T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
3
,
J
u
n
e
2021
,
p
p
.
8
1
7
~8
2
8
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Ke
m
e
n
r
is
te
k
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
KA
.
v
1
9
i3
.
1
6
9
0
7
817
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Para
m
e
ter
tuning
of so
ft
w
a
re e
ff
o
rt
esti
m
a
tion
m
o
del
s usin
g
a
ntlion
o
pti
m
i
z
a
ti
o
n
M
a
rr
w
a
Abd
-
AlK
a
re
e
m
Ala
ba
j
ee
,
Na
j
la
Ak
ra
m
Al
S
a
a
t
i
,
T
a
g
hree
d Riy
a
dh
Alre
f
f
a
ee
De
p
a
rtme
n
t
o
f
S
o
f
twa
re
,
Co
ll
e
g
e
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
M
a
t
h
e
m
a
ti
c
s,
Un
iv
e
rsit
y
o
f
M
o
su
l,
Ira
q
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
n
6
,
2020
R
ev
i
s
ed
Oct
5
,
2
0
2
0
A
cc
ep
ted
Oct
1
5
,
2
0
2
0
In
th
is
w
o
rk
,
th
e
a
n
tl
io
n
o
p
t
i
m
iz
a
ti
o
n
(AL
O)
is
e
m
p
lo
y
e
d
d
u
e
to
it
s
e
ff
ici
e
n
c
y
a
n
d
w
id
e
a
p
p
li
c
a
b
i
li
ty
to
e
stim
a
te
th
e
p
a
ra
m
e
ter
s
o
f
f
o
u
r
m
o
d
if
ied
m
o
d
e
ls
o
f
th
e
b
a
sic
c
o
n
stru
c
ti
v
e
c
o
st
m
o
d
e
l
(
COCO
M
O
)
m
o
d
e
l.
T
h
re
e
tes
ts
a
re
c
a
rried
o
u
t
t
o
sh
o
w
th
e
e
f
f
e
c
t
iv
e
n
e
ss
o
f
AL
O:
f
irst,
it
is
u
se
d
w
it
h
Ba
il
e
y
a
n
d
Ba
sili
d
a
tas
e
t
f
o
r
th
e
b
a
sic
COCO
M
O
M
o
d
e
l
a
n
d
S
h
e
ta’s
M
o
d
e
l
1
a
n
d
2
,
a
n
d
is
c
o
m
p
a
re
d
w
it
h
th
e
f
ire
f
l
y
a
lg
o
rit
h
m
(F
A
),
g
e
n
e
ti
c
a
lg
o
rit
h
m
s
(
GA
),
a
n
d
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
iza
ti
o
n
(
P
S
O).
S
e
c
o
n
d
,
p
a
ra
m
e
ters
o
f
S
h
e
ta’s
M
o
d
e
l
1
a
n
d
2
,
Uy
s
a
l’s
M
o
d
e
l
1
a
n
d
2
a
re
o
p
ti
m
ize
d
u
sin
g
Ba
il
e
y
a
n
d
Ba
sili
d
a
tas
e
t;
re
su
lt
s
a
re
c
o
m
p
a
re
d
w
it
h
d
irec
ted
a
rti
f
icia
l
b
e
e
c
o
lo
n
y
a
lg
o
ri
th
m
(DA
BC
A),
GA
,
a
n
d
sim
u
late
d
a
n
n
e
a
li
n
g
(S
A
).
T
h
ird
,
A
L
O
is
u
se
d
w
it
h
Ba
sic
COCO
M
O
m
o
d
e
l
a
n
d
f
o
u
r
larg
e
d
a
ta
se
ts,
re
su
lt
s
a
r
e
c
o
m
p
a
re
d
w
it
h
h
y
b
rid
b
a
t
in
sp
ired
g
ra
v
it
a
ti
o
n
a
l
se
a
rc
h
a
l
g
o
rit
h
m
(h
B
A
TG
S
A
),
im
p
ro
v
e
d
BAT
(IB
AT
),
a
n
d
BAT
a
lg
o
rit
h
m
s.
Re
su
lt
s
o
f
T
e
st1
a
n
d
T
e
st
2
sh
o
w
th
a
t
AL
O
o
u
tp
e
rf
o
rm
e
d
o
th
e
rs,
a
s
f
o
r
T
e
st3
,
A
LO
is
b
e
tt
e
r
t
h
a
n
BA
T
a
n
d
IBA
T
u
sin
g
M
A
E
a
n
d
th
e
n
u
m
b
e
r
o
f
b
e
st
e
sti
m
a
ti
o
n
s.
