stuff
like
that.
That
needs
to
be
cleaned
up.
But
at
this
point
in
time,
like
I,
think
it's
sort
of
it's
good
enough
to
you
know,
and
then
we
can
refactor
still
because
I
think
it'll
make
more
sense
to
refactor
it
with
the
simulation
class
in
mind,
anyways
yeah
yeah.
So
that
is
the
biggest
changes
there.
B
B
A
B
There
is
a
fair
bit
of
my
own
code.
It's
also
like
a
very
nice
example,
and
so
like
at
this
point.
I,
don't
actually
think
it
necessarily
makes
sense
to
like
go
the
effort
of
bringing
it
into
stem
page,
because
the
number
of
people
who
are
looking
at
solving
casing
problems
with
the
Emma's
like
before
I
think
there's
probably
more
high
impact
ones
that,
like
we
could
play
around
with
CKC
or
something
like
that.
B
B
A
B
Okay,
then,
the
last
item
that
I
have
is
there's
this
conference
in
Montreal.
That's
coming
up,
I
knew
Gigi
I,
don't
actually
know
what
that
stands
for
Oh
international
union
of
geodesy
and
geophysics,
so
I
think
DOM
is
gonna,
submit
something
on
MBI
I'm
gonna
submit
something
on
casings.
B
A
The
attestation
on
the
new
technologies
for
mining
exploration,
so
I,
might
submit
to
that
with
a
picture.
Physical
exam
for
us
to
think
which
example
I
want
to
do
yeah
but
like
I,
would
like
to
do
a
new
example,
but
that's
an
example.
I
would
not
have
started
yet
so,
but
I
mean
the
code.
Is
there,
and
so
just
like?
Okay,
final
data
set
I
think
is
good
enough,
for
that
and
I
can
that
I
can
publish
so.
C
B
B
A
Get
in
there
like
I'm
running
out
things
with
Doug
like
we
had
a
discussion
so
hopefully
maybe
we
were
Amy
like
maybe
try
to
have
something
by
the
end
of
that
week,
but
I
feel
that
we
had
big
questions
delaying
of
hanging
around
yesterday.
So
maybe
maybe
in
the
end
of
next
week,
I
mean
all
the
simulation.
Is
there.
So
most
of
the
text
is
there,
but
we
have
still
some
justification
to
give
a
slight
a
slight
yeah
improvement
or
yes
al.
But
it's
I
mean
it's
it's
all
there.
C
A
B
A
Was
thinking
to
just
to
take
my
hard
drive
and
to
get
it
floating
in
your
in
your
tower,
so
it's
like
so
I,
don't
lose
anything
of
my
thing
that
not
have
to
install
anything
in
just
like
everything
or
the
other
way
around
otherwise
I
Drive
change,
but
everything
stays
the
same.
All
the
way
go
for
it.
A
Like
he
started
yesterday,
I
think
he's
still
getting
things
ready
like
getting
his
key
and
access
to
internet.
He
might,
he
might
take
over
so
his
computer,
that
has
to
check
with
to
charge
that
yeah,
what's
up
depends
mostly
12
and
11,
mag
and
I.
Think
it's
really
interesting
to
like.
Also
like
all
who's
like
we
were
talking
this
morning,
and
it
has
a
couple
of
paper
of
like.