I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
15
,
No
.
4
,
A
u
g
u
s
t
20
26
,
p
p
.
3
4
1
1
~
3
4
2
0
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
4
.
p
p
3
4
1
1
-
3
4
2
0
3411
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
SH
A
P
-
en
ha
nced
ensem
ble learn
in
g
f
o
r yield pr
edict
io
n:
insig
hts
from CY
-
Bench
d
a
ta o
n India
n wh
ea
t
So
m
a
G
up
t
a
1
,
Da
y
a
l K
u
m
a
r
B
eher
a
2
,
Sa
t
a
rupa
M
o
ha
nty
2
,
Su
bh
ra
Swet
a
ni
s
ha
3
,
Rit
ik
M
a
llik
2
,
Na
m
it
a
P
a
nd
a
2
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
R
a
v
e
n
sh
a
w
U
n
i
v
e
r
si
t
y
,
C
u
t
t
a
c
k
,
I
n
d
i
a
2
S
c
h
o
o
l
o
f
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
K
I
I
T
D
e
e
med
t
o
b
e
U
n
i
v
e
r
si
t
y
,
B
h
u
b
a
n
e
sw
a
r
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
S
i
l
i
c
o
n
U
n
i
v
e
r
s
i
t
y
,
B
h
u
b
a
n
e
sw
a
r
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ap
r
25
,
2
0
2
5
R
ev
is
ed
May
19
,
2
0
2
6
Acc
ep
ted
J
u
n
19
,
2
0
2
6
P
re
d
ictin
g
c
ro
p
y
ield
a
c
c
u
ra
tely
i
s
e
ss
e
n
ti
a
l
fo
r
p
ro
v
id
in
g
fo
o
d
se
c
u
rit
y
a
n
d
imp
ro
v
i
n
g
a
g
ric
u
lt
u
ra
l
p
ra
c
ti
c
e
s.
Th
is
stu
d
y
e
x
a
m
in
e
s
t
h
e
u
se
o
f
e
n
se
m
b
le
m
a
c
h
in
e
lea
rn
in
g
m
o
d
e
ls
c
o
m
b
in
e
d
with
S
h
a
p
ley
a
d
d
it
iv
e
e
x
p
lan
a
ti
o
n
s
(S
HA
P
)
fe
a
tu
re
se
lec
ti
o
n
to
imp
ro
v
e
wh
e
a
t
y
iel
d
p
re
d
icti
o
n
i
n
I
n
d
ia.
T
h
e
stu
d
y
u
se
s
CY
-
Be
n
c
h
d
a
ta,
i
n
c
o
rp
o
ra
ti
n
g
n
o
rm
a
li
z
e
d
d
i
ffe
re
n
c
e
v
e
g
e
tatio
n
in
d
e
x
(
NDVI
)
,
m
e
teo
r
o
lo
g
ica
l
d
a
ta,
a
n
d
so
il
m
o
ist
u
re
d
a
ta
fo
r
y
iel
d
p
re
d
ictio
n
.
Va
rio
u
s
e
n
se
m
b
le
te
c
h
n
iq
u
e
s,
i
n
c
lu
d
in
g
v
o
t
in
g
,
sta
c
k
in
g
,
a
n
d
b
o
o
st
in
g
a
re
e
v
a
lu
a
ted
.
B
o
o
st
m
1
e
n
se
m
b
le
m
o
d
e
l
c
o
n
siste
n
tl
y
o
u
t
p
e
rfo
rm
s
o
th
e
r
m
o
d
e
ls
i
n
th
e
p
re
d
ictio
n
.
Ad
d
it
i
o
n
a
l
ly
,
t
h
e
in
teg
ra
ti
o
n
o
f
S
H
AP
-
b
a
se
d
fe
a
tu
re
se
lec
ti
o
n
with
th
e
b
e
st
e
n
se
m
b
le
m
o
d
e
l
sig
n
ifi
c
a
n
tl
y
imp
ro
v
e
s
th
e
m
o
d
e
l
a
c
c
u
ra
c
y
a
n
d
i
n
terp
re
tab
il
it
y
b
y
id
e
n
ti
fy
i
n
g
th
e
m
o
st
i
n
flu
e
n
ti
a
l
fe
a
tu
re
s
a
ffe
c
ti
n
g
y
ield
.
Th
e
re
su
lt
s
sh
o
w
th
e
e
ffe
c
ti
v
e
n
e
ss
o
f
e
n
se
m
b
le
b
o
o
st
in
g
m
o
d
e
l,
in
c
a
p
t
u
rin
g
t
h
e
c
o
m
p
lex
re
latio
n
sh
i
p
s
with
i
n
a
g
ricu
lt
u
ra
l
d
a
ta
p
a
rti
c
u
larly
wh
e
n
c
o
m
b
in
e
d
wit
h
fe
a
tu
re
se
lec
ti
o
n
.
T
h
i
s
m
e
th
o
d
imp
ro
v
e
s
t
h
e
tran
s
p
a
re
n
c
y
a
n
d
a
c
ti
o
n
a
b
il
it
y
o
f
m
a
c
h
in
e
lea
rn
i
n
g
m
o
d
e
ls
fo
r
a
g
ro
n
o
m
ists,
p
o
li
c
y
m
a
k
e
rs,
a
n
d
f
a
rm
e
rs.
K
ey
w
o
r
d
s
:
B
o
o
s
tin
g
E
n
s
em
b
le
m
o
d
els
F
ea
tu
r
e
s
elec
tio
n
N
o
r
m
a
l
i
z
e
d
d
i
f
f
e
r
e
n
c
e
v
e
g
et
a
tio
n
i
n
d
e
x
R
em
o
te
s
en
s
in
g
Sh
ap
ley
ad
d
itiv
e
e
x
p
lan
atio
n
s
Y
ield
p
r
ed
ictio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
So
m
a
Gu
p
ta
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
,
R
av
en
s
h
aw
Un
iv
er
s
ity
C
u
ttack
,
I
n
d
ia
E
m
ail:
s
o
m
ag
u
p
ta2
9
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
W
h
ea
t
is
an
im
p
o
r
tan
t
a
g
r
icu
lt
u
r
al
cr
o
p
o
f
I
n
d
ia
an
d
is
ex
tr
e
m
ely
im
p
o
r
tan
t
f
o
r
its
ec
o
n
o
m
y
,
cu
ltu
r
e
,
an
d
f
o
o
d
s
ec
u
r
ity
.
Desp
ite
its
im
p
o
r
tan
ce
wh
ea
t
p
r
o
d
u
ctio
n
in
I
n
d
ia
f
ac
es
s
ev
er
al
ch
allen
g
es,
in
clu
d
in
g
clim
ate
ch
an
g
e,
wate
r
s
ca
r
city
,
s
o
il
d
eg
r
ad
atio
n
,
p
ests
,
an
d
d
is
ea
s
es.
Sev
er
al
s
tep
s
ar
e
b
ei
n
g
tak
en
to
a
d
d
r
ess
th
ese
ch
allen
g
es
an
d
en
s
u
r
e
th
e
f
u
tu
r
e
o
f
wh
ea
t
in
I
n
d
ia,
in
clu
d
in
g
d
e
v
elo
p
in
g
clim
ate
-
r
esil
ien
t
v
ar
ieties,
p
r
o
m
o
tin
g
s
u
s
tain
ab
le
f
ar
m
in
g
p
r
ac
tices,
an
d
in
v
esti
n
g
in
r
esear
ch
an
d
tech
n
o
lo
g
y
th
a
t
ar
e
ess
en
tial
f
o
r
in
cr
ea
s
in
g
y
ield
s
a
n
d
ad
d
r
es
s
n
ew
p
r
o
b
lem
s
.
Ad
d
r
ess
in
g
th
ese
p
r
o
b
lem
s
an
d
in
v
esti
n
g
in
s
u
s
tain
ab
le
p
r
ac
tices,
it
will
b
e
p
o
s
s
ib
le
t
o
en
s
u
r
e
th
e
im
p
o
r
tan
ce
o
f
w
h
ea
t
to
s
u
p
p
o
r
t
I
n
d
ia’
s
g
r
o
win
g
p
o
p
u
latio
n
an
d
its
ag
r
icu
ltu
r
al
ec
o
n
o
m
y
.
Pre
v
io
u
s
ly
h
is
to
r
ical
d
ata
an
d
ex
p
e
r
t
k
n
o
wled
g
e
wer
e
u
s
ed
b
y
f
a
r
m
er
s
f
o
r
p
r
e
d
ictin
g
cr
o
p
y
ield
.
C
r
o
p
y
ield
p
r
e
d
ictio
n
aim
s
to
co
n
tr
i
b
u
te
to
a
m
o
r
e
s
u
s
tain
ab
le,
r
esil
ien
t,
an
d
f
o
o
d
-
s
ec
u
r
e
f
u
tu
r
e
b
y
in
f
o
r
m
e
d
d
ec
is
io
n
-
m
a
k
in
g
s
u
p
p
o
r
tin
g
i
n
d
iv
id
u
al
f
ar
m
e
r
s
to
p
o
licy
m
ak
er
s
.
Yield
p
r
ed
ictio
n
s
till
f
ac
es c
h
a
llen
g
es d
esp
ite
tr
em
en
d
o
u
s
ad
v
an
ce
m
en
ts
.
Fo
r
lar
g
e
-
s
ca
le
ag
r
icu
ltu
r
al
o
p
er
atio
n
s
,
f
ield
s
u
r
v
ey
s
an
d
m
an
u
al
d
ata
co
llectio
n
tech
n
iq
u
es
ar
e
la
b
o
r
-
i
n
ten
s
iv
e,
ti
m
e
-
co
n
s
u
m
in
g
,
a
n
d
co
s
tly
[
1
]
.
T
h
e
s
tu
d
ies
s
h
o
w
a
s
h
if
t
f
r
o
m
tr
a
d
itio
n
al
s
tatis
tical
m
o
d
els
to
m
ac
h
i
n
e
lear
n
i
n
g
an
d
d
ee
p
lear
n
i
n
g
ap
p
r
o
ac
h
es
f
o
r
y
ield
p
r
e
d
ictio
n
.
Acc
o
r
d
in
g
to
Sh
ar
m
a
et
a
l.
[
2
]
,
au
to
r
e
g
r
ess
iv
e
in
teg
r
a
ted
m
o
v
in
g
av
er
a
g
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
4
1
1
-
3
4
2
0
3412
(
AR
I
MA
)
with
d
if
f
er
en
cin
g
f
o
r
s
tatio
n
ar
ity
a
n
d
q
u
an
til
e
r
eg
r
ess
io
n
f
o
r
ests
is
ap
p
lied
f
o
r
u
n
ce
r
tain
t
y
esti
m
atio
n
s
.
Nu
m
er
o
u
s
s
tu
d
ie
s
h
ig
h
lig
h
t
th
e
im
p
o
r
tan
ce
o
f
m
eteo
r
o
lo
g
ical
f
ac
to
r
s
[
3
]
–
[
5
]
,
s
o
il
m
o
is
tu
r
e
[
6
]
,
m
u
ltis
p
ec
tr
al
r
em
o
te
s
en
s
in
g
d
ata,
v
eg
etatio
n
in
d
ices
[
7
]
,
[
8
]
,
an
d
h
y
p
er
s
p
ec
tr
al
b
an
d
s
[
9
]
to
im
p
r
o
v
e
y
ield
p
r
ed
ictio
n
ac
cu
r
ac
y
.
Ag
r
icu
lt
u
r
al
E
R
A5
(
Ag
E
R
A5
)
[
1
0
]
s
u
p
p
o
r
t
ag
r
icu
ltu
r
al
an
aly
s
is
b
y
p
r
o
v
id
i
n
g
r
ea
d
y
to
u
s
e
m
eteo
r
o
lo
g
ical
d
ata.
Dee
p
lear
n
in
g
m
o
d
els
d
o
m
i
n
ates
th
e
p
r
ed
ictio
n
o
f
wh
ea
t,
m
aize
,
co
r
n
,
s
o
y
b
ea
n
,
tea
,
an
d
s
u
g
ar
ca
n
e
.
Su
n
et
a
l.
[
1
1
]
u
s
es
a
m
u
ltil
ev
el
d
ee
p
lear
n
in
g
d
ee
p
lear
n
in
g
n
etwo
r
k
with
tim
e
s
er
ies
s
atellite
d
ata
f
o
r
c
o
r
n
y
ield
p
r
e
d
ic
tio
n
wh
ile
o
th
er
s
tu
d
ies
e
m
p
lo
y
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
C
NN
)
-
t
r
an
s
f
o
r
m
er
[
1
2
]
,
h
y
b
r
id
C
NN
-
r
ec
u
r
r
e
n
t
n
eu
r
al
n
etwo
r
k
(
R
NN
)
m
o
d
els
[
1
3
]
,
B
ay
esian
o
p
ti
m
ized
d
ee
p
n
eu
r
al
n
etwo
r
k
(
DNN
)
[
1
4
]
,
a
n
d
c
o
n
v
o
l
u
ti
o
n
al
l
o
n
g
s
h
o
r
t
-
t
e
r
m
m
e
m
o
r
y
–
v
is
i
o
n
tr
an
s
f
o
r
m
er
(
C
o
n
v
L
ST
M
-
V
iT
)
[
1
5
]
to
ca
p
tu
r
e
s
p
atial
a
n
d
tem
p
o
r
al
p
atter
n
s
.
Sh
a
r
m
a
et
a
l.
[
1
6
]
co
m
p
ar
es
r
e
g
r
ess
io
n
a
n
d
d
e
ep
lear
n
i
n
g
m
o
d
els
u
s
in
g
m
u
ltip
le
ev
alu
atio
n
m
etr
ics,
an
d
Gu
p
ta
et
a
l.
[
1
7
]
co
n
f
ir
m
s
d
ee
p
lear
n
in
g
m
o
d
els o
u
t
p
er
f
o
r
m
tr
ad
itio
n
al
ap
p
r
o
ac
h
es in
lear
n
in
g
co
m
p
lex
n
o
n
li
n
ea
r
r
elatio
n
s
h
ip
s
.