A
LO
p
ro
o
f
e
d
a
c
h
iev
in
g
b
e
tt
e
r
re
su
lt
s
th
a
n
h
BA
TG
S
A
f
o
r
d
a
tas
e
t
s
2
a
n
d
4
o
u
t
o
f
th
e
f
o
u
r
d
a
tas
e
ts
e
x
p
lo
re
d
in
term
s
o
f
M
A
E
a
n
d
th
e
n
u
m
b
e
r
o
f
b
e
st es
ti
m
a
tes
.
K
ey
w
o
r
d
s
:
An
tlio
n
o
p
ti
m
izatio
n
alg
o
r
it
h
m
P
ar
am
eter
tu
n
i
n
g
So
f
t
w
ar
e
ef
f
o
r
t e
s
ti
m
atio
n
T
h
e
C
OC
OM
O
m
o
d
el
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
BY
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Ma
r
r
w
a
A
b
d
-
A
lKar
ee
m
A
lab
a
j
ee
Dep
ar
t
m
en
t o
f
So
f
t
w
ar
e
Mo
s
u
l U
n
i
v
er
s
it
y
A
l M
aj
m
o
aa
Stre
et,
Mo
s
u
l,
I
r
aq
E
m
ail:
m
ar
r
w
a_
ze
d
an
@
u
o
m
o
s
u
l.e
d
u
.
iq
1.
I
NT
RO
D
UCT
I
O
N
W
ith
o
v
er
t
h
e
la
s
t
f
e
w
d
ec
ad
e
s
,
s
o
f
t
w
ar
e
s
y
s
te
m
s
h
av
e
s
p
r
ea
d
o
u
ts
ta
n
d
in
g
l
y
a
n
d
th
i
s
s
p
r
ea
d
is
v
er
y
lik
el
y
to
g
o
o
n
m
u
ch
s
w
i
f
ter
in
t
h
e
f
u
tu
r
e.
T
h
is
is
d
u
e
t
o
th
e
r
ap
id
g
r
o
w
t
h
in
s
cie
n
ti
f
ic,
i
n
d
u
s
tr
ial,
a
n
d
co
m
m
er
cial
tech
n
o
lo
g
ie
s
all
ar
o
u
n
d
t
h
e
w
o
r
ld
[
1
]
.
Star
tin
g
a
p
r
o
j
ec
t
w
it
h
a
n
ac
ce
p
tab
l
y
es
ti
m
ated
s
ize
o
r
ef
f
o
r
t
e
n
ab
les
t
h
e
m
a
n
g
er
to
g
ain
co
n
f
id
en
ce
r
eg
ar
d
i
n
g
f
u
tu
r
e
ac
tiv
itie
s
,
th
i
s
is
d
u
e
to
t
h
e
f
ac
t
th
at
m
o
s
t
o
f
t
h
e
d
ec
is
io
n
s
m
ad
e
t
h
r
o
u
g
h
o
u
t
th
e
d
ev
elo
p
m
e
n
t
ar
e
a
f
f
ec
ted
b
y
th
e
in
i
tial
e
s
ti
m
atio
n
s
.
So
,
t
h
e
p
r
o
ce
s
s
o
f
co
s
t
o
r
ef
f
o
r
t
es
ti
m
atio
n
s
is
co
n
s
id
er
ed
as
o
n
e
o
f
th
e
m
o
s
t
d
ec
i
s
iv
e
s
ta
g
es
o
f
p
lan
n
i
n
g
a
n
d
m
an
a
g
i
n
g
s
o
f
t
w
ar
e
p
r
o
j
ec
ts
[
2
]
.
T
h
e
im
p
o
r
tan
ce
o
f
esti
m
ati
n
g
s
o
f
t
w
ar
e
ef
f
o
r
t
d
u
r
in
g
ea
r
l
y
s
ta
g
es
o
f
s
o
f
t
w
ar
e
d
ev
elo
p
m
e
n
t
h
a
s
b
ee
n
p
o
in
ted
o
u
t
s
o
m
an
y
ti
m
es.
B
u
t
th
e
n
ag
ai
n
,
esti
m
ati
n
g
th
e
s
o
f
t
w
ar
e
e
f
f
o
r
t
is
co
n
s
id
e
r
ed
to
b
e
v
er
y
h
ar
d
at
th
e
b
eg
i
n
n
in
g
o
f
a
p
r
o
j
ec
t,
th
is
i
s
b
ec
au
s
e
p
eo
p
le
p
r
o
d
u
ctiv
it
y
d
if
f
er
s
,
s
o
v
er
y
li
ttle
i
s
k
n
o
w
n
ab
o
u
t
t
h
e
s
o
f
t
w
ar
e
b
ein
g
d
ev
elo
p
ed
[
1
]
.
