►
From YouTube: A Quick Tour of the ML.NET CLI
Description
A quick tour of using the ML.NET CLI to explore machine learning models with the AutoML engine.
Data - https://www.kaggle.com/camnugent/california-housing-prices
CLI Documentation - https://docs.microsoft.com/en-us/dotnet/machine-learning/reference/ml-net-cli-reference
Contact:
Twitter: https://twitter.com/JWood/
Blog: https://jonwood.co/
A
Here,
everyone
with
the
release
of
mo
dotnet
version
1.
They
also
released
a
command-line
interface
to
use
the
auto
mo
engine,
and
so
in
this
video
I'm
going
to
show
you
how
to
use
it
and
some
of
the
options
that
you
can
use
with
it
so
to
be
able
to
install
it.
You'll
need
to
use
the
net
seola
and
with
the
tool,
install
commands
g
for
global
and
just
do
mo
net.
If
you
don't
have
the
dotnet
see
a
lot
itself,
you'll
need
to
download
the
undead
core
sdk.
A
You
would
do
that
and
already
have
the
tool
installed
and
real
quick
look
at
the
data
already
have
in
this
directory
here
and
it's
the
same
housing
data
that
we
looked
at
in
the
previous
aldo
ml
videos.
But
if
you
haven't
seen
those
we'll
take
a
quick
look
at
it
and
it
is
the
California
housing
data
set
and
just
some
attributes
on
the
house,
it
houses
their
latitude
longitude.
A
And
to
use
the
email
doesn't
add,
CL
on
that
were
just
installed
with
the
mo
net.
It's
a
auto
train
and
since
we're
going
to
predict
a
continuous
value
here
of
the
median
house
bail,
you
will
give
it
a
task
to
use
of
regression
and
it
will
give
it
the
data
set
that
we
have
in
this
directory
housing
that
CSV
and
they
also
give
it
the
column
name
that
has
our
label.
So
this
label
column
name
option,
and
this
could
be
that
median
that
so
let
this
run
for
a
bit,
it's
gonna
take
a
while.
A
A
A
So
a
couple
of
things
that
you
can
add
to
it
that
you
can
give
this.
You
can
specify
the
max
exploration
time
and
I'll,
give
it
15
here
and
as
you'll
see
here
that
that's
giving
us
the
current
best
algorithm
and
and
its
current
r-squared,
then
I
ran
for
15
seconds,
and
you
see
our
R
squared,
especially
not
too
much
different
from
the
30
minute
run,
is
86
percent
instead
of
88
or
89
that
it
was
before
now.
A
A
A
So
that
was
kind
of
just
a
quick
tour
of
the
in
London
Etsy
ela
and,
as
you
can
see
it's,
it's
probably
not
gonna
be
used
most
for
production,
workflows
and
deployments
of
models.
But,
as
you
can
see,
it's
gonna
be
a
good
way
to
get
started
with
getting
a
model
going
and
I.
Think
the
best
thing
that
it's
going
to
be
useful
is
to
do
exploration,
especially
for
the
feature
engineering
that
we
just
saw.