Nu
m
er
o
u
s
s
tu
d
ies
d
is
cu
s
s
ed
e
n
s
em
b
le
m
o
d
els
[
1
8
]
,
[
1
9
]
,
p
r
in
cip
al
co
m
p
o
n
en
t
a
n
aly
s
is
(
PC
A
)
an
d
Sh
ap
ley
ad
d
itiv
e
ex
p
lan
atio
n
s
(
SHAP)
[
2
0
]
,
B
ay
esian
h
y
p
er
p
ar
a
m
eter
o
p
tim
izatio
n
[
2
1
]
to
im
p
r
o
v
e
r
o
b
u
s
tn
ess
an
d
in
te
r
p
r
etab
ilit
y
.
T
h
ese
s
tu
d
ies
s
h
o
w
en
s
e
m
b
le
m
o
d
els
ac
h
ie
v
e
b
etter
ac
cu
r
ac
y
th
a
n
s
in
g
le
m
o
d
els.
T
h
e
au
th
o
r
s
[
2
2
]
,
[
2
3
]
ap
p
lies
m
u
ltip
le
m
ac
h
i
n
e
le
ar
n
in
g
m
o
d
els
f
o
r
y
ield
p
r
ed
i
ctio
n
an
d
id
e
n
tifie
s
th
e
k
ey
f
ac
to
r
s
af
f
ec
tin
g
it
.
Ho
wev
er
,
wh
ile
d
ee
p
lear
n
in
g
d
o
m
in
ates
m
aize
an
d
wh
ea
t
p
r
ed
ictio
n
,
in
ter
p
r
etab
le
en
s
em
b
le
m
o
d
el
s
r
em
ain
less
ex
p
lo
r
ed
f
o
r
w
h
ea
t
y
ield
f
o
r
ec
asti
n
g
,
m
o
tiv
atin
g
th
e
n
ee
d
f
o
r
h
y
b
r
id
an
d
e
x
p
lain
ab
le
f
r
am
e
wo
r
k
s
.
T
o
ad
d
r
ess
th
ese,
th
is
s
tu
d
y
p
r
o
p
o
s
es
th
e
d
e
v
elo
p
m
e
n
t
o
f
a
r
eliab
le,
in
ter
p
r
eta
b
le
y
ield
p
r
ed
ictio
n
f
r
am
ewo
r
k
u
s
in
g
n
o
r
m
alize
d
d
if
f
er
en
ce
v
eg
etatio
n
i
n
d
ex
(
NDVI
)
,
m
eteo
r
o
lo
g
ical,
an
d
s
o
il
m
o
is
tu
r
e
d
ata
b
y
co
m
b
in
in
g
en
s
em
b
le
lear
n
in
g
m
eth
o
d
s
with
SHAP
-
b
ase
d
f
ea
tu
r
e
s
elec
tio
n
.
T
h
e
u
s
e
o
f
SHAP
f
ea
tu
r
e
s
elec
tio
n
an
d
m
o
d
el
in
ter
p
r
eta
b
ilit
y
ad
d
s
an
im
p
o
r
ta
n
t
d
eg
r
e
e
o
f
n
o
v
elty
.
I
t
im
p
r
o
v
es
m
o
d
el
tr
an
s
p
ar
en
cy
b
y
id
en
tify
in
g
m
o
s
t
in
f
lu
en
tial
en
v
ir
o
n
m
en
tal
f
ac
to
r
s
af
f
ec
ti
n
g
y
ield
an
d
h
elp
s
to
b
r
id
g
e
th
e
g
ap
b
etwe
en
m
ac
h
in
e
lear
n
in
g
a
n
d
d
o
m
a
in
-
s
p
ec
if
ic
d
ec
is
io
n
-
m
ak
in
g
.
T
h
is
s
tu
d
y
d
is
tin
g
u
is
h
es
it
s
elf
b
y
im
p
r
o
v
in
g
ex
p
lain
ab
ilit
y
alo
n
g
with
ac
cu
r
ac
y
,
m
a
k
in
g
th
e
r
esu
lts
m
o
r
e
ac
tio
n
a
b
le
an
d
r
eliab
le
f
o
r
ag
r
o
n
o
m
is
ts
,
p
o
licy
m
ak
er
s
,
a
n
d
f
a
r
m
er
s
.
T
h
e
s
tr
u
ctu
r
e
o
f
th
is
p
ap
er
is
o
r
g
an
ized
as
f
o
llo
ws
.
Sectio
n
1
in
tr
o
d
u
ce
s
th
e
im
p
o
r
ta
n
ce
o
f
ac
cu
r
ate
cr
o
p
y
ield
p
r
ed
ictio
n
a
n
d
th
e
lim
itatio
n
s
o
f
ex
is
tin
g
m
eth
o
d
s
.
I
t
h
ig
h
lig
h
ts
th
e
u
s
e
o
f
en
s
em
b
le
m
ac
h
i
n
e
lear
n
in
g
m
eth
o
d
s
an
d
SHAP
f
ea
tu
r
e
s
elec
tio
n
to
ad
d
r
ess
th
ese
lim
itatio
n
s
.
Me
th
o
d
o
lo
g
y
u
s
ed
in
th
e
p
ap
e
r
is
ex
p
lain
ed
in
s
ec
tio
n
2
.
Sectio
n
3
d
is
cu
s
s
es
th
e
p
er
f
o
r
m
a
n
ce
m
etr
ics
o
f
v
ar
io
u
s
en
s
em
b
le
m
eth
o
d
s
alo
n
g
with
SHAP f
ea
tu
r
e
s
elec
tio
n
f
o
r
th
e
b
est en
s
em
b
le
m
o
d
el.
Sectio
n
4
co
n
cl
u
d
es th
e
p
a
p
er
.
2.
M
E
T
H
O
D
2
.
1
.
Da
t
a
s
et
des
cr
iptio
n
I
n
th
is
s
tu
d
y
,
th
e
wh
ea
t
y
ield
d
ataset
is
o
b
tain
e
d
f
o
r
I
n
d
ia
u
s
in
g
C
Y
-
B
en
ch
d
ata
[
2
4
]
t
h
at
co
v
e
r
s
wh
ea
t
an
d
m
aize
cr
o
p
s
f
o
r
2
9
an
d
4
2
co
u
n
tr
ies.
A
b
en
ch
m
a
r
k
d
ataset
f
o
r
s
u
b
n
atio
n
al
cr
o
p
y
ield
p
r
ed
ictio
n
s
,
CY
-
B
en
ch
d
escr
ib
ed
in
T
ab
l
e
1
th
at
f
ac
ilit
ates
m
o
d
el
i
n
ter
co
m
p
ar
is
o
n
ac
r
o
s
s
d
i
v
er
s
e
ag
r
icu
ltu
r
al
s
y
s
tem
wo
r
ld
wid
e.
C
Y
-
B
en
ch
d
ataset
is
o
r
g
an
ized
as
a
co
llectio
n
o
f
C
SV
f
iles
,
wh
er
e
ea
ch
f
ile
r
ep
r
esen
ts
a
s
p
ec
if
ic
ca
teg
o
r
y
o
f
v
ar
iab
le
f
o
r
a
s
p
ec
if
ic
co
u
n
tr
y
.
T
h
e
C
SV f
ile
i
s
n
am
ed
b
ased
o
n
th
e
ca
teg
o
r
y
a
n
d
co
u
n
tr
y
,
h
elp
i
n
g
in
ea
s
y
id
e
n
tific
atio
n
a
n
d
r
etr
i
ev
al.
C
Y
-
B
en
ch
o
f
f
er
s
a
r
o
b
u
s
t,
m
u
lti
-
co
u
n
tr
y
d
ataset
f
o
r
y
ield
p
r
ed
ictio
n
b
u
t
f
ac
es
lim
itatio
n
s
r
eg
ar
d
in
g
s
p
atial
-
tem
p
o
r
al
r
eso
lu
tio
n
m
is
m
atch
an
d
d
ata
co
r
r
elati
o
n
s
.
I
t
f
o
cu
s
es
o
n
s
u
b
n
atio
n
al
b
u
t
n
o
t
f
ield
-
lev
e
l
d
ata
th
at
r
eq
u
ir
es
p
r
ep
r
o
ce
s
s
in
g
to
alig
n
th
e
v
ar
ied
p
r
e
d
i
cto
r
r
eso
lu
tio
n
s
th
at
m
ay
in
tr
o
d
u
ce
u
n
ce
r
tain
ty
.
T
ab
le
1
.
C
Y
-
B
en
ch
s
u
m
m
ar
y
tab
le
C
a
t
e
g
o
r
y
D
e
s
c
r
i
p
t
i
o
n
s
P
r
i
mary
c
r
o
p
s
M
a
i
z
e
,
w
h
e
a
t
S
p
a
t
i
a
l
c
o
v
e
r
a
g
e
M
a
i
z
e
(
3
8
c
o
u
n
t
r
i
e
s)
,
w
h
e
a
t
(
2
9
c
o
u
n
t
r
i
e
s)
S
p
a
t
i
a
l
r
e
s
o
l
u
t
i
o
n
S
u
b
n
a
t
i
o
n
a
l
(
h
i
g
h
e
st
a
v
a
i
l
a
b
l
e
r
e
s
o
l
u
t
i
o
n
a
t
a
d
mi
n
i
s
t
r
a
t
i
v
e
l
e
v
e
l
)
T
e
m
p
o
r
a
l
c
o
v
e
r
a
g
e
V
a
r
i
e
s
N
o
.
o
f
r
e
c
o
r
d
s
5
7
,
0
0
0
r
o
w
s
×
1
2
c
o
l
u
m
n
s
T
ab
le
2
d
escr
ib
es
th
e
d
ata
u
s
ed
in
th
is
s
tu
d
y
,
wh
ic
h
f
o
c
u
s
es
o
n
th
e
I
n
d
ian
wh
ea
t
cr
o
p
.
NDVI
d
ata
o
b
tain
ed
f
r
o
m
MO
D0
9
C
MG
[
2
5
]
h
as
a
s
p
atial
r
eso
lu
tio
n
o
f
5
0
0
m
eter
s
×
5
0
0
m
eter
s
with
a
wee
k
ly
tem
p
o
r
a
l
r
eso
lu
tio
n
.
Me
teo
r
o
l
o
g
ical
d
at
a
co
llected
f
r
o
m
Ag
E
R
A5
[
2
6
]
,
p
r
o
v
id
es
a
s
p
atial
r
eso
lu
tio
n
o
f
0
.
1
°
×
0
.
1
°
with
d
aily
tem
p
o
r
al
r
eso
lu
tio
n
.
So
il
m
o
is
tu
r
e
o
b
tain
ed
f
r
o
m
GL
DAS
[
2
7
]
h
as
a
s
p
atial
r
eso
lu
tio
n
1
°
an
d
1
/4
°
with
d
aily
tem
p
o
r
al
r
eso
lu
tio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
S
HA
P
-
en
h
a
n
ce
d
e
n
s
emb
le
lea
r
n
in
g
fo
r
yield
p
r
ed
ictio
n
:
in
s
ig
h
ts
fr
o
m
C
Y
-
B
en
ch
d
a
ta
o
n
…
(
S
o
ma
Gu
p
ta
)
3413
T
ab
le
2
.
Deta
iled
d
ata
d
escr
ip
t
io
n
D
a
t
a
D
e
s
c
r
i
p
t
i
o
n
V
a
r
i
a
b
l
e
s (u
n
i
t
s)
Te
mp
o
r
a
l
r
e
s
o
l
u
t
i
o
n
ndvi
N
o
r
m
a
l
i
z
e
d
d
i
f
f
e
r
e
n
c
e
v
e
g
e
t
a
t
i
o
n
i
n
d
e
x
-
w
e
e
k
l
y
m
e
t
e
o
Te
mp
e
r
a
t
u
r
e
,
p
r
e
c
i
p
i
t
a
t
i
o
n
,
r
a
d
i
a
t
i
o
n
,
p
o
t
e
n
t
i
a
l
e
v
a
p
o
t
r
a
n
s
p
i
r
a
t
i
o
n
,
c
l
i
ma
t
i
c
w
a
t
e
r
b
a
l
a
n
c
e
t
mi
n
(
◦
C
)
,
t
ma
x
(
◦
C
)
,
t
a
v
g
(
◦
C
)
,
p
r
e
c
(
mm
)
,
e
t
0
(
mm
)
,
c
w
b
(
mm
)
,
r
a
d
(J
m
-
2
d
a
y
-
1)
d
a
i
l
y
so
i
l
m
o
i
s
t
u
r
e
S
u
r
f
a
c
e
a
n
d
r
o
o
t
z
o
n
e
s
o
i
l
m
o
i
s
t
u
r
e
ssm
(
k
g
m
−
2
)
,
r
sm (
k
g
m
−
2
)
d
a
i
l
y
y
i
e
l
d
E
nd
-
of
-
se
a
so
n
y
i
e
l
d
y
i
e
l
d
(t
ha
−
1
)
y
e
a
r
l
y
2
.
2
.
Da
t
a
prepro
ce
s
s
ing
a
nd
y
ield pre
dict
io
n f
ra
m
ewo
rk
T
h
e
y
ield
p
r
e
d
ictio
n
s
tu
d
y
u
s
es
a
s
tan
d
alo
n
e
co
m
p
u
ter
s
y
s
t
em
with
J
u
p
y
ter
n
o
teb
o
o
k
,
Py
th
o
n
3
.
1
2
u
s
in
g
a
s
et
o
f
Py
th
o
n
lib
r
a
r
ies
f
o
r
d
ata
p
r
ep
r
o
ce
s
s
in
g
,
m
o
d
e
lin
g
,
ev
alu
atio
n
,
an
d
in
ter
p
r
et
ab
ilit
y
d
escr
ib
ed
in
T
ab
le
3
.