As
th
e
s
o
f
t
w
ar
e
d
ev
elo
p
m
en
t
co
s
t
o
r
ef
f
o
r
t
is
p
r
o
g
r
ess
iv
el
y
r
i
s
in
g
,
s
o
f
t
w
ar
e
h
as
g
r
o
w
n
to
b
e
th
e
m
aj
o
r
co
s
t ta
k
en
b
y
a
s
y
s
te
m
.
Ma
n
y
p
r
o
j
ec
ts
s
u
ch
a
s
Natio
n
al
A
er
o
n
au
t
ics a
n
d
Sp
ac
e
A
d
m
i
n
is
tr
atio
n
(
N
A
S
A
)
o
r
A
ir
Fo
r
ce
esti
m
ate
s
s
o
f
t
w
ar
e
d
ev
elo
p
m
en
t
co
s
t
to
r
ea
ch
5
0
%
o
f
th
e
to
tal
co
s
t
r
eq
u
ir
ed
.
T
h
is
is
d
u
e
t
h
e
h
i
g
h
co
m
p
le
x
it
y
e
n
co
u
n
ter
ed
in
t
h
e
N
A
S
A
s
o
f
t
w
ar
e
p
r
o
j
ec
ts
’
s
y
s
te
m
s
,
e
ith
er
h
ar
d
w
ar
e
o
r
s
o
f
t
war
e
[
3
]
.
A
ll
t
h
is
led
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
19
,
No
.
3
,
J
u
n
e
2
0
2
1
:
8
1
7
-
8
2
8
818
th
e
w
a
y
to
w
ar
d
s
th
e
d
ev
elo
p
m
en
t
o
f
a
m
o
d
el
ca
p
ab
le
o
f
p
r
ed
ictin
g
s
o
f
t
w
ar
e
ef
f
o
r
t,
an
d
alth
o
u
g
h
th
er
e
ar
e
v
ar
io
u
s
ac
c
u
r
ate
m
o
d
els
f
o
r
co
m
p
u
ti
n
g
p
r
o
j
ec
t’
s
e
f
f
o
r
t,
th
er
e
is
s
till
a
s
tr
o
n
g
ch
alle
n
g
e
f
o
r
m
o
r
e
ac
cu
r
ate
an
d
p
r
ec
is
e
m
o
d
els.
Of
t
h
ese
p
o
w
er
f
u
l
m
o
d
els
is
t
h
e
co
n
s
tr
u
cti
v
e
co
s
t
m
o
d
el
(
C
OC
OM
O
)
t
h
at
esti
m
ate
s
th
e
e
f
f
o
r
t
o
f
an
y
p
r
o
je
c
t
d
u
r
in
g
th
e
e
a
r
ly
s
t
ag
e
o
f
th
e
p
r
o
je
c
t
.
T
h
is
c
an
l
ea
d
t
o
m
in
im
i
z
e
t
h
e
o
v
e
r
al
l
c
o
s
t
o
f
th
e
p
r
o
je
c
t
[
1
]
.
T
h
e
p
r
o
ce
s
s
o
f
ef
f
o
r
t
esti
m
ati
o
n
is
af
f
ec
ted
b
y
a
lar
g
e
n
u
m
b
er
o
f
p
ar
am
eter
s
;
t
h
ese
p
ar
am
eter
s
ar
e
esti
m
ated
u
s
i
n
g
d
if
f
er
en
t
tec
h
n
iq
u
es.
T
h
e
ai
m
o
f
th
is
w
o
r
k
is
to
u
s
e
th
e
a
n
tlio
n
o
p
ti
m
izatio
n
alg
o
r
ith
m
,
attr
ib
u
tab
le
to
its
e
f
f
icien
c
y
,
i
n
tu
n
i
n
g
th
e
p
ar
a
m
eter
s
o
f
t
h
e
C
OC
O
MO
m
o
d
el
v
ar
ian
t
s
,
th
e
f
ir
s
t
i
s
th
e
b
as
ic
C
OC
OM
O
m
o
d
el
an
d
th
e
o
t
h
er
f
o
u
r
ar
e
its
ex
te
n
s
io
n
s
.
Af
ter
p
ar
a
m
eter
tu
n
i
n
g
,
t
h
e
o
b
tain
ed
r
esu
lts
ar
e
co
m
p
ar
ed
w
it
h
t
h
e
o
t
h
er
m
et
h
o
d
s
.
