►
From YouTube: SimPEG meeting May 16
Description
We do a dry run of the presentation that we will be giving at the BCGS.
B
A
C
C
D
F
B
A
C
A
Alright
cool,
so
I
will
introduce
presenting
on
stand
egg,
and
so
this
is
an
open
source
software
project
that
we've
been
working
on
at
UBC
and
building
up
collaborations
elsewhere
now,
and
so
this
is
really
targeted
at
building
software
for
research.
So
we're
want
to
be
able
to
explore
and
experiment
and
wrap
you
prototype
ideas.
So
that's
the
space
which
are
you
giving
this.
A
I
can
post
the
link
later
on
about
McGinnis
and
pigs
dogs
actually,
and
so
the
point
is
making
here
is
that
okay,
we've
got
stacks,
it's
called
time
than
anything
diverging
away
from
this
axis
is
effort,
and
so
you
know
when
we
look
at
a
multidisciplinary
problem,
something
like
pizza
here,
where
we
want
to
be
characterizing
a
mineral
of
the
puzzle,
but
Nolan
physical
property
is
completely
diagnostic.
You
need
to
be
able
to
integrate
social
fields.
D
A
Geologic
buddha
under
make
that
it's
to
actually
characterize
this
and
spot
a
location
area
online,
and
in
order
to
do
that,
we
also
need
access
to
Algar
and
a
mission
techniques,
and
so
both
of
these
things
are
diverging
away
from
this
axis
with
time
is
because
we
are
dealing
with
these
large
mafia-style
projects,
but
also
when
you
look
at
something
like
airborne
kind.
The
main
electromagnetics
I
mean
that
home
has
been
solved,
but
it
is
a
non-trivial
problem
to
solve
and
if
you're
looking
at
somebody
who
wants
to
actually
go
and.
D
A
A
So
this
is
what
we're
trying
to
do,
which
also,
let's
start
by
showing
you
the
ahead
layout
and
how
we've
broken
down
inverse
balls.
So
we
start
up
here
with
our
inputs.
We
have
our
data
and
uncertainty
estimates.
We
know
something
about
our
governing
equations
with
physics
covering
the
situation
and
we
have
some
prior
knowledge
and
assumptions.
A
So
these
might
be
things
like
you
know
how
concept
of
what
you
have
to
be
reasonable,
but
they
also
might
be
sort
of
stronger
constraints
and
that
if
you
have
building
a
connection,
how
wormhole,
logs
or
things
like
that,
those
are
pieces
of
data
that
you
can
bring
in,
and
so
this
is
the
box
that
we're
going
to
focus
on.
This
is
actually
then
going
through
Asian
and
I
walk
through
each
of
these
pieces.
A
But
the
main
thing
to
highlight
here
we
thought
green
and
we
have
red,
and
so
the
green
part
here
is
the
Ford's
innovation.
So
that
is
how
we
go
through
and
actually
are
going
to
produce
simulated
data
give
any
mall,
and
then
the
red
are
all
of
the
pieces
that
we'll
use
to
actually
construct
an
inversion
and
go
through
and
find
a
a
single
model
that
fits
arguing
and
then
obviously
this
process
is
iterative.
You
don't
always
we
often
start
with
the
blind
inversion
and
want
to
go
back
in
memory.
A
So
the
way
we've
chosen
to
implement
this
is
in
Python
there's
a
few
reasons
for
that.
There
is
a
big
open
source
community
around
Python,
so
there's
a
lot
of
resource
or
thing.
It's
also
nice,
because
it
is
an
object-oriented
language,
which
means
it's
in
general,
in
kernels,
being
explicit
rather
than
implicit.
I
was
important
because
that,
like
promotes
sort
of
the
writing
of.
E
A
Fun-
and
so
in
this
case
it
helps
promote
organization
making
building
blocks
rather
than
building
Guyana
scripts,
so
that
you
can
then
go
so,
let's
start
by
looking
at
the
Ford
simulation
side
of
things,
that's
composed
of
two
pieces
here,
and
so
we've
lookin
this
into
the
survey,
which
contains
all
of
the
geometry.
Where
are
your
sources?
What
types
of
sources
are
they
I?
Where
are
your
secrets
more
than
terms,
so
we
have
a
DC
survey.
A
We
want
to
look
at
where
are
our
electrodes
and
then,
where
are
where
our
potential
edges
and
then
in
this
case
we
have
the
physics
which
is
then
the
DC
resistivity
equations.