A
cr
u
cial
s
tep
in
th
e
m
ac
h
in
e
lear
n
in
g
p
ip
elin
e
is
p
r
ep
r
o
ce
s
s
in
g
th
e
d
ata
b
ec
au
s
e
r
aw
d
ata
co
llected
f
r
o
m
r
ea
l
-
wo
r
l
d
s
o
u
r
ce
s
is
o
f
ten
m
ess
y
an
d
n
o
t
im
m
ed
iately
s
u
itab
le
f
o
r
b
u
ild
in
g
ef
f
ec
t
iv
e
m
o
d
els.
B
ef
o
r
e
tr
ain
in
g
th
e
p
r
o
p
o
s
ed
m
o
d
els,
a
s
er
ies
o
f
f
ea
tu
r
e
en
g
in
ee
r
in
g
an
d
p
r
ep
r
o
ce
s
s
in
g
p
r
o
ce
d
u
r
es
wer
e
ap
p
lied
to
th
e
ac
q
u
ir
ed
d
ata
s
et.
I
n
th
is
s
tu
d
y
s
o
il
m
o
is
tu
r
e
an
d
m
etr
o
lo
g
ical
d
atasets
wer
e
g
r
o
u
p
e
d
o
n
ad
m
i
n
id
with
s
o
r
ted
d
ate
an
d
am
alg
am
ated
th
ese
d
atasets
h
av
in
g
d
aily
t
em
p
o
r
al
r
eso
lu
tio
n
.
Var
io
u
s
a
g
g
r
eg
ate
f
u
n
ctio
n
s
s
u
ch
as
m
ax
(
)
,
m
in
(
)
,
an
d
m
e
an
(
)
wer
e
ap
p
lied
o
n
th
e
co
lu
m
n
s
o
f
th
e
f
o
r
m
ed
d
ataset
to
ca
lcu
late
th
e
wee
k
ly
v
alu
es.
Me
an
ag
g
r
eg
atio
n
ca
p
tu
r
ed
th
e
tem
p
o
r
al
tr
en
d
s
,
r
ed
u
ce
d
n
o
is
e
in
d
aily
o
b
s
er
v
atio
n
s
,
p
r
o
v
id
e
d
m
o
r
e
s
tab
le
p
er
f
o
r
m
a
n
ce
,
lo
wer
p
r
ed
ictio
n
er
r
o
r
co
m
p
ar
e
d
to
m
ax
(
)
/m
in
(
)
an
d
is
th
u
s
ch
o
s
en
o
v
er
m
ax
(
)
/m
i
n
(
)
.
W
ee
k
ly
f
o
r
m
e
d
d
ataset
h
av
in
g
s
o
il
an
d
m
etr
o
lo
g
ical
f
ea
tu
r
es
ar
e
th
e
n
m
e
r
g
ed
with
N
DVI
d
ataset
h
av
in
g
wee
k
ly
tem
p
o
r
al
r
eso
lu
tio
n
.
NDVI
is
tem
p
o
r
ally
ag
g
r
eg
ate
d
u
s
in
g
s
u
m
m
ar
y
s
tatis
tics
an
d
u
s
ed
as
an
in
p
u
t
f
o
r
y
ield
p
r
ed
ictio
n
as
th
is
ap
p
r
o
a
ch
r
ed
u
ce
s
d
im
en
s
i
o
n
ality
a
n
d
n
o
is
e.
Fin
ally
,
th
e
r
esu
ltin
g
d
ataset
was
m
er
g
ed
with
y
iel
d
d
ataset.
On
am
alg
am
atin
g
co
lu
m
n
s
s
u
ch
as
co
u
n
tr
y
co
d
e,
s
ea
s
o
n
n
a
m
e,
p
lan
tin
g
,
y
ea
r
,
p
lan
ted
ar
e
a
wer
e
id
en
tifie
d
as
ir
r
elev
an
t
to
th
e
r
esear
ch
an
d
is
elim
in
ated
as
th
e
s
tu
d
y
d
o
m
ain
is
I
n
d
ia.
T
h
e
i
d
en
tific
atio
n
a
n
d
h
an
d
lin
g
o
f
n
u
ll
v
alu
es
is
a
cr
u
cial
s
tep
i
n
th
e
p
r
ep
ar
atio
n
an
d
an
aly
s
i
s
o
f
d
ata.
Fo
llo
win
g
a
th
o
r
o
u
g
h
an
aly
s
is
o
f
n
u
ll
v
alu
es,
th
e
d
ataset
was
d
eter
m
in
ed
to
b
e
n
u
ll
v
alu
e
-
f
r
ee
,
en
ab
lin
g
u
s
to
m
o
v
e
o
n
with
th
is
n
u
ll
-
v
alu
e
-
f
r
ee
d
ata.
An
ess
en
tial
p
r
ep
r
o
ce
s
s
in
g
s
tep
in
d
ata
an
aly
s
is
an
d
m
ac
h
in
e
lear
n
in
g
is
id
en
tify
i
n
g
an
d
elim
in
ati
n
g
d
u
p
licate
r
ec
o
r
d
s
.
I
n
th
is
s
tu
d
y
n
o
d
u
p
licate
r
ec
o
r
d
s
wer
e
f
o
u
n
d
.
No
r
m
aliza
tio
n
p
r
e
p
r
o
ce
s
s
in
g
ap
p
r
o
ac
h
g
u
ar
an
tees
th
at
ev
er
y
f
ea
tu
r
e
m
ak
es
an
eq
u
al
co
n
tr
ib
u
tio
n
to
th
e
m
o
d
el’
s
lear
n
i
n
g
p
r
o
ce
s
s
.
T
h
e
r
esu
ltan
t
d
ataset
co
m
p
r
is
es
twelv
e
co
lu
m
n
s
,
with
‘
Yield
’
as
th
e
tar
g
e
t
an
d
th
e
r
em
ain
in
g
elev
e
n
as
f
ea
tu
r
es.
Fig
u
r
e
1
s
h
o
ws th
e
p
air
wis
e
co
r
r
elatio
n
s
am
o
n
g
th
e
ag
r
o
clim
atic
v
ar
i
ab
les,
s
o
il v
ar
iab
les,
v
eg
etatio
n
in
d
ices,
an
d
y
ield
s
h
o
win
g
th
e
k
e
y
r
elatio
n
s
h
ip
r
elev
an
t to
y
ield
p
r
ed
ictio
n
.
T
ab
le
3
.
So
f
twar
e
e
n
v
ir
o
n
m
en
t
S
y
st
e
m
s
p
e
c
i
f
i
c
a
t
i
o
n
D
e
s
c
r
i
p
t
i
o
n
s
O
p
e
r
a
t
i
n
g
s
y
st
e
m
W
i
n
d
o
w
s 1
1
P
r
o
c
e
ss
o
r
I
n
t
e
l
C
o
r
e
i
5
IDE
P
y
C
h
a
r
m
p
r
o
f
e
ssi
o
n
a
l
2
0
2
4
.
3
.
4
P
y
t
h
o
n
v
e
r
s
i
o
n
3
.
1
2
D
a
t
a
p
r
o
c
e
s
si
n
g
p
a
n
d
a
s,
N
u
mP
y
,
s
k
l
e
a
r
n
V
i
su
a
l
i
z
a
t
i
o
n
mat
p
l
o
t
l
i
b
,
sea
b
o
r
n
Fig
u
r
e
1
.
Featu
r
e
co
r
r
elatio
n
h
ea
tm
ap
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
4
1
1
-
3
4
2
0
3414
Fig
u
r
e
2
illu
s
tr
ates
th
e
c
o
m
p
r
eh
en
s
iv
e
wo
r
k
f
lo
w
ad
o
p
ted
f
o
r
y
ield
p
r
e
d
ictio
n
u
s
in
g
th
e
C
Y
-
B
en
ch
d
ataset.
T
h
e
p
r
o
ce
s
s
b
eg
in
s
with
d
ata
ac
q
u
is
itio
n
f
r
o
m
th
e
C
Y
-
B
en
ch
d
ataset,
wh
ich
co
m
p
r
is
es
f
o
u
r
k
ey
co
m
p
o
n
en
ts
m
eteo
r
o
lo
g
ical
d
ata,
s
o
il
m
o
is
tu
r
e
d
ata,
NDV
I
,
an
d
cr
o
p
y
ield
d
ata.
Af
ter
p
r
ep
r
o
ce
s
s
in
g
o
f
s
ev
er
al
p
ar
titi
o
n
in
g
r
atio
s
co
n
s
id
er
ed
,
p
ar
titi
o
n
in
g
r
atio
o
f
6
0
%,
20
%,
an
d
2
0
%
f
o
r
tr
ai
n
in
g
,
v
alid
atio
n
,
an
d
test
in
g
is
tak
en
,
as
it
r
esu
lted
in
lo
w
v
a
r
ian
ce
o
n
1
0
r
ep
etitio
n
s
.
Par
titi
o
n
in
g
en
s
u
r
es
th
e
m
o
d
els
ar
e
tr
ai
n
ed
,
v
alid
ated
,
an
d
test
ed
o
n
s
ep
ar
ate
s
u
b
s
ets.
Her
e,
5
-
f
o
ld
c
r
o
s
s
v
alid
atio
n
tech
n
iq
u
e
is
em
p
l
o
y
ed
,
wh
e
r
e
th
e
d
ata
is
p
ar
titi
o
n
ed
in
to
f
iv
e
d
is
jo
in
t f
o
ld
s
.
I
n
ea
ch
iter
atio
n
,
o
n
e
-
f
o
ld
is
u
s
ed
as test an
d
th
e
r
em
ain
in
g
f
o
u
r
ar
e
u
s
ed
f
o
r
tr
ain
in
g
.
T
h
e
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
ar
e
ca
r
r
ied
o
u
t
with
in
ea
ch
tr
ain
in
g
f
o
ld
o
n
ly
,
th
er
e
b
y
en
s
u
r
in
g
th
e
test
f
o
ld
r
em
ain
s
co
m
p
letely
u
n
s
e
en
d
u
r
in
g
tr
ain
in
g
.
T
h
is
ap
p
r
o
ac
h
s
tr
ictly
en
s
u
r
es
th
at
th
er
e
is
n
o
d
ata
leak
ag
e
an
d
th
e
in
te
g
r
ity
o
f
ev
alu
atio
n
p
r
o
ce
s
s
is
m
ain
tain
ed
.
Fig
u
r
e
2
.
C
o
m
p
r
eh
en
s
iv
e
f
r
am
ewo
r
k
f
o
r
y
ield
p
r
ed
ictio
n
Sev
er
al
b
aselin
e
m
o
d
els,
s
u
ch
as
m
u
ltip
le
lin
ea
r
r
eg
r
ess
io
n
(
L
R
)
,
d
ec
is
io
n
tr
ee
(
DT
)
,
r
an
d
o
m
f
o
r
est
(
R
F)
,
k
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
,
s
u
p
p
o
r
t
v
ec
to
r
r
eg
r
e
s
s
io
n
(
SVR
)
,
an
d
ex
tr
em
e
g
r
ad
ien
t
b
o
o
s
tin
g
(
XGBo
o
s
t)
,
ar
e
u
s
ed
to
d
ev
el
o
p
en
s
em
b
le
m
o
d
els.
Var
i
o
u
s
en
s
em
b
le
m
ac
h
i
n
e
lear
n
i
n
g
t
ec
h
n
iq
u
es
s
u
ch
as
v
o
tin
g
,
s
tack
in
g
an
d
,
b
o
o
s
tin
g
en
s
em
b
le
ar
e
u
s
ed
.
T
h
e
g
o
al
o
f
u
s
in
g
a
n
en
s
em
b
le
is
to
im
p
r
o
v
e
p
er
f
o
r
m
an
ce
(
lik
e
ac
cu
r
ac
y
o
r
r
o
b
u
s
tn
ess
)
co
m
p
ar
ed
t
o
an
y
s
in
g
le
m
o
d
el.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
el
is
ev
alu
ated
u
s
in
g
th
e
p
er
f
o
r
m
an
ce
m
etr
ics,
n
a
m
ely
m
ea
n
s
q
u
ar
e
d
er
r
o
r
(
MSE
)
,
r
o
o
t
m
ea
n
s
q
u
ar
ed
e
r
r
o
r
(
R
MSE
)
,
m
ea
n
ab
s
o
lu
te
er
r
o
r
(
MA
E
)
,
an
d
co
ef
f
icien
t
o
f
d
eter
m
in
ati
o
n
(
R²
)
.
T
r
an
s
p
a
r
en
cy
an
d
in
ter
p
r
etab
ilit
y
o
f
t
h
e
m
o
d
el
p
r
ed
ictio
n
s
is
en
h
an
ce
d
u
s
in
g
SHAP
in
th
e
ex
p
lain
a
b
ilit
y
p
h
ase
[
2
8
]
.
T
h
e
d
ataset
is
d
iv
id
e
d
in
to
t
r
ain
in
g
an
d
test
d
ata
to
av
o
id
d
ata
leak
ag
e.
T
h
e
m
o
d
els
ar
e
tr
ain
ed
u
s
in
g
tr
ain
in
g
d
ata,
an
d
SHAP
v
alu
es
ar
e
co
m
p
u
ted
u
s
in
g
o
n
ly
tr
ain
e
d
m
o
d
els
an
d
tr
ain
in
g
s
am
p
les.
Featu
r
e
im
p
o
r
tan
ce
a
n
d
s
elec
tio
n
is
d
o
n
e
u
s
in
g
th
e
tr
ain
in
g
SHAP
v
alu
es.
T
h
e
s
elec
ted
f
ea
tu
r
es
ar
e
th
e
n
ap
p
lied
to
th
e
test
d
ata
f
o
r
m
o
d
el
ev
al
u
atio
n
en
s
u
r
in
g
n
o
in
f
o
r
m
atio
n
f
r
o
m
th
e
test
d
ata
in
f
lu
en
ce
s
f
ea
tu
r
e
s
elec
tio
n
o
r
m
o
d
el
tr
ain
in
g
.