Mo
s
t
o
f
t
h
e
w
o
r
k
r
elate
d
to
th
is
f
ield
g
o
es
b
ac
k
to
2
0
1
0
,
w
h
e
n
A
lj
ah
d
ali
an
d
Sh
e
ta
[
3
]
s
u
g
g
e
s
ted
th
e
u
s
e
o
f
d
if
f
er
e
n
tial
e
v
o
lu
t
io
n
(
D
E
)
to
esti
m
ate
t
h
e
C
OC
OM
O
m
o
d
el
p
ar
a
m
eter
s
,
th
ese
m
o
d
els
w
er
e
test
ed
u
s
i
n
g
N
A
S
A
s
o
f
t
w
ar
e
p
r
o
j
ec
t
d
ata
s
et
.
I
n
2
0
1
2
,
Sin
g
h
an
d
M
is
r
a
[
4
]
ex
p
lo
r
ed
th
e
cr
i
s
p
i
n
p
u
t
e
f
f
ec
t
w
it
h
g
en
e
tic
al
g
o
r
ith
m
s
(
GAs),
an
d
ap
p
lied
a
m
o
d
i
f
ied
v
er
s
io
n
o
f
th
e
C
OC
OM
O
m
o
d
el
to
N
A
S
A
d
ata
s
et
.
I
n
t
h
e
s
a
m
e
y
ea
r
,
Ku
n
d
u
a
n
d
Set
h
i
[
5
]
r
ec
o
m
m
e
n
d
ed
th
e
u
s
e
o
f
s
i
m
u
lated
an
n
ea
lin
g
(
S
A
)
to
o
p
ti
m
i
ze
th
e
co
e
f
f
icien
ts
o
f
C
OC
O
MO
I
I
m
o
d
el
ai
m
i
n
g
f
o
r
m
o
r
e
ac
cu
r
ate
e
f
f
o
r
t
esti
m
atio
n
s
.
L
ater
i
n
2
0
1
3
,
Dh
i
m
a
n
an
d
Di
w
a
k
er
[
6
]
also
u
s
ed
G
As
to
o
p
ti
m
ize
th
e
co
ef
f
icie
n
t
s
o
f
C
OC
OM
O
I
I
m
o
d
el
i
n
o
r
d
er
to
ac
q
u
ir
e
ac
cu
r
ate
esti
m
at
io
n
s
.
Gh
atas
h
e
h
et
a
l
.
[
7
]
,
ex
p
lo
r
ed
th
e
u
s
e
o
f
th
e
Fire
f
l
y
al
g
o
r
ith
m
i
n
2
0
1
5
to
o
p
tim
ize
t
h
e
p
ar
a
m
eter
s
o
f
th
r
ee
C
OC
O
MO
b
ased
m
o
d
el
s
.
A
ls
o
,
i
n
2
0
1
5
,
Gu
p
ta
a
n
d
S
h
ar
m
a
[
8
]
s
u
b
m
itted
a
n
e
w
ca
l
ib
r
ated
in
ter
m
ed
iate
C
OC
OM
O
m
o
d
el
d
e
v
elo
p
ed
u
s
i
n
g
t
h
e
b
at
alg
o
r
it
h
m
.
T
h
e
m
o
d
el
g
e
n
er
ated
n
e
w
o
p
ti
m
iz
ed
co
ef
f
icie
n
t
s
,
an
d
r
esu
lt
s
s
h
o
w
ed
th
at
t
h
e
o
p
tim
ized
co
ef
f
icien
ts
g
a
v
e
b
ett
er
r
esu
lts
f
o
r
all
p
r
o
j
ec
t
ty
p
es
in
ter
m
s
o
f
m
ea
n
m
ag
n
it
u
d
e
o
f
r
elati
v
e
er
r
o
r
(
M
MRE
)
w
h
e
n
co
m
p
ar
ed
to
co
ef
f
icie
n
ts
g
ai
n
ed
u
s
i
n
g
r
eg
r
es
s
io
n
an
al
y
s
is
.
Kh
u
at
an
d
L
e
[
9
]
in
2
0
1
6
u
s
ed
ar
tif
icial
b
ee
co
lo
n
y
al
g
o
r
it
h
m
f
o
r
p
ar
am
eter
t
u
n
i
n
g
ac
co
r
d
in
g
to
th
e
ac
tu
al
e
f
f
o
r
t.
T
h
eir
w
o
r
k
w
a
s
v
er
i
f
ied
w
it
h
N
A
S
A
s
o
f
t
war
e
d
ataset
an
d
w
a
s
co
m
p
ar
e
d
to
s
o
m
e
e
x
is
t
in
g
m
o
d
el
s
.
