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
4
1
~
3
4
5
1
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
4
1
-
3
4
5
1
3441
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
M
a
chine learning
t
echni
ques
for ra
infall predi
ction:
a
sy
stema
tic
lite
ra
t
ure re
v
iew
Dee
pa
Sh
a
rm
a
1
,
Ana
nd
K
um
a
r
Sh
uk
la
1
,
P
un
a
m Ra
t
t
a
n
2
1
S
c
h
o
o
l
o
f
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s,
L
o
v
e
l
y
P
r
o
f
e
ssi
o
n
a
l
U
n
i
v
e
r
si
t
y
,
P
h
a
g
w
a
r
a
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s,
M
a
n
a
v
R
a
c
h
n
a
U
n
i
v
e
r
s
i
t
y
,
F
a
r
i
d
a
b
a
d
,
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
Sep
23
,
2
0
2
4
R
ev
is
ed
May
17
,
2
0
2
6
Acc
ep
ted
J
u
n
19
,
2
0
2
6
Th
e
re
a
re
n
u
m
e
ro
u
s
a
sp
e
c
ts
o
f
h
u
m
a
n
li
fe
in
wh
ich
k
n
o
win
g
h
o
w
m
u
c
h
ra
in
to
e
x
p
e
c
t
m
ig
h
t
b
e
b
e
n
e
ficia
l.
He
a
v
y
ra
i
n
fa
ll
e
v
e
n
ts
s
u
c
h
a
s
f
las
h
f
lo
o
d
s
a
n
d
lan
d
slid
e
s,
a
s
we
ll
a
s
d
r
o
u
g
h
ts,
c
a
n
b
e
p
re
d
icte
d
wi
th
e
ffe
c
ti
v
e
ra
in
fa
ll
fo
re
c
a
stin
g
.
Be
c
a
u
se
o
f
re
li
a
b
le
w
e
a
th
e
r
fo
re
c
a
sts,
th
e
in
fra
stru
c
tu
r
e
re
q
u
ired
to
c
a
p
t
u
re
ra
in
wa
ter
a
n
d
c
u
lt
i
v
a
t
e
c
ro
p
s
m
a
y
b
e
p
lan
n
e
d
a
h
e
a
d
o
f
ti
m
e
.
A
v
a
riety
o
f
m
a
c
h
in
e
lea
rn
in
g
(M
L)
a
n
d
d
e
e
p
lea
rn
in
g
(DL)
a
lg
o
r
it
h
m
s
e
n
a
b
le
a
c
c
u
ra
te
we
a
th
e
r
fo
re
c
a
stin
g
.
Th
i
s
wo
rk
se
e
k
s
t
o
p
r
o
v
id
e
a
fu
ll
o
v
e
rv
iew
o
f
th
e
n
u
m
e
ro
u
s
ML
a
lg
o
rit
h
m
s
u
se
d
fo
r
ra
in
fa
ll
p
re
d
icti
o
n
b
y
fo
c
u
si
n
g
o
n
th
e
tec
h
n
iq
u
e
,
in
p
u
t
p
a
ra
m
e
ters
,
a
n
d
se
v
e
ra
l
p
e
rfo
rm
a
n
c
e
m
e
a
su
re
s.
T
h
e
re
v
iew
c
o
n
sists
o
f
5
1
wo
rk
s
d
i
v
i
d
e
d
in
t
o
th
re
e
se
c
ti
o
n
s.
It
is
fo
u
n
d
th
a
t
l
o
n
g
sh
o
rt
-
term
m
e
m
o
ry
(
LS
TM
)
,
o
n
e
o
f
th
e
DL
a
l
g
o
rit
h
m
s
,
is
m
o
stl
y
u
se
d
b
y
re
se
a
rc
h
e
rs
fo
r
d
e
v
e
lo
p
i
n
g
th
e
m
o
d
e
l
,
b
u
t
in
re
c
e
n
t
y
e
a
rs
,
e
n
se
m
b
l
e
lea
rn
in
g
a
n
d
h
y
b
ri
d
lea
rn
i
n
g
h
a
v
e
a
lso
g
a
in
e
d
p
o
p
u
larity
a
m
o
n
g
re
se
a
rc
h
e
rs
a
s
th
e
y
g
iv
e
m
o
re
a
c
c
u
ra
te res
u
lt
s.
Th
e
se
m
e
th
o
d
s
n
e
e
d
to
b
e
e
x
p
lo
re
d
fu
r
t
h
e
r.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
E
n
s
em
b
le
lear
n
in
g
Hy
b
r
id
m
o
d
el
L
ong
s
h
o
r
t
-
t
er
m
m
e
m
o
r
y
Ma
ch
in
e
lear
n
in
g
R
ain
f
all
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
:
Pu
n
am
R
attan
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
A
p
p
licatio
n
s
,
Ma
n
av
R
ac
h
n
a
Un
iv
er
s
ity
Far
id
ab
ad
,
Har
y
an
a,
I
n
d
ia
E
m
ail:
p
u
n
am
r
attan
@
g
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
am
o
u
n
t
o
f
r
ain
f
all
o
n
t
h
e
p
lan
et
is
th
e
s
o
u
r
ce
o
f
f
r
esh
w
ater
an
d
th
u
s
h
as
a
m
aj
o
r
im
p
ac
t
o
n
th
e
s
u
r
v
iv
al
o
f
p
lan
ts
an
d
a
n
im
als.
Flo
o
d
s
[
1
]
an
d
la
n
d
s
lid
es
[
2
]
,
[
3
]
r
esu
lt
f
r
o
m
e
x
ce
s
s
iv
e
p
r
e
cip
itatio
n
,
wh
er
ea
s
d
r
o
u
g
h
ts
an
d
cr
o
p
f
ailu
r
es
[
4
]
r
esu
lt
f
r
o
m
in
s
u
f
f
icien
t
p
r
ec
i
p
itatio
n
d
u
e
to
in
f
r
eq
u
e
n
t
o
r
n
o
n
-
ex
is
ten
t
r
ain
f
all.
B
o
th
s
itu
atio
n
s
ar
e
u
n
f
o
r
tu
n
a
te
f
o
r
th
e
r
esid
en
ts
co
n
ce
r
n
e
d
.
T
h
e
W
o
r
ld
B
an
k
(
2
0
2
3
)
r
ep
o
r
ts
th
at
7
0
%
o
f
n
atu
r
al
d
is
aster
s
ar
e
ca
u
s
ed
b
y
h
y
d
r
o
m
eteo
r
o
lo
g
ical
e
v
en
ts
th
at
lead
to
g
lo
b
al
ec
o
n
o
m
ic
lo
s
s
es
o
f
USD
3
0
0
b
illi
o
n
an
n
u
ally
.
T
h
e
u
n
p
r
ed
ic
tab
le
n
atu
r
e
o
f
m
o
n
s
o
o
n
p
atte
r
n
s
in
I
n
d
ia
h
as
r
esu
lted
in
ag
r
icu
ltu
r
al
lo
s
s
es
o
f
m
o
r
e
th
an
I
NR
3
0
,
0
0
0
cr
o
r
e
an
n
u
ally
,
as
p
er
th
e
r
e
p
o
r
ts
o
f
th
e
I
n
d
ian
Me
teo
r
o
lo
g
ical
Dep
ar
tm
en
t
(
I
MD
)
2
0
2
2
.
T
h
ese
f
ig
u
r
es
u
n
d
er
s
co
r
e
th
e
cr
itical
n
ec
ess
ity
f
o
r
p
r
ec
is
e
an
d
p
r
o
m
p
t
r
ai
n
f
all
f
o
r
ec
asti
n
g
,
as
ev
en
m
in
o
r
i
n
ac
cu
r
ac
ies
in
p
r
ed
ic
tio
n
s
ca
n
s
ig
n
i
f
ican
tly
a
f
f
ec
t
f
o
o
d
s
ec
u
r
ity
,
d
is
aster
m
itig
atio
n
,
a
n
d
wate
r
r
eso
u
r
ce
m
an
a
g
em
en
t.
C
o
n
v
e
n
ti
o
n
al
s
t
atis
t
ic
al
a
n
d
n
u
m
er
ic
al
m
o
d
e
ls
f
o
r
r
ai
n
f
al
l
p
r
e
d
i
cti
o
n
f
r
e
q
u
e
n
t
ly
e
n
c
o
u
n
t
e
r
d
i
f
f
ic
u
lt
ies
in
ac
c
u
r
ate
ly
r
e
p
r
ese
n
t
in
g
t
h
e
n
o
n
li
n
ea
r
a
n
d
h
i
g
h
l
y
d
y
n
am
ic
ch
ar
ac
t
er
is
ti
cs
o
f
t
h
e
r
ai
n
f
a
ll
p
r
o
ce
s
s
es.
H
o
we
v
er
,
r
e
ce
n
t
p
r
o
g
r
ess
in
m
ac
h
i
n
e
l
ea
r
n
in
g
(
ML
)
a
n
d
d
e
ep
l
ea
r
n
in
g
(
D
L
)
h
as
e
n
h
a
n
ce
d
p
r
e
d
i
cti
o
n
ac
c
u
r
a
cy
b
y
ex
t
r
ac
ti
n
g
i
n
t
r
ic
ate
s
p
at
ial
–
t
e
m
p
o
r
al
p
a
tte
r
n
s
d
i
r
e
ctl
y
f
r
o
m
d
a
ta
.
Nu
m
e
r
o
u
s
s
t
u
d
i
es
h
av
e
u
t
iliz
e
d
ar
ti
f
ic
ial
n
e
u
r
al
n
e
two
r
k
s
(
ANN
)
,
c
o
n
v
o
l
u
ti
o
n
al
n
e
u
r
a
l
n
e
tw
o
r
k
(
C
N
N
)
,
lo
n
g
s
h
o
r
t
-
te
r
m
m
e
m
o
r
y
(
L
STM
)
,
a
n
d
h
y
b
r
i
d
en
s
em
b
l
e
ar
ch
ite
ct
u
r
es
,
y
i
el
d
i
n
g
p
r
o
m
is
in
g
o
u
t
c
o
m
es
ac
r
o
s
s
v
ar
i
o
u
s
cli
m
a
tic
r
e
g
i
o
n
s
.
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
4
1
-
3
4
5
1
3442
M
a
n
y
e
x
i
s
t
in
g
r
ev
i
e
w
s
ar
e
e
i
t
h
e
r
o
u
t
d
a
te
d
o
r
n
a
r
r
o
w
ly
f
o
cu
s
e
d
o
n
s
p
ec
i
f
i
c
m
o
d
e
l
f
a
m
i
l
i
e
s
o
r
g
e
o
g
r
ap
h
i
c
a
r
e
a
s
,
wh
i
ch
r
e
s
u
l
t
s
i
n
a
l
i
m
i
t
ed
u
n
d
er
s
t
a
n
d
i
n
g
o
f
h
o
w
cu
r
r
e
n
t
M
L
a
n
d
D
L
m
o
d
e
l
s
c
o
m
p
a
r
e
i
n
t
e
r
m
s
o
f
a
r
ch
i
t
e
c
tu
r
e,
d
a
t
a
s
e
t
s
,
a
n
d
e
v
a
lu
a
t
i
o
n
m
e
t
r
ic
s
.
T
h
i
s
r
ev
i
e
w
s
ee
k
s
to
b
r
id
g
e
t
h
i
s
g
ap
b
y
s
y
s
t
e
m
a
t
i
c
a
l
ly
in
v
e
s
t
ig
a
t
i
n
g
t
h
e
r
e
c
en
t
p
r
o
g
r
e
s
s
i
n
r
a
in
f
a
l
l
f
o
r
e
ca
s
t
i
n
g
u
s
in
g
M
L
a
n
d
D
L
ap
p
r
o
a
ch
e
s
.
I
t
p
r
e
s
e
n
t
s
:
i
)
a
co
m
p
a
r
a
t
iv
e
a
n
a
l
y
s
i
s
o
f
p
r
o
m
in
e
n
t
m
o
d
e
ls
a
n
d
a
l
g
o
r
i
th
m
s
,
i
i
)
a
s
y
n
t
h
es
i
s
o
f
p
e
r
f
o
r
m
an
ce
m
e
t
r
i
c
s
a
n
d
m
e
t
h
o
d
o
l
o
g
i
c
a
l
tr
e
n
d
s
,
an
d
i
i
i
)
i
d
e
n
t
if
i
c
a
t
i
o
n
o
f
c
u
r
r
en
t
ch
a
l
l
en
g
e
s
a
n
d
r
e
s
e
a
r
c
h
o
p
p
o
r
tu
n
i
t
ie
s
f
o
r
c
r
ea
t
i
n
g
r
o
b
u
s
t
,
d
a
t
a
-
d
r
iv
en
r
a
in
f
a
l
l
p
r
e
d
ic
t
i
o
n
s
y
s
t
e
m
s
.
T
h
is
r
ev
iew
is
s
tr
u
ctu
r
ed
ar
o
u
n
d
th
e
f
o
llo
win
g
r
esear
ch
q
u
esti
o
n
s
:
−
W
h
ich
ML
an
d
DL
tec
h
n
iq
u
e
s
,
in
clu
d
in
g
ANN,
C
NN,
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
(
R
NN
)
,
L
STM
,
g
ate
d
r
ec
u
r
r
en
t
u
n
it
(
GR
U
)
,
an
d
h
y
b
r
id
e
n
s
em
b
les,
h
a
v
e
p
r
o
v
e
n
th
e
m
o
s
t
ef
f
ec
tiv
e
f
o
r
p
r
e
d
ictin
g
r
ain
f
all
ac
r
o
s
s
d
if
f
er
en
t tim
ef
r
am
es a
n
d
g
eo
g
r
ap
h
ical
a
r
ea
s
?
−
W
h
at
ty
p
es
o
f
in
p
u
t
p
a
r
am
eter
s
,
s
u
ch
as
m
eteo
r
o
l
o
g
ical,
h
y
d
r
o
lo
g
ical,
a
n
d
s
atellite
-
d
er
iv
ed
f
ea
tu
r
es,
ar
e
ty
p
ically
u
s
ed
,
an
d
h
o
w
d
o
t
h
e
y
in
f
lu
en
ce
th
e
p
e
r
f
o
r
m
an
ce
o
f
th
e
m
o
d
els?
−
Ho
w
d
o
v
ar
io
u
s
p
er
f
o
r
m
a
n
ce
m
etr
ics
an
d
c
o
m
p
ar
ativ
e
an
aly
s
es
h
ig
h
lig
h
t
th
e
s
tr
en
g
th
s
,
we
ak
n
ess
es,
an
d
g
en
er
aliza
tio
n
a
b
ilit
ies o
f
d
if
f
er
en
t m
o
d
elin
g
s
tr
ateg
ies?
T
h
is
r
ev
iew
aim
s
to
o
f
f
er
a
th
o
r
o
u
g
h
s
y
n
th
esis
o
f
t
h
e
cu
r
r
en
t
m
eth
o
d
s
,
b
en
c
h
m
ar
k
o
u
tco
m
es,
an
d
n
ew
tr
e
n
d
s
th
at
ca
n
in
f
o
r
m
th
e
c
r
ea
tio
n
o
f
m
o
r
e
p
r
ec
is
e
an
d
ad
a
p
tab
le
r
a
in
f
all
f
o
r
ec
asti
n
g
s
y
s
tem
s
.
2.
RE
S
E
ARCH
M
E
T
H
O
DO
L
O
G
Y
AND
SYS
T
E
M
A
T
I
C
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
2
.
1
.
F
o
cus
o
f
re
v
iew
Fo
r
ea
ch
p
a
p
er
th
e
f
o
cu
s
is
o
n
th
e
ca
teg
o
r
y
o
f
ML
a
n
d
DL
t
ec
h
n
iq
u
e
u
s
ed
f
o
r
d
ev
elo
p
i
n
g
th
e
m
o
d
el,
d
ataset
p
ar
am
eter
s
an
d
s
o
u
r
ce
,
an
d
n
atu
r
e
o
f
th
e
d
ataset
(
n
u
m
er
ic/im
ag
e
)
.
T
h
e
f
o
c
u
s
also
in
clu
d
es
th
e
tim
e
p
er
io
d
an
d
d
ata
f
r
eq
u
en
cy
(
d
a
ily
,
m
o
n
t
h
ly
)
o
f
th
e
d
ataset
co
llected
,
as
well
as
th
e
p
er
f
o
r
m
an
ce
m
etr
ics
u
s
ed
f
o
r
c
o
m
p
ar
i
n
g
th
e
ac
cu
r
ac
y
o
f
th
e
m
o
d
el
with
o
th
er
m
o
d
els
.