T
h
is
s
tep
p
r
o
v
id
es
in
s
ig
h
ts
in
to
th
e
co
n
tr
ib
u
tio
n
o
f
in
d
i
v
id
u
al
f
ea
tu
r
es to
m
o
d
el
o
u
tp
u
ts
,
s
u
p
p
o
r
tin
g
i
n
f
o
r
m
ed
d
ec
is
io
n
-
m
a
k
in
g
.
T
h
e
d
etailed
co
m
p
o
s
itio
n
o
f
b
ase
lear
n
er
s
,
en
s
em
b
le
s
tr
ateg
ies
,
an
d
th
e
p
ar
a
m
eter
s
ettin
g
u
s
ed
in
th
e
s
tu
d
y
is
d
escr
ib
ed
as
f
o
llo
ws.
Vo
tin
g
m
o
d
els
u
s
e
s
o
f
t
v
o
tin
g
with
co
m
b
in
atio
n
s
o
f
L
R
,
DT
,
R
F,
KN
N,
SV
R
,
an
d
XGB
o
o
s
t
.
Key
p
ar
am
eter
s
f
o
r
R
F
an
d
XGB
o
o
s
t
in
clu
d
e
1
0
0
esti
m
ato
r
s
,
DT
d
ep
th
o
f
4
,
KNN
(
k
=5
)
,
an
d
SVR
wi
th
r
ad
ial
b
asis
f
u
n
ctio
n
(
R
B
F)
k
er
n
el
(
C
=1
0
0
,
γ
=
0
.
1
)
.
Stack
in
g
m
o
d
els
u
s
es
s
im
ilar
b
ase
lear
n
er
s
with
m
eta
-
m
o
d
el
o
f
1
0
0
esti
m
ato
r
s
,
lear
n
in
g
r
ate
o
f
0
.
1
,
a
n
d
m
ax
im
u
m
d
e
p
th
o
f
3
.
B
o
o
s
tin
g
m
o
d
els
ap
p
lies
s
eq
u
en
tial
b
o
o
s
tin
g
with
s
el
ec
ted
lear
n
er
s
c
o
n
f
i
g
u
r
ed
wi
th
1
0
0
esti
m
ato
r
s
,
lear
n
in
g
r
ate
o
f
0
.
0
5
,
a
n
d
m
ax
im
u
m
d
ep
th
o
f
3
.
Alg
o
r
ith
m
1
to
3
d
escr
ib
es th
e
en
s
em
b
le
r
eg
r
ess
io
n
tech
n
i
q
u
es
[
2
9
]
th
at
im
p
r
o
v
e
th
e
p
er
f
o
r
m
an
ce
b
y
co
m
b
in
in
g
th
e
o
u
tp
u
ts
o
f
m
u
lt
ip
le
b
ase
lear
n
er
s
.
Vo
tin
g
g
en
er
ates f
in
al
p
r
ed
ictio
n
b
y
a
v
er
ag
in
g
th
e
o
u
tp
u
ts
o
f
s
ev
er
al
in
d
ep
e
n
d
en
t
r
eg
r
ess
o
r
s
th
u
s
r
ed
u
ci
n
g
v
ar
ian
ce
an
d
im
p
r
o
v
in
g
r
o
b
u
s
tn
ess
.
Stack
i
n
g
em
p
lo
y
s
a
m
eta
r
eg
r
ess
o
r
to
lea
r
n
th
e
o
p
tim
al
co
m
b
in
atio
n
o
f
p
r
ed
ictio
n
s
f
r
o
m
m
u
ltip
le
b
ase
m
o
d
els.
B
o
o
s
tin
g
s
eq
u
en
tially
co
n
s
tr
u
cts
a
s
tr
o
n
g
p
r
ed
ictiv
e
m
o
d
el
b
y
s
eq
u
e
n
tially
tr
ain
i
n
g
wea
k
lear
n
er
s
wh
er
e
ea
c
h
s
u
cc
ess
iv
e
lear
n
er
f
o
cu
s
es o
n
co
r
r
ec
tin
g
t
h
e
er
r
o
r
s
m
ad
e
b
y
its
p
r
e
d
ec
ess
o
r
s
.
Alg
o
r
ith
m
1
.
Vo
tin
g
r
eg
r
ess
o
r
alg
o
r
ith
m
I
n
p
u
t:
g
iv
en
M
r
eg
r
ess
io
n
m
o
d
els
1
,
2
,
…,
;
d
ataset
D
.
(
)
=
(
)
is
m
o
d
el
p
r
ed
ictio
n
o
f
o
n
in
p
u
t
.
Ou
tp
u
t:
av
er
ag
e
o
f
p
r
ed
ictio
n
s
ac
r
o
s
s
all
k
m
o
d
els.
Step
1
:
f
o
r
ea
c
h
=
1
to
k
,
tr
ain
m
o
d
el
on
D
.
Step
2
:
p
r
ed
ict
u
s
in
g
ea
ch
m
o
d
el.
Fo
r
n
ew
in
p
u
t
x
,
c
o
m
p
u
te
p
r
ed
ictio
n
s
(
)
=
(
)
f
o
r
=
1
,
2
,
…
,
.
Step
3
:
co
m
p
u
te
th
e
f
in
al
p
r
ed
ictio
n
u
s
in
g
av
e
r
ag
e
(
s
im
p
le
v
o
tin
g
)
:
(
)
=
1
∑
(
)
=
1
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
S
HA
P
-
en
h
a
n
ce
d
e
n
s
emb
le
lea
r
n
in
g
fo
r
yield
p
r
ed
ictio
n
:
in
s
ig
h
ts
fr
o
m
C
Y
-
B
en
ch
d
a
ta
o
n
…
(
S
o
ma
Gu
p
ta
)
3415
Alg
o
r
ith
m
2
.
Stack
in
g
r
e
g
r
ess
o
r
alg
o
r
ith
m
I
n
p
u
t:
=
{
(
,
)
}
=
1
b
e
th
e
d
ataset.
1
,
2
,
…,
:
b
ase
r
eg
r
ess
o
r
s
(
lev
el
0
)
.
:
m
eta
r
eg
r
ess
o
r
(
lev
el
1
)
.
(
)
: m
o
d
el
M
j
p
r
ed
ictio
n
f
o
r
i
n
p
u
t
x
i
.
Ou
tp
u
t:
p
r
ed
ictio
n
g
en
e
r
ated
b
y
th
e
m
eta
-
r
e
g
r
ess
o
r
u
s
in
g
c
o
m
b
in
ed
p
r
ed
ictio
n
s
o
f
all
b
ase
r
eg
r
ess
o
r
s
.
Step
1
:
tr
ain
th
e
b
ase
m
o
d
els.
E
ac
h
b
ase
m
o
d
el
M
k
is
tr
ain
ed
o
n
t
h
e
tr
ain
in
g
d
ataset:
(
)
=
(
)
f
o
r
all
=
1
,
…
,
an
d
=
1
,
…
,
.
Step
2
:
f
o
r
ea
ch
in
p
u
t
x
i
,
co
lle
ct
p
r
ed
ictio
n
f
r
o
m
all
b
ase
m
o
d
els:
=
[
(
1
)
,
(
2
)
,
…
,
(
)
]
wh
ich
f
o
r
m
s
a
n
ew
d
ataset
=
{
(
,
)
}
=
1
Step
3
:
tr
ain
th
e
m
eta
r
eg
r
ess
o
r
o
n
d
ataset
Z
.
Step
4
:
m
ak
e
f
in
al
p
r
ed
ictio
n
s
:
i)
Get
b
ase
m
o
d
el
p
r
e
d
ictio
n
s
f
o
r
x
new
:
(
)
=
(
)
,
=
1
,
…
,
.
ii)
C
r
ea
te
m
eta
f
ea
tu
r
e
v
ec
to
r
=
[
(
1
)
,
…
,
(
)
]
.
iii)
Pre
d
ict
with
m
eta
m
o
d
el:
=
(
)
.
Alg
o
r
ith
m
3
.
B
o
o
s
tin
g
r
eg
r
ess
o
r
alg
o
r
ith
m
I
n
p
u
t:
tr
ain
in
g
d
ata
=
{
(
1
,
1
)
,
(
2
,
2
)
,
…
,
(
,
)
}
.
W
ea
k
lear
n
er
h
(
x
,
θ
)
.
Nu
m
b
er
o
f
b
o
o
s
tin
g
r
o
u
n
d
s
M
.
L
ea
r
n
in
g
r
ate
v
.
Ou
tp
u
t:
a
s
tr
o
n
g
p
r
ed
ictiv
e
m
o
d
el
(
)
f
o
r
m
e
d
b
y
s
eq
u
e
n
tially
co
m
b
in
i
n
g
all
t
h
e
wea
k
lear
n
er
s
wit
h
o
p
tim
ized
weig
h
ts
.
Step
1
:
in
itialize
th
e
m
o
d
el
wi
th
co
n
s
tan
t p
r
e
d
icto
r
:
0
(
)
=
min
∑
(
,
)
=
1
.
Step
2
:
f
o
r
m
=
1
to
M
d
o
:
i)
C
o
m
p
u
te
r
esid
u
als:
r
i
=
−
∂
L
(
y
i
,
F
(
x
i
)
∂
F
(
x
i
)
.
i
i
)
Fit
a
wea
k
lear
n
er
h
m
(
x
)
to
p
r
ed
ict
r
esid
u
als:
h
m
(
x
)
~
r
i
i
i
i
)
C
o
m
p
u
te
s
tep
s
ize:
γ
m
=
min
γ
∑
L
(
y
i
,
F
m
−
1
(
x
i
)
+
γ
h
m
(
x
i
)
)
n
i
=
1
.
i
v
)
Up
d
ate
th
e
m
o
d
el:
F
m
(
x
)
=
F
m
−
1
(
x
)
+
v
γ
m
h
m
(
x
)
.
Step
3
:
o
u
tp
u
t
f
in
al
b
o
o
s
ted
m
o
d
el
F
M
(
x
).
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
F
i
g
u
r
e
3
i
l
l
u
s
tr
a
t
e
th
e
p
er
f
o
r
m
a
n
c
e
o
f
th
e
e
n
s
e
m
b
l
e
m
o
d
e
l
s
.
F
i
g
u
r
e
3
(
a
)
p
r
e
s
e
n
t
s
th
e
a
c
t
u
a
l
a
n
d
p
r
ed
i
c
t
e
d
y
i
e
l
d
v
a
l
u
e
s
o
b
t
a
in
e
d
u
s
i
n
g
t
h
e
b
o
o
s
t
m
1
m
o
d
e
l,
w
h
i
l
e
F
ig
u
r
e
s
3
(
b
)
an
d
3
(
c
)
c
o
m
p
a
r
e
th
e
R
M
S
E
a
n
d
R
²
v
a
lu
e
s
o
f
a
l
l
e
n
s
em
b
l
e
m
o
d
e
l
s
,
r
e
s
p
ec
t
i
v
e
ly
.
B
o
o
s
t
m
1
ac
h
i
ev
e
d
t
h
e
l
o
w
e
s
t
R
M
S
E
(
0
.
2
8
9
)
a
n
d
M
A
E
(
0
.
2
1
)
,
w
i
th
a
h
i
g
h
R
²
s
c
o
r
e
(
0
.
9
2
)
,
i
n
d
i
c
a
t
in
g
s
t
r
o
n
g
a
c
c
u
r
a
cy
a
n
d
g
en
e
r
a
l
i
z
at
i
o
n
ab
i
l
i
t
y
o
f
t
h
e
m
o
d
e
l
.
A
m
o
n
g
a
l
l
t
h
e
v
o
t
in
g
m
o
d
e
l
s
,
v
o
t
e
m
3
h
a
s
t
h
e
b
e
s
t
(
R
M
S
E
=
0
.
3
7
,
R
²
=0
.
8
6
2
)
.
s
t
a
c
k
m
1
s
h
o
w
s
th
e
b
e
s
t
(
R
M
S
E
=
0
.
4
1
,
R
²
=
0
.
8
3
)
,
w
h
i
l
e
s
t
a
ck
m
3
e
x
h
i
b
i
t
ed
t
h
e
w
o
r
s
t
(
R
M
S
E
=
0
.
4
4
,
R
²
=
0
.
8
1
)
o
f
a
l
l
t
h
e
s
t
a
c
k
i
n
g
m
o
d
e
l
s
.
T
h
e
r
e
s
u
l
ts
s
h
o
w
v
o
t
in
g
a
n
d
s
t
a
c
k
in
g
e
n
s
e
m
b
l
e
s
p
r
o
v
i
d
e
co
m
p
e
t
it
i
v
e
p
e
r
f
o
r
m
a
n
c
e
,
b
o
o
s
t
in
g
w
i
l
l
p
r
o
v
id
e
m
o
r
e
r
el
i
a
b
l
e
s
o
lu
t
i
o
n
i
n
co
m
p
l
e
x
ag
r
i
c
u
l
tu
r
a
l
p
r
ed
i
c
t
i
o
n
s
c
en
a
r
io
s
.
T
a
b
l
e
4
p
r
o
v
id
e
s
R
M
S
E
,
M
A
E
,
R
²,
an
d
M
S
E
p
e
r
f
o
r
m
a
n
c
e
m
e
t
r
i
c
s
f
o
r
v
a
r
io
u
s
en
s
e
m
b
l
e
m
o
d
e
l
ap
p
l
i
ed
f
o
r
y
i
e
ld
p
r
ed
i
c
t
io
n
.
5
-
f
o
l
d
c
r
o
s
s
v
a
l
i
d
a
t
io
n
i
s
p
er
f
o
r
m
ed
an
d
m
ea
n
±
s
t
a
n
d
a
r
d
d
e
v
i
a
t
io
n
f
o
r
d
if
f
er
e
n
t
ev
al
u
a
t
i
o
n
m
e
t
r
i
ce
s
is
h
i
g
h
l
i
g
h
t
e
d
in
T
a
b
l
e
5
.