So
we
know
how
to
go
about
and
actually
simulate
that
will
set
up
our
dip,
Sigma
graphs.
So
with
those
two
things
together,
we
can
then
go
ahead
and
see
and
the
way
we
actually
go
about,
and
thus
all
these
equations
is
using
a
finite
volume.
A
This
is
free
standard
in
the
UV
seeker,
we've
framed
out
the
world
interviews,
or
maybe
cylinders
or
perviness,
and
place
our
variables
on
cell
centers
for
physical
properties,
so
we
think
of
them
as
space
filling
but
flux.
It's
your
faces
and
fields
on
edges
on
to
the
way
you
would
work
with
this
in
synthetic
is
you
can
go
ahead
and
you
define
your
mesh,
and
so
we
defined
here
just
the
geometry.
This
is
a
2d
mesh.
A
So
you
can
look
at
that
the
DC
to
skip
the
equation,
so
we
have
our
tip
Sigma
grad
system.
We
discretize
that
and
if
we
look
at
the
code-
and
this
is
where
meant
by
implicit
versions
it's
here-
we
are
really
looking
at
and
explicit
that
the
math
looks
like
because
the
math
looks
like
the
code.
So
here,
when
we
look
at
our
a
matrix
with
G,
transpose
mu,
Sigma
Chi
master,
they
have
a
right
hand
side,
and
then
we
go
ahead
and
solve
that.
A
A
D
A
A
So
when
we
State
our
inverse
problem,
it's
just
minimizing
this
objective
function.
We
want
to
make
sure
we
fit
our
data
sets
are
talking
criteria
and
then
it's
perhaps
some
bounds
that
you
want
to
apply,
and
so
you
can
go
ahead
and
we
can
do
like
a
beta
cooling
schedule
any
of
those
types
of
tricks
that
you're
perhaps
familiar
with.
B
C
D
C
E
C
C
E
C
Like
to
point
to
that,
the
you
know,
in
order
to
contribute,
you
need
to
be
an
expert
in
all
of
these
Indian
expert,
the
only
one
of
those
right
with
go
to
module
and
still
an
important
contributor
and
advanced
science,
because
you
don't
need
to
learn
thing
to
be
able
to
get
to
a
problem.
Put
a
little
bit
more,
inviting
it's
like
that
square.
Can.
E
E
A
E
E
You're
solving
it
like
that
in
all
of
your
black
box
algorithm,
they
saw
that
yeah,
but
like
what
you're
seeing
is
just
a
few
points
of
data
yeah
but
like
if
you
have
a
specific
questions
like
over
the
Carm
flow
yeah
or
like
this
tools
are
kind
of
design
that
you
can
see.
Every
piece
is
that
you
computed
yeah.
A
E
I
A
Yeah,
we
have
quite
a
few
methods,
so
we've
got
DC
EMS,
intra
fields,
a
little
bit
of
work.
That's
been
done
on
seismic
and
we
will
locally
in
this
modular
way.
It
really
does
enable
people
to
bring
their
specific
expertise
to
the
table.
You
know
all
of
the
pieces
to
be
able
that
value-
and
that's
also
true
of
you,
know
technical
and
user
experience
as
well.
A
C
C
C
You
can
model
all
of
it
in
one
spot
and
ideally
you'd
be
able
to
just
set
up
here
your
earth,
my
mother
laboratory
and
just
swap
here,
swap
your
Ford
Armour
and
then
just
like
go
from
one,
what
physical
properties
the
other,
fairly
seamlessly
and
be
able
to
experiment
that
way.
So
that's
what
we'll
try
to
demonstrate
with
this
since
light
17
and
just
a
brief
overview
of
what
it
is.
We
just
started
looking
into
into
this
together
that's
about
two
months
ago,
so
we're
finding.
D
C
Experts
and
they
give
it's
a
deposit,
but
we
will
try
to
poke
Adam
through
through
different
physical
properties.
So
discovery
was
in
the
70s
first
month
on
minerals,
headed
for
a
long
time
and
the
diplom
very
generous
to
release
all
their
data
and
the
public
domain,
and
just
before
they
got
the
solder
property
to
bully
them
to
sixty,
and
they
they
also
let
us
use
that
they
test
thanks
to
both
of
them,
and
so
it's
a
disseminated.