I
n
ad
d
itio
n
,
th
e
f
o
cu
s
is
o
n
th
e
tim
e
p
er
io
d
f
o
r
wh
ich
th
e
p
r
e
d
ictio
n
is
d
o
n
e
(
v
er
y
s
h
o
r
t,
s
h
o
r
t,
m
ed
iu
m
,
lo
n
g
)
,
an
d
g
eo
g
r
ap
h
ical
ar
ea
o
f
th
e
r
esear
ch
.
2
.
2
.
Na
t
ure
a
nd
s
o
urce
o
f
inp
ut
pa
ra
m
et
er
s
us
ed
in t
he
d
a
t
a
s
et
T
h
e
m
o
d
el
u
s
es
a
d
ataset
th
at
ca
n
b
e
o
f
eith
e
r
n
u
m
er
ic
o
r
im
ag
e
ty
p
e.
R
NN
/LST
M
m
o
d
els
u
s
e
n
u
m
er
ical
m
eteo
r
o
lo
g
ical
d
at
a,
wh
er
ea
s
C
NN
m
o
d
els
u
s
e
s
atellite
im
ag
e
d
ata.
All
a
u
th
o
r
s
u
s
ed
d
ata
f
r
o
m
m
eteo
r
o
lo
g
ical
d
e
p
ar
tm
en
ts
,
wea
th
er
s
tatio
n
s
,
o
r
o
p
en
clim
ate
r
ep
o
s
ito
r
ies.
R
esear
ch
er
s
h
av
e
ex
am
in
e
d
b
o
th
lo
ca
l
(
e.
g
.
,
tem
p
er
atu
r
e,
win
d
s
p
ee
d
,
h
u
m
id
ity
)
a
n
d
g
l
o
b
al
f
ac
to
r
s
e.
g
.
,
E
l
Nin
o
,
I
n
d
ian
Oce
an
Dip
o
le
(
I
OD)
,
s
ea
lev
el
p
r
ess
u
r
e
,
th
at
af
f
ec
t
r
ain
f
all.
2
.
3
.
Sea
rc
h
s
t
ra
t
eg
y
a
nd
t
imef
ra
m
e
An
ex
ten
s
iv
e
liter
atu
r
e
r
ev
ie
w
was
p
er
f
o
r
m
ed
to
lo
ca
te
s
t
u
d
ies
co
n
ce
r
n
in
g
ML
m
eth
o
d
o
lo
g
ies
f
o
r
p
r
ed
ictin
g
r
ain
f
all.
T
o
en
s
u
r
e
th
o
r
o
u
g
h
c
o
v
er
a
g
e
o
f
th
e
e
x
is
tin
g
liter
atu
r
e,
f
o
u
r
p
r
i
n
cip
al
s
cien
tific
d
atab
ases
wer
e
q
u
er
ied
u
s
in
g
p
r
ed
eter
m
in
ed
B
o
o
lean
s
ea
r
c
h
ex
p
r
ess
io
n
s
.
T
ab
le
1
p
r
o
v
id
es
a
s
u
m
m
a
r
y
o
f
th
e
d
atab
ases
s
ea
r
ch
ed
alo
n
g
with
th
e
r
esp
ec
tiv
e
B
o
o
lean
s
ea
r
ch
s
tr
in
g
s
.
T
ab
le
1
.
Sear
ch
s
tr
ateg
y
an
d
k
ey
wo
r
d
s
u
s
ed
f
o
r
th
e
s
y
s
tem
atic
r
ev
iew
A
sp
e
c
t
D
e
scri
p
t
i
o
n
D
a
t
a
b
a
s
e
s s
e
a
r
c
h
e
d
G
o
o
g
l
e
S
c
h
o
l
a
r
,
S
c
i
e
n
c
e
D
i
r
e
c
t
,
I
EEE
X
p
l
o
r
e
,
a
n
d
S
c
o
p
u
s
B
o
o
l
e
a
n
q
u
e
r
y
u
s
e
d
(
“
r
a
i
n
f
a
l
l
f
o
r
e
c
a
st
i
n
g
”
O
R
“
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
”
O
R
”
p
r
e
c
i
p
i
t
a
t
i
o
n
p
r
e
d
i
c
t
i
o
n
”
)
A
N
D
(
”
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
”
O
R
“
d
e
e
p
l
e
a
r
n
i
n
g
”
O
R
“
n
e
u
r
a
l
n
e
t
w
o
r
k
s
”
O
R
“
e
n
s
e
mb
l
e
l
e
a
r
n
i
n
g
”
O
R
”
h
y
b
r
i
d
m
o
d
e
l
”
)
S
u
p
p
l
e
me
n
t
a
r
y
k
e
y
w
o
r
d
s
“
S
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
s”,
“
d
e
c
i
s
i
o
n
t
r
e
e
s
”
,
“
t
i
me
ser
i
e
s
a
n
a
l
y
si
s”,
“
w
e
a
t
h
e
r
f
o
r
e
c
a
s
t
i
n
g
”
,
“
met
e
o
r
o
l
o
g
y
”
,
“
r
e
m
o
t
e
s
e
n
s
i
n
g
”
,
“
d
a
t
a
m
i
n
i
n
g
”
,
“
b
i
g
d
a
t
a
”
,
”
d
e
e
p
l
e
a
r
n
i
n
g
”
,
“
c
l
i
ma
t
e
c
h
a
n
g
e
”
P
u
b
l
i
c
a
t
i
o
n
t
i
me
p
e
r
i
o
d
Jan
u
a
r
y
2
0
1
9
-
A
p
r
i
l
2
0
2
4
La
n
g
u
a
g
e
a
n
d
t
y
p
e
O
n
l
y
p
e
e
r
-
r
e
v
i
e
w
e
d
j
o
u
r
n
a
l
s
a
n
d
c
o
n
f
e
r
e
n
c
e
p
a
p
e
r
s i
n
t
h
e
E
n
g
l
i
sh
La
n
g
u
a
g
e
P
u
r
p
o
se
To
i
d
e
n
t
i
f
y
o
r
i
g
i
n
a
l
s
t
u
d
i
e
s
a
p
p
l
y
i
n
g
ML
/
DL
/
e
n
s
e
mb
l
e
me
t
h
o
d
s fo
r
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
2
.
4
.
I
nclus
io
n a
nd
ex
clus
io
n
cr
it
er
ia
Pre
d
ef
in
ed
in
clu
s
io
n
a
n
d
ex
cl
u
s
io
n
cr
iter
ia
wer
e
estab
lis
h
ed
p
r
io
r
to
th
e
s
cr
ee
n
in
g
p
r
o
ce
s
s
to
en
s
u
r
e
th
e
r
elev
an
ce
an
d
q
u
ality
o
f
th
e
s
elec
ted
s
tu
d
ies.
T
h
ese
cr
iter
ia
wer
e
co
n
s
is
ten
tly
ap
p
lied
d
u
r
in
g
th
e
s
cr
ee
n
in
g
s
tag
es o
f
titl
es,
ab
s
tr
ac
ts
,
an
d
f
u
ll tex
ts
.
T
h
e
in
clu
s
io
n
a
n
d
ex
clu
s
io
n
cr
iter
ia
ar
e
d
etailed
as
f
o
llo
ws:
−
L
an
g
u
ag
e
:
an
E
n
g
lis
h
lan
g
u
a
g
e
ab
s
tr
ac
t w
as r
eq
u
ir
ed
f
o
r
th
e
r
ev
iew.
−
R
ain
f
all
p
r
ed
ictio
n
cr
iter
io
n
:
o
n
ly
s
tu
d
ies
o
n
r
ain
f
all
p
r
ed
i
ctio
n
m
o
d
els
u
s
in
g
ML
,
DL
,
an
d
en
s
em
b
le
lear
n
in
g
wer
e
in
clu
d
ed
.
Pre
d
ictio
n
s
f
o
r
r
ain
f
all
r
u
n
o
f
f
,
win
d
s
p
ee
d
,
p
o
wer
lo
ad
,
an
d
tem
p
er
atu
r
e
wer
e
elim
in
ated
.
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
Ma
ch
in
e
lea
r
n
in
g
tec
h
n
iq
u
es f
o
r
r
a
in
fa
ll p
r
ed
ictio
n
:
a
s
ystema
tic
liter
a
tu
r
e
r
ev
ie
w
(
Dee
p
a
S
h
a
r
ma
)
3443
−
Ou
tco
m
e
cr
iter
io
n
:
s
tu
d
ies
th
a
t
co
m
p
ar
ed
d
if
f
er
e
n
t
m
o
d
els
a
n
d
alg
o
r
ith
m
s
f
o
r
th
eir
p
r
ed
ict
io
n
ac
cu
r
ac
y
wer
e
in
clu
d
ed
.
T
h
is
h
elp
s
in
a
th
o
r
o
u
g
h
u
n
d
er
s
tan
d
in
g
o
f
th
e
wo
r
k
f
lo
w
o
f
th
e
m
o
d
els
p
r
o
v
id
in
g
an
i
d
ea
o
f
th
e
b
est tec
h
n
iq
u
es to
b
e
u
s
ed
f
o
r
p
r
ed
ictio
n
.
−
Or
ig
in
al
r
esear
ch
:
in
clu
d
es
o
r
i
g
in
al
r
esear
ch
,
s
k
ip
p
in
g
r
ev
ie
ws,
p
an
el
talk
s
,
an
d
o
t
h
er
s
tu
d
ies.
2
.
5
.
Scre
ening
a
nd
P
RIS
M
A
f
lo
w
Stu
d
ies
wer
e
id
en
tifie
d
,
s
cr
ee
n
ed
an
d
i
n
clu
d
ed
i
n
ac
co
r
d
an
ce
with
p
r
ef
er
r
ed
r
ep
o
r
tin
g
item
s
f
o
r
s
y
s
tem
atic
r
ev
iews
an
d
m
et
a
-
an
aly
s
es
(
PR
I
SMA
)
2
0
2
0
g
u
id
elin
es
to
en
s
u
r
e
b
o
th
tr
an
s
p
ar
en
cy
an
d
r
ep
r
o
d
u
cib
ilit
y
.
Fig
u
r
e
1
p
r
o
v
id
es
a
v
is
u
al
r
ep
r
esen
tatio
n
o
f
th
e
PR
I
SMA
-
b
ased
s
ele
ctio
n
p
r
o
ce
s
s
[
5
]
.
I
n
itially
,
1
9
0
r
ec
o
r
d
s
we
r
e
g
a
th
er
ed
f
r
o
m
Go
o
g
le
Sch
o
lar
,
Scien
ce
Dir
ec
t,
I
E
E
E
Xp
l
o
r
e,
an
d
Sco
p
u
s
.
Af
ter
elim
in
atin
g
5
0
d
u
p
licate
e
n
tr
i
es,
1
4
0
p
a
p
er
s
r
e
m
ain
ed
.
Du
r
i
n
g
th
e
s
cr
ee
n
in
g
p
r
o
ce
s
s
,
titl
e
s
an
d
a
b
s
tr
ac
ts
wer
e
r
ev
iewe
d
to
d
eter
m
in
e
t
h
eir
r
elev
an
ce
t
o
r
ain
f
all
p
r
ed
ic
tio
n
u
s
in
g
ML
/DL
m
eth
o
d
s
;
4
0
r
ec
o
r
d
s
wer
e
d
is
m
is
s
ed
b
ec
au
s
e
th
ey
wer
e
eith
er
u
n
r
elate
d
o
r
n
o
n
-
t
ec
h
n
ical.
T
h
e
r
em
ain
i
n
g
1
0
0
f
u
ll
-
tex
t
ar
ticles
wer
e
ev
alu
ated
f
o
r
elig
ib
ilit
y
.
At
th
e
elig
ib
ilit
y
s
tag
e,
4
9
a
r
ticles
wer
e
ex
clu
d
ed
b
ec
au
s
e
o
f
th
e
ab
s
en
ce
o
f
r
ain
f
all
-
s
p
ec
if
ic
an
aly
s
is
,
m
is
s
in
g
p
er
f
o
r
m
an
ce
m
etr
ics,
o
r
b
ec
au
s
e
th
ey
wer
e
r
ev
iews
o
r
co
n
ce
p
t
u
al
p
ap
e
r
s
.
Ultim
ately
,
5
1
s
tu
d
ies m
et
all
cr
iter
ia
an
d
wer
e
in
cl
u
d
ed
in
t
h
e
s
tu
d
y
.
Fig
u
r
e
1
.
PR
I
SMA
f
r
am
ewo
r
k
illu
s
tr
atin
g
th
e
s
elec
tio
n
p
r
o
c
ess
o
f
th
e
p
ap
er
s
2
.
6
.
T
ec
hn
iqu
e
a
nd
a
lg
o
rit
hm
s
us
ed
f
o
r
t
he
m
o
del dev
el
o
pm
ent
T
h
e
L
STM
,
a
f
o
r
m
o
f
R
NN
,
was
u
tili
ze
d
f
o
r
m
eteo
r
o
lo
g
ic
al
n
u
m
er
ic
d
ata
,
an
d
a
C
NN
f
o
r
im
ag
e
d
ata
(
r
ad
ar
/s
atellite)
.
R
ec
en
t
s
tu
d
ies
h
av
e
u
s
ed
e
n
s
em
b
le/h
y
b
r
id
m
o
d
els
co
m
b
in
in
g
b
o
th
i
m
ag
e
an
d
n
u
m
e
r
ic
d
ata
to
en
h
an
ce
ac
c
u
r
ac
y
.
B
ig
d
ata
an
d
th
e
in
ter
n
et
o
f
th
i
n
g
s
(
I
o
T
)
p
r
o
v
id
e
a
wea
lth
o
f
d
ata
f
o
r
DL
m
o
d
els
ca
p
ab
le
o
f
ca
p
tu
r
in
g
co
m
p
lex
n
o
n
lin
ea
r
p
atter
n
s
.
2
.
7
.
Det
er
m
ini
ng
m
o
del pa
r
a
m
et
er
s
R
eg
io
n
al
r
ain
f
all
is
af
f
ec
ted
b
y
b
o
th
lo
ca
l
a
n
d
g
l
o
b
al
clim
at
ic
f
ac
to
r
s
s
u
ch
as
tem
p
er
atu
r
e,
h
u
m
id
ity
,
s
u
n
s
h
in
e,
win
d
,
E
l
Niñ
o
,
an
d
I
OD
m
o
m
en
t.
R
esear
ch
er
s
u
s
e
th
e
co
v
ar
ia
n
ce
b
etwe
en
p
a
r
a
m
eter
s
to
d
eter
m
in
e
th
e
lin
k
b
etwe
en
r
ain
f
all
an
d
o
th
er
f
ac
to
r
s
to
tr
ain
th
e
m
o
d
el
with
o
n
ly
s
ig
n
i
f
ican
t
p
ar
am
eter
s
.
Fo
r
in
s
tan
ce
,
L
o
p
ez
et
a
l.
[
6
]
f
o
u
n
d
a
C
lau
s
iu
s
–
C
lap
ey
r
o
n
r
elatio
n
s
h
ip
b
etwe
en
atm
o
s
p
h
er
ic
p
r
ess
u
r
e,
d
ew
p
o
in
t,
an
d
h
u
m
id
ity
th
at
p
r
ed
icted
p
r
ec
ip
itatio
n
,
wh
er
ea
s
Path
an
et
a
l.
[
7
]
f
o
u
n
d
t
h
at
win
d
s
p
ee
d
an
d
m
in
im
u
m
tem
p
er
atu
r
e
ar
e
s
tr
o
n
g
i
n
d
icat
o
r
s
o
f
r
ain
f
all.
2
.
8
.
Da
t
a
e
x
t
ra
ct
i
o
n a
nd
s
y
nthesis
Data
f
r
o
m
ea
ch
s
tu
d
y
wer
e
o
r
g
an
ized
in
to
ca
teg
o
r
ies
s
u
ch
as
th
e
ty
p
e
o
f
m
o
d
el
u
s
ed
(
A
NN,
C
NN
,
R
NN/L
STM
,
h
y
b
r
id
,
e
n
s
em
b
le)
,
d
etails
ab
o
u
t
t
h
e
d
at
aset
(
s
o
u
r
ce
,
d
u
r
atio
n
,
f
r
eq
u
en
cy
,
d
ata
t
y
p
e)
,
p
er
f
o
r
m
an
ce
in
d
icato
r
s
(
r
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
,
m
ea
n
ab
s
o
lu
te
er
r
o
r
(
M
AE
)
,
co
ef
f
icien
t
o
f
d
eter
m
in
atio
n
(
R
²)
,
an
d
ac
cu
r
ac
y
)
,
an
d
th
e
s
tu
d
y
'
s
g
eo
g
r
ap
h
ical
f
o
cu
s
a
n
d
p
r
ed
ictio
n
tim
ef
r
am
e.