T
h
e
s
e
t
e
s
t
s
w
e
r
e
ap
p
l
ie
d
o
n
f
o
ld
-
w
i
s
e
r
e
s
u
l
t
s
s
o
th
a
t
ea
c
h
c
o
m
p
a
r
i
s
o
n
u
s
e
s
p
a
i
r
ed
o
b
s
e
r
v
a
t
io
n
s
(
s
am
e
f
o
ld
s
)
.
(
a)
(
b
)
(
c)
Fig
u
r
e
3
.
Per
f
o
r
m
an
c
e
m
etr
ice
s
p
lo
ts
o
f
(
a)
ac
tu
al
v
/s
p
r
ed
ict
ed
y
ield
,
(
b
)
R
MSE
,
an
d
(
c)
R
2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
4
1
1
-
3
4
2
0
3416
T
ab
le
4
.
Per
f
o
r
m
an
ce
o
f
e
n
s
em
b
le
m
eth
o
d
s
M
o
d
e
l
s
R
M
S
E
M
A
E
R²
M
S
E
v
o
t
e
m1
0
.
3
9
0
.
2
9
0
.
8
5
0
.
1
5
v
o
t
e
m2
0
.
4
2
0
.
3
1
0
.
8
2
0
.
1
8
v
o
t
e
m3
0
.
3
7
0
.
2
9
0
.
8
6
0
.
1
4
st
a
c
k
m
1
0
.
4
1
0
.
3
3
0
.
8
3
0
.
1
7
st
a
c
k
m
2
0
.
4
2
0
.
3
2
0
.
8
2
0
.
1
7
st
a
c
k
m
3
0
.
4
4
0
.
3
5
0
.
8
1
0
.
1
9
b
o
o
s
t
m
1
0
.
2
8
0
.
2
1
0
.
9
2
0
.
0
8
b
o
o
s
t
m
2
0
.
4
0
.
3
0
.
8
4
0
.
1
6
T
ab
le
5
.
5
-
f
o
ld
C
V
r
esu
lts
(
m
ea
n
±
s
tan
d
ar
d
d
ev
iatio
n
)
M
o
d
e
l
s
R
M
S
E
M
A
E
R²
v
o
t
e
m
1
0
.
3
9
±
0
.
0
3
0.
29±
0.
02
0.
85±
0.
02
v
o
t
e
m
2
0
.
4
2
±
0
.
0
2
0.
31±
0.
03
0.
82±
0.
02
v
o
t
e
m
3
0
.
3
7
±
0
.
0
2
0.
29±
0.
01
0.
86±
0.
02
st
ac
k
m1
0
.
4
1
±
0
.
0
3
0.
33±
0.
02
0.
83±
0.
02
st
ac
k
m2
0
.
4
2
±
0
.
0
2
0.
32±
0.
02
0.
82±
0.
02
st
ac
k
m3
0
.
4
4
±
0
.
0
3
0.
35±
0.
03
0.
81±
0.
02
b
o
o
s
t
m
1
0
.
2
8
±
0
.
0
1
0.
21±
0.
01
0.
92±
0.
01
b
o
o
s
t
m
2
0
.
4
0
±
0
.
0
3
0.
30±
0.
01
0.
84±
0.
02
T
o
o
b
s
er
v
e
a
n
d
d
eter
m
in
e
th
e
im
p
ac
ts
o
f
ex
p
lain
ab
le
f
ea
t
u
r
e
s
elec
tio
n
,
th
e
b
est
en
s
em
b
le
m
o
d
el
(
b
o
o
s
t
m
1
)
is
en
h
an
ce
d
b
y
ad
d
in
g
a
f
ea
t
u
r
e
F1
-
s
co
r
e
,
o
b
tai
n
ed
th
r
o
u
g
h
SHAP
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
.
T
h
e
r
esu
ltin
g
m
o
d
el,
b
o
o
s
t
m
1
+
F1
,
s
h
o
ws
n
o
tab
le
im
p
r
o
v
em
en
ts
ac
r
o
s
s
all
t
h
e
e
v
alu
atio
n
m
et
r
ics,
as
s
u
m
m
ar
ized
in
T
ab
le
6
.
T
h
e
p
er
f
o
r
m
an
ce
ev
alu
atio
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
els
s
h
o
ws
th
e
im
p
ac
t
o
f
f
ea
tu
r
e
s
elec
tio
n
o
n
p
r
ed
ictiv
e
ac
cu
r
ac
y
.
T
h
e
b
aselin
e
b
o
o
s
t
m
1
m
o
d
el
ex
h
ib
ited
s
tr
o
n
g
p
er
f
o
r
m
an
ce
,
ac
h
iev
in
g
a
R
MSE
o
f
0
.
2
8
,
MA
E
o
f
0
.
2
1
,
an
d
R
²
o
f
0
.
9
2
.
T
h
ese
r
esu
lts
in
d
icate
th
at
th
e
en
s
em
b
le
b
o
o
s
tin
g
ap
p
r
o
ac
h
is
ef
f
ec
tiv
e
in
ca
p
t
u
r
in
g
th
e
n
o
n
l
in
ea
r
r
elatio
n
s
h
ip
s
with
in
th
e
i
n
p
u
t
f
ea
tu
r
es
co
m
p
r
is
in
g
NDVI
,
m
eteo
r
o
l
o
g
ical,
an
d
s
o
il
m
o
is
tu
r
e
d
ata.
Fo
ll
o
win
g
th
e
a
p
p
licatio
n
o
f
F
1
-
b
ased
f
ea
t
u
r
e
s
elec
tio
n
,
t
h
e
en
h
an
ce
d
m
o
d
el
(
b
o
o
s
t
m
1
+
F1
)
s
h
o
wed
a
m
ar
k
ed
im
p
r
o
v
e
m
en
t
ac
r
o
s
s
all
ev
alu
atio
n
m
etr
ics.
R
MSE
an
d
MA
E
wer
e
s
ig
n
if
ican
tly
r
ed
u
ce
d
to
0
.
1
7
an
d
0
.
1
5
,
r
esp
ec
tiv
ely
,
wh
ile
th
e
MSE
d
ec
r
ea
s
ed
f
r
o
m
0
.
0
8
to
0
.
0
3
.
T
h
e
R
²
v
alu
e
also
in
cr
ea
s
ed
to
0
.
9
6
,
in
d
icatin
g
a
s
tr
o
n
g
er
f
it
an
d
im
p
r
o
v
e
d
g
en
er
aliza
tio
n
c
ap
ab
ilit
y
.
Fig
u
r
e
4
illu
s
tr
ates
h
o
w
SHAP
tr
ee
ex
p
lain
er
is
u
s
e
d
to
co
m
p
u
te
SHAP
v
alu
es
f
r
o
m
m
o
d
el
p
r
e
d
ictio
n
s
en
ab
lin
g
g
lo
b
al
an
d
lo
ca
l f
ea
tu
r
e
attr
ib
u
tio
n
an
d
s
u
m
m
ar
izin
g
f
ea
tu
r
e
im
p
o
r
t
an
ce
.
T
ab
le
6
.
Per
f
o
r
m
an
ce
m
etr
ics f
o
r
b
est en
s
em
b
le
m
o
d
el
with
an
d
with
o
u
t
f
ea
tu
r
e
s
elec
tio
n
M
e
t
r
i
c
b
o
o
s
t
m
1
b
o
o
s
t
m
1
+
F
1
(
F
1
=
s
e
l
e
c
t
e
d
f
e
a
t
u
r
e
s
u
si
n
g
S
H
A
P
:
p
r
o
d
u
c
t
i
o
n
,
h
a
r
v
e
st
a
r
e
a
,
t
mi
n
,
N
D
V
I
,
t
a
v
g
)
R
M
S
E
0
.
2
8
0
.
1
7
M
A
E
0
.
2
1
0
.
1
5
M
S
E
0
.
0
8
0
.
0
3
R2
0
.
9
2
0
.
9
6
Fig
u
r
e
4
.
SHAP
-
b
ased
m
o
d
el
in
ter
p
r
etab
ilit
y
wo
r
k
f
lo
w
F
i
g
u
r
e
5
p
r
e
s
e
n
t
s
th
e
S
H
A
P
-
b
a
s
e
d
e
x
p
l
a
i
n
a
b
i
l
i
ty
r
e
s
u
l
t
s
f
o
r
th
e
p
r
o
p
o
s
e
d
m
o
d
e
l
.
S
u
m
m
a
r
y
p
lo
t
s
h
o
w
n
i
n
F
i
g
u
r
e
5
(
a)
in
t
e
r
p
r
e
t
s
t
h
e
i
m
p
a
c
t
o
f
f
e
a
tu
r
e
s
o
n
m
o
d
e
l
o
u
tp
u
t.
B
a
r
p
l
o
t
s
h
o
w
n
i
n
F
i
g
u
r
e
5
(
b
)
d
e
s
c
r
ib
e
s
t
h
e
av
e
r
a
g
e
i
m
p
a
ct
(
m
ea
n
a
b
s
o
l
u
t
e
S
H
A
P
v
a
lu
e
)
o
f
e
ac
h
f
e
a
tu
r
e
o
n
m
o
d
e
l
p
r
e
d
ic
t
i
o
n
s
w
i
t
h
p
r
o
d
u
c
t
i
o
n
t
h
e
m
o
s
t
i
m
p
a
c
tf
u
l
w
i
t
h
a
S
H
A
P
v
a
l
u
e
o
f
+1
.
0
8
.
W
a
t
e
r
f
a
l
l
p
l
o
t
in
F
i
g
u
r
e
5
(
c)
s
h
o
w
s
h
o
w
i
n
d
iv
i
d
u
a
l
f
e
a
tu
r
e
s
c
o
n
t
r
i
b
u
te
s
t
o
a
s
i
n
g
l
e
p
r
e
d
i
c
t
io
n
.
T
h
e
s
e
r
e
s
u
l
t
s
i
n
d
i
c
a
t
e
s
th
a
t
F
1
-
b
a
s
e
d
f
ea
t
u
r
e
s
e
l
e
c
t
i
o
n
e
f
f
e
c
t
iv
e
l
y
f
i
l
te
r
ed
o
u
t
l
e
s
s
r
e
l
ev
a
n
t
f
ea
t
u
r
e
s
,
a
l
l
o
w
in
g
t
h
e
m
o
d
e
l
to
f
o
c
u
s
o
n
t
h
e
m
o
s
t
i
n
f
o
r
m
a
t
iv
e
f
e
a
tu
r
e
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
S
HA
P
-
en
h
a
n
ce
d
e
n
s
emb
le
lea
r
n
in
g
fo
r
yield
p
r
ed
ictio
n
:
in
s
ig
h
ts
fr
o
m
C
Y
-
B
en
ch
d
a
ta
o
n
…
(
S
o
ma
Gu
p
ta
)
3417
(
a)
(
b
)
(
c)
Fig
u
r
e
5
.
Var
io
u
s
SHAP p
lo
ts
o
f
(
a)
s
u
m
m
ar
y
p
lo
t,
(
b
)
b
a
r
p
l
o
t,
an
d
(
c)
wate
r
f
all
p
l
o
t
T
h
is
r
esu
lted
in
b
etter
p
r
e
d
ictiv
e
ac
cu
r
ac
y
a
n
d
m
o
d
el
ef
f
icie
n
cy
.
T
h
e
r
esu
lts
s
h
o
w
th
e
in
te
g
r
atio
n
o
f
f
ea
tu
r
e
s
elec
tio
n
ap
p
r
o
ac
h
es
,
s
u
ch
as
SHAP
-
b
ased
an
aly
s
is
,
with
en
s
em
b
le
lear
n
in
g
en
h
an
ce
s
th
e
p
er
f
o
r
m
an
ce
an
d
in
ter
p
r
etab
il
ity
o
f
y
ield
p
r
e
d
ictio
n
m
o
d
els
.
T
h
e
n
u
ll
h
y
p
o
th
esis
(
H0
)
is
s
et
as
:
th
er
e
is
n
o
s
ig
n
if
ican
t
d
if
f
er
en
ce
b
etwe
en
b
o
o
s
t
m
1
an
d
o
th
er
e
n
s
em
b
le
m
eth
o
d
s
.
Pair
ed
t
-
test
s
an
d
W
i
lco
x
o
n
s
ig
n
ed
-
r
an
k
test
s
co
n
f
ir
m
e
d
t
h
at
th
e
r
e
d
u
ctio
n
in
R
MSE
a
n
d
th
e
in
cr
ea
s
e
in
R²
a
f
ter
S
HAP
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
wer
e
s
ig
n
if
ican
t
at
th
e
9
5
%
c
o
n
f
id
en
ce
lev
el
(
p
<
0
.
0
5
)
.
A
c
o
m
p
ar
is
o
n
with
p
r
e
v
io
u
s
s
tu
d
ies
is
also
h
ig
h
lig
h
ted
in
T
ab
le
7
.
T
ab
le
7
.
Qu
a
n
titativ
e
co
m
p
ar
i
s
o
n
with
p
r
ev
io
u
s
s
tu
d
ies
S
t
u
d
y
C
r
o
p
/
R
e
g
i
o
n
D
a
t
a
s
e
t
t
y
p
e
M
o
d
e
l
s
u
se
d
E
x
p
l
a
i
n
a
b
i
l
i
t
y
B
e
st
p
e
r
f
o
r
ma
n
c
e
S
e
i
r
e
g
e
t
a
l
.
[
2
3
]
W
i
l
d
B
l
u
e
b
e
r
r
y
/
U
S
A
S
i
mu
l
a
t
i
o
n
+
w
e
a
t
h
e
r
L
i
g
h
t
g
r
a
d
i
e
n
t
b
o
o
s
t
i
n
g
m
a
c
h
i
n
e
(
L
G
B
M
)
,
g
r
a
d
i
e
n
t
b
o
o
s
t
r
e
g
r
e
s
s
i
o
n
(
G
B
R
)
,
X
G
B
o
o
s
t
+
s
t
a
c
k
i
n
g
r
e
g
r
e
s
s
i
o
n
S
H
A
P
(
i
n
t
e
r
p
r
e
t
a
t
i
o
n
o
n
l
y
)
R²
=
0
.