The
poor
freestyle
deposits,
so
grades
are
fairly
low,
but
it's
huge
huge
tonnage.
C
The
dark
gray
that
they
were
seeing
in
that
on
the
on
the
right
side.
This
is
the
ultramarine,
fake
intrusion
and
then
the
light
gray
on
top
of
it.
This
is
the
main
thing
Avro,
so
both
of
them
have
intruded
into
like
sequences
of
volcanic
sand,
green
and
then
sedimentary
units
shells,
I
have
to
say
that
within
the
shelter
is
black
shells
and
with
with
graphite,
so
some
come
conductor
and
there
for
sure
and
yeah.
C
My
new
switch
next
slides
so
data
that
we
put
our
hands
on
this
pretty
broad,
there's
a
tons
of
borehole
log
with
the
densities
of
the
voting
on
activity.
We
have
ground
gravity
survey,
ground
mag,
which
pretty
higher
solution
than
your
batteries
per
the
time
domain
to
serve
based
on
the
main
airborne,
and
obviously
they
both
also
have
mag
magnet
on
hangers
on
board,
there's
about
five
lines
of
dcpip
and
NT
and
I
think
another
10
or
15
lines
of
SP
right
that
are
not
on
these.
C
Mt
has
been
inverted,
the
2016
there's
an
abstract
by
Kwang
and
find
out,
which
is
a
neat
paper.
Look
at
it
and
the
most
orgasmic
done.
Can
you
slide
19.
The
most
work
that
has
been
done
already
is
on
the
2d
and
3d
seismic,
so
Amelia
Kobus
to
the
University
of
well,
since
we
processed
and
analyzed
those
those
seismic
clients
with
Warhol
and
came
up
with
a
series
of
surfaces
in
a
go
cat
project,
and
then
she
she's
been
very
generous.
She
gave
us
her
go
cat
project,
so
that
was
adamant.
C
The
air
strong
starting
point
for
us
to
start
looking
to
go
and
so
20
we've
we've
simplified,
but
expanded
yeah,
the
George
go
model,
so
we
can
start
testing,
and
here
this
is
just
we
highlight
we
identified
the
primary
contacts
between
the
volcanics
intrusion
and
the
cementery
units
and
then
build
a
3d
50
meter
cell
size
resolution
autovod
with
these
populated
with
the
densities
of
the
world
economy
chargeability,
and
I
will
see
how
symposium
scanner
I
will
be
able
to
investigate
this
model.
It's
not
bad,
so
21
I
will
start
off
with
the
potential
field.
C
Speak
is
part
of
my
research
topic
so,
as
you
can
see
it,
just
I
having
this
nice
George,
going
on
I'll
be
able
to
have
each
units
can
assign
some
physical
properties
to
it.
Now
we're
going
to
start
with
density.
That
sounds
really
long.
So
then
CT,
hey
anyone,
know
Nelson
asleep.
So
it's
in
Prague
we
can
easily
afford
model.
C
It's
always
a
maybe
good
question
and
then
potential
fails
with
the
debt
resolution.
So
we
will
convert
the
cheese
out
the
G
set
theta
and
which
is
shown
inside
slide.
23
I
would
same
peg.
We
we
made
some
some
fencing
arms
to
be
able
to
get,
and
things
are
black.
He
more
compact-
and
this
is
what
we're
showing
you
here
is
slices
through
the
inverted
simulated
gravity
data,
and
the
take-home
message
here
is
that
we
can
probably
hope
to
see
the
to
see
the
bottom
boundary
of
the
of
the
intrusion.
C
C
To
the
real
data
set
probably
have
a
good
confident
that
we
can
map
out.
Those
is
this
unit.
I
will
switch
rapidly
to
susceptibility,
so
the
body
is
pretty
complicated.
Give
it
sir,
and
that's
the
main
reason
why
I
got
interested
in
the
deposit.
You
notice
there's
a
right
in
the
center
of
the
intrusion.
There
was
a
negative
and
that's
a
going
to
look
at
we're
going
to
look
at
into
into
this
now.
C
So
we
assigned
some
susceptibility
we're
getting
the
important
parts
here
they
pair
the
type
which
is
susceptible
it's
a
big
arc
that
goes
around
and
then
the
intrusion
itself
seems
to
be
relatively
susceptible
we're
missing
one
unit,
the
big
red
that
was
a
big
red
on
the
year
that
I
found
here.