I
n
t
h
e
s
a
m
e
y
ea
r
,
B
a
r
d
s
ir
i
an
d
Do
r
o
s
ti
[
1
0
]
in
cr
ea
s
ed
th
e
ac
c
u
r
ac
y
o
f
C
OC
O
M
O
esti
m
atio
n
b
y
a
h
y
b
r
id
m
o
d
el
t
h
at
co
m
b
i
n
ed
b
ee
co
lo
n
y
al
g
o
r
ith
m
w
it
h
t
h
e
C
OC
OM
O
est
i
m
a
tio
n
m
et
h
o
d
.
T
h
eir
m
et
h
o
d
g
a
v
e
m
o
r
e
ef
f
icie
n
t
co
e
f
f
icie
n
t
c
o
m
p
ar
ati
v
e
to
t
h
e
b
as
ic
C
O
C
OM
O,
f
i
n
d
in
g
b
etter
co
e
f
f
icien
t
s
ca
n
g
r
ea
t
l
y
m
ax
i
m
izes
t
h
e
m
et
h
o
d
’
s
ef
f
ic
ien
c
y
.
Als
o
,
i
n
2
0
1
6
Gir
o
tr
a
an
d
S
h
ar
m
a
[
1
1
]
co
n
s
id
er
ed
co
s
t
d
r
iv
er
an
d
t
h
e
is
s
u
e
o
f
in
ac
c
u
r
ate
an
d
a
m
b
i
g
u
o
u
s
s
elec
tio
n
o
f
v
al
u
es
w
h
ic
h
lead
s
to
in
ac
cu
r
ate
e
f
f
o
r
t
es
ti
m
atio
n
s
,
an
d
t
h
e
y
s
h
o
w
ed
th
a
t
a
s
m
all
c
h
a
n
g
e
i
n
th
e
s
elec
tio
n
o
f
C
O
C
OM
O
c
o
s
t
d
r
iv
er
s
ca
n
ca
u
s
e
s
i
g
n
if
ica
n
t
i
m
p
r
o
v
e
m
e
n
t
s
i
n
m
etr
i
cs
s
u
c
h
as
MM
R
E
.
W
h
e
r
ea
s
in
2
0
1
7
,
A
-
Sr
h
an
,
et
a
l.
[
1
]
estab
lis
h
ed
a
h
y
b
r
id
c
u
ck
o
o
s
ea
r
ch
al
g
o
r
ith
m
an
d
g
e
n
etic
al
g
o
r
ith
m
ca
l
led
(
C
SG
A
)
f
o
r
p
ar
a
m
eter
esti
m
ati
o
n
.
A
N
A
S
A
s
o
f
t
w
ar
e
p
r
o
j
ec
t d
ataset
w
as
u
s
ed
i
n
th
e
ex
p
er
i
m
en
ts
.
R
e
s
u
l
ts
s
h
o
w
th
at
C
SG
A
en
h
a
n
ce
d
th
e
e
f
f
o
r
t e
s
ti
m
at
io
n
ac
cu
r
ac
y
.
I
n
2
0
1
8
Nan
d
al
an
d
San
g
w
a
n
[
1
2
]
in
tr
o
d
u
ce
d
a
h
y
b
r
id
b
a
t
in
s
p
ir
ed
g
r
av
itatio
n
a
l
s
ea
r
c
h
alg
o
r
it
h
m
m
et
h
o
d
ca
lled
(
B
A
T
GS
A
)
to
o
p
ti
m
ize
t
h
e
C
O
C
OM
O
m
o
d
el.
I
n
t
h
e
s
a
m
e
y
ea
r
,
K
h
atib
i
an
d
B
ar
d
s
ir
i
[
1
3
]
s
u
g
g
e
s
ted
a
co
m
b
in
ed
m
o
d
el
f
o
r
ef
f
o
r
t
es
ti
m
a
tio
n
.
T
h
e
m
o
d
el
w
as
b
ased
o
n
p
ar
ticle
s
w
ar
m
o
p
ti
m
izat
io
n
alg
o
r
ith
m
w
i
th
a
li
n
ea
r
r
eg
r
ess
io
n
m
et
h
o
d
to
o
p
ti
m
all
y
d
is
co
v
er
co
ef
f
icie
n
t.
L
ater
i
n
2
0
1
8
Diza
j
an
d
Gh
ar
eh
c
h
o
p
o
g
h
[
1
4
]
i
m
p
r
o
v
e
d
GA
s
w
it
h
b
at
alg
o
r
it
h
m
to
s
tu
d
y
t
h
e
i
n
f
lu
e
n
ce
o
f
q
u
alitat
i
v
e
f
ac
to
r
s
an
d
f
alse
v
ar
iab
les
o
n
t
h
e
to
tal
co
s
t
est
i
m
atio
n
.