9
8
5
,
R
M
S
E
=
1
7
9
.
9
W
a
q
a
r
e
t
a
l
.
[
2
1
]
M
u
l
t
i
-
c
r
o
p
/
G
l
o
b
a
l
R
e
a
l
+
s
y
n
t
h
e
t
i
c
(
P
r
o
p
h
e
t
)
R
F
,
D
T
,
X
G
B
,
K
N
N
+
o
p
t
i
m
i
z
e
d
st
a
c
k
i
n
g
e
n
s
e
mb
l
e
mo
d
e
l
(
O
S
EM
)
N
o
t
r
e
p
o
r
t
e
d
R²
=
0
.
9
9
6
,
M
A
E
=
0
.
1
8
5
t
/
h
a
P
r
o
p
o
se
d
s
t
u
d
y
W
h
e
a
t
/
I
n
d
i
a
CY
-
B
e
n
c
h
(
r
e
a
l
,
su
b
-
n
a
t
i
o
n
a
l
)
V
o
t
i
n
g
,
st
a
c
k
i
n
g
,
b
o
o
s
t
i
n
g
S
H
A
P
-
b
a
se
d
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
R²
=
0
.
9
2
,
R
M
S
E
=
0
.
2
8
4.
CO
NCLU
SI
O
N
E
n
s
em
b
le
m
o
d
els
co
m
b
in
ed
with
SHAP
f
ea
tu
r
e
s
elec
ti
o
n
co
u
l
d
b
e
a
p
o
wer
f
u
l
a
p
p
r
o
ac
h
f
o
r
in
cr
ea
s
in
g
ac
cu
r
ac
y
an
d
in
ter
p
r
etab
ilit
y
in
y
ield
p
r
ed
ictio
n
m
o
d
els.
T
h
ese
m
eth
o
d
s
co
m
b
in
e
th
e
s
tr
en
g
th
s
o
f
d
if
f
er
en
t
lear
n
in
g
alg
o
r
ith
m
s
t
o
im
p
r
o
v
e
p
r
e
d
ictio
n
ac
c
u
r
ac
y
an
d
g
e
n
er
aliza
tio
n
.
SHAP
v
alu
es
id
en
tifie
s
th
e
m
o
s
t
in
f
lu
en
tial
f
ea
tu
r
es
f
o
r
y
ield
p
r
ed
ictio
n
an
d
cr
ea
te
m
o
r
e
ef
f
icien
t
an
d
ac
cu
r
ate
m
o
d
els.
B
o
o
s
t
m
1
p
er
f
o
r
m
s
b
etter
u
s
in
g
F1
-
b
ase
d
f
ea
tu
r
e
s
elec
tio
n
with
im
p
r
o
v
ed
ev
alu
atio
n
m
etr
ics,
th
u
s
co
n
f
ir
m
in
g
f
ea
tu
r
e
s
elec
tio
n
f
ilter
s
o
u
t
less
r
elev
an
t
f
ea
tu
r
es
ef
f
ec
tiv
ely
.
C
o
m
b
in
in
g
SHAP
-
b
ased
an
aly
s
is
with
en
s
em
b
le
lear
n
in
g
a
g
r
icu
ltu
r
al
y
ield
p
r
ed
ictio
n
m
o
d
els
ca
n
ac
h
ie
v
e
b
etter
p
er
f
o
r
m
a
n
ce
an
d
in
ter
p
r
etab
ilit
y
.
Featu
r
e
s
elec
tio
n
im
p
r
o
v
e
d
th
e
p
e
r
f
o
r
m
an
ce
,
s
ig
n
if
ican
tly
r
ed
u
cin
g
R
MSE
,
MA
E
,
MSE
to
0
.
1
7
,
0
.
1
5
,
0
.
0
3
an
d
in
cr
ea
s
in
g
R
²
to
0
.
9
6
.
T
h
ese
f
in
d
in
g
s
in
d
icate
th
e
e
f
f
ec
tiv
en
ess
o
f
en
s
em
b
le
b
o
o
s
tin
g
ap
p
r
o
ac
h
,
in
ca
p
tu
r
in
g
th
e
n
o
n
lin
ea
r
r
elatio
n
s
h
ip
s
with
in
th
e
in
p
u
t f
ea
t
u
r
es c
o
m
p
r
is
in
g
NDVI
,
m
eteo
r
o
l
o
g
ical,
an
d
s
o
il m
o
is
tu
r
e
d
ata
wh
en
co
m
b
in
e
d
with
f
ea
tu
r
e
s
elec
tio
n
.
Un
lik
e
p
r
io
r
s
tu
d
ies
th
at
ap
p
ly
SHAP
s
o
l
ely
f
o
r
p
o
s
t
-
h
o
c
in
ter
p
r
etab
ilit
y
,
th
is
s
tu
d
y
is
th
e
f
ir
s
t
attem
p
t
to
ap
p
ly
SH
AP
-
d
r
iv
en
f
ea
tu
r
e
s
elec
tio
n
o
n
C
Y
-
B
en
ch
I
n
d
ian
w
h
ea
t
d
ata
in
a
n
en
s
em
b
le
b
o
o
s
tin
g
f
r
am
ewo
r
k
.
NDVI
,
Ag
E
R
A5
m
eteo
r
o
lo
g
y
,
a
n
d
s
o
il
m
o
is
tu
r
e
at
d
is
tr
ict
lev
el
h
av
e
also
b
ee
n
in
teg
r
ate
d
.
T
h
is
d
u
al
f
o
c
u
s
o
n
p
r
ed
ictiv
e
p
e
r
f
o
r
m
an
ce
en
h
a
n
ce
m
en
t
an
d
ex
p
lain
ab
ilit
y
o
r
ien
ted
f
ea
t
u
r
e
s
elec
tio
n
d
i
f
f
er
en
tiate
th
e
p
r
o
p
o
s
ed
f
r
a
m
ewo
r
k
f
r
o
m
av
ailab
le
liter
atu
r
es.
Per
f
o
r
m
a
n
ce
im
p
r
o
v
em
e
n
ts
ar
e
s
tatis
tically
v
alid
ated
th
r
o
u
g
h
c
r
o
s
s
-
v
alid
atio
n
an
d
h
y
p
o
th
esis
test
in
g
.
Alth
o
u
g
h
th
is
s
tu
d
y
s
h
o
ws
th
e
ef
f
ec
tiv
en
ess
o
f
e
n
s
em
b
le
m
o
d
els
co
m
b
in
ed
w
ith
SHAP
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
f
o
r
cr
o
p
y
ield
p
r
ed
ictio
n
,
b
u
t
th
er
e
ar
e
s
ev
er
al
ar
ea
s
y
et
to
b
e
e
x
p
lo
r
ed
f
o
r
f
u
tu
r
e
r
esear
c
h
.
A
tem
p
o
r
al
e
x
ten
s
io
n
o
f
SHAP
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
4
1
1
-
3
4
2
0
3418
v
alu
es
co
u
ld
b
e
e
x
p
lo
r
e
d
to
u
n
d
er
s
tan
d
h
o
w
f
ea
tu
r
e
im
p
o
r
t
an
ce
ev
o
lv
es
d
u
r
in
g
th
e
g
r
o
wi
n
g
s
ea
s
o
n
,
o
f
f
e
r
in
g
m
o
r
e
ac
tio
n
a
b
le
in
s
ig
h
ts
f
o
r
d
y
n
am
ic
d
ec
is
io
n
-
m
ak
in
g
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
r
esear
ch
r
ec
eiv
ed
n
o
g
r
an
t f
r
o
m
a
n
y
f
u
n
d
in
g
ag
e
n
cy
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
So
m
a
Gu
p
ta
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Day
al
Ku
m
ar
B
eh
er
a
✓
✓
✓
✓
✓
✓
✓
✓
Satar
u
p
a
Mo
h
an
t
y
✓
✓
✓
✓
✓
✓
✓
✓
Su
b
h
r
a
Swetan
is
h
a
✓
✓
✓
✓
✓
✓
✓
R
itik
Ma
llik
✓
✓
✓
✓
✓
✓
✓
Nam
ita
Pan
d
a
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
d
ec
lar
e
n
o
co
n
f
lict o
f
in
ter
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
o
p
en
ly
av
ailab
le
with
th
e
Ag
ML
(
h
ttp
s
://www.
ag
m
l.o
r
g
/)
at
h
tt
p
s
://d
o
i.o
r
g
/1
0
.
5
2
8
1
/zen
o
d
o
.
1
3
8
3
8
9
1
2
,
r
ef
e
r
en
ce
[
2
4
]
.
T
h
e
f
u
ll
d
ataset
ca
n
b
e
d
o
wn
lo
ad
e
d
d
ir
ec
tly
f
r
o
m
Z
en
o
d
o
o
r
u
s
in
g
t
h
e
“z
en
o
d
o
g
et”
lib
r
ar
y
.
RE
F
E
R
E
NC
E
S
[
1
]
X
.
Zh
u
,
R
.
G
u
o
,
T.
Li
u
,
a
n
d
K
.
X
u
,
“
C
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
b
a
se
d
o
n
a
g
r
o
m
e
t
e
o
r
o
l
o
g
i
c
a
l
i
n
d
e
x
e
s
a
n
d
r
e
mo
t
e
se
n
s
i
n
g
d
a
t
a
,
”
Re
m
o
t
e
S
e
n
s
i
n
g
,
v
o
l
.
1
3
,
n
o
.
1
0
,
M
a
y
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
r
s1
3
1
0
2
0
1
6
.
[
2
]
C
.
S
h
a
r
ma,
P
.
J
h
a
,
P
.
A
n
a
n
d
,
a
n
d
M
.
S
h
a
r
ma
,
“
P
r
e
d
i
c
t
i
o
n
a
n
d
t
i
me
s
e
r
i
e
s
a
n
a
l
y
s
i
s
o
f
w
h
e
a
t
,
r
i
c
e
a
n
d
ma
i
z
e
y
i
e
l
d
s
u
s
i
n
g
A
R
I
M
A
mo
d
e
l
s
,
”
i
n
2
0
2
3
3
r
d
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
T
e
c
h
n
o
l
o
g
i
c
a
l
A
d
v
a
n
c
e
m
e
n
t
s
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
S
c
i
e
n
c
e
s
(
I
C
T
A
C
S
)
,
N
o
v
.
2
0
2
3
,
p
p
.
1
0
5
5
–
1
0
6
1
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
TA
C
S
5
9
8
4
7
.
2
0
2
3
.
1
0
3
8
9
9
1
9
.
[
3
]
M
.
A
sh
f
a
q
,
I
.
K
h
a
n
,
A
.
A
l
z
a
h
r
a
n
i
,
M
.
U
.
Ta
r
i
q
,
H
.
K
h
a
n
,
a
n
d
A
.
G
h
a
n
i
,
“
A
c
c
u
r
a
t
e
w
h
e
a
t
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
u
si
n
g
mac
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
c
l
i
mat
e
-
N
D
V
I
d
a
t
a
f
u
s
i
o
n
,
”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
4
0
9
4
7
–
4
0
9
6
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
3
7
6
7
3
5
.
[
4
]
M
.
J.
H
o
q
u
e
e
t
a
l
.
,
“
I
n
c
o
r
p
o
r
a
t
i
n
g
m
e
t
e
o
r
o
l
o
g
i
c
a
l
d
a
t
a
a
n
d
p
e
st
i
c
i
d
e
i
n
f
o
r
m
a
t
i
o
n
t
o
f
o
r
e
c
a
st
c
r
o
p
y
i
e
l
d
s
u
s
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
4
7
7
6
8
–
4
7
7
8
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
3
8
3
3
0
9
.
[
5
]
M
.
J
.
H
o
q
u
e
,
M
.
S
.
I
sl
a
m,
A
.
A
.
N
o
m
a
n
,
M
.
A
.
H
o
q
u
e
,
I
.
A
.
C
h
o
w
d
h
u
r
y
,
a
n
d
M
.
S
a
i
f
u
d
d
i
n
,
“
O
p
t
i
m
i
z
i
n
g
p
o
t
a
t
o
c
r
o
p
p
r
o
d
u
c
t
i
v
i
t
y
:
a
met
e
o
r
o
l
o
g
i
c
a
l
a
n
a
l
y
si
s
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
I
AE
S
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
1
4
,
n
o
.
2
,
A
p
r
.
2
0
2
5
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
1
4
.
i
2
.
p
p
1
1
1
6
-
1
1
2
9
.
[
6
]
R
.
M
a
i
,
Q
.
X
i
n
,
J.
Q
i
u
,
Q
.
W
a
n
g
,
a
n
d
P
.
Zh
u
,
“
H
i
g
h
s
p
a
t
i
a
l
r
e
s
o
l
u
t
i
o
n
so
i
l
mo
i
st
u
r
e
i
mp
r
o
v
e
s
c
r
o
p
y
i
e
l
d
e
s
t
i
m
a
t
i
o
n
,
”
I
EE
E
J
o
u
rn
a
l
o
f
S
e
l
e
c
t
e
d
T
o
p
i
c
s
i
n
A
p
p
l
i
e
d
Ea
r
t
h
O
b
s
e
rv
a
t
i
o
n
s
a
n
d
R
e
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
7
,
p
p
.
1
9
0
6
7
–
1
9
0
7
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
JS
TA
R
S
.
2
0
2
4
.
3
4
1
7
4
2
4
.
[
7
]
U
.
S
h
a
f
i
e
t
a
l
.
,
“
Ta
c
k
l
i
n
g
f
o
o
d
i
n
se
c
u
r
i
t
y
u
s
i
n
g
r
e
m
o
t
e
se
n
s
i
n
g
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
se
d
c
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
1
0
8
6
4
0
–
1
0
8
6
5
7
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
2
1
0
2
0
.