That's
a
hydrothermal
hydrothermal
intrusion,
which
we
won't
focus
right
now,
but
it's
typically
a
interesting
for
us.
You
go
on
so
yeah.
The
remnants
has
been
studied
by
another
researcher
and
master's
thesis
specifically
on
on
the
Dnieper
intrusion
in
the
center.
C
So,
let's
see,
if
we
can
simulate
hitting
with
that,
I
would
send
back.
That's
the
neat
thing.
Is
it
fairly
rapidly?
We
can
design
a
fourth
foreign
modeler
here,
because
I'm
a
little
app
that
we
can
vary
the
susceptibility
inclination
diggnation
with
on
to
do
it
within
the
derive
to
be
able
to
test
if
those
numbers
that
were
published
make
any
sense
right
then
we'll
make
this
this
app
available
online.
C
Not
the
it's
not
in
the
main
framework.
At
this
point,
it's
still
on
my
research
branch,
but
it's
the
idea
of
working
invert.
The
magnetization
back
there
come
with
a
compact
body
using
more
robust
way,
and
what
is
hinting
us
is
that
what
Amelia
could
be
still
identified
as
deeper
didn't
deny
clean.
That's
this
most
likely
true,
most
likely
real,
because
well,
I
has
been
confirmed
by
drilling,
but
the
inversion
seems
to
confirm
this
is
that
the
the
mag
anomaly
is
actually
much
deeper
than
we
were
thinking
and
that's
it
four
PS
that's
lying.
E
A
E
Yeah
and
I
can
I
can
work
on
like
I,
probably
like
it
using
different
color
map,
basically
same
color
map
that
you
provide
at
indigo
20s
like
20
yeah.
We
have
that
color,
so
I
think
we're
probably
when
we're
showing
that
your
logical
pictures
yeah
because
I
it's
kind
of
like
that
I
was
having
heart.
So
no
myself
I
was
having
our
tech
where's
water,
don't
like
Edna
me
like
at
least
music
same
colors
for
good,
welcome
awful
roof.
One
yeah.
E
A
I
feel
we
can
travel
this
more
afterwards,
too.
I
think
like
in
the
intro
like
when
I'm
introducing.
It
would
be
helpful
to
just
perhaps
here's
a
quick
overview
of
things
that
you
guys
are
going
to
talk
about,
and
it
just
like.
Maybe
when
I'm
talking
about
the
framework
like
pointing
to
pieces
that
you're
going
to
free
off,
so
it
says
we
can
try
to
put
it
at
the
end.
Okay,.
E
A
D
E
Then,
generally
I'm,
just
thinking
about
where,
where
do
we
need
to
pull
up
that
special
properties?
Because
probably
18
is
good
because,
like
you're
showing
the
data
that
Moses
where
who
could
actually
like
a
pull
up
the
physical
property
people
again
and
and
they're
like
okay?
What's
the
kind
of
key
units
that
bird
enough
to
focus
on
for
each
12?
It's
physics
you
for.
A
E
D
A
E
F
E
E
C
E
A
C
I
C
A
C
E
Like
it
and
I
hoped
was
like
it's
a
question,
question
based
in
like
just
like:
try
to
use
two
screws
to
posture
that
yeah
I
am
like
not
hunters,
person.
Sure
whatever
is,
must
construe
this,
but
it
seems
like
it
like
I
think
Dom's.
That
ran
an
example.
That
was
great.
It's
like.
Oh,
here's
like
you,
have
big
boobs
negative
gurus,
and
what
is
that?
Are
you
explain
that
you
can
L
can
think
about
it.
Rent
show
the
silhouette
images,
but
we
can
match
that
those.
E
Great
that's
sort
of
what
I
want,
but
the
looks
like
it's
like
28
still,
not
sure
what
is
right
there,
because
so
it's
a
quote
from
not
sure
guru,
Wiseman
was
1997.
There
was
a
report
and
that
they're
like
if
they
actually
look
good
introduced
what
challenges
they
have
so
that
I'm,
focusing
on
the
first
one.
The
percentage
of
the
certified
is
too
low.
So
it's
a
disseminate
certified,
and
so
it's
not
very
like
that.
High
concentration,
it
doesn't
generate
really
high
physical
property
contrast.
E
So,
okay,
so
then
work
today,
I
think
seems
like
they
did
inductive
surveys
like
large
surveys
or
smaller
in
surveys,
but
it
wasn't
very
yeah
but
the
it's
not
unless
something,
what
I'm
not
sure.