T
h
eir
m
o
d
el
w
as
e
x
p
lo
r
ed
an
d
test
ed
u
s
in
g
f
o
u
r
d
atasets
w
it
h
s
e
v
en
cr
iter
ia;
r
esu
lt
s
s
h
o
w
ed
t
h
at
th
e
m
o
d
el
i
m
p
r
o
v
ed
th
e
ac
cu
r
ac
y
o
f
co
s
t e
s
t
i
m
a
tio
n
.
T
o
g
iv
e
a
p
r
ec
is
e
es
ti
m
ated
c
o
s
t
f
o
r
p
r
o
j
ec
t
d
e
v
elo
p
m
e
n
t,
i
n
2
0
1
9
Ven
k
ata
iah
et
a
l
.
[
1
5
]
s
u
g
g
ested
th
e
i
m
p
le
m
e
n
tatio
n
o
f
h
y
b
r
id
m
et
h
o
d
o
lo
g
y
f
o
r
tu
n
in
g
p
ar
a
m
eter
s
o
f
C
OC
OM
O
m
o
d
el.
T
o
ch
ec
k
th
e
ef
f
icien
c
y
o
f
th
e
p
r
ese
n
ted
m
o
d
el,
t
h
e
y
u
s
ed
I
B
MD
P
S,
C
OC
OM
O
N
AS
A
2
an
d
DE
SH
A
R
N
A
I
S
an
d
C
OC
OM
O
8
1
.
A
s
f
o
r
th
e
e
m
p
lo
y
ed
o
p
ti
m
izatio
n
m
et
h
o
d
,
AL
O
al
g
o
r
ith
m
h
as
b
ee
n
s
u
c
ce
s
s
f
u
l
l
y
ap
p
lied
i
n
m
an
y
ar
ea
s
s
u
ch
as
t
u
n
in
g
th
e
p
ar
am
eter
o
f
co
n
tr
o
l
d
ev
ices
in
s
y
s
te
m
s
o
f
g
en
er
ato
r
s
’
ex
ci
tatio
n
f
o
r
th
e
m
o
d
el
o
f
T
A
FM
i
n
2
0
1
8
b
y
Šp
o
lj
ar
ić
an
d
P
av
ić
[
1
6
]
.
I
n
th
is
w
o
r
k
,
t
h
e
AL
O
al
g
o
r
ith
m
is
e
m
p
lo
y
e
d
alo
n
g
w
it
h
h
ea
v
y
co
m
p
ar
is
o
n
s
i
n
o
p
p
o
s
ite
to
r
elate
d
w
o
r
k
ai
m
i
n
g
to
co
v
er
p
o
s
s
ib
le
g
ap
s
;
u
s
i
n
g
f
i
v
e
m
o
d
el
s
i
n
s
tead
o
f
o
n
l
y
o
n
e
o
r
th
r
ee
,
t
h
e
s
a
m
e
also
g
o
es
f
o
r
d
atasets
e
n
g
a
g
ed
i
n
test
i
n
g
an
d
co
m
p
ar
i
s
o
n
s
,
as
f
i
v
e
d
if
f
er
en
t
s
ized
d
atasets
we
r
e
u
s
ed
to
ca
r
r
y
o
u
t i
n
cl
u
s
i
v
e
co
m
p
ar
is
o
n
s
a
m
o
n
g
AL
O
an
d
th
e
o
th
er
m
et
h
o
d
s
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
is
s
ec
tio
n
i
n
cl
u
d
es
p
r
o
b
lem
d
escr
ip
tio
n
o
f
s
o
f
t
w
ar
e
e
f
f
o
r
t
est
i
m
a
tio
n
m
o
d
els
,
t
h
er
e
ar
e
m
a
n
y
m
o
d
el
s
s
u
g
g
e
s
ted
f
o
r
esti
m
ati
n
g
s
o
f
t
w
ar
e
ef
f
o
r
t
,
all
o
f
t
h
e
m
w
er
e
d
er
iv
ed
f
r
o
m
t
h
e
C
O
C
OM
O
Mo
d
el,
in
t
h
is
s
ec
tio
n
,
t
h
e
s
o
f
t
w
ar
e
ef
f
o
r
t
es
t
i
m
atio
n
m
o
d
el
s
t
h
at
w
er
e
u
ti
li
ze
d
in
t
h
i
s
p
ap
er
w
ill
b
e
d
escr
ib
ed
as
w
e
ll
a
s
t
h
e
m
at
h
e
m
a
tical
eq
u
at
io
n
o
f
th
e
m
.
