►
From YouTube: Freeman42's anomaly model params question part 1
Description
https://discourse.numenta.org/t/issue-with-getting-anomaly-parameters-instead-of-swarming/5754/30 -- Watch live at https://www.twitch.tv/rhyolight_
A
B
A
Old
bold
new
pick
Python
2,
so
before
I
get
started,
doing
the
Python
3
thing,
I
wanted
to
try
and
figure
out
why
this
was
why
he
was
having
a
problem
and
he
pasted
his
code
at
some
point
here.
It
is
so
I
should
be
able
to
take
this
code,
copy
paste
it
and
then
figure
out
what's
going
on
and
if
I
can
get
a
piece
of
his
data.
You
know
I,
but
I
think
I
can
I
think.
B
A
A
B
A
A
B
A
Is
just
making
sure
that
it
matches
the
file?
What
I
really
don't
care
about
in
this
all
I
want
is
a
input
file,
so
whatever
the
gym
name
is
I,
don't
care
I'm,
just
gonna
hard
code,
this
I
don't
care
about
it.
I'm
just
gonna
have
an
info
file
open
something
somewhere
wherever
the
CSV
file
is
and
read
it.
A
No,
no!
No!
No!
That's
fine.
I,
don't
know
if
lots
of
em
work,
I'm,
not
gonna,
use
plot,
it
doesn't
matter,
run
gym,
name,
plot,
equals
plot
sure,
so
so
I'm
just
going
to
hard-code
the
CSV.
So
really
all
I
need
to
test
this
out
is
a
CSV
file.
So
let's
go
back
to
this
and
see
if
we
talked
about
data
code
with
changes.
Is
this
the
same
gist.
B
A
A
A
A
A
A
A
A
In
a
moment,
but
this
right
now
I'm
trying
to
answer
a
forum
question
about
an
HTM
example.
It's
like
the
typical
HTM
example.
It's
called
hot
gym.
It
just
takes
this
gym
data,
the
temporal
sequence
of
temperatures
from
a
gym
and
passes
it
through
with
the
HTM
and
the
HTM
spits
out
predictions
and
anomaly
scores
and
he's
having
trouble
getting
it
running
and
he's
hey
it's
basically
his
code
and
I'm
trying
to
run
it.
So,
let's
see
what
happens
this
time.
A
B
A
A
Input
data
is
the
flippin
seus
path
to
the
CSV
file.
Why
it's
called
info
data
I
do
not
know
why.
B
B
A
Is
just
an
example:
input
data
this
guy
is
gonna,
take
the
potatoes,
gonna
open
it.
Okay,.
A
A
B
A
A
B
A
B
A
B
A
A
A
A
A
A
Names
that
are
kinda
wait
give
a
watch
on
this.
How
can
I
like
right?
How
can
I
definition
I
want
to
like
bring
up
this
this
in
a
bigger
window,
so
I
can
inspect
it
easier
inspect.
This
thing
go
to
definition,
peak
definition,
rename
it
should
be
like
an
execute
or
a
watch
or
run
selection
align
in
terminal
I.
A
A
A
Parameter
row,
column,
1,
column,
2,
etc,
so
there's
c1,
for
example,
column
1
time
of
day.
These
are
the
sensor
parameters,
so
these
are,
there
are
encoders.
So
when
you
call
this
function,
it
is
going
to
come
back
with
or
it's
going
to
can
the
names
of
the
fields
2,
C,
0,
C,
1,
C
0
is
going
to
be
the
type
of
day,
and
so
this
is
3.
So
it's
only
got
time
of
day
and
but
you're
gonna.
You
have
to
translate
this
to
either
call
your
so
the
easiest
thing.
A
B
A
B
A
A
A
A
B
A
A
A
A
A
A
A
B
A
A
A
Input
predicted
filled,
Auto
predicted
field
is
c1
prediction.
Steps
is
one,
so
we've
got
a
one
step
ahead
of
fiction
and
the
model
params.
So
we
want
to
change
the
inference
tight
from
temporal
anomoly
to
temporal
multi-step.
Ok,
so
this
is
where
you
come
in
here
or
where
someone
comes
in.
You
want
to
go
to
in
these
grams.
A
B
A
A
A
A
B
A
A
A
A
A
A
A
A
A
B
A
A
A
A
A
B
B
A
B
A
A
A
A
A
A
A
B
A
B
A
A
A
B
B
A
A
A
B
A
A
Shift
whatever,
whatever,
whatever
whatever
we
should
get
the
timestamp
in
a
consumption?
Yes,
nice
Road
data
and
then
we're
gonna
go
let's
dive
in
C,
1
and
C.
2
here
are
the
values
and
now
we're
in
the
okiya
and
now
I
have
to
go
so
we'll
start
right
back
here
in
the
middle
of
this
debugging
session
right
after
my
standup,
so
I'm
gonna
have
to
mute
both
of
my
cameras
and
you're
gonna
have
to
miss
out
on
all
this
music.
Sorry,
I'm
gonna
meet
my
other
camera
I'll.