Nar
r
ativ
e
an
d
th
em
atic
s
y
n
th
esis
ca
teg
o
r
ized
th
e
r
esu
lts
b
ased
o
n
th
e
ty
p
e
o
f
alg
o
r
ith
m
a
n
d
th
e
n
at
u
r
e
o
f
th
e
d
ata.
T
h
e
q
u
an
titativ
e
f
in
d
i
n
g
s
ar
e
d
etailed
in
T
ab
les 2
t
o
4
,
e
m
p
h
asizin
g
co
m
p
ar
ativ
e
ac
cu
r
ac
y
a
n
d
t
r
en
d
s
.
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
4
1
-
3
4
5
1
3444
T
ab
le
2
.
Su
m
m
a
r
y
o
f
m
o
d
els d
ev
elo
p
e
d
f
o
r
r
ain
f
all
p
r
ed
icti
o
n
u
s
in
g
d
if
f
er
e
n
t n
eu
r
al
n
etw
o
r
k
s
Ref
.
NN
u
se
d
Ty
p
e
O
t
h
e
r
mo
d
e
l
s f
o
r
c
o
m
p
a
r
i
so
n
Ti
me
p
e
r
i
o
d
(
d
a
t
a
s
e
t
)
/
d
a
t
a
se
t
t
y
p
e
/
s
o
u
r
c
e
s
P
e
r
f
o
r
ma
n
c
e
me
t
r
i
c
s
(
v
a
l
u
e
)
I
n
p
u
t
p
a
r
a
m
e
t
e
r
s
*
[
8
]
Te
mp
o
r
a
l
C
N
N
H
y
b
r
i
d
A
R
I
M
A
,
M
LP
,
J
M
A
EC
M
W
F
,
B
P
N
N
,
S
V
M
LST
M
2
0
1
5
-
2
0
1
7
(
n
u
m
e
r
i
c
)
1
1
met
r
o
l
o
g
i
c
a
l
s
t
a
t
i
o
n
s
TR
(
7
8
.
6
4
,
7
6
.
1
2
)
,
M
A
E
(
1
.
8
1
)
,
M
S
E
(
4
.
9
4
)
T
,
R
H
,
W
S
,
S
LP
[
9
]
R
N
N
,
LST
M
NA
H
o
l
t
-
w
i
n
t
e
r
s
e
x
t
r
e
me
l
e
a
r
n
i
n
g
1
9
8
0
-
2
0
1
3
(
3
4
y
e
a
r
s)
N
u
meri
c
R
M
S
E
,
a
c
c
u
r
a
c
y
(
8
8
%),
Ep
o
c
h
s,
l
o
ss,
LR
T,
R
H
,
W
S
,
SR
[
1
0
]
R
N
N
,
LST
M
NA
NA
A
u
g
u
st
2
0
2
0
(
n
u
me
r
i
c
)
B
M
D
1
A
c
c
u
r
a
c
y
(
7
6
%)
T,
R
H
,
W
S
,
S
LP
[
1
1
]
A
N
N
(
F
F
N
N
)
,
M
LP (
P
S
O
)
H
y
b
r
i
d
M
LP
2
0
0
9
(
n
u
m
e
r
i
c
)
a
u
t
o
m
a
t
i
c
w
e
a
t
h
e
r
st
a
t
i
o
n
R
M
S
E
PSO
(
0
.
1
4
)
R
M
S
E
LM
(
0
.
1
8
)
T,
R
H
,
W
S
[
1
2
]
R
N
N
B
LS
TM
-
G
R
U
H
y
b
r
i
d
M
LP,
LST
M
,
B
LST
M
1
9
9
7
-
2
0
1
7
(
n
u
meri
c
)
N
C
H
M
D
B
2
M
S
E
(
0
.
0
0
7
5
)
,
M
S
E
(
0
.
8
7
)
,
R
2
(
0
.
8
7
)
,
C
o
r
r
(
0
.
9
3
8
)
T,
R
H
,
W
S
,
S
R
,
R
F
[
1
3
]
ANN
,
EEM
D
-
ANN
H
y
b
r
i
d
H
i
n
d
c
a
st
f
o
r
e
c
a
st
m
o
d
e
l
1
8
7
1
–
2
0
1
6
(
n
u
meri
c
)
I
I
TM
P
u
n
e
a
n
d
I
M
D
R
,
M
A
E,
N
R
M
S
E
I
A
(
I
n
d
e
x
o
f
a
g
r
e
e
me
n
t
)
RF
[
1
4
]
C
N
N
,
C
o
n
v
1
D
-
M
LP
H
y
b
r
i
d
D
e
e
p
M
LR
S
V
R
1
9
4
1
-
2
0
0
5
(
N
u
meri
c
)
I
M
D
R
M
S
E
,
r
,
N
S
E
T,
R
H
,
W
S
,
S
R
,
S
LP
,
RF
[
1
5
]
R
N
N
,
LST
M
NA
C
o
m
p
a
r
i
so
n
(
d
i
f
f
e
r
e
n
t
p
a
r
a
m
e
t
e
r
s)
D
e
c
2
0
1
4
-
A
u
g
2
0
1
9
(
n
u
m
e
r
i
c
)
met
e
o
r
o
l
o
g
i
c
a
l
st
a
t
i
o
n
M
A
A
P
E
(
0
.
9
6
4
4
)
C
(
I
)
,
R
F
(
I
I
)
[
1
6
]
DNN
,
C
o
n
v
3
D
-
G
R
U
H
y
b
r
i
d
C
o
n
v
2
D
C
o
n
v
2
D
-
G
R
U
R
a
i
n
y
d
a
y
s (
r
a
d
a
r
i
ma
g
e
s)
m
e
t
e
o
r
o
l
o
g
i
c
a
l
st
a
t
i
o
n
M
S
E(
2
5
9
6
)
,
M
A
E(
6
7
9
0
)
,
B
-
M
S
E,
B
-
M
A
E
I
M
G
[
1
7
]
R
N
N
,
LST
M
NA
F
F
N
N
R
N
N
1
9
0
1
-
2
0
1
7
(
n
u
meri
c)
w
w
w
.
d
a
t
a
.
g
o
v
.
i
n
M
A
P
E(
7
9
)
R
M
S
E(
1
3
5
.
4
)
RF
[
1
8
]
R
N
N
,
LST
M
NA
ANN
1
9
0
1
-
2
0
1
7
(
n
u
meri
c)
I
M
D
P
u
n
e
M
S
E,
R
M
S
E(
1
2
6
.
0
0
6
)
,
M
A
D
,
R
,
C
S
RF
[
1
9
]
R
N
N
,
I
R
F
-
LSTM
H
y
b
r
i
d
A
N
N
,
S
V
M
,
LST
M
,
C
N
N
-
G
R
U
1
9
8
1
-
2
0
2
0
(
n
u
meri
c
)
st
a
t
e
d
e
v
e
l
o
p
m
e
n
t
p
l
a
n
n
i
n
g
N
S
E(
2
3
.
5
6
)
,
R
M
S
E(
1
.
5
)
,
M
A
E(
0
.
5
8
)
,
r
(
0
.
7
1
)
T,
R
H
,
W
S
,
S
R
,
R
F
[
2
0
]
DNN
,
F
M
C
G
EP
-
DNN
H
y
b
r
i
d
M
LP,
B
P
N
N
S
V
M
,
R
a
n
d
o
m f
o
r
e
s
t
1
9
5
7
-
2
0
1
8
(
n
u
meri
c
)
I
S
I
3
,
R
N
M
I
4
a
n
d
A
W
S
5
M
S
E,
R
M
S
E,
M
A
E,
R
2
RF
[
2
1
]
R
N
N
,
GA
-
O
LSTM
H
y
b
r
i
d
LSTM
1
9
0
1
-
2
0
1
7
(
n
u
meri
c
)
I
MD
P
u
n
e
M
S
E(
0
.
0
0
4
)
,
R
M
S
E(
0
.
0
7
8
)
,
C
S
,
R
RF
[
2
2
]
C
EE
M
D
-
C
M
S
E
-
s
t
a
c
k
i
n
g
H
y
b
r
i
d
C
EE
M
D
-
LST
M
st
a
c
k
i
n
g
1
9
6
0
-
2
0
1
9
(
n
u
m
e
r
i
c
)
NA
R
M
S
E
,
M
A
E,
R
2
RF
[
2
3
]
R
N
N
,
R
N
N
-
G
R
U
NA
LSTM
1
9
6
9
-
2
0
2
1
(
n
u
meri
c
)
I
M
D
R
M
S
E
(
1
2
.
7
1
)
,
M
A
E
(
1
3
.
7
2
)
R
F
,
T
,
W
S
[
2
4
]
R
N
N
,
LST
M
NA
A
R
I
M
A
,
M
LP
,
S
V
M
1
9
8
5
-
2
0
1
7
(
n
u
meri
c
)
N
M
S
A
6
R
M
S
E
,
M
S
E,
N
S
E,
M
A
E,
M
A
P
E
,
R
2
(
9
9
.
7
2
)
T,
R
H
,
S
R
,
W
S
,
R
F
[
2
5
]
LSTM
,
S
V
M
-
B
i
LST
M
H
y
b
r
i
d
LSTM
,
B
i
LST
M
S
V
M
-
LSTM
1
9
8
1
-
2
0
2
0
(
n
u
meri
c
)
met
e
o
r
o
l
o
g
i
c
a
l
B
u
r
e
a
u
M
S
E,
N
S
E
,
a
n
d
M
A
E
RF
[
2
6
]
K
N
N
,
X
G
B
,
S
V
R
,
A
N
N
S
t
a
c
k
i
n
g
M
o
d
e
l
En
se
mb
l
e
B
a
se
l
i
n
e
m
o
d
e
l
s
1
9
6
1
-
2
0
1
9
(
n
u
meri
c
)
c
h
i
n
a
m
e
t
e
o
r
o
l
o
g
i
c
a
l
d
a
t
a
s
e
r
v
i
c
e
c
e
n
t
r
e
R
M
S
E
,
M
A
E
,
R
2
C
,
S
LP
,
W
S
,
T,
R
H
,
SR
[
2
7
]
R
N
N
-
LSTM
,
Bi
-
LST
M
,
C
N
N
En
se
mb
l
e
B
a
se
l
i
n
e
m
o
d
e
l
s
2
0
1
5
-
2
0
2
1
(
n
u
mer
i
c
)
I
M
D
R
M
S
E
,
M
A
E,
M
S
E
R
F
,
R
H
,
T
[
2
8
]
R
N
N
,
LST
M
NA
R
a
n
d
o
m f
o
r
e
s
t
1
9
8
0
-
2
0
2
0
(
m
o
n
t
h
l
y
)
n
u
m
e
r
i
c
me
t
e
o
r
o
l
o
g
i
c
a
l
st
a
t
i
o
n
s
R
M
S
E
,
R
S
R
,
L
M
I
,
R
2
,
N
S
E,
Ta
y
l
o
r
a
n
d
V
i
o
l
i
n
d
i
a
g
r
a
ms
RF
[
2
9
]
DNN
,
C
N
N
-
LSTM
H
y
b
r
i
d
M
I
M
Tr
a
j
G
R
U
C
o
n
v
LST
M
Ju
n
e
-
O
c
t
2
0
1
8
-
1
9
(
n
u
m
e
r
i
c
)
w
e
a
t
h
e
r
st
a
t
i
o
n
s
C
S
I
,
F
A
R
,
P
O
D
a
n
d
H
S
S
RF
[
3
0
]
R
N
N
,
LST
M
,
P
C
A
NA
K
N
N
,
L
o
g
i
s
t
i
c
S
V
M
,
r
a
n
d
o
m
f
o
r
e
s
t
,
NB
,
NN
2
0
1
2
-
1
8
(
n
u
m
e
r
i
c
)
B
M
D
1
A
c
c
u
r
a
c
y
(
9
7
.
1
4
)
T,
R
H
,
W
S
,
S
LP
[
3
1
]
DNN
,
C
o
n
v
LST
M
H
y
b
r
i
d
NA
M
a
y
2
0
2
1
a
n
d
D
e
c
2
0
2
1
H
i
maw
a
r
i
r
e
a
l
-
t
i
me
i
ma
g
e
s
si
t
e
,
JM
A
7
S
S
I
M
,
M
S
E
A
c
c
u
r
a
c
y
(
0
.
9
9
0
7
10m
i
n
,
.
9
7
1
7
30
m
i
n
0
.
9
2
0
1
60
m
i
n
)
I
M
G
[
3
2
]
R
N
N
,
LST
M
NA
NA
1
9
0
1
-
2
0
1
5
(
n
u
meri
c
)
I
I
TM
,
P
u
n
e
R
M
S
E(
5
8
.
4
6
)
f
o
r
t
e
st
d
a
t
a
RF
[
3
3
]
R
N
N
,
C
EE
M
D
A
N
w
i
t
h
LST
M
En
se
mb
l
e
LSTM
(
EE
M
D
,
C
EE
M
D
,
C
EE
M
D
A
N
)
1
9
5
1
-
2
0
2
0
(
n
u
meri
c
)
C
h
i
n
a
M
e
t
e
o
r
o
l
o
g
i
c
a
l
D
a
t
a
N
e
t
w
o
r
k
M
A
P
E,
R
M
S
E,
R
2
RF
[
3
4
]
DNN
,
D
e
e
p
En
s
-
R
E
M
En
se
m
b
l
e
S
V
R
,
D
T
b
o
o
s
t
i
n
g
R
a
n
d
o
m f
o
r
e
s
t
Ju
l
y
-
D
e
c
2
0
1
6
(
n
u
mer
i
c
,
i
ma
g
e
)
I
TC
P
8
C
S
I
,
F
A
R
,
P
O
D
,
a
n
d
M
S
E
RF
1
B
a
nglade
s
h
M
e
teor
ologi
c
a
l
De
pa
r
t
ment,
2
Na
ti
ona
l
C
e
nter
of
Hyd
r
olog
y
a
nd
M
e
teor
ology
De
pa
r
tm
e
n
t
B
hutan,
3
I
ndian
S
tati
s
ti
c
a
l
I
ns
ti
tut
e
,
4
R
oya
l
Ne
ther
lands
M
e
teor
ologi
c
a
l
I
ns
ti
tut
e
,
5
Aus
tr
a
li
a
n
W
e
a
ther
S
tation,
6
Na
ti
ona
l
M
e
teor
ologi
c
a
l
S
e
r
vice
Age
nc
y,
7
J
a
pa
n
M
e
teor
ologi
c
a
l
Age
nc
y,
8
I
talian
De
pa
r
tm
e
nt
of
C
ivi
l
P
r
otec
ti
on.
P
a
r
a
mete
r
c
ode
s
:
T
=
tempe
r
a
tur
e
va
r
iable
s
(
e
.
g.
,
T
m
a
x,
T
mi
n,
mea
n
tempe
r
a
tur
e
)
,
R
H
=
r
e
lative
humi
dit
y/dew
-
poin
t
tempe
r
a
tu
r
e
,
W
S
=
W
ind
s
pe
e
d
a
nd/
or
wi
n
d
dir
e
c
ti
o
n,
S
R
=
S
ola
r
r
a
diat
ion/
s
uns
hine/
longwa
ve
r
a
d
iation/
e
va
p
otr
a
ns
pi
r
a
ti
o
n,
S
L
P
=
a
tm
os
phe
r
ic
or
s
e
a
leve
l
pr
e
s
s
ur
e
/geopo
tential
he
ig
ht,
R
F
=
r
a
inf
a
ll
/p
r
e
c
ipi
ta
ti
on/
r
a
inf
a
ll
int
e
ns
it
y,
C
=
c
li
mate
os
c
il
lati
on
indi
c
e
s
(
e
.
g.
,
E
l
Niño,
I
OD
,
S
OI
,
S
HA
M
,
W
P
pa
tt
e
r
ns
)
,
I
M
G
=
r
a
da
r
or
s
a
telli
te
im
a
ge
-
ba
s
e
d
s
pa
ti
a
l
input
s
,
P
C
A
=
pr
incipa
l
c
omponent
a
na
lys
is
us
e
d
f
or
dim
e
ns
ionalit
y
r
e
d
uc
ti
on
.