I
n
ad
d
itio
n
,
th
i
s
s
ec
t
io
n
i
n
clu
d
es
a
n
ex
p
la
n
atio
n
o
f
t
h
e
an
tlio
n
o
p
ti
m
iza
tio
n
alg
o
r
ith
m
a
n
d
d
escr
ib
es
th
e
m
et
h
o
d
o
lo
g
y
o
f
t
h
is
al
g
o
r
ith
m
an
d
ex
p
lai
n
s
h
o
w
ca
n
t
h
e
tr
ap
s
o
f
th
e
an
tlio
n
s
af
f
ec
t
t
h
e
r
an
d
o
m
w
al
k
o
f
an
t
an
d
h
o
w
ca
n
an
t
io
n
b
u
ild
in
g
t
h
e
tr
ap
s
to
m
a
k
e
t
h
e
tr
ap
p
ed
an
t
s
lid
in
g
d
o
wn
to
w
ar
d
s
t
h
e
ce
n
ter
o
f
t
h
e
p
its
,
th
i
s
m
ec
h
a
n
i
s
m
m
o
d
eled
b
y
m
at
h
e
m
a
tical
eq
u
atio
n
s
d
escr
i
b
ed
b
elo
w
.
Fi
n
all
y
,
th
e
last
s
u
b
s
ec
ti
o
n
r
ep
r
esen
t
s
t
h
e
m
ec
h
a
n
is
m
o
f
eli
tis
m
of
t
h
i
s
alg
o
r
ith
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
P
a
r
a
mete
r
tu
n
in
g
o
f so
ftw
a
r
e
effo
r
t e
s
tima
tio
n
mo
d
els u
s
in
g
…
(
Ma
r
r
w
a
A
b
d
-
A
lK
a
r
ee
m
A
l
a
b
a
jee
)
819
2
.
1
.
So
f
t
w
a
re
ef
f
o
rt
estim
a
t
i
o
n
m
o
del
s
So
f
t
w
ar
e
co
s
t
esti
m
atio
n
is
t
h
e
p
r
ac
tice
o
f
p
r
ed
ictin
g
th
e
r
eq
u
ir
ed
ef
f
o
r
t
f
o
r
d
ev
elo
p
in
g
a
p
r
o
j
ec
t.
Su
c
h
es
ti
m
atio
n
g
i
v
es
th
e
i
m
p
r
ess
io
n
o
f
s
i
m
p
licit
y
,
b
u
t
i
n
r
ea
lit
y
,
it
is
v
er
y
d
if
f
ic
u
lt
an
d
co
m
p
lex
.
C
o
s
t
s
f
o
r
s
o
f
t
w
ar
e
p
r
o
j
ec
ts
d
ep
en
d
lar
g
el
y
o
n
th
e
p
r
o
j
ec
t’
s
n
at
u
r
e
a
n
d
ch
ar
ac
ter
is
tic
s
,
w
h
ile
t
h
e
es
ti
m
atio
n
ac
c
u
r
ac
y
d
ep
en
d
s
m
er
el
y
u
p
o
n
th
e
a
m
o
u
n
t o
f
r
eliab
le
in
f
o
r
m
atio
n
g
ai
n
ed
r
eg
ar
d
in
g
t
h
e
d
ev
elo
p
ed
p
r
o
d
u
ct
[
5
]
.
Scien
ti
f
ic
e
f
f
o
r
ts
ar
e
b
ein
g
c
ar
r
ied
o
u
t
f
o
r
d
ev
elo
p
in
g
n
e
w
tec
h
n
iq
u
es
to
es
ti
m
ate
s
o
f
t
w
ar
e
co
s
t
.
Nea
r
l
y
all
e
s
ti
m
atio
n
m
o
d
el
s
f
o
r
s
o
f
t
w
ar
e
co
s
t
ar
e
al
g
o
r
ith
m
ic
an
d
e
x
p
er
t
j
u
d
g
m
en
t
b
ased
.
A
cc
u
r
ac
y
m
o
d
ell
in
g
af
f
ec
t
s
esti
m
atio
n
ac
cu
r
ac
y
,
w
h
ic
h
is
w
h
y
f
i
n
d
in
g
g
o
o
d
m
o
d
els
f
o
r
s
o
f
t
w
ar
e
esti
m
atio
n
is
th
e
g
r
ea
test
s
i
g
n
i
f
ica
n
t
o
b
j
ec
tiv
e
f
o
r
s
o
f
t
w
ar
e
en
g
i
n
ee
r
s
.