It's
not
quite
that,
like
a
matching
with
our
stories,
is
if
I
look
at
my.
You
are
a
model
they're
very
hard
time,
giving
the
contrast.
When
you
get
older
response,
yeah.
E
E
E
E
E
Or
mineralization
we're
not
100%
sure
that's
really
conductive,
but
that's
locations
about
here,
H
like
9,000
meter
but
they're,
probably
song
I,
guess
we're
not
sure,
and
then
we
could
form.
Also
we
pick.
No.
You
can
see
the
time
decay
curves
at
back
to
location,
so
this
location
is
where
you
got
like
this
sediment
conductive
sediments.
So
if
you
look
at
the
data,
it's
very
high
and
then
slowly
became
response
which
is
expected
if.
E
Here,
where
you
got
that
is
mineralization,
it's
not
that
bad
compared
to
the
sediments,
but
like
still
so,
you
see
bump
but
smaller.
So
it's
probably
can
guess
some
of
you
I
guess
what's
happening
here
and
that
another
thing
so
we
now
we
know
this
like
a
curved
bone,
which
is
few
lights
and
those
sediments
units
are
very
adaptive,
but
we
have
a
smaller
scale
data
which
is
you
can
see.
E
E
E
E
This
room,
okay,
so
we
assign
a
comic.
You
model
everything
that,
but
they
were
interested
like
okay,
we
have
to
method
method,
a
method
where
which
is
airborne,
Ian
data
a
no
see
so
we're
going
to
start
with
airborne
beta,
and
so
let's
do
a
for
modeling.
So
here's
my
simple
model
so
I
want
to
see,
became
data
is
enough
to
see
like
this
conductive
mineralized
young,
but
one
case
what
we
could
do.
We
could
do
one
II
earth-like,
which
is
sort
of
like
a
typical
tool.
E
They
interpret
this
data
but
assign
the
conductivity
like
a
thousand
millimeter
resistivity
and
then
assigned
panel
meter
resistivity,
and
that
thinking
is
that
the
layer
is
one
hundred
year
and
then
it's
embedded
at
two
hundred
year
and
the
black
client
is
that
response,
which
I
am
slowly
became.
It's
kinda
made
sense
but
like
let's
but
like
the
earth
is
3d
and
the
also
mineralization
is
not
in
1d
but
sort
of
three.
Let's
make
it
as
a
cylinder.
E
So
it's
a
it's
a
block,
but
actually
I
get
to
like
a
cylinder,
and
so
the
length
of
the
cylinder
is
three
hundred
meter.
So
diameter
is
a
three
hundred
meters
and
the
thickness
is
hundred
the
same
reason.
To
be
sorry
like
do
this
for
mowing,
but
I
use
the
sort
of
like
a
practical
way
from
day
after
using
that
example-
and
this
is
much
smaller
compared
to
the
layer
case-
it's
much
smaller,
so
it's
sort
of
expected
but
kind
of
surprising
and
for
like
yeah
we
could
see
what's
so.
E
This
is
the
voltage
that
we
measured,
but
what
generates
it?
It's
an
electrical
field
for
and
years.
So,
if
we
plot
a
certain
time
where
a
times
snapshot
of
the
electrical
field
that
this
plane
so
electric
it's
an
amplitude
of
electric
fields,
they
like
to
feel
this
rotating,
but
I
got
a
conductors
that
stayed.
It
stays
there
for.
But
if
you
look
at
the
other
case,
where
you
got
the
so
longer,
your
footprint
is
pretty
different,
so
probably
like
just
using
Monday
code,
it
may
be
okay
and
I.
D
E
E
The
synthetic
resistivity
section
and
then
they
think
that
the
piping
system
is
a
cold
light,
babe
I'm,
basically
using
the
same
survey
geometry,
but
using
this
impact
model
to
generate
to
do
the
fast
innovation.
So
here
we
got
the
like
a
current
electrodes.
Here
we
are
a
reference
electrodes
far
away
and
those
greens
are
like
the
potential
electrodes
that
live
nothing
data,
but
when
we're
injecting
the
currents
like
the
kind
of
flows
in
everywhere
in
the
earth.
E
D
E
High
resistivity,
like
your
background,
is
like
really
existed,
but
as
you
go
further,
it
goes
down.
So
it's
getting
conductive
because
of
this
guy
and
increase.