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
Ma
ch
in
e
lea
r
n
in
g
tec
h
n
iq
u
es f
o
r
r
a
in
fa
ll p
r
ed
ictio
n
:
a
s
ystema
tic
liter
a
tu
r
e
r
ev
ie
w
(
Dee
p
a
S
h
a
r
ma
)
3445
2
.
9
.
B
ia
s
a
nd
lim
it
a
t
i
o
ns
So
m
e
p
o
ten
tial
b
iases
in
clu
d
e
th
e
lim
itatio
n
to
E
n
g
lis
h
-
lan
g
u
a
g
e
s
o
u
r
ce
s
an
d
f
o
u
r
p
r
im
a
r
y
d
atab
ases
,
in
co
n
s
is
ten
t
m
etr
ic
r
ep
o
r
tin
g
ac
r
o
s
s
s
tu
d
ies,
an
d
ex
clu
s
io
n
o
f
u
n
p
u
b
lis
h
ed
o
r
p
r
ep
r
in
t
wo
r
k
.
So
m
e
r
elev
an
t stu
d
ies m
ig
h
t h
av
e
b
e
en
lef
t o
u
t d
u
e
to
th
ese
lim
itatio
n
s
,
wh
ich
co
u
ld
im
p
ac
t h
o
w
b
r
o
ad
ly
th
e
r
ev
iew
f
in
d
in
g
s
ca
n
b
e
a
p
p
lied
.
Fu
r
t
h
er
m
o
r
e
,
th
e
wid
e
r
an
g
e
o
f
d
atasets
,
ev
alu
atio
n
m
eth
o
d
s
,
an
d
ex
p
er
im
e
n
tal
s
etu
p
s
u
s
ed
in
th
e
s
tu
d
ies
m
ad
e
it
d
if
f
icu
lt
to
co
n
d
u
ct
d
i
r
ec
t
q
u
a
n
titativ
e
co
m
p
ar
is
o
n
s
.
Nev
er
th
eless
,
th
is
r
ev
iew
h
ig
h
lig
h
ts
th
e
m
o
s
t sig
n
if
ican
t a
d
v
a
n
ce
m
en
ts
in
r
ai
n
f
all
p
r
ed
ictio
n
f
r
o
m
2
0
1
9
to
2
0
2
4
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
f
in
d
in
g
s
o
f
th
e
5
1
s
tu
d
ies
r
ev
iew
ed
.
T
h
e
m
o
d
els
wer
e
ass
ess
e
d
b
ased
o
n
th
eir
ar
ch
itectu
r
es,
in
p
u
t
d
ata,
an
d
p
er
f
o
r
m
a
n
ce
m
etr
ics.
A
clea
r
tr
en
d
em
e
r
g
ed
,
s
h
o
win
g
a
p
r
ef
er
en
ce
f
o
r
D
L
tech
n
iq
u
es
,
p
ar
ticu
lar
ly
L
ST
M,
GR
U,
an
d
h
y
b
r
id
/en
s
em
b
l
e
m
o
d
els,
wh
ich
g
en
er
ally
o
u
tp
er
f
o
r
m
tr
a
d
itio
n
al
ML
an
d
s
tatis
tical
m
eth
o
d
s
o
win
g
to
th
eir
ab
ilit
y
to
ca
p
tu
r
e
n
o
n
lin
ea
r
an
d
tem
p
o
r
al
p
att
er
n
s
in
r
ain
f
all
d
ata.
A
d
etailed
co
m
p
ar
is
o
n
o
f
t
h
e
m
o
d
els u
s
ed
in
th
e
s
elec
ted
s
tu
d
ies is
p
r
o
v
id
e
d
in
s
u
b
s
ec
tio
n
3
.
1
.
3
.
1
.
O
v
er
a
ll
co
m
pa
riso
ns
o
f
m
a
chine le
a
rning
/deep le
a
rn
ing
m
o
dels
T
ab
le
2
o
f
f
er
s
a
co
m
p
r
eh
en
s
iv
e
co
m
p
a
r
is
o
n
o
f
th
e
ML
a
n
d
DL
m
o
d
els
u
tili
ze
d
in
th
e
r
ev
iewe
d
s
tu
d
ies,
d
etailin
g
th
e
m
o
d
el
t
y
p
es,
d
ataset
f
ea
tu
r
es,
an
d
p
e
r
f
o
r
m
a
n
ce
ev
al
u
atio
n
m
et
r
ics.
T
ab
le
2
illu
s
tr
ates
th
at
r
ec
u
r
r
en
t
m
o
d
el
s
,
s
u
ch
as
L
STM
an
d
GR
U
,
ex
ce
lled
in
h
an
d
lin
g
tim
e
-
s
er
ies
r
ain
f
all
d
ata,
wh
er
ea
s
C
NN
an
d
c
o
n
v
o
lu
tio
n
al
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
C
o
n
v
L
STM
)
m
o
d
els
wer
e
ef
f
ec
tiv
e
w
h
en
wo
r
k
in
g
with
s
p
atial
r
ad
ar
o
r
s
atellite
im
ag
es.
Hy
b
r
id
an
d
e
n
s
em
b
le
m
o
d
els
ty
p
i
ca
lly
ac
h
iev
e
th
e
h
ig
h
est
ac
cu
r
ac
y
b
y
co
m
b
i
n
in
g
s
p
atial
an
d
tem
p
o
r
al
f
ea
tu
r
e
r
e
p
r
esen
tatio
n
s
.
3
.
2
.
Co
m
pa
riso
n o
f
a
lg
o
rit
h
m
s
o
n sa
m
e
da
t
a
s
et
B
ey
o
n
d
th
e
g
en
er
al
m
o
d
el
co
m
p
ar
is
o
n
s
,
n
u
m
er
o
u
s
s
tu
d
ies
h
av
e
s
p
ec
if
ically
ass
es
s
e
d
v
ar
io
u
s
alg
o
r
ith
m
s
o
n
id
en
tical
d
atasets
u
s
in
g
th
e
s
am
e
p
er
f
o
r
m
an
ce
m
etr
ics
to
id
e
n
tify
th
e
m
o
s
t
ef
f
ec
tiv
e
m
o
d
el.
T
ab
le
3
d
ea
ls
with
c
o
m
p
ar
is
o
n
s
d
o
n
e
b
y
r
esear
ch
er
s
b
etwe
en
v
ar
io
u
s
lear
n
in
g
alg
o
r
it
h
m
s
b
ased
o
n
th
e
s
am
e
d
ataset
an
d
th
e
s
am
e
p
er
f
o
r
m
a
n
ce
m
etr
ics in
o
r
d
er
to
f
in
d
th
e
b
est o
n
e.
As s
h
o
wn
in
T
ab
le
3
,
class
if
ier
s
b
a
s
ed
o
n
en
s
em
b
le
an
d
b
o
o
s
tin
g
tech
n
iq
u
es,
s
u
ch
as
r
an
d
o
m
f
o
r
est
,
XGBo
o
s
t,
an
d
C
atB
o
o
s
t
,
ty
p
ically
s
u
r
p
ass
in
d
iv
id
u
al
class
if
ier
s
s
u
ch
as
n
aïv
e
B
ay
es
,
k
-
n
ea
r
est
n
eig
h
b
o
r
s
(
KNN)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
an
d
d
ec
is
io
n
tr
ee
s
(
DT
)
.
T
h
is
is
b
ec
au
s
e
th
ey
ca
n
co
m
b
in
e
m
u
ltip
le
d
ec
is
io
n
b
o
u
n
d
a
r
ies
an
d
m
in
im
ize
v
ar
ian
ce
.
No
n
eth
eless
,
wh
en
d
atasets
d
is
p
lay
p
r
o
n
o
u
n
ce
d
n
o
n
lin
ea
r
tem
p
o
r
al
p
atter
n
s
,
DL
m
o
d
els
lik
e
L
STM
o
f
te
n
d
eliv
er
s
u
p
er
io
r
p
r
e
d
ictiv
e
ac
c
u
r
ac
y
co
m
p
ar
e
d
with
tr
ad
itio
n
al
ML
m
eth
o
d
s
.
T
ab
le
3
.
C
o
m
p
a
r
is
o
n
o
f
m
u
lti
p
le
lear
n
in
g
al
g
o
r
ith
m
s
ev
alu
a
ted
o
n
th
e
s
am
e
d
ataset
Ref
.
A
l
g
o
r
i
t
h
m
u
s
e
d
D
a
t
a
s
e
t
u
s
e
d
P
e
r
f
o
r
ma
n
c
e
me
t
r
i
c
s a
n
d
c
o
n
c
l
u
s
i
o
n
[
3
5
]
C
5
.
0
(
w
i
t
h
a
n
d
w
i
t
h
o
u
t
S
M
O
TE)
R
a
i
n
f
a
l
l
,
SR
,
t
e
m
p
,
h
u
mi
d
i
t
y
,
e
v
a
p
o
r
a
t
i
o
n
,
w
i
n
d
s
p
e
e
d
(
2
0
0
5
-
2
0
1
7
)
C
o
n
f
u
s
i
o
n
ma
t
r
i
x
,
i
mp
r
o
v
e
d
a
c
c
u
r
a
c
y
w
h
e
n
S
M
O
TE
i
s
u
se
d
[
3
6
]
S
V
R
-
f
i
r
e
f
l
y
m
o
d
e
l
R
a
i
n
f
a
l
l
(
1
9
9
0
-
2
0
1
4
)
R
M
S
E
a
n
d
N
S
E
(
b
e
t
t
e
r
a
c
c
u
r
a
c
y
t
h
a
n
S
V
R
)
[
3
7
]
EM
L
R
M
,
WA
-
S
V
M
,
A
N
N
,
a
n
d
n
o
n
-
l
i
n
e
a
r
r
e
g
r
e
s
si
o
n
4
7
c
l
i
m
a
t
e
v
a
r
i
a
b
l
e
s
(
m
e
a
n
v
a
l
u
e
s)
R
M
S
E
,
M
A
E,
a
n
d
R
2
,
b
e
t
t
e
r
p
e
r
f
o
r
ma
n
c
e
w
i
t
h
t
h
e
M
a
p
R
e
d
u
c
e
a
l
g
o
r
i
t
h
m
[
3
8
]
D
T
mo
d
e
l
p
r
o
d
u
c
e
d
b
y
J
4
8
M
i
n
,
m
ax
,
a
n
d
av
g
t
e
mp
,
av
g
h
u
m
i
d
i
t
y
,
r
a
i
n
f
a
l
l
,
s
u
n
e
x
p
o
s
u
r
e
t
i
me
,
m
a
x
a
n
d
a
v
g
w
i
n
d
s
p
e
e
d
(
2
0
1
3
-
2
0
1
9
)
A
c
c
u
r
a
c
y
:
7
7
.
8
%
(
t
r
a
i
n
i
n
g
d
a
t
a
)
;
8
6
%
(
t
e
st
i
n
g
d
a
t
a
)
[
3
9
]
R
a
n
d
o
m f
o
r
e
s
t
a
n
d
l
o
g
i
s
t
i
c
r
e
g
r
e
ss
i
o
n
2
0
1
5
-
18
C
o
n
f
u
s
i
o
n
ma
t
r
i
x
,
a
c
c
u
r
a
c
y
l
o
g
i
s
t
i
c
r
e
g
r
e
ss
i
o
n
[
4
0
]
N
a
ï
v
e
B
a
y
e
s
H
u
mi
d
i
t
y
,
r
a
i
n
f
a
l
l
,
w
i
n
d
s
p
e
e
d
,
a
n
d
p
r
e
c
i
p
i
t
a
t
i
o
n
C
o
n
f
u
s
i
o
n
ma
t
r
i
x
,
a
c
c
u
r
a
c
y
,
e
r
r
o
r
r
a
t
e
N
B
C
g
i
v
e
s a
n
a
c
c
u
r
a
c
y
o
f
9
5
.
9
1
%
[
4
1
]
XG
B
o
o
st
M
i
n
,
ma
x
,
a
n
d
a
v
g
t
e
mp
,
av
g
h
u
m
i
d
i
t
y
,
r
a
i
n
f
a
l
l
,
s
u
n
e
x
p
o
s
u
r
e
t
i
me
ma
x
,
a
n
d
a
v
g
w
i
n
d
s
p
e
e
d
(
2
0
1
3
-
2
0
1
9
)
R
M
S
E
,
M
A
E
,
X
G
B
o
o
s
t
i
n
c
r
e
a
s
e
s
a
c
c
u
r
a
c
y
a
n
d
r
e
d
u
c
e
s
o
v
e
r
f
i
t
t
i
n
g
,
a
n
d
A
v
g
h
u
m
i
d
i
t
y
a
n
d
mi
n
t
e
m
p
i
n
f
l
u
e
n
c
e
r
a
i
n
f
a
l
l
[
4
2
]
M
L
R
,
r
a
n
d
o
m fo
r
e
st
,
a
n
d
X
G
B
o
o
st
D
a
t
e
,
e
v
a
p
o
r
a
t
i
o
n
,
su
n
s
h
i
n
e
,
m
a
x
a
n
d
mi
n
t
e
mp
,
h
u
m
i
d
i
t
y
,
w
i
n
d
s
p
e
e
d
a
n
d
r
a
i
n
f
a
l
l
(
1
9
9
9
-
2
0
1
8
)
R
M
S
E
,
M
A
E
,
a
n
d
X
G
B
o
o
s
t
[
4
3
]
C
a
t
b
o
o
st
c
l
a
ssi
f
i
e
r
,
p
e
r
c
e
p
t
r
o
n
c
l
a
ss
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
ms
2
3
p
a
r
a
me
t
e
r
s
N
o
v
2
0
0
7
-
Ju
n
e
2
0
1
7
A
c
c
u
r
a
c
y
,
c
o
n
f
u
si
o
n
ma
t
r
i
x
c
a
t
b
o
o
st
c
l
a
ssi
f
i
e
r
a
c
c
u
r
a
c
y
=
8
1
.
3
6
[
4
4
]
Lo
g
i
s
t
i
c
r
e
g
l
i
n
e
a
r
DA
q
u
a
d
r
a
t
i
c
DA
,
K
N
N
,
D
T,
g
r
a
d
i
e
n
t
b
o
o
s
t
e
d
t
r
e
e
s
,
r
a
n
d
o
m f
o
r
e
s
t
b
e
r
n
o
u
l
l
i
N
B
,
a
n
d
d
e
e
p
NN
2
5
p
a
r
a
me
t
e
rs
2
0
0
7
-
17
p
r
e
c
i
si
o
n
,
F1
-
sc
o
r
e
,
a
n
d
DL
m
o
d
e
l
(
9
8
.
2
6
%
,
8
8
.
6
1
%)
[
4
5
]
Li
n
e
a
r
a
n
d
l
a
ss
o
r
e
g
,
r
i
d
g
e
m
o
d
e
l
,
K
N
N
,
a
n
d
r
a
n
d
o
m fo
r
e
st
R
a
i
n
f
a
l
l
1
9
0
1
-
2
0
1
5
M
A
E
,
M
S
E
,
R
M
S
E
K
N
N
,
a
n
d
r
a
n
d
o
m fo
r
e
st
p
e
r
f
o
r
m
b
e
t
t
e
r
[
4
6
]
D
T,
K
N
N
,
a
n
d
l
o
g
i
s
t
i
c
r
e
g
r
e
ss
i
o
n
max
a
n
d
mi
n
t
e
m
p
,
w
i
n
d
sp
e
e
d
,
h
u
m
i
d
i
t
y
,
p
r
e
ss
u
r
e
,
a
n
d
c
l
o
u
d
M
A
E
,
M
S
E
,
R
M
S
E
,
K
N
N
(
8
4
.
1
8
3
%)
,
a
n
d
DT
(
8
3
.
7
6
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
4
1
-
3
4
5
1
3446
3
.
3
.
Co
m
pa
riso
n o
f
m
o
del a
rc
hite
ct
ures
I
n
ad
d
itio
n
to
c
o
m
p
ar
i
n
g
al
g
o
r
ith
m
s
,
n
u
m
er
o
u
s
s
tu
d
ies
h
av
e
ass
ess
ed
v
ar
io
u
s
m
o
d
el
ar
c
h
itectu
r
es
to
id
en
tify
th
e
m
o
s
t
ef
f
ec
tiv
e
m
e
th
o
d
f
o
r
r
ain
f
all
p
r
ed
ictio
n
.
T
ab
le
4
p
r
esen
ts
a
co
m
p
ar
is
o
n
o
f
d
if
f
e
r
en
t
r
ain
f
all
p
r
ed
ictio
n
m
o
d
els
t
o
d
eter
m
i
n
e
th
e
b
est
m
o
d
el
b
ased
o
n
t
h
e
p
er
f
o
r
m
an
ce
m
et
r
ics.