[
8
]
N
.
A
l
i
e
t
a
l
.
,
“
F
i
e
l
d
-
s
c
a
l
e
p
r
e
c
i
si
o
n
:
p
r
e
d
i
c
t
i
n
g
g
r
a
i
n
y
i
e
l
d
o
f
d
i
v
e
r
s
e
w
h
e
a
t
b
r
e
e
d
i
n
g
l
i
n
e
s
u
s
i
n
g
h
i
g
h
-
t
h
r
o
u
g
h
p
u
t
U
A
V
mu
l
t
i
s
p
e
c
t
r
a
l
i
ma
g
i
n
g
,
”
I
E
EE
J
o
u
r
n
a
l
o
f
S
e
l
e
c
t
e
d
T
o
p
i
c
s
i
n
A
p
p
l
i
e
d
Ea
rt
h
O
b
ser
v
a
t
i
o
n
s
a
n
d
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
7
,
p
p
.
1
1
4
1
9
–
1
1
4
3
3
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
JS
TA
R
S
.
2
0
2
4
.
3
4
1
1
9
9
4
.
[
9
]
E.
C
h
e
n
g
e
t
a
l
.
,
“
A
G
T
-
LST
M
s
p
a
t
i
o
-
t
e
mp
o
r
a
l
a
p
p
r
o
a
c
h
f
o
r
w
i
n
t
e
r
w
h
e
a
t
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
:
f
r
o
m
t
h
e
f
i
e
l
d
s
c
a
l
e
t
o
c
o
u
n
t
y
sca
l
e
,
”
I
EEE
T
r
a
n
s
a
c
t
i
o
n
s
o
n
G
e
o
sc
i
e
n
c
e
a
n
d
R
e
m
o
t
e
S
e
n
s
i
n
g
,
v
o
l
.
6
2
,
p
p
.
1
–
1
8
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
TG
R
S
.
2
0
2
4
.
3
4
1
8
0
4
6
.
[
1
0
]
E
.
C
h
o
i
,
V
.
C
.
D
.
S
o
u
z
a
,
J
.
A
.
D
i
l
l
o
n
,
E
.
K
e
b
r
e
a
b
,
a
n
d
N
.
D
.
M
u
e
l
l
e
r
,
“
C
o
m
p
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s
o
f
t
h
e
r
m
a
l
i
n
d
i
c
e
s
f
o
r
m
o
d
e
l
i
n
g
c
o
l
d
a
n
d
h
e
a
t
s
t
r
e
s
s
i
n
U
S
d
a
i
r
y
s
y
s
t
e
m
s
,
”
J
o
u
r
n
a
l
o
f
D
a
i
r
y
S
c
i
e
n
c
e
,
v
o
l
.
1
0
7
,
n
o
.
8
,
p
p
.
5
8
1
7
–
5
8
3
2
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
3
1
6
8
/
j
d
s
.
2
0
2
3
-
2
4
4
1
2
.
[
1
1
]
J.
S
u
n
,
Z.
La
i
,
L.
D
i
,
Z.
S
u
n
,
J.
Ta
o
,
a
n
d
Y
.
S
h
e
n
,
“
M
u
l
t
i
l
e
v
e
l
d
e
e
p
l
e
a
r
n
i
n
g
n
e
t
w
o
r
k
f
o
r
c
o
u
n
t
y
-
l
e
v
e
l
c
o
r
n
y
i
e
l
d
e
s
t
i
m
a
t
i
o
n
i
n
t
h
e
U
.
S
.
c
o
r
n
b
e
l
t
,
”
I
EEE
J
o
u
r
n
a
l
o
f
S
e
l
e
c
t
e
d
T
o
p
i
c
s
i
n
Ap
p
l
i
e
d
E
a
rt
h
O
b
ser
v
a
t
i
o
n
s
a
n
d
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
3
,
p
p
.
5
0
4
8
–
5
0
6
0
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
JS
TA
R
S
.
2
0
2
0
.
3
0
1
9
0
4
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
S
HA
P
-
en
h
a
n
ce
d
e
n
s
emb
le
lea
r
n
in
g
fo
r
yield
p
r
ed
ictio
n
:
in
s
ig
h
ts
fr
o
m
C
Y
-
B
en
ch
d
a
ta
o
n
…
(
S
o
ma
Gu
p
ta
)
3419
[
1
2
]
J.
D
u
,
Y
.
Z
h
a
n
g
,
P
.
W
a
n
g
,
K
.
Ta
n
se
y
,
J.
L
i
u
,
a
n
d
S
.
Z
h
a
n
g
,
“
E
n
h
a
n
c
i
n
g
w
i
n
t
e
r
w
h
e
a
t
y
i
e
l
d
e
s
t
i
m
a
t
i
o
n
w
i
t
h
a
C
N
N
-
t
r
a
n
sf
o
r
mer
h
y
b
r
i
d
f
r
a
mew
o
r
k
u
t
i
l
i
z
i
n
g
m
u
l
t
i
p
l
e
r
e
mo
t
e
l
y
se
n
s
e
d
p
a
r
a
met
e
r
s,”
I
EE
E
T
ra
n
sa
c
t
i
o
n
s
o
n
G
e
o
s
c
i
e
n
c
e
a
n
d
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
6
3
,
p
p
.
1
–
1
3
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
TG
R
S
.
2
0
2
5
.
3
5
4
4
4
3
4
.
[
1
3
]
S
.
K
h
a
k
i
,
L.
W
a
n
g
,
a
n
d
S
.
V
.
A
r
c
h
o
n
t
o
u
l
i
s
,
“
A
C
N
N
-
R
N
N
f
r
a
mew
o
r
k
f
o
r
c
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
,
”
F
ro
n
t
i
e
rs
i
n
Pl
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
0
,
Ja
n
.
2
0
2
0
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s.
2
0
1
9
.
0
1
7
5
0
.
[
1
4
]
Z.
R
a
mz
a
n
,
H
.
M
.
S
.
A
s
i
f
,
I
.
Y
o
u
su
f
,
a
n
d
M
.
S
h
a
h
b
a
z
,
“
A
m
u
l
t
i
m
o
d
a
l
d
a
t
a
f
u
s
i
o
n
a
n
d
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
s
b
a
sed
t
e
c
h
n
i
q
u
e
f
o
r
t
e
a
y
i
e
l
d
e
st
i
ma
t
i
o
n
i
n
P
a
k
i
s
t
a
n
u
s
i
n
g
sat
e
l
l
i
t
e
i
m
a
g
e
r
y
,
”
I
E
EE
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
4
2
5
7
8
–
4
2
5
9
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
2
7
1
4
1
0
.
[
1
5
]
S
.
M
.
M
.
N
e
j
a
d
,
D
.
A
.
-
M
o
g
h
a
d
a
m,
a
n
d
A
.
S
h
a
r
i
f
i
,
“
C
o
n
v
LSTM
-
V
i
T:
a
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
c
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
u
si
n
g
e
a
r
t
h
o
b
s
e
r
v
a
t
i
o
n
s
a
n
d
r
e
m
o
t
e
l
y
se
n
se
d
d
a
t
a
,
”
I
EEE
J
o
u
r
n
a
l
o
f
S
e
l
e
c
t
e
d
T
o
p
i
c
s
i
n
A
p
p
l
i
e
d
E
a
rt
h
O
b
s
e
rv
a
t
i
o
n
s
a
n
d
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
7
,
p
p
.
1
7
4
8
9
–
1
7
5
0
2
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
JS
TA
R
S
.
2
0
2
4
.
3
4
6
4
4
1
1
.
[
1
6
]
P
.
S
h
a
r
m
a
,
P
.
D
a
d
h
e
e
c
h
,
N
.
A
n
e
j
a
,
a
n
d
S
.
A
n
e
j
a
,
“
P
r
e
d
i
c
t
i
n
g
a
g
r
i
c
u
l
t
u
r
e
y
i
e
l
d
s
b
a
se
d
o
n
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
u
si
n
g
r
e
g
r
e
s
si
o
n
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
EE
E
A
c
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
1
1
1
2
5
5
–
1
1
1
2
6
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
3
.
3
3
2
1
8
6
1
.
[
1
7
]
S
.
G
u
p
t
a
,
S
.
M
o
h
a
n
t
y
,
a
n
d
D
.
K
.
B
e
h
e
r
a
,
“
A
I
-
b
a
se
d
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
:
a
t
h
o
r
o
u
g
h
r
e
v
i
e
w
,
”
I
n
d
i
a
n
J
o
u
r
n
a
l
o
f
S
c
i
e
n
c
e
A
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
8
,
n
o
.
1
0
,
p
p
.
8
2
2
–
8
3
8
,
M
a
r
.
2
0
2
5
,
d
o
i
:
1
0
.
1
7
4
8
5
/
I
JS
T/
v
1
8
i
1
0
.
1
7
5
.
[
1
8
]
R
.
S
h
a
m
i
m
a
n
d
T.
A
g
a
r
w
a
l
,
“
O
p
t
i
m
i
z
i
n
g
c
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
u
si
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,
”
i
n
S
m
a
rt
A
g
ri
t
e
c
h
:
R
o
b
o
t
i
c
s
,
AI
,
a
n
d
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
(
I
o
T
)
i
n
Ag
ri
c
u
l
t
u
r
e
,
W
i
l
e
y
,
2
0
2
4
,
p
p
.
4
4
3
–
4
8
7
,
d
o
i
:
1
0
.
1
0
0
2
/
9
7
8
1
3
9
4
3
0
2
9
9
4
.
c
h
1
7
.
[
1
9
]
L.
S
.
C
e
d
r
i
c
e
t
a
l
.
,
“
C
r
o
p
s
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
b
a
se
d
o
n
mac
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
s
:
c
a
se
o
f
W
e
s
t
A
f
r
i
c
a
n
c
o
u
n
t
r
i
e
s,”
S
m
a
r
t
Ag
ri
c
u
l
t
u
r
a
l
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
t
e
c
h
.
2
0
2
2
.
1
0
0
0
4
9
.
[
2
0
]
S
.
K
u
mar,
M
.
P
a
n
t
,
a
n
d
A
.
N
a
g
a
r
,
“
F
o
r
e
c
a
st
i
n
g
t
h
e
s
u
g
a
r
c
a
n
e
y
i
e
l
d
s
b
a
s
e
d
o
n
m
e
t
e
o
r
o
l
o
g
i
c
a
l
d
a
t
a
t
h
r
o
u
g
h
e
n
semb
l
e
l
e
a
r
n
i
n
g
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
1
7
6
5
3
9
–
1
7
6
5
5
3
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
5
0
2
5
4
7
.
[
2
1
]
M
.
W
a
q
a
r
,
Y
.
-
W
.
K
i
m,
a
n
d
Y
.
-
C
.
B
y
u
n
,
“
A
s
t
a
c
k
i
n
g
e
n
sem
b
l
e
f
r
a
m
e
w
o
r
k
l
e
v
e
r
a
g
i
n
g
sy
n
t
h
e
t
i
c
d
a
t
a
f
o
r
a
c
c
u
r
a
t
e
a
n
d
s
t
a
b
l
e
c
r
o
p
y
i
e
l
d
f
o
r
e
c
a
st
i
n
g
,
”
I
EE
E
A
c
c
e
ss
,
v
o
l
.
1
3
,
p
p
.
1
3
6
9
0
9
–
1
3
6
9
2
6
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
5
.
3
5
9
1
8
0
2
.
[
2
2
]
S
.
P
.
R
a
j
a
,
B
.
S
a
w
i
c
k
a
,
Z.
S
t
a
me
n
k
o
v
i
c
,
a
n
d
G
.
M
a
r
i
a
mm
a
l
,
“
C
r
o
p
p
r
e
d
i
c
t
i
o
n
b
a
se
d
o
n
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
o
f
t
h
e
a
g
r
i
c
u
l
t
u
r
a
l
e
n
v
i
r
o
n
m
e
n
t
u
si
n
g
v
a
r
i
o
u
s
f
e
a
t
u
r
e
sel
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
s
a
n
d
c
l
a
ss
i
f
i
e
r
s,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
2
3
6
2
5
–
2
3
6
4
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
1
5
4
3
5
0
.
[
2
3
]
H
.
R
.
S
e
i
r
e
g
,
Y
.
M
.
K
.
O
m
a
r
,
F
.
E.
A
.
El
-
S
a
m
i
e
,
A
.
S
.
E
l
-
F
i
sh
a
w
y
,
a
n
d
A
.
El
ma
h
a
l
a
w
y
,
“
En
s
e
m
b
l
e
mac
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
u
si
n
g
c
o
m
p
u
t
e
r
s
i
mu
l
a
t
i
o
n
d
a
t
a
f
o
r
w
i
l
d
b
l
u
e
b
e
r
r
y
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
6
4
6
7
1
–
6
4
6
8
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
1
8
1
9
7
0
.
[
2
4
]
M
.
K
a
l
l
e
n
b
e
r
g
e
t
a
l
.
,
“
C
Y
-
B
e
n
c
h
:
a
c
o
mp
r
e
h
e
n
s
i
v
e
b
e
n
c
h
mar
k
d
a
t
a
s
e
t
f
o
r
su
b
-
n
a
t
i
o
n
a
l
c
r
o
p
y
i
e
l
d
f
o
r
e
c
a
s
t
i
n
g
(
1
.
1
0
)
,
”
A
g
ML
,
J
u
n
.
2
0
2
6
,
d
o
i
:
1
0
.
5
2
8
1
/
z
e
n
o
d
o
.
1
3
8
3
8
9
1
2
.
[
2
5
]
E.
V
e
r
mo
t
e
,
“
M
O
D
I
S
/
t
e
r
r
a
su
r
f
a
c
e
r
e
f
l
e
c
t
a
n
c
e
8
-
d
a
y
L
3
g
l
o
b
a
l
5
0
0
m
S
I
N
g
r
i
d
V
0
0
6
,
”
N
A
S
A
L
a
n
d
Pr
o
c
e
sses
D
i
st
r
i
b
u
t
e
d
Ac
t
i
v
e
Arc
h
i
v
e
C
e
n
t
e
r
,
2
0
1
5
,
d
o
i
:
1
0
.