So
you
see
the
anomalies
from
this
mineralized
zone
at
that
source,
location,
so
kind
of
nice,
but
we
have
another
source
location.
Let's
say
our
social
occasion
is
here
where
we've
got
the
sediments
and
well
similar
voltages.
E
But
you
see
this
like
a
sharp
increase
in
the
parent
resistivity
and
then
decays
because
of
this
conductive
gods
and
then
we're
actually
injecting
the
current
at
that
conductive
unit
very
close
to
sources.
You've
got
really
small
and
we're
so
exciting
this
guy
as
well,
but
we
may
not
see
that
much
on
our
beta
location.
D
D
E
These
conductors,
as
well,
just
nice
and
then
like,
if
you
sum
that
up
you
can
build
this
kind
of
pseudo
section.
Where
you
see
this
panel.
Ack
phenomena
stop
this
stuff
under
time.
Okay
and
then
that's
quite
not
exactly
similar
but
like
it's
quite
it
actually
captures
what's
happening
in
the
field.
They
guys.
We've
got
this
Cadillac
from
this
near
surface
conductor,
which
is
a
sediment
and
also
not
the
response
from
this.
Your
lives
normal
nice,
a
Chauhan.
You
can
invert
that
and
recover
like
common
theory
model,
it's
a
2d
inversion
to.
E
C
E
G
E
C
E
A
E
E
A
Did
what
I
would
do,
though,
because
I
think,
but
you
just
sort
of
jumped
right
into
doing
things
here
is
I
know
you
know
you
were
even
this.
Give
you
a
bit
of
context,
but
I
would
just
pose
a
question
like
start
with
question
and
easy
it
right,
and
then
here
okay,
not
to
answer
that
question
I'm
going
to
go
through
and
from
the
V
time
data
and
from
the
synthetically
how
we're
going
to
build
up
a
conductivity
model
and
then
simulate
and
see.
A
E
C
C
C
E
D
A
A
G
G
E
A
A
E
E
I
A
Yeah
and
then
here
you
need
a
little
bit
of
a
transition,
the
villager,
so
you
could
actually
even
start
like
basically
duplicating
the
slide
and
only
showing
the
resistivity
section
and
just
it
like.
So
do
a
quick
summary
on
like
what
did
we
get
from
telling
people
to
see
it,
but
we
also
know
that
if
DC
survey
is
conducted,
yes,.
E
C
E
C
I
I
E
C
H
C
A
A
C
A
D
A
And
then
what
I
was,
maybe
thinking
just
for
summary
is
actually
whatever
we
do
for
what
is
yeah?
Okay,
whatever
you
do
for
this
slide
it
just
like
what
do
we
have
I'll
just
bring
that
up,
because
I
think
we're
pretty
tight
on
time,
we're
going
to
be
ton
of
time,
so
I
don't
want
to
switch
and
like
bring
up
the
website
officers,
don't
skanky.
If
people
are
curious
want
a
cat,
we
can
do
that
full
look.
I
also
I,
might
try
and
trim
a
few
things
to
give
you
guys
a
little
more
time.
A
C
A
A
A
A
C
E
E
A
C
C
A
A
C
I
haven't
thought
of
them?
Oh,
so
our
don't
notice
like
there's
problem
I.
Do
you
mean
to
this
rather
than
if
eyes
work
you
can't
you
can't
use
this
space,
then
you
might
be
able
to
let's
we'll
try
to
and
see
unless
you,
why
not
yeah,
even
if
it's
temporary
that
pushes
the
other
musician,
yeah
yeah.
A
And
obvious,
like
a
really
short
update
and
it
worth
running
over
stare
down,
so
the
width
where
computation
is
right
now
what
I've
run
into
is
that
the
Sphinx
gallery
is
a
very
specific
formatting
for
Python
scripts,
and
so
you
can't
refuse
nd
convert
to
go
to
our
II,
we'll
just
fun,
so
we're
not
necessarily
going
to
use
the
gallery
for
this
okay.
So
what
I
was
planning
on
doing
and
I've
already
implemented?
A
But
half
of
this
is
basically
we
have
the
notebooks
and
then
all
I
will
do
is
like
basically
just
walk
through
all
of
those
holders
and
enemy
convert
them
just
to
RS
key.
So
you
get
it
a
paid
felt
right
away,
okay
and
then
similar
to
how
we
build
the
case
histories
page
on
the
infuse
islet
there's
just
well
as
far
as
our
own
custom
gallery
right.