T
ab
le
4
s
h
o
ws
th
at
th
e
ef
f
ec
tiv
en
ess
o
f
th
e
r
ain
f
all
p
r
ed
ictio
n
m
o
d
el
d
ep
e
n
d
s
o
n
th
e
n
etwo
r
k
a
r
ch
itectu
r
e.
R
ec
u
r
r
en
t
DL
m
o
d
els
lik
e
L
STM
,
Stack
ed
L
STM
,
an
d
b
id
ir
ec
tio
n
al
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
B
iLST
M
)
o
u
tp
e
r
f
o
r
m
tr
ad
itio
n
al
ANN
an
d
R
NN
m
o
d
els
b
y
ca
p
tu
r
in
g
lo
n
g
-
ter
m
tem
p
o
r
al
d
e
p
en
d
e
n
cies
in
r
ain
f
all
d
ata.
Mo
d
els
w
ith
m
em
o
r
y
g
atin
g
m
ec
h
an
is
m
s
ex
ce
l
at
h
an
d
lin
g
s
eq
u
en
tial
r
ain
f
all
ev
en
ts
d
u
r
in
g
m
u
lti
-
s
tep
f
o
r
ec
asti
n
g
.
Hy
b
r
id
m
o
d
els
co
m
b
in
in
g
L
STM
with
C
NN
o
r
o
p
tim
izatio
n
co
m
p
o
n
en
ts
s
h
o
w
im
p
r
o
v
e
m
en
ts
b
y
in
teg
r
a
tin
g
tem
p
o
r
al
an
d
s
p
atial
lear
n
in
g
.
T
h
ese
f
in
d
in
g
s
s
h
o
w
th
at
s
u
cc
ess
f
u
l
r
ain
f
a
ll
p
r
ed
ictio
n
r
eq
u
ir
es
m
o
d
elin
g
co
m
p
lex
tem
p
o
r
al
in
ter
ac
tio
n
s
r
ath
er
th
a
n
u
s
in
g
s
tatic
r
ep
r
esen
tatio
n
s
.
T
ab
le
4
.
C
o
m
p
a
r
is
o
n
o
f
d
if
f
e
r
en
t r
ain
f
all
p
r
e
d
ictio
n
m
o
d
el
a
r
ch
itectu
r
es
Ref
.
C
o
m
p
a
r
i
so
n
d
o
n
e
b
e
t
w
e
e
n
B
e
st
one
P
e
r
f
o
r
ma
n
c
e
me
t
r
i
c
s
u
se
d
[
4
7
]
M
LN
P
P
(
5
0
h
i
d
d
e
n
a
n
d
Ta
n
g
e
n
t
)
a
n
d
M
LN
P
P
(
1
0
0
h
i
d
d
e
n
a
n
d
S
i
g
m
o
i
d
)
M
LN
P
P
w
i
t
h
5
0
h
i
d
d
e
n
n
e
u
r
o
n
s S
C
G
-
t
a
n
g
e
n
t
M
A
E
a
n
d
R
M
S
E
[
4
8
]
Ec
h
o
s
t
a
t
e
n
e
t
w
o
r
k
(
ESN
)
a
n
d
d
e
e
p
E
SN
D
e
e
p
ESN
R
M
S
E
,
N
R
M
S
E
,
a
n
d
r
[
4
9
]
P
S
O
A
N
F
I
S
,
ANN
,
a
n
d
S
V
M
S
V
M
R,
M
A
E,
P
O
D
,
C
S
I
,
F
A
R
,
a
n
d
r
o
b
u
st
n
e
ss
[
5
0
]
ANN
a
n
d
LST
M
LSTM
M
S
E,
R
M
S
E,
a
n
d
M
A
E
[
5
1
]
B
D
TR
,
D
F
R
,
NNR
,
a
n
d
B
LR
B
D
TR
M
A
E,
R
M
S
E,
R
A
E
,
R
S
E,
a
n
d
R
2
[
5
2
]
NN
,
S
V
M
,
N
B
,
r
a
n
d
o
m f
o
r
e
st
,
a
n
d
GA
NN
A
c
c
u
r
a
c
y
=
9
6
.
4
4
[
5
3
]
B
r
a
z
i
l
i
a
n
A
t
m
o
s
p
h
e
r
i
c
m
o
d
e
l
,
NN
-
Te
n
s
o
r
F
l
o
w
,
a
n
d
NN
-
M
P
C
A
N
N
(
o
t
h
e
r
)
a
n
d
NN
-
M
P
C
A
(
S
p
r
i
n
g
)
R
M
S
E
C
O
V
-
c
o
v
a
r
i
a
n
c
e
mea
n
e
r
r
o
r
(
M
E)
[
5
4
]
M
L
R
,
S
V
M
,
K
N
N
,
a
n
d
r
a
n
d
o
m f
o
r
e
st
R
a
n
d
o
m f
o
r
e
s
t
A
c
c
u
r
a
c
y
(
8
9
.
1
6
)
,
p
r
e
c
i
si
o
n
,
a
n
d
r
e
c
a
l
l
[
5
5
]
M
L
R
,
K
N
N
,
S
V
M
,
ANN
,
DT
,
a
n
d
r
a
n
d
o
m
f
o
r
e
s
t
R
a
n
d
o
m f
o
r
e
s
t
A
c
c
u
r
a
c
y
(
9
6
.
1
)
,
p
r
e
c
i
s
i
o
n
,
r
e
c
a
l
l
,
R
M
S
E,
Er
r
o
r
,
M
A
E,
a
n
d
F
-
mea
s
u
r
e
s
[
5
6
]
M
L
R
,
K
N
N
,
S
V
M
,
ANN
,
DT
,
a
n
d
r
a
n
d
o
m
f
o
r
e
s
t
R
a
n
d
o
m f
o
r
e
s
t
A
c
c
u
r
a
c
y
(
9
6
.
1
)
,
c
o
n
f
u
si
o
n
ma
t
r
i
x
,
a
n
d
R
O
C
[
5
7
]
K
N
N
a
n
d
DT
DT
A
c
c
u
r
a
c
y
(
9
3
.
1
3
)
[
5
8
]
LSTM
,
s
t
a
c
k
e
d
LSTM
,
B
i
LS
TM
,
X
G
B
,
a
n
d
e
n
s
e
mb
le
S
t
a
c
k
e
d
LSTM
a
n
d
B
i
LST
M
Lo
ss,
R
M
S
E,
a
n
d
R
M
S
E
L
3
.
4
.
T
im
e
perio
d f
o
r
which
t
he
da
t
a
s
et
is
co
llect
ed
Data
s
et
len
g
th
s
in
th
e
r
ev
iewe
d
s
tu
d
ies
r
an
g
ed
f
r
o
m
m
o
n
th
s
to
o
v
er
a
ce
n
tu
r
y
,
with
m
o
s
t
s
p
an
n
in
g
5
-
2
0
y
ea
r
s
to
p
r
o
v
id
e
tem
p
o
r
al
d
iv
er
s
ity
wh
ile
m
in
im
izin
g
h
is
to
r
ical
d
ata
is
s
u
es.
T
em
p
o
r
al
r
eso
lu
tio
n
s
elec
tio
n
d
ep
en
d
s
o
n
f
o
r
ec
ast
in
g
g
o
als:
h
o
u
r
ly
d
ata/d
aily
d
ata/we
ek
ly
o
r
m
o
n
th
ly
d
ata
f
o
r
v
er
y
s
h
o
r
t
-
ter
m
/
s
h
o
r
t/ m
ed
iu
m
-
r
an
g
e
r
ain
f
all
i
n
ten
s
ity
.
Fig
u
r
e
2
s
h
o
ws th
e
d
ataset
d
u
r
atio
n
d
is
tr
ib
u
tio
n
ac
r
o
s
s
s
tu
d
ies.
Fig
u
r
e
2
.
Dis
tr
ib
u
tio
n
o
f
th
e
d
ataset
tim
e
p
er
io
d
3
.
5
.
F
re
qu
ency
o
f
i
np
ut
pa
ra
m
et
er
s
us
ed
C
h
o
o
s
in
g
in
p
u
t p
ar
am
eter
s
is
cr
u
cial
f
o
r
ef
f
ec
tiv
e
r
ain
f
all
p
r
ed
ictio
n
m
o
d
els,
as th
ese
d
ir
ec
tly
im
p
ac
t
th
e
lear
n
in
g
p
r
o
ce
s
s
.
I
n
r
ev
ie
wed
s
tu
d
ies,
k
ey
m
eteo
r
o
lo
g
i
ca
l
p
ar
am
eter
s
wer
e
f
r
eq
u
en
tl
y
u
s
ed
d
u
e
to
th
eir
s
tr
o
n
g
c
o
r
r
elatio
n
with
r
ai
n
f
a
ll.
As
s
h
o
wn
in
Fig
u
r
e
3
,
th
e
m
o
s
t
co
m
m
o
n
in
p
u
ts
ar
e
r
ai
n
f
all,
tem
p
e
r
atu
r
e,
h
u
m
id
ity
,
an
d
win
d
s
p
ee
d
,
f
o
l
lo
wed
b
y
s
o
lar
r
a
d
iatio
n
,
p
r
es
s
u
r
e,
clo
u
d
c
o
v
er
,
a
n
d
clim
ate
in
d
ices
lik
e
E
NSO
an
d
I
OD.
T
h
ese
in
d
icate
th
at
b
o
th
lo
ca
l a
n
d
g
lo
b
al
clim
ate
co
n
d
itio
n
s
in
f
l
u
en
ce
r
ain
f
all
v
a
r
iab
ilit
y
.
22
7
3
3
0
6
0
1
0
5
10
15
20
25
0
-
2
0
2
1
-
4
0
4
1
-
6
0
6
1
-
8
0
8
1
-
1
0
0
1
0
1
-
1
2
0
1
2
1
-
1
4
0
1
4
1
-
1
6
0
F
r
e
q
u
e
n
c
y
T
i
me
P
e
r
i
o
d
(
R
a
n
g
e
i
n
Y
e
a
r
s)
D
a
t
a
se
t
T
i
me
P
e
r
i
o
d
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
Ma
ch
in
e
lea
r
n
in
g
tec
h
n
iq
u
es f
o
r
r
a
in
fa
ll p
r
ed
ictio
n
:
a
s
ystema
tic
liter
a
tu
r
e
r
ev
ie
w
(
Dee
p
a
S
h
a
r
ma
)
3447
Fig
u
r
e
3
.
W
o
r
d
clo
u
d
r
ep
r
esen
tatio
n
o
f
th
e
m
o
s
t f
r
eq
u
en
tly
u
s
ed
in
p
u
t p
a
r
am
eter
s
3
.
6
.
P
er
f
o
r
m
a
nce
m
e
t
rics us
ed
Per
f
o
r
m
an
ce
m
etr
ics
ar
e
ess
en
tial
f
o
r
ev
alu
atin
g
r
ai
n
f
all
p
r
ed
ictio
n
m
o
d
els
’
ac
cu
r
ac
y
an
d
s
tab
ilit
y
.
Fig
u
r
e
4
s
h
o
ws
th
e
f
r
eq
u
e
n
c
y
o
f
co
m
m
o
n
p
e
r
f
o
r
m
an
ce
m
etr
ics.
T
h
e
R
MSE
i
s
m
o
s
t
wid
ely
u
s
ed
as
it
p
en
alize
s
lar
g
er
er
r
o
r
s
,
wh
ile
MA
E
is
v
alu
e
d
f
o
r
in
ter
p
r
e
tab
ilit
y
.
C
o
r
r
elatio
n
-
b
ased
m
etr
ics
lik
e
R
²
an
d
P
ea
r
s
o
n
co
r
r
elatio
n
c
o
ef
f
icien
t
(
r
)
ev
alu
ate
p
r
ed
ictio
n
alig
n
m
en
t
with
o
b
s
er
v
ed
p
atter
n
s
.
C
las
s
if
icatio
n
-
b
ased
s
tu
d
ies
u
s
e
co
n
f
u
s
io
n
m
at
r
ix
m
etr
ics
,
in
clu
d
in
g
ac
cu
r
a
cy
,
p
r
ec
is
io
n
,
r
ec
all
,
an
d
F
-
m
ea
s
u
r
e
.
Hy
b
r
id
f
o
r
ec
asti
n
g
u
s
ed
Nash
–
Su
tcli
f
f
e
ef
f
icien
cy
(
NSE)
a
n
d
m
e
an
ab
s
o
lu
te
p
er
ce
n
tag
e
er
r
o
r
(
MA
PE)
.
Fig
u
r
e
4
s
h
o
ws th
e
d
is
tr
ib
u
tio
n
,
with
R
MSE
an
d
MA
E
as d
o
m
in
a
n
t i
n
d
icato
r
s
.
Fig
u
r
e
4
.
W
o
r
d
clo
u
d
r
ep
r
esen
tatio
n
o
f
d
i
f
f
er
en
t
p
er
f
o
r
m
an
c
e
m
etr
ics u
s
ed
3
.
7
.
I
ncre
a
s
ing
us
e
o
f
ens
em
ble le
a
rning
f
o
r
m
a
k
ing
predict
io
n m
o
dels
R
ec
en
t
tr
en
d
s
in
r
ain
f
all
p
r
e
d
ictio
n
r
esear
c
h
h
a
v
e
s
h
if
te
d
to
war
d
s
h
y
b
r
id
a
n
d
en
s
em
b
le
lear
n
in
g
ap
p
r
o
ac
h
es.
W
h
ile
in
d
iv
id
u
al
m
o
d
els
s
u
ch
as
L
STM
,
C
NN,
o
r
r
an
d
o
m
f
o
r
est
ca
p
tu
r
e
s
p
ec
if
ic
p
atter
n
s
,
h
y
b
r
id
f
r
am
ewo
r
k
s
co
m
b
in
e
th
e
s
tr
en
g
th
s
o
f
m
u
ltip
le
alg
o
r
ith
m
s
to
im
p
r
o
v
e
th
e
p
r
ed
ictio
n
ac
cu
r
ac
y
.
E
n
s
em
b
le
m
o
d
els
s
u
ch
as
s
ta
ck
in
g
,
b
o
o
s
tin
g
,
an
d
b
a
g
g
in
g
r
ed
u
ce
m
o
d
el
v
ar
ia
n
ce
b
y
a
g
g
r
eg
atin
g
o
u
tp
u
ts
,
wh
ile
h
y
b
r
id
DL
ar
ch
itectu
r
e
s
(
e.
g
.
,
C
NN
-
L
STM
,
C
o
n
v
L
STM
,
an
d
GR
U
-
L
STM
co
m
b
in
atio
n
s
)
in
teg
r
ate
s
p
atial
an
d
tem
p
o
r
al
lear
n
in
g
.
Fro
m
1
9
9
0
to
2
0
1
9
,
th
e
n
u
m
b
er
o
f
p
ap
er
s
o
n
“
en
s
em
b
le
lear
n
i
n
g
”
th
at
wer
e
r
elea
s
ed
in
th
e
c
o
r
e
s
et
o
f
W
eb
o
f
Scien
ce
h
as
g
r
o
wn
at
a
r
ap
id
p
ac
e
[
5
9
]
.
Gan
aie
et
a
l.
[
6
0
]
ad
d
r
ess
ed
th
e
ef
f
ec
tiv
en
ess
o
f
en
s
em
b
l
e
lear
n
in
g
an
d
t
h
e
m
an
y
s
tr
ateg
ies
u
s
ed
to
en
h
an
ce
it,
i
n
clu
d
in
g
b
ag
g
in
g
,
b
o
o
s
tin
g
,
an
d
h
eter
o
g
en
eo
u
s
en
s
em
b
les.
Fig
u
r
e
5
s
h
o
ws
an
in
cr
ea
s
in
g
tr
en
d
in
u
s
in
g
th
e
h
y
b
r
i
d
m
o
d
el
o
r
en
s
em
b
le
m
o
d
el
f
o
r
p
r
ed
ictio
n
co
m
p
ar
e
d
to
ea
r
lier
tim
es.
R
ec
en
t
s
tu
d
ies
h
av
e
f
u
r
th
er
r
ein
f
o
r
ce
d
th
is
tr
en
d
.
Kan
an
i
et
a
l.
[
6
1
]
in
tr
o
d
u
ce
d
an
e
n
s
em
b
le
m
et
h
o
d
b
ased
o
n
L
STM
t
h
at
g
r
ea
tly
e
n
h
an
ce
d
th
e
ac
cu
r
ac
y
o
f
r
ain
f
all
class
if
icatio
n
an
d
r
eg
r
ess
io
n
.