5
0
6
7
/
M
O
D
I
S
/
M
O
D
0
9
A
1
.
0
0
6
.
[
2
6
]
H
.
B
o
o
g
a
a
r
d
,
J
.
S
c
h
u
b
e
r
t
,
A
.
D
.
W
i
t
,
J.
La
z
e
b
n
i
k
,
R
.
H
u
t
j
e
s,
a
n
d
G
.
V
.
D
.
G
r
i
j
n
,
“
A
g
r
o
m
e
t
e
o
r
o
l
o
g
i
c
a
l
i
n
d
i
c
a
t
o
r
s
f
r
o
m
1
9
7
9
t
o
p
r
e
se
n
t
d
e
r
i
v
e
d
f
r
o
m
r
e
a
n
a
l
y
si
s
,
”
C
o
p
e
rn
i
c
u
s
C
l
i
m
a
t
e
C
h
a
n
g
e
S
e
rvi
c
e
(
C
3
S
)
C
l
i
m
a
t
e
D
a
t
a
S
t
o
re
(
C
D
S
)
,
2
0
2
0
,
d
o
i
:
1
0
.
2
4
3
8
1
/
c
d
s
.
6
c
6
8
c
9
b
b
.
[
2
7
]
M
.
R
o
d
e
l
l
e
t
a
l
.
,
“
Th
e
g
l
o
b
a
l
l
a
n
d
d
a
t
a
a
ssi
m
i
l
a
t
i
o
n
s
y
st
e
m,
”
Bu
l
l
e
t
i
n
o
f
t
h
e
Am
e
r
i
c
a
n
Me
t
e
o
ro
l
o
g
i
c
a
l
S
o
c
i
e
t
y
,
v
o
l
.
8
5
,
n
o
.
3
,
p
p
.
3
8
1
–
3
9
4
,
M
a
r
.
2
0
0
4
,
d
o
i
:
1
0
.
1
1
7
5
/
B
A
M
S
-
85
-
3
-
3
8
1
.
[
2
8
]
Y
.
W
a
n
g
,
P
.
W
a
n
g
,
K
.
Ta
n
se
y
,
J
.
Li
u
,
B
.
D
e
l
a
n
e
y
,
a
n
d
W
.
Q
u
a
n
,
“
A
n
i
n
t
e
r
p
r
e
t
a
b
l
e
a
p
p
r
o
a
c
h
c
o
m
b
i
n
i
n
g
S
h
a
p
l
e
y
a
d
d
i
t
i
v
e
e
x
p
l
a
n
a
t
i
o
n
s
a
n
d
L
i
g
h
t
G
B
M
b
a
s
e
d
o
n
d
a
t
a
a
u
g
me
n
t
a
t
i
o
n
f
o
r
i
m
p
r
o
v
i
n
g
w
h
e
a
t
y
i
e
l
d
e
s
t
i
m
a
t
e
s,
”
C
o
m
p
u
t
e
rs
a
n
d
El
e
c
t
ro
n
i
c
s
i
n
Ag
ri
c
u
l
t
u
r
e
,
v
o
l
.
2
2
9
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
mp
a
g
.
2
0
2
4
.
1
0
9
7
5
8
.
[
2
9
]
D
.
B
a
n
i
k
,
R
.
P
a
u
l
,
R
.
S
.
R
a
t
h
o
r
e
,
a
n
d
R
.
H
.
Jh
a
v
e
r
i
,
“
I
mp
r
o
v
e
d
r
e
g
r
e
ssi
o
n
a
n
a
l
y
si
s
w
i
t
h
e
n
sem
b
l
e
p
i
p
e
l
i
n
e
a
p
p
r
o
a
c
h
f
o
r
a
p
p
l
i
c
a
t
i
o
n
s
a
c
r
o
ss
m
u
l
t
i
p
l
e
d
o
ma
i
n
s,”
AC
M
T
ra
n
s
a
c
t
i
o
n
s
o
n
Asi
a
n
a
n
d
L
o
w
-
Re
s
o
u
r
c
e
L
a
n
g
u
a
g
e
I
n
f
o
rm
a
t
i
o
n
Pr
o
c
e
s
si
n
g
,
v
o
l
.
2
3
,
n
o
.
3
,
p
p
.
1
–
1
3
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
4
5
/
3
6
4
5
1
1
0
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Dr
.
S
o
m
a
G
u
p
t
a
is
p
re
se
n
tl
y
wo
rk
in
g
in
t
h
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
,
Ra
v
e
n
sh
a
w
Un
i
v
e
rsity
,
C
u
tt
a
c
k
,
In
d
ia
.
S
h
e
h
a
s
o
v
e
r
1
2
y
e
a
rs
o
f
e
x
p
e
rien
c
e
in
tea
c
h
i
n
g
u
n
d
e
r
g
ra
d
u
a
te
a
n
d
p
o
st
g
ra
d
u
a
te
stu
d
e
n
ts.
S
h
e
h
a
s
o
n
e
p
a
ten
t
a
n
d
a
u
t
h
o
re
d
se
v
e
ra
l
re
se
a
rc
h
p
u
b
li
c
a
ti
o
n
s
in
re
p
u
ted
j
o
u
r
n
a
ls
a
n
d
in
ter
n
a
ti
o
n
a
l
c
o
n
fe
re
n
c
e
s.
He
r
re
se
a
r
c
h
in
tere
sts
in
c
lu
d
e
m
a
c
h
in
e
lea
rn
in
g
,
d
a
ta
sc
ien
c
e
,
d
e
sig
n
a
n
d
a
n
a
ly
sis
o
f
a
l
g
o
r
it
h
m
s,
a
n
d
d
a
tab
a
se
m
a
n
a
g
e
m
e
n
t
sy
ste
m
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
so
m
a
g
u
p
ta
2
9
@
g
m
a
il
.
c
o
m
.
Dr
.
Da
y
a
l
K
u
m
a
r
Be
h
e
r
a
is
wo
rk
i
n
g
a
s
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
in
t
h
e
S
c
h
o
o
l
o
f
Co
m
p
u
ter
En
g
in
e
e
rin
g
,
KIIT
D
e
e
m
e
d
to
b
e
Un
iv
e
rsity
,
Bh
u
b
a
n
e
sw
a
r,
In
d
ia
.
He
is
a
n
e
n
th
u
sia
stic,
p
r
o
a
c
ti
v
e
fa
c
u
lt
y
m
e
m
b
e
r
a
n
d
re
se
a
rc
h
e
r
with
m
o
re
th
a
n
2
0
y
e
a
rs
o
f
e
x
p
e
rien
c
e
tea
c
h
in
g
c
o
u
rse
s
a
t
th
e
u
n
d
e
rg
ra
d
u
a
te
a
n
d
p
o
st
g
ra
d
u
a
te
lev
e
ls.
S
u
p
e
rv
ise
d
m
a
n
y
IEDC
-
fu
n
d
e
d
p
ro
jec
ts,
P
h
.
D
.
a
n
d
M
.
Tec
h
.
t
h
e
sis,
a
n
d
B
.
Tec
h
.
p
ro
jec
ts.
He
h
a
s
two
p
a
ten
ts
a
n
d
p
u
b
li
s
h
e
d
m
o
re
th
a
n
4
0
re
se
a
rc
h
p
a
p
e
rs
in
v
a
rio
u
s
re
p
u
ted
,
p
e
e
r
-
re
v
iew
e
d
re
fe
re
e
d
jo
u
rn
a
ls
,
a
n
d
in
tern
a
ti
o
n
a
l
c
o
n
fe
re
n
c
e
s.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
d
a
y
a
lb
e
h
e
ra
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
4
,
Au
g
u
s
t
20
26
:
3
4
1
1
-
3
4
2
0
3420
Dr
.
S
a
ta
r
u
p
a
Mo
h
a
n
t
y
is
c
u
rre
n
tl
y
wo
r
k
in
g
a
s
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
in
S
c
h
o
o
l
of
Co
m
p
u
ter
E
n
g
i
n
e
e
rin
g
,
KIIT
d
e
e
m
e
d
to
b
e
Un
i
v
e
rsity
,
B
h
u
b
a
n
e
sw
a
r,
In
d
ia
a
n
d
h
a
v
i
n
g
m
o
re
th
a
n
2
0
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
S
h
e
h
a
s
a
u
th
o
re
d
m
o
re
th
a
n
5
0
re
se
a
rc
h
p
a
p
e
rs
,
e
d
it
e
d
se
v
e
ra
l
b
o
o
k
s,
a
n
d
g
u
id
e
d
m
u
lt
ip
le
P
h
.
D
.
stu
d
e
n
ts.
He
r
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
m
a
c
h
in
e
lea
rn
in
g
,
b
i
o
i
n
fo
rm
a
ti
c
s,
a
n
d
c
o
m
p
u
ter
g
ra
p
h
ics
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
sa
taru
p
a
fc
s@
k
ii
t.
a
c
.
in
.
Dr
.
S
u
b
h
r
a
S
we
ta
n
ish
a
re
c
e
iv
e
d
th
e
P
h
.
D
.
d
e
g
re
e
in
C
o
m
p
u
ter
S
c
ie
n
c
e
En
g
i
n
e
e
rin
g
fro
m
KIIT
d
e
e
m
e
d
to
b
e
Un
iv
e
rsity
,
Bh
u
b
a
n
e
sw
a
r,
In
d
ia
.
S
h
e
h
a
s
b
e
e
n
wo
r
k
in
g
a
s
a
n
a
ss
o
c
iate
p
ro
fe
ss
o
r
in
th
e
D
e
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
a
t
S
il
ic
o
n
Un
iv
e
rsity
,
Od
is
h
a
.
S
h
e
h
a
s
c
o
m
p
lete
d
M
.
Tec
h
.
d
e
g
re
e
i
n
C
o
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
ri
n
g
fro
m
KIIT
De
e
m
e
d
to
b
e
Un
iv
e
rsity
.
He
r
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
m
a
c
h
in
e
lea
rn
in
g
,
d
a
ta
sc
ien
c
e
,
ima
g
e
p
ro
c
e
ss
in
g
,
a
n
d
re
m
o
te
se
n
sin
g
.
S
h
e
h
a
s
n
in
e
tee
n
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
a
n
d
p
u
b
li
sh
e
d
m
o
re
t
h
a
n
fift
e
e
n
S
c
o
p
u
s/
S
CI
i
n
d
e
x
e
d
re
se
a
rc
h
a
rti
c
les
.
S
h
e
is
a
li
fe
m
e
m
b
e
r
o
f
th
e
IS
TE
a
n
d
IAENG
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
ss
we
tan
ish
a
@g
m
a
il
.
c
o
m
.
Ritik
M
a
ll
i
k
is
c
u
rre
n
tl
y
a
re
s
e
a
rc
h
sc
h
o
lar
i
n
S
c
h
o
o
l
o
f
Co
m
p
u
ter
E
n
g
i
n
e
e
rin
g
,
KIIT
d
e
e
m
e
d
t
o
b
e
Un
i
v
e
rsity
,
Bh
u
b
a
n
e
sw
a
r,
In
d
ia
.
He
h
a
s
so
fa
r
p
u
b
li
s
h
e
d
3
c
o
n
fe
re
n
c
e
p
a
p
e
r’s
wit
h
k
e
e
n
in
tere
st
i
n
m
a
c
h
in
e
lea
rn
in
g
,
a
rti
ficia
l
in
telli
g
e
n
c
e
,
p
a
tt
e
rn
re
c
o
g
n
it
i
o
n
,
a
n
d
d
a
ta m
in
n
i
n
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
rit
i
k
m
a
ll
ik
@g
m
a
il
.
c
o
m
.
Dr
.
Na
m
it
a
Pa
n
d
a
is
c
u
rre
n
tl
y
wo
r
k
in
g
a
s
a
ss
o
c
iate
p
r
o
fe
ss
o
r
i
n
S
c
h
o
o
l
o
f
Co
m
p
u
ter
E
n
g
i
n
e
e
rin
g
,
KIIT
d
e
e
m
e
d
to
b
e
Un
i
v
e
rsity
,
Bh
u
b
a
n
e
s
wa
r,
In
d
ia
.
T
h
e
re
a
re
m
o
re
th
a
n
3
0
n
u
m
b
e
r
o
f
p
u
b
li
c
a
ti
o
n
s
t
o
h
e
r
c
re
d
it
i
n
m
a
n
y
in
tern
a
t
io
n
a
l
a
n
d
n
a
ti
o
n
a
l
lev
e
l
jo
u
rn
a
l
s
a
n
d
c
o
n
fe
re
n
c
e
p
r
o
c
e
e
d
in
g
s.
S
h
e
h
a
s
2
0
y
e
a
rs
o
f
e
x
p
e
rien
c
e
in
t
e
a
c
h
in
g
u
n
d
e
rg
ra
d
u
a
te
a
n
d
p
o
stg
ra
d
u
a
te
c
o
u
rse
s.
He
r
re
se
a
rc
h
i
n
tere
sts
in
c
l
u
d
e
o
b
jec
t
-
o
rien
te
d
sy
ste
m
s,
s
o
ftwa
re
tes
ti
n
g
,
se
c
u
rit
y
tes
ti
n
g
,
d
a
ta
m
in
i
n
g
,
o
p
e
ra
ti
n
g
sy
ste
m
s,
a
n
d
c
o
m
p
u
ter
a
rc
h
it
e
c
tu
re
.
S
h
e
is
a
m
e
m
b
e
r
o
f
IS
TE
a
n
d
a
re
v
iew
e
r
o
f
m
a
n
y
c
o
n
fe
re
n
c
e
s
a
n
d
jo
u
rn
a
ls.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
n
p
a
n
d
a
fc
s@
k
ii
t.
a
c
.
in
.
Evaluation Warning : The document was created with Spire.PDF for Python.