We
that
and
so
and
eventually
like
we
can
arrange
valve.
A
So
it
has
two
sensor
covers
or
what
took
off
buddy
I
think
about
Easter,
though
that's
a
trajectory
through
thickness.
It
is
lower
overhead
for
us,
because
this
shapes
gallery
the
get
works.
For
this
can
pick
examples.
That's
fine
but
yeah
tailoring
some
scripts
in
that
way.
You
store
something
why
it's
not
an
intuitive
way
to
write
them,
so
I
don't
really
trial,
but
us
on
that
trajectory
about.
E
Yeah,
can
you
help
a
little
bit
like
that?
Getting
the
book
phenomenal
data
because
you
use
you
have
it
and
then
the
salinity?
You
need
an
example
for
paper
or
weaker
and
just
think
about,
like
a
small
chunk
up
with
a
number
that
we're
probably
just
want
to
first
and
providing
a
wall
sounding
sound
like
a
sounding
in
version
of
your
song
and
good
skycam
Lydia
did
you
have
to
you,
have
to
dig
I've.
G
C
A
There's
a
few
things
that
we
could
do
and
this
just
works
like
in
here.
Oh
this
is
a
comedian
paper
if
we
go
have
reviews
and
they
to
the
reviewers
that
hey
it'd
be
great.
If
you
have
a
field
example,
which
is
totally
fair,
cuz
everything
that's
in
there
right
now
is
so
I
think,
like
a
1d
example
is
actually
probably
enough,
but
if
we
actually
have
a
3d
that
could
point
to
that
would
be.
They
don't
have.
A
lot
of
impact
I
think
is
for
realizing.
You
can
in
fact,
do
three
divisions.
E
E
E
C
C
A
B
C
C
Will
look
yet
just
look
at
the
sequence:
sequence
inversion
and
an
indica
pick
a
spot
that
it's
not
3d,
I'm,
feeling
itchy
on
top
of
the
on
that
of
the
bank,
you'll
be
able
to
just
you're
just
a
real
cool
sounding
yeah.
If
you
are
right,
yeah
so
being
over
to
relearn
they're,
always
we
get
site,
but
when
you're.
E
A
F
F
F
E
E
F
F
A
F
F
G
E
A
J
I
guess
I
thought
about
we
do
the
presentation
is.
It
would
be
kind
of
interesting
to
tie
some
of
the
case.
Studies
back
in
the
simple
a
little
bit
and
I
think
that'd
be
like
not
a
very
big
deal.
Just
like
I,
don't
know,
I.
Think
someone
at
some
point
said
using
the
same
few
physical
survey
for
different
for
different
inversions
or
something
I
forget
exactly
what
was
but
I.
Think
just
like
saying
that
that's
like
this
is
using
the
same
object
in
simple
or
something
might
be.
A
J
And
I
don't
think
it
would
be
any
like
major
major
change
at
all.
It's
just
kind
of,
like
briefly
mentioning
little
things,
thats
impact
with,
because
I
think,
if
you
do,
if,
if
I
like
came
into
this
talk
after
you
gave
your
your
bit,
Lindsey
yeah
and
I
I
wouldn't
have
even
realized
that
it
was
a
simple
talk.
I
would
have
just
thought.
It
was
a
really
interesting
case
study
with
a
bunch
of
different
methods.
B
J
I,
don't
know
coming
from
a
like
a
enjoying
coding
point
of
view.
I
think
would
be
really
interested,
so
I
think
the
one
thing
you
said
was
having
like
sliders
on
some
figures.
Even
if
it's
just
like
a
totally
toy
thing,
that's
not
that's,
not
real
I,
think
that
would
be
kind
of
cool
just
to
take
it
back
to
be
like.
J
J
A
J
A
E
J
I
I'm
not
sure
if
this
this
makes
sense,
but
they're
a
few
of
the
little
synthetic
like
on
slide,
30,
33
and
34,
like
really
tiny
little
synthetic
things
like
that,
yeah
I
think
it
would
also
be
kind
of
neat
to
be
like
okay.
Well,
this
is
like
some
toy
model,
but
the
actual
physics
or
the
actual
problem
and
then
like
on
the
next
slide.
It's
like
okay!
Well
now
we
have
our
our
actual
data,
but
we're
still
using
the
same
physics
or
problem
from
that
little
toy
example.