I
n
a
s
im
ilar
v
ein
,
Pu
tr
a
et
a
l.
[
6
2
]
u
tili
ze
d
m
u
ltis
en
s
o
r
d
ata
alo
n
g
with
en
s
em
b
le
lear
n
in
g
to
im
p
r
o
v
e
h
ig
h
-
r
eso
lu
tio
n
r
ain
f
all
esti
m
atio
n
,
h
i
g
h
lig
h
tin
g
th
e
in
cr
ea
s
in
g
m
o
v
e
t
o
war
d
s
h
y
b
r
id
an
d
e
n
s
em
b
le
-
b
ased
s
tr
ateg
ies.
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
4
1
-
3
4
5
1
3448
Fig
u
r
e
5
.
Yea
r
-
wis
e
f
r
eq
u
en
c
y
o
f
h
y
b
r
id
/e
n
s
em
b
le
o
r
s
im
p
le
m
o
d
els u
s
ed
3
.
8
.
F
ind
ing
t
he
bes
t
m
o
del
T
h
e
co
m
p
a
r
ativ
e
an
aly
s
is
s
h
o
wed
th
at
r
ec
u
r
r
en
t
DL
a
r
c
h
itectu
r
es
o
u
tp
er
f
o
r
m
e
d
tr
ad
i
tio
n
al
ML
m
eth
o
d
s
in
r
ain
f
all
p
r
e
d
ictio
n
.
L
STM
o
u
tp
er
f
o
r
m
ed
o
th
er
r
ec
u
r
r
en
t
n
etwo
r
k
alg
o
r
ith
m
s
an
d
co
u
ld
tack
le
co
m
p
licated
p
r
o
b
lem
s
with
s
u
b
s
tan
tial
tim
e
lag
s
[
6
3
]
.
Hy
b
r
id
m
o
d
els,
s
u
c
h
as
C
NN
-
L
STM
an
d
C
o
n
v
L
STM
,
en
h
an
ce
p
r
ed
ictio
n
b
y
in
teg
r
at
in
g
tem
p
o
r
al
an
d
s
p
atial
f
ea
tu
r
e
ex
tr
ac
tio
n
,
lead
in
g
to
r
o
b
u
s
t
p
r
ed
ictio
n
s
u
n
d
er
v
ar
y
in
g
co
n
d
itio
n
s
.
E
n
s
em
b
l
e
-
b
ased
d
ee
p
-
lear
n
i
n
g
f
r
am
e
wo
r
k
s
th
at
co
m
b
i
n
e
m
u
ltip
le
ar
ch
itectu
r
es
s
h
o
w
im
p
r
o
v
e
d
g
e
n
er
aliza
tio
n
an
d
r
ed
u
ce
d
f
o
r
ec
ast
u
n
ce
r
tain
ty
.
T
h
ese
f
i
n
d
in
g
s
in
d
icate
t
h
a
t
ef
f
ec
tiv
e
r
ain
f
all
p
r
ed
ictio
n
m
o
d
els
m
u
s
t
ca
p
tu
r
e
tem
p
o
r
al
r
ain
f
all
ev
o
lu
tio
n
an
d
s
p
atial
clim
atic
in
ter
ac
ti
o
n
s
.
L
STM
-
b
ased
h
y
b
r
id
an
d
en
s
em
b
le
m
o
d
els
h
av
e
em
e
r
g
ed
as
s
tr
o
n
g
ca
n
d
id
ates
f
o
r
o
p
er
atio
n
al
f
o
r
ec
asti
n
g
o
win
g
to
th
eir
s
tab
ilit
y
an
d
ac
cu
r
ac
y
ac
r
o
s
s
m
eteo
r
o
lo
g
ical
c
o
n
tex
ts
.
3
.
9
.
F
ind
ing
t
he
bes
t
a
lg
o
rit
hm
T
ab
le
3
s
h
o
ws
alg
o
r
ith
m
ch
o
ice
s
ig
n
if
ican
tly
im
p
ac
ts
r
ain
f
all
p
r
ed
ictio
n
ac
cu
r
a
cy
.
C
las
s
ica
l
alg
o
r
ith
m
s
lik
e
KNN,
SVM,
DT
,
an
d
lo
g
is
tic
r
eg
r
ess
io
n
p
er
f
o
r
m
ad
e
q
u
ately
with
lim
ited
d
ata
b
u
t
s
tr
u
g
g
le
with
co
m
p
lex
atm
o
s
p
h
e
r
ic
r
el
atio
n
s
h
ip
s
.
R
an
d
o
m
f
o
r
est
o
u
t
p
er
f
o
r
m
s
in
d
iv
id
u
al
class
if
ier
s
lik
e
n
aïv
e
B
a
y
es,
m
u
ltil
ay
er
p
e
r
ce
p
tr
o
n
(
MLP
)
,
s
eq
u
en
tial
m
i
n
im
al
o
p
tim
izatio
n
(
SMO
)
,
DT
,
KNN,
a
n
d
SVM
b
y
m
o
d
elin
g
n
o
n
lin
ea
r
in
ter
ac
tio
n
s
[
6
4
]
.
L
o
g
is
tic
r
eg
r
ess
io
n
s
u
r
p
ass
ed
r
an
d
o
m
f
o
r
est
with
lin
ea
r
r
elatio
n
s
h
ip
s
[
3
9
]
,
w
h
ile
XGBo
o
s
t
s
h
o
wed
h
ig
h
er
ac
c
u
r
ac
y
,
d
em
o
n
s
tr
atin
g
b
o
o
s
ted
en
s
em
b
le
ad
v
a
n
tag
es
[
4
2
]
.
Fo
r
tim
e
-
d
ep
e
n
d
en
t
r
ain
f
all,
L
STM
an
d
GR
U
o
u
tp
er
f
o
r
m
class
ical
m
o
d
els
b
y
r
etain
in
g
lo
n
g
-
ter
m
tem
p
o
r
al
i
n
f
o
r
m
atio
n
th
r
o
u
g
h
m
em
o
r
y
m
ec
h
an
is
m
s
.
No
alg
o
r
ith
m
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
s
o
th
er
s
ac
r
o
s
s
co
n
d
itio
n
s
.
Selectio
n
s
h
o
u
ld
b
e
b
ased
o
n
d
ataset
c
h
ar
ac
ter
is
tics
:
en
s
em
b
le
m
eth
o
d
s
s
u
it
n
o
n
lin
ea
r
d
ata,
wh
ile
L
STM
m
o
d
els
e
x
ce
l
in
tim
e
-
s
er
ies f
o
r
ec
asti
n
g
.
4.
CO
NCLU
SI
O
N
AND
F
U
T
U
RE
WO
RK
T
h
is
r
ev
iew
h
i
g
h
lig
h
ts
s
ig
n
if
ican
t
alg
o
r
ith
m
ic
d
ef
icie
n
cies
an
d
p
r
o
p
o
s
es
f
u
t
u
r
e
r
esear
c
h
p
ath
s
i
n
m
u
lti
-
m
o
d
al,
h
y
b
r
id
,
a
n
d
r
ea
l
-
tim
e
r
ain
f
all
f
o
r
ec
asti
n
g
.
Ac
cu
r
ate
r
ain
f
all
p
r
ed
ictio
n
is
cr
itical
f
o
r
m
itig
atin
g
ex
tr
em
e
wea
th
er
im
p
ac
ts
s
u
ch
as
f
lo
o
d
s
,
lan
d
s
lid
es,
an
d
d
r
o
u
g
h
ts
.
Stu
d
ies
h
av
e
s
h
o
wn
th
at
m
o
d
el
p
er
f
o
r
m
an
ce
im
p
r
o
v
es
th
r
o
u
g
h
th
e
ca
r
ef
u
l
s
elec
tio
n
o
f
m
et
eo
r
o
lo
g
ical
p
ar
a
m
eter
s
,
n
etwo
r
k
ar
c
h
itectu
r
e,
an
d
lear
n
in
g
s
tr
ateg
ies.
W
h
ile
DL
m
o
d
els,
p
ar
ticu
lar
ly
L
STM
-
b
ased
an
d
h
y
b
r
id
f
r
am
ew
o
r
k
s
,
s
h
o
w
s
u
p
er
io
r
ca
p
ab
ilit
y
in
ca
p
tu
r
in
g
r
ai
n
f
all
p
atter
n
s
,
r
ef
in
em
e
n
t is r
eq
u
ir
e
d
f
o
r
r
eliab
ilit
y
u
n
d
e
r
v
ar
ia
b
le
co
n
d
itio
n
s
.
F
u
t
u
r
e
r
es
ea
r
c
h
s
h
o
u
l
d
f
o
cu
s
o
n
h
y
b
r
id
a
n
d
f
u
s
i
o
n
-
b
as
ed
en
s
em
b
le
tec
h
n
i
q
u
es
,
m
u
l
ti
-
s
o
u
r
ce
d
at
ase
ts
,
r
e
al
-
ti
m
e
p
r
ed
ictio
n
s
,
an
d
in
teg
r
atio
n
o
f
p
h
y
s
ical
clim
ate
m
o
d
els with
DL
f
o
r
r
o
b
u
s
t r
ain
f
all
f
o
r
ec
ast
in
g
s
y
s
tem
s
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
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
.
0
4
2
11
1
1
2
6
0
2
4
6
8
10
12
2
0
1
9
2
0
2
0
2
0
2
1
2
0
2
2
Fr
e
q
u
e
n
c
y
Y
e
a
r
H
Y
B
R
I
D
/
EN
S
EM
B
LE
S
I
M
P
L
E
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
Ma
ch
in
e
lea
r
n
in
g
tec
h
n
iq
u
es f
o
r
r
a
in
fa
ll p
r
ed
ictio
n
:
a
s
ystema
tic
liter
a
tu
r
e
r
ev
ie
w
(
Dee
p
a
S
h
a
r
ma
)
3449
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Dee
p
a
Sh
ar
m
a
✓
✓
✓
✓
✓
✓
✓
✓
An
an
d
Ku
m
a
r
Sh
u
k
la
✓
✓
✓
✓
Pu
n
am
R
attan
✓
✓
✓
✓
✓
✓
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
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
a
v
aila
b
i
lit
y
is
n
o
t
a
p
p
li
ca
b
le
t
o
t
h
is
p
ap
er
as
n
o
n
e
w
d
a
t
a
w
er
e
cr
ea
te
d
o
r
an
al
y
z
e
d
i
n
t
h
is
s
t
u
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
B
a
b
a
r
,
M
.
R
a
n
i
,
a
n
d
I
.
A
l
i
,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
se
d
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
f
o
r
f
l
o
o
d
m
a
n
a
g
e
me
n
t
,
”
i
n
2
0
2
2
1
7
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Em
e
r
g
i
n
g
T
e
c
h
n
o
l
o
g
i
e
s,
I
C
ET
2
0
2
2
,
2
0
2
2
,
p
p
.
1
9
6
–
1
9
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
ET5
6
6
0
1
.
2
0
2
2
.
1
0
0
0
4
6
6
3
.
[
2
]
R
.
T
i
c
h
a
v
s
k
ý
,
J
.
A
.
B
.
-
C
á
n
o
v
a
s
,
K
.
Š
i
l
h
á
n
,
R
.
T
o
l
a
s
z
,
a
n
d
M
.
S
t
o
f
f
e
l
,
“
D
r
y
s
p
e
l
l
s
a
n
d
e
x
t
r
e
m
e
p
r
e
c
i
p
i
t
a
t
i
o
n
a
r
e
t
h
e
m
a
i
n
t
r
i
g
g
e
r
o
f
l
a
n
d
s
l
i
d
e
s
i
n
C
e
n
t
r
a
l
E
u
r
o
p
e
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
r
t
s
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
1
–
1
0
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
1
9
-
5
1
1
4
8
-
2.
[
3
]
A
.
C
.
M
o
n
d
i
n
i
,
F
.
G
u
z
z
e
t
t
i
,
a
n
d
M
.
M
e
l
i
l
l
o
,
“
D
e
e
p
l
e
a
r
n
i
n
g
f
o
r
e
c
a
st
o
f
r
a
i
n
f
a
l
l
-
i
n
d
u
c
e
d
s
h
a
l
l
o
w
l
a
n
d
sl
i
d
e
s,
”
N
a
t
u
r
e
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
1
4
,
n
o
.
1
,
p
p
.
1
–
1
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
023
-
3
8
1
3
5
-
y.
[
4
]
S
.
P
o
o
r
n
i
m
a
,
M
.
P
u
s
h
p
a
l
a
t
h
a
,
R
.
B
.
J
a
n
a
,
a
n
d
L
.
A
.
P
a
t
t
i
,
“
R
a
i
n
f
a
l
l
f
o
r
e
c
a
s
t
a
n
d
d
r
o
u
g
h
t
a
n
a
l
y
si
s
f
o
r
r
e
c
e
n
t
a
n
d
f
o
r
t
h
c
o
mi
n
g
y
e
a
r
s
i
n
I
n
d
i
a
,
”
W
a
t
e
r
,
v
o
l
.
1
5
,
n
o
.
3
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
w
1
5
0
3
0
5
9
2
.
[
5
]
D
.
M
o
h
e
r
,
D
.
G
.
A
l
t
m
a
n
,
A
.
Li
b
e
r
a
t
i
,
a
n
d
J.
T
e
t
z
l
a
f
f
,
“
P
R
I
S
M
A
st
a
t
e
me
n
t
,
”
E
p
i
d
e
m
i
o
l
o
g
y
,
v
o
l
.
2
2
,
n
o
.
1
,
2
0
1
1
,
d
o
i
:
1
0
.
1
0
9
7
/
ED
E.
0
b
0
1
3
e
3
1
8
1
f
e
7
8
2
5
.
[
6
]
A
.
G
.
-
L
o
p
e
z
,
I
.
C
.
-
P
a
z
,
a
n
d
M
.
M
.
M
a
n
d
u
j
a
n
o
,
“
A
l
g
o
r
i
t
h
m
t
o
p
r
e
d
i
c
t
t
h
e
r
a
i
n
f
a
l
l
s
t
a
r
t
i
n
g
p
o
i
n
t
a
s
a
f
u
n
c
t
i
o
n
o
f
a
t
m
o
s
p
h
e
r
i
c
p
r
e
ss
u
r
e
,
h
u
mi
d
i
t
y
,
a
n
d
d
e
w
p
o
i
n
t
,
”
C
l
i
m
a
t
e
,
v
o
l
.
7
,
n
o
.
1
1
,
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/
c
l
i
7
1
1
0
1
3
1
.
[
7
]
M
.
S
.
P
a
t
h
a
n
,
J
.
W
u
,
Y
.
H
.
Le
e
,
J.
Y
a
n
,
a
n
d
S
.
D
e
v
,
“
A
n
a
l
y
z
i
n
g
t
h
e
i
mp
a
c
t
o
f
m
e
t
e
o
r
o
l
o
g
i
c
a
l
p
a
r
a
me
t
e
r
s
o
n
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
,
”
i
n
2
0
2
1
I
EE
E
U
S
N
C
-
U
RS
I
R
a
d
i
o
S
c
i
e
n
c
e
M
e
e
t
i
n
g
(
J
o
i
n
t
w
i
t
h
A
P
-
S
S
y
m
p
o
s
i
u
m
)
,
U
S
N
C
-
U
RS
I
2
0
2
1
,
2
0
2
1
,
p
p
.
1
0
0
–
1
0
1
,
d
o
i
:
1
0
.
2
3
9
1
9
/
U
S
N
C
-
U
R
S
I
5
1
8
1
3
.
2
0
2
1
.
9
7
0
3
6
6
4
.
[
8
]
P
.
Z
h
a
n
g
,
W
.
C
a
o
,
a
n
d
W
.
Li
,
“
S
u
r
f
a
c
e
a
n
d
h
i
g
h
-
a
l
t
i
t
u
d
e
c
o
m
b
i
n
e
d
r
a
i
n
f
a
l
l
f
o
r
e
c
a
st
i
n
g
u
s
i
n
g
a
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
Pe
e
r
-
to
-
P
e
e
r
N
e
t
w
o
rk
i
n
g
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
4
,
n
o
.
3
,
p
p
.
1
7
6
5
–
1
7
7
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
2
0
8
3
-
0
2
0
-
0
0
9
3
8
-
x.
[
9
]
S
.
P
o
o
r
n
i
ma
a
n
d
M
.
P
u
s
h
p
a
l
a
t
h
a
,
“
P
r
e
d
i
c
t
i
o
n
o
f
r
a
i
n
f
a
l
l
u
s
i
n
g
i
n
t
e
n
si
f
i
e
d
LSTM
-
b
a
s
e
d
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
w
i
t
h
w
e
i
g
h
t
e
d
l
i
n
e
a
r
u
n
i
t
s,
”
A
t
m
o
s
p
h
e
re
,
v
o
l
.
1
0
,
n
o
.
1
1
,
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/
a
t
m
o
s1
0
1
1
0
6
6
8
.
[
1
0
]
I
.
S
a
l
e
h
i
n
,
I
.
M
.
T
a
l
h
a
,
M
.
M
.
H
a
s
a
n
,
S
.
T
.
D
i
p
,
M
.
S
a
i
f
u
z
z
a
m
a
n
,
a
n
d
N
.
N
.
M
o
o
n
,
“
A
n
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
b
a
s
e
d
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
LS
T
M
a
n
d
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
2
0
I
EE
E
I
n
t
e
r
n
a
t
i
o
n
a
l
W
o
m
e
n
i
n
E
n
g
i
n
e
e
r
i
n
g
(
W
I
E
)
C
o
n
f
e
r
e
n
c
e
o
n
E
l
e
c
t
r
i
c
a
l
a
n
d
C
o
m
p
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
WI
E
C
O
N
-
E
C
E
2
0
2
0
,
2
0
2
0
,
p
p
.
5
–
8
,
d
o
i
:
1
0
.
1
1
0
9
/
W
I
E
C
O
N
-
E
C
E
5
2
1
3
8
.
2
0
2
0
.
9
3
9
8
0
2
2
.
[
1
1
]
H
.
A
.
-
K
a
d
e
r
,
M
.
A
.
-
E
l
sa
l
a
m,
a
n
d
M
.
M
o
h
a
me
d
,
“
H
y
b
r
i
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
r
a
i
n
f
a
l
l
f
o
r
e
c
a
st
i
n
g
,
”
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
s
t
e
m
s
a
n
d
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
,
v
o
l
.
1
,
n
o
.
1
,
p
p
.
5
–
1
2
,
2
0
2
0
,
d
o
i
:
1
0
.
5
4
2
1
6
/
j
i
si
o
t
.
0
1
0
1
0
1
.
[
1
2
]
M
.
C
h
h
e
t
r
i
,
S
.
K
u
m
a
r
,
P
.
P
.
R
o
y
,
a
n
d
B
.
G
.
K
i
m
,
“
D
e
e
p
B
LST
M
-
G
R
U
mo
d
e
l
f
o
r
mo
n
t
h
l
y
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
:
a
c
a
s
e
s
t
u
d
y
o
f
S
i
mt
o
k
h
a
,
B
h
u
t
a
n
,
”
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
1
2
,
n
o
.
1
9
,
p
p
.
1
–
1
3
,
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
r
s
1
2
1
9
3
1
7
4
.
[
1
3
]
K
.
Jo
h
n
y
,
M
.
L
.
P
a
i
,
a
n
d
S
.
A
d
a
r
s
h
,
“
A
d
a
p
t
i
v
e
EE
M
D
-
A
N
N
h
y
b
r
i
d
mo
d
e
l
f
o
r
I
n
d
i
a
n
s
u
mm
e
r
m
o
n
s
o
o
n
r
a
i
n
f
a
l
l
f
o
r
e
c
a
st
i
n
g
,
”
T
h
e
o
re
t
i
c
a
l
a
n
d
Ap
p
l
i
e
d
C
l
i
m
a
t
o
l
o
g
y
,
v
o
l
.
1
4
1
,
n
o
.
1
–
2
,
p
p
.
1
–
1
7
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s0
0
7
0
4
-
020
-
0
3
1
7
7
-
5.
[
1
4
]
M
.
I
.
K
h
a
n
a
n
d
R
.
M
a
i
t
y
,
“
H
y
b
r
i
d
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
f
o
r
m
u
l
t
i
-
s
t
e
p
-
a
h
e
a
d
d
a
i
l
y
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
G
C
M
si
mu
l
a
t
i
o
n
s
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
8
,
p
p
.
5
2
7
7
4
–
5
2
7
8
4
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
2
9
8
0
9
7
7
.
[
1
5
]
D
.
Z.
H
a
q
e
t
a
l
.
,
“
L
o
n
g
s
h
o
r
t
-
t
e
r
m
m
e
mo
r
y
a
l
g
o
r
i
t
h
m
f
o
r
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
b
a
se
d
o
n
E
l
-
N
i
n
o
a
n
d
I
O
D
d
a
t
a
,
”
Pr
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
7
9
,
p
p
.
8
2
9
–
8
3
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
1
.
0
1
.
0
7
1
.
[
1
6
]
D
.
S
u
n
,
J.
W
u
,
H
.
H
u
a
n
g
,
R
.
W
a
n
g
,
F
.
L
i
a
n
g
,
a
n
d
H
.
X
i
n
h
u
a
,
“
P
r
e
d
i
c
t
i
o
n
o
f
s
h
o
r
t
-
t
i
m
e
r
a
i
n
f
a
l
l
b
a
s
e
d
o
n
d
e
e
p
l
e
a
r
n
i
n
g
,
”
Ma
t
h
e
m
a
t
i
c
a
l
Pr
o
b
l
e
m
s i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
0
2
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
6
6
6
4
4
1
3
.
[
1
7
]
P
.
K
a
n
c
h
a
n
,
“
R
a
i
n
f
a
l
l
a
n
a
l
y
s
i
s
a
n
d
f
o
r
e
c
a
s
t
i
n
g
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
,
”
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
c
s
El
e
c
t
ri
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
s
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
,
n
o
.
2
,
p
p
.
1
–
1
1
,
2
0
2
1
,
d
o
i
:
1
0
.
5
4
0
6
0
/
j
i
e
e
e
/
0
0
2
.
0
2
.
0
1
5
.
[
1
8
]
N
.
T
h
a
k
u
r
a
n
d
S
.
K
a
r
ma
k
a
r
,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
u
si
n
g
l
o
n
g
s
h
o
r
t
t
e
r
m
me
mo
r
y
t
e
c
h
n
i
q
u
e
f
o
r
m
o
n
t
h
l
y
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
i
n
C
h
h
a
t
t
i
sg
a
r
h
,
I
n
d
i
a
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
S
c
i
e
n
t
i
f
i
c
R
e
se
a
rc
h
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
8
–
1
3
,
2
0
2
1
,
d
o
i
:
1
0
.
2
6
4
3
8
/
i
j
sr
c
se
/
v
9
i
1
.
8
1
3
.
[
1
9
]
U
.
B
h
i
ma
v
a
r
a
p
u
,
“
I
R
F
-
LSTM
:
e
n
h
a
n
c
e
d
r
e
g
u
l
a
r
i
z
a
t
i
o
n
f
u
n
c
t
i
o
n
i
n
LS
TM
t
o
p
r
e
d
i
c
t
t
h
e
r
a
i
n
f
a
l
l
,
”
N
e
u
r
a
l
C
o
m
p
u
t
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
4
,
n
o
.
2
2
,
p
p
.
2
0
1
6
5
–
2
0
1
7
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
5
2
1
-
0
2
2
-
0
7
5
7
7
-
8.
[
2
0
]
Y
.
P
e
n
g
,
D
.
G
o
n
g
,
C
.
D
e
n
g
,
H
.
L
i
,
H
.
C
a
i
,
a
n
d
H
.
Z
h
a
n
g
,
“
A
n
a
u
t
o
mat
i
c
h
y
p
e
r
p
a
r
a
me
t
e
r
o
p
t
i
mi
z
a
t
i
o
n
D
N
N
mo
d
e
l
f
o
r
p
r
e
c
i
p
i
t
a
t
i
o
n
p
r
e
d
i
c
t
i
o
n
,
”
A
p
p
l
i
e
d
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
5
2
,
n
o
.
3
,
p
p
.
2
7
0
3
–
2
7
1
9
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
4
8
9
-
0
2
1
-
0
2
5
0
7
-
y.
[
2
1
]
N
.
Th
a
k
u
r
,
S
.
K
a
r
ma
k
a
r
,
a
n
d
S
.
S
o
n
i
,
“
Ti
me
s
e
r
i
e
s
f
o
r
e
c
a
st
i
n
g
f
o
r
u
n
i
-
v
a
r
i
a
n
t
d
a
t
a
u
si
n
g
h
y
b
r
i
d
G
A
-
O
LSTM
m
o
d
e
l
a
n
d
p
e
r
f
o
r
m
a
n
c
e
e
v
a
l
u
a
t
i
o
n
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
4
,
n
o
.
4
,
p
p
.
1
9
6
1
–
1
9
6
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
1
8
7
0
-
022
-
0
0
9
1
4
-
z.
[
2
2
]
X
.
Z
h
a
n
g
,
K
.
W
a
n
g
,
a
n
d
Z
.
Z
h
e
n
g
,
“
A
n
o
v
e
l
i
n
t
e
g
r
a
t
e
d
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
C
EE
M
D
-
F
C
M
S
E
-
st
a
c
k
i
n
g
,
”
E
a
rt
h
S
c
i
e
n
c
e
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
5
,
n
o
.
3
,
p
p
.
1
9
9
5
–
2
0
0
5
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
2
1
4
5
-
0
2
2
-
0
0
8
1
9
-
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
4
1
-
3
4
5
1
3450
[
2
3
]
G
.
S
.
J
e
b
a
,
P
.
C
h
i
t
r
a
,
a
n
d
U
.
M
.
R
a
j
a
s
e
k
a
r
a
n
,
“
Ti
m
e
-
seri
e
s
a
n
a
l
y
si
s
a
n
d
f
l
o
o
d
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
a
d
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
i
n
2
0
2
2
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
W
i
re
l
e
ss
C
o
m
m
u
n
i
c
a
t
i
o
n
s,
S
i
g
n
a
l
Pr
o
c
e
ss
i
n
g
a
n
d
N
e
t
w
o
r
k
i
n
g
,
W
i
S
PN
ET
2
0
2
2
,
2
0
2
2
,
p
p
.
1
3
9
–
142
,
d
o
i
:
1
0
.
1
1
0
9
/
W
i
S
P
N
ET
5
4
2
4
1
.
2
0
2
2
.
9
7
6
7
1
0
2
.
[
2
4
]
D
.
En
d
a
l
i
e
,
G
.
H
a
i
l
e
,
a
n
d
W
.
Ta
y
e
,
“
D
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
d
a
i
l
y
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
:
c
a
s
e
st
u
d
y
o
f
Ji
mm
a
,
Et
h
i
o
p
i
a
,
”
W
a
t
e
r
S
u
p
p
l
y
,
v
o
l
.
2
2
,
n
o
.
3
,
p
p
.
3
4
4
8
–
3
4
6
1
,
2
0
2
2
,
d
o
i
:
1
0
.
2
1
6
6
/
W
S
.
2
0
2
1
.
3
9
1
.
[
2
5
]
X
.
L
i
u
,
X
.
S
a
n
g
,
J.
C
h
a
n
g
,
Y
.
Z
h
e
n
g
,
a
n
d
Y
.
H
a
n
,
“
R
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
o
p
t
i
mi
z
a
t
i
o
n
m
o
d
e
l
i
n
t
e
n
-
d
a
y
t
i
me
st
e
p
b
a
s
e
d
o
n
s
l
i
d
i
n
g
w
i
n
d
o
w
me
c
h
a
n
i
sm
a
n
d
z
e
r
o
-
su
m
g
a
me,
”
A
q
u
a
W
a
t
e
r
I
n
f
r
a
st
r
u
c
t
u
re,
E
c
o
sys
t
e
m
s
a
n
d
S
o
c
i
e
t
y
,
v
o
l
.
7
1
,
n
o
.
1
,
p
p
.
1
–
1
8
,
2
0
2
2
,
d
o
i
:
1
0
.
2
1
6
6
/
a
q
u
a
.
2
0
2
1
.
0
8
6
.
[
2
6
]
J.
G
u
,
S
.
L
i
u
,
Z.
Z
h
o
u
,
S
.
R
.
C
h
a
l
o
v
,
a
n
d
Q
.
Z
h
u
a
n
g
,
“
A
s
t
a
c
k
i
n
g
e
n
sem
b
l
e
l
e
a
r
n
i
n
g
m
o
d
e
l
f
o
r
m
o
n
t
h
l
y
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
i
n
t
h
e
Ta
i
h
u
B
a
si
n
,
C
h
i
n
a
,
”
W
a
t
e
r
,
v
o
l
.
1
4
,
n
o
.
3
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
w
1
4
0
3
0
4
9
2
.
[
2
7
]
V
.
C
.
M
o
u
l
i
,
P
.
C
h
i
t
r
a
,
M
.
H
.
S
u
b
r
a
m
a
n
i
a
n
,
a
n
d
S
.
A
b
i
r
a
mi
,
“
A
d
e
e
p
l
e
a
r
n
i
n
g
e
n
sem
b
l
e
m
o
d
e
l
f
o
r
s
h
o
r
t
-
t
e
r
m
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
,
”
i
n
2
0
2
2
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Wi
r
e
l
e
ss
C
o
m
m
u
n
i
c
a
t
i
o
n
s,
S
i
g
n
a
l
Pro
c
e
ss
i
n
g
a
n
d
N
e
t
w
o
rk
i
n
g
,
W
i
S
P
N
ET
2
0
2
2
,
2
0
2
2
,
p
p
.
1
3
5
–
138
,
d
o
i
:
1
0
.
1
1
0
9
/
W
i
S
P
N
ET
5
4
2
4
1
.
2
0
2
2
.
9
7
6
7
1
6
3
.
[
2
8
]
C
.
C
h
e
n
e
t
a
l
.
,
“
F
o
r
e
c
a
st
o
f
r
a
i
n
f
a
l
l
d
i
st
r
i
b
u
t
i
o
n
b
a
se
d
o
n
f
i
x
e
d
s
l
i
d
i
n
g
w
i
n
d
o
w
l
o
n
g
s
h
o
r
t
-
t
e
r
m
mem
o
r
y
,
”
E
n
g
i
n
e
e
ri
n
g
Ap
p
l
i
c
a
t
i
o
n
s
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
F
l
u
i
d
Me
c
h
a
n
i
c
s
,
v
o
l
.
1
6
,
n
o
.
1
,
p
p
.
2
4
8
–
2
6
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
8
0
/
1
9
9
4
2
0
6
0
.
2
0
2
1
.
2
0
0
9
3
7
4
.
[
2
9
]
S
.
C
h
e
n
,
X
.
X
u
,
Y
.
Zh
a
n
g
,
D
.
S
h
a
o
,
S
.
Zh
a
n
g
,
a
n
d
M
.
Ze
n
g
,
“
Tw
o
-
st
r
e
a
m
c
o
n
v
o
l
u
t
i
o
n
a
l
LSTM
f
o
r
p
r
e
c
i
p
i
t
a
t
i
o
n
n
o
w
c
a
st
i
n
g
,
”
N
e
u
ra
l
C
o
m
p
u
t
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
4
,
n
o
.
1
6
,
p
p
.
1
3
2
8
1
–
1
3
2
9
0
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s0
0
5
2
1
-
0
2
1
-
0
6
8
7
7
-
9.
[
3
0
]
M
.
B
i
l
l
a
h
,
M
.
N
.
A
d
n
a
n
,
M
.
R
.
A
k
h
o
n
d
,
R
.
R
.
Em
a
,
M
.
A
.
H
o
ssa
i
n
,
a
n
d
S
.
M
d
.
G
a
l
i
b
,
“
R
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
s
y
s
t
e
m
f
o
r
B
a
n
g
l
a
d
e
s
h
u
si
n
g
l
o
n
g
sh
o
r
t
-
t
e
r
m m
e
m
o
r
y
,
”
O
p
e
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
2
,
n
o
.
1
,
p
p
.
3
2
3
–
3
3
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
5
1
5
/
c
o
m
p
-
2
0
2
2
-
0
2
5
4
.
[
3
1
]
B
.
C
.
M
a
j
a
n
g
,
N
.
Za
i
n
i
,
a
n
d
L
.
M
a
z
a
l
a
n
,
“
R
a
i
n
f
a
l
l
n
o
w
c
a
s
t
i
n
g
b
a
se
d
o
n
sat
e
l
l
i
t
e
i
m
a
g
e
s
u
s
i
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
l
o
n
g
-
s
h
o
r
t
t
e
r
m
memo
r
y
,
”
i
n
1
2
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
y
s
t
e
m
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
I
C
S
ET
2
0
2
2
,
2
0
2
2
,
p
p
.
6
7
–
71
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
ET
5
7
5
4
3
.
2
0
2
2
.
1
0
0
1
0
8
0
6
.
[
3
2
]
S
h
w
e
t
a
,
H
.
S
e
h
r
a
w
a
t
,
a
n
d
V
.
S
i
w
a
c
h
,
“
M
o
n
s
o
o
n
a
l
r
a
i
n
f
a
l
l
f
o
r
e
c
a
st
i
n
g
u
s
i
n
g
LST
M
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
2
2
1
0
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
R
e
l
i
a
b
i
l
i
t
y
,
I
n
f
o
c
o
m
T
e
c
h
n
o
l
o
g
i
e
s
a
n
d
O
p
t
i
m
i
z
a
t
i
o
n
(
T
r
e
n
d
s
a
n
d
F
u
t
u
r
e
D
i
re
c
t
i
o
n
s)
,
I
C
RI
T
O
2
0
2
2
,
2
0
2
2
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
R
I
TO
5
6
2
8
6
.
2
0
2
2
.
9
9
6
4
8
1
6
.
[
3
3
]
S
.
G
u
o
,
Y
.
W
e
n
,
X
.
Z
h
a
n
g
,
G
.
Z
h
u
,
a
n
d
J.
H
u
a
n
g
,
“
R
e
sea
r
c
h
o
n
p
r
e
c
i
p
i
t
a
t
i
o
n
p
r
e
d
i
c
t
i
o
n
b
a
se
d
o
n
a
c
o
mp
l
e
t
e
e
n
semb
l
e
e
m
p
i
r
i
c
a
l
mo
d
e
d
e
c
o
m
p
o
si
t
i
o
n
w
i
t
h
a
d
a
p
t
i
v
e
n
o
i
se
–
l
o
n
g
s
h
o
r
t
-
t
e
r
m me
m
o
r
y
c
o
u
p
l
e
d
mo
d
e
l
,
”
Wa
t
e
r
S
u
p
p
l
y
,
v
o
l
.
2
2
,
n
o
.
1
2
,
p
p
.
9
0
6
1
–
9
0
6
2
,
2
0
2
2
,
d
o
i
:
1
0
.
2
1
6
6
/
w
s.
2
0
2
2
.
4
1
2
.
[
3
4
]
G
.
F
o
l
i
n
o
,
M
.
G
u
a
r
a
s
c
i
o
,
a
n
d
F
.
C
h
i
a
r
a
v
a
l
l
o
t
i
,
“
Le
a
r
n
i
n
g
e
n
se
mb
l
e
s
o
f
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
s fo
r
e
x
t
r
e
me
r
a
i
n
f
a
l
l
e
v
e
n
t
d
e
t
e
c
t
i
o
n
,
”
N
e
u
ra
l
C
o
m
p
u
t
i
n
g
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
3
5
,
n
o
.
1
4
,
p
p
.
1
0
3
4
7
–
1
0
3
6
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s0
0
5
2
1
-
0
2
3
-
0
8
2
3
8
-
0.
[
3
5
]
E.
K
u
r
n
i
a
w
a
n
,
F
.
N
h
i
t
a
,
A
.
A
d
i
t
sa
n
i
a
,
a
n
d
D
.
S
a
e
p
u
d
i
n
,
“
C
5
.
0
a
l
g
o
r
i
t
h
m
a
n
d
sy
n
t
h
e
t
i
c
mi
n
o
r
i
t
y
o
v
e
r
sa
mp
l
i
n
g
t
e
c
h
n
i
q
u
e
(
S
M
O
TE)
f
o
r
r
a
i
n
f
a
l
l
f
o
r
e
c
a
s
t
i
n
g
i
n
b
a
n
d
u
n
g
r
e
g
e
n
c
y
,
”
2
0
1
9
7
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
I
C
o
I
C
T
2
0
1
9
.
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
o
I
C
T.
2
0
1
9
.
8
8
3
5
3
2
4
.
[
3
6
]
A
.
D
.
M
e
h
r
,
V
.
N
o
u
r
a
n
i
,
V
.
K
.
K
h
o
sr
o
w
s
h
a
h
i
,
a
n
d
M
.
A
.
G
h
o
r
b
a
n
i
,
“
A
h
y
b
r
i
d
s
u
p
p
o
r
t
v
e
c
t
o
r
r
e
g
r
e
ssi
o
n
–
f
i
r
e
f
l
y
mo
d
e
l
f
o
r
mo
n
t
h
l
y
r
a
i
n
f
a
l
l
f
o
r
e
c
a
st
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
En
v
i
r
o
n
m
e
n
t
a
l
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
6
,
n
o
.
1
,
p
p
.
3
3
5
–
3
4
6
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
7
6
2
-
018
-
1
6
7
4
-
2.
[
3
7
]
P
.
C
.
S
.
R
e
d
d
y
a
n
d
A
.
S
u
r
e
s
h
b
a
b
u
,
“
A
n
a
d
a
p
t
i
v
e
m
o
d
e
l
f
o
r
f
o
r
e
c
a
st
i
n
g
se
a
so
n
a
l
r
a
i
n
f
a
l
l
u
s
i
n
g
p
r
e
d
i
c
t
i
v
e
a
n
a
l
y
t
i
c
s,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
E
n
g
i
n
e
e
r
i
n
g
a
n
d
S
y
s
t
e
m
s
,
v
o
l
.
1
2
,
n
o
.
5
,
p
p
.
2
2
–
3
2
,
2
0
1
9
,
d
o
i
:
1
0
.
2
2
2
6
6
/
i
j
i
e
s2
0
1
9
.
1
0
3
1
.
0
3
.
[
3
8
]
M
.
T
.
A
n
w
a
r
,
S
.
N
u
g
r
o
h
a
d
i
,
V
.
Ta
n
t
r
i
y
a
t
i
,
a
n
d
V
.
A
.
W
i
n
d
a
r
n
i
,
“
R
a
i
n
p
r
e
d
i
c
t
i
o
n
u
si
n
g
r
u
l
e
-
b
a
se
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
,
”
Ad
v
a
n
c
e
S
u
s
t
a
i
n
a
b
l
e
S
c
i
e
n
c
e
,
E
n
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
,
n
o
.
1
,
p
p
.
1
–
6
,
2
0
2
0
,
d
o
i
:
1
0
.
2
6
8
7
7
/
a
sse
t
.
v
2
i
1
.
6
0
1
9
.
[
3
9
]
S
.
M
.
G
o
w
t
h
a
m
,
Y
.
S
.
G
a
n
e
s
h
,
a
n
d
M
.
M
.
A
l
i
,
“
Ef
f
i
c
i
e
n
t
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
a
n
d
a
n
a
l
y
s
i
s
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,”
T
u
rk
i
sh
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
a
n
d
M
a
t
h
e
m
a
t
i
c
s
Ed
u
c
a
t
i
o
n
,
v
o
l
.
1
2
,
n
o
.
6
,
p
p
.
3
4
6
7
–
3
4
7
4
,
2
0
2
1
.
[
4
0
]
A
.
U
.
A
z
m
i
,
A
.
F
.
H
a
d
i
,
D
.
A
n
g
g
r
a
e
n
i
,
a
n
d
A
.
R
i
s
k
i
,
“
N
a
i
v
e
B
a
y
e
s
me
t
h
o
d
s
f
o
r
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
c
l
a
ssi
f
i
c
a
t
i
o
n
i
n
B
a
n
y
u
w
a
n
g
i
,
”
i
n
J
o
u
r
n
a
l
o
f
P
h
y
s
i
c
s:
C
o
n
f
e
r
e
n
c
e
S
e
r
i
e
s
,
2
0
2
1
,
v
o
l
.
1
8
7
2
,
n
o
.
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
4
2
-
6
5
9
6
/
1
8
7
2
/
1
/
0
1
2
0
2
8
.
[
4
1
]
M
.
T.
A
n
w
a
r
,
E
.
W
i
n
a
r
n
o
,
W
.
H
a
d
i
k
u
r
n
i
a
w
a
t
i
,
a
n
d
M
.
N
o
v
i
t
a
,
“
R
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
u
si
n
g
e
x
t
r
e
me
g
r
a
d
i
e
n
t
b
o
o
s
t
i
n
g
,
”
i
n
J
o
u
r
n
a
l
o
f
Ph
y
si
c
s:
C
o
n
f
e
re
n
c
e
S
e
r
i
e
s
,
2
0
2
1
,
v
o
l
.
1
8
6
9
,
n
o
.
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
4
2
-
6
5
9
6
/
1
8
6
9
/
1
/
0
1
2
0
7
8
.
[
4
2
]
C
.
M
.
Li
y
e
w
a
n
d
H
.
A
.
M
e
l
e
s
e
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
t
o
p
r
e
d
i
c
t
d
a
i
l
y
r
a
i
n
f
a
l
l
a
mo
u
n
t
,
”
J
o
u
r
n
a
l
o
f
B
i
g
D
a
t
a
,
v
o
l
.
8
,
n
o
.
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s4
0
5
3
7
-
0
2
1
-
0
0
5
4
5
-
4.
[
4
3
]
A
.
M
a
h
a
d
w
a
r
e
,
A
.
S
a
i
g
i
r
i
d
h
a
r
i
,
A
.
M
i
s
h
r
a
,
A
.
T
u
p
e
,
a
n
d
N
.
M
a
r
a
t
h
e
,
“
R
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
u
si
n
g
d
i
f
f
e
r
e
n
t
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
ms,”
i
n
2
0
2
2
2
n
d
As
i
a
n
C
o
n
f
e
r
e
n
c
e
o
n
I
n
n
o
v
a
t
i
o
n
i
n
T
e
c
h
n
o
l
o
g
y
,
AS
I
A
N
C
O
N
2
0
2
2
,
2
0
2
2
,
p
p
.
1
–
8
.
d
o
i
:
1
0
.
1
1
0
9
/
A
S
I
A
N
C
O
N
5
5
3
1
4
.
2
0
2
2
.
9
9
0
8
8
5
7
.
[
4
4
]
M
.
R
a
v
a
l
,
P
.
S
i
v
a
s
h
a
n
mu
g
a
m,
V
.
P
h
a
m,
H
.
G
o
h
e
l
,
A
.
K
a
u
sh
i
k
,
a
n
d
Y
.
W
a
n
,
“
A
u
t
o
m
a
t
e
d
p
r
e
d
i
c
t
i
v
e
a
n
a
l
y
t
i
c
s
t
o
o
l
f
o
r
r
a
i
n
f
a
l
l
f
o
r
e
c
a
st
i
n
g
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
r
t
s
,
v
o
l
.
1
1
,
n
o
.
1
,
p
p
.
1
–
1
3
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
0
2
1
-
9
5
7
3
5
-
8.
[
4
5
]
K
.
R
a
j
a
n
d
T.
D
.
A
m
i
n
,
“
R
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,”
Re
se
a
rc
h
S
q
u
a
re
,
M
a
y
1
1
,
2
0
2
3
,
d
o
i
:
1
0
.
2
1
2
0
3
/
r
s
.
3
.
r
s
-
2
7
9
0
9
6
7
/
v
1
.
[
4
6
]
A
.
M
.
S
.
V
i
g
i
l
,
R
.
S
h
w
e
t
h
a
,
S
.
S
i
n
g
h
,
a
n
d
R
.
S
h
r
u
t
h
i
,
“
E
v
a
l
u
a
t
i
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
mo
d
e
l
s
t
o
e
n
h
a
n
c
e
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
,
”
Re
v
i
st
a
E
l
e
c
t
r
o
n
i
c
a
d
e
Ve
t
e
r
i
n
a
r
i
a
,
v
o
l
.
2
5
,
n
o
.
1
,
p
p
.
6
4
6
–
6
5
2
,
2
0
2
4
,
d
o
i
:
1
0
.
6
9
9
8
0
/
r
e
d
v
e
t
.
v
2
5
i
1
s
.
8
0
5
.
[
4
7
]
L
.
C
.
P
.
V
e
l
a
s
c
o
,
R
.
P
.
S
e
r
q
u
i
ñ
a
,
M
.
S
.
A
.
A
.
Z
a
m
a
d
,
B
.
F
.
J
u
a
n
i
c
o
,
a
n
d
J
.
C
.
L
o
m
o
c
s
o
,
“
W
e
e
k
-
a
h
e
a
d
r
a
i
n
f
a
l
l
f
o
r
e
c
a
s
t
i
n
g
u
s
i
n
g
m
u
l
t
i
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
P
r
o
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
6
1
,
p
p
.
3
8
6
–
3
9
7
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
1
9
.
1
1
.
1
3
7
.
[
4
8
]
M
.
H
.
Y
e
n
,
D
.
W
.
L
i
u
,
Y
.
C
.
H
s
i
n
,
C
.
E.
L
i
n
,
a
n
d
C
.
C
.
C
h
e
n
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
t
h
e
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
t
h
e
p
r
e
d
i
c
t
i
o
n
o
f
r
a
i
n
f
a
l
l
i
n
S
o
u
t
h
e
r
n
Ta
i
w
a
n
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
r
t
s
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
1
–
9
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
3
8
/
s4
1
5
9
8
-
019
-
4
9
2
4
2
-
6.
[
4
9
]
B
.
T
.
P
h
a
m
e
t
a
l
.
,
“
D
e
v
e
l
o
p
me
n
t
o
f
a
d
v
a
n
c
e
d
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
m
o
d
e
l
s
f
o
r
d
a
i
l
y
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
,
”
At
m
o
s
p
h
e
ri
c
R
e
se
a
rc
h
,
v
o
l
.
2
3
7
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
t
mo
sr
e
s.2
0
2
0
.
1
0
4
8
4
5
.
[
5
0
]
A
.
S
a
m
a
d
,
B
h
a
g
y
a
n
i
d
h
i
,
V
.
G
a
u
t
a
m,
P
.
Ja
i
n
,
S
a
n
g
e
e
t
a
,
a
n
d
K
.
S
a
r
k
a
r
,
“
A
n
a
p
p
r
o
a
c
h
f
o
r
r
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
l
o
n
g
s
h
o
r
t
t
e
r
m
memo
r
y
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
2
0
I
EEE
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
i
n
g
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
A
u
t
o
m
a
t
i
o
n
,
I
C
C
C
A
2
0
2
0
,
2
0
2
0
,
p
p
.
1
9
0
–
1
9
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
C
A
4
9
5
4
1
.
2
0
2
0
.
9
2
5
0
8
0
9
.
[
5
1
]
W
.
M
.
R
i
d
w
a
n
,
M
.
S
a
p
i
t
a
n
g
,
A
.
A
z
i
z
,
K
.
F
.
K
u
s
h
i
a
r
,
A
.
N
.
A
h
m
e
d
,
a
n
d
A
.
El
-
S
h
a
f
i
e
,
“
R
a
i
n
f
a
l
l
f
o
r
e
c
a
s
t
i
n
g
m
o
d
e
l
u
si
n
g
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
me
t
h
o
d
s:
c
a
se
s
t
u
d
y
T
e
r
e
n
g
g
a
n
u
,
M
a
l
a
y
s
i
a
,
”
A
i
n
S
h
a
m
s
En
g
i
n
e
e
ri
n
g
J
o
u
r
n
a
l
,
v
o
l
.
1
2
,
n
o
.
2
,
p
p
.
1
6
5
1
–
1
6
6
3
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
se
j
.
2
0
2
0
.
0
9
.
0
1
1
.
[
5
2
]
M
.
La
k
s
h
mi
t
h
a
,
G
.
A
n
i
s
h
a
,
S
.
R
a
j
e
s
h
,
a
n
d
B
.
V
e
n
k
a
t
e
sa
n
,
“
R
a
i
n
f
a
l
l
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
Re
se
a
rc
h
J
o
u
r
n
a
l
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
8
,
n
o
.
1
0
,
p
p
.
1
1
7
9
–
1
1
8
2
,
2
0
2
1
.
[
5
3
]
J.
A
.
A
n
o
c
h
i
,
V
.
A
.
D.
A
l
mei
d
a
,
a
n
d
H
.
F
.
D.
C
.
V
e
l
h
o
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
f
o
r
c
l
i
ma
t
e
p
r
e
c
i
p
i
t
a
t
i
o
n
p
r
e
d
i
c
t
i
o
n
m
o
d
e
l
i
n
g
o
v
e
r
S
o
u
t
h
A
meri
c
a
,
”
Re
m
o
t
e
S
e
n
s
i
n
g
,
v
o
l
.
1
3
,
n
o
.
1
3
,
p
p
.
1
–
1
8
,
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
r
s
1
3
1
3
2
4
6
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.