I
n
th
e
co
r
e
o
f
t
h
ese
m
o
d
els
is
t
h
e
C
OC
OM
O
(
co
n
s
tr
u
cti
v
e
co
s
t
m
o
d
el
)
,
th
is
m
o
d
el
i
s
t
h
e
m
o
s
t
f
r
eq
u
e
n
tl
y
u
s
ed
d
u
e
it
s
s
i
m
p
licit
y
in
es
ti
m
atin
g
t
h
e
p
er
s
o
n
-
m
o
n
t
h
e
f
f
o
r
t f
o
r
p
r
o
j
ec
ts
at
v
ar
io
u
s
d
ev
elo
p
m
e
n
t sta
g
es [
5
]
.
T
h
e
C
OC
OM
O
m
o
d
el
in
(
1
)
w
a
s
f
ir
s
t
d
e
v
elo
p
ed
in
1
9
8
4
b
y
B
o
eh
m
[
1
7
]
.
I
t
h
as
b
ee
n
co
n
s
id
er
ed
to
b
e
e
m
p
ir
ical
d
u
e
to
th
e
en
o
r
m
o
u
s
a
m
o
u
n
t
o
f
d
ata
u
s
ed
i
n
it
s
d
ev
elo
p
m
e
n
t;
t
h
ese
d
ata
ar
e
t
ak
en
f
r
o
m
s
e
v
er
al
p
r
o
j
ec
ts
.
I
n
ad
d
itio
n
,
it
is
f
o
u
n
d
th
at
m
a
n
y
p
r
o
j
ec
t
m
an
a
g
er
s
e
m
p
lo
y
t
h
e
C
O
C
OM
O
m
o
d
el;
th
is
i
s
b
ec
au
s
e
i
ts
d
etails ar
e
av
ailab
le
u
n
li
k
e
o
th
er
m
o
d
els [
1
8
,
1
9
]
.
E
=
a
(
SIZ
E
)
b
(
1
)
P
ar
am
eter
v
al
u
e
s
(
a
)
an
d
(
b
)
,
r
el
y
p
r
in
cip
all
y
o
n
th
e
s
o
f
t
w
ar
e
p
r
o
j
ec
t
class
.
So
f
t
w
ar
e
p
r
o
j
e
cts
w
er
e
class
i
f
ied
b
ased
o
n
th
e
co
m
p
le
x
it
y
o
f
t
h
e
p
r
o
j
ec
t
in
to
th
r
ee
ca
teg
o
r
i
es:
o
r
g
a
n
ic,
s
e
m
id
etac
h
ed
,
an
d
e
m
b
ed
d
ed
.
T
h
e
m
o
d
el
h
elp
s
is
d
e
f
i
n
i
n
g
m
at
h
e
m
atica
l
eq
u
atio
n
s
t
h
at
id
e
n
ti
f
y
th
e
co
s
t,
s
c
h
ed
u
le
an
d
q
u
alit
y
o
f
a
s
o
f
t
w
ar
e
p
r
o
d
u
ct
[
1
8
]
.
T
h
is
w
o
r
k
r
ep
r
esen
ts
an
at
te
m
p
t
to
o
p
ti
m
ize
th
e
p
ar
a
m
e
t
er
s
o
f
f
i
v
e
v
ar
iat
io
n
s
o
f
th
e
C
OC
OM
O
m
o
d
el.
T
h
e
f
ir
s
t
i
s
th
e
b
asic
C
OC
OM
O
m
o
d
el
g
i
v
e
n
in
(
1
)
.
T
h
e
o
t
h
er
t
w
o
ar
e
m
o
d
i
f
ic
atio
n
s
o
f
t
h
e
b
asic
C
OC
OM
O
m
o
d
el
p
r
o
p
o
s
ed
b
y
Sh
eta
[
2
0
]
,
b
o
th
m
o
d
if
ie
d
m
o
d
els
co
n
s
id
er
th
e
m
et
h
o
d
o
lo
g
ies
(
ME
)
to
lin
ea
r
l
y
a
f
f
ec
t
ef
f
o
r
t.
O
n
e
o
f
t
h
e
m
is
S
h
eta
’
s
Mo
d
el
1
g
iv
e
n
in
(
2
)
an
d
is
n
a
m
ed
(
Mo
d
el
I
)
in
t
h
i
s
w
o
r
k
,
t
h
e
o
th
er
is
S
h
eta’
s
Mo
d
el
2
g
iv
e
n
in
(
3
)
,
an
d
n
a
m
ed
(
Mo
d
el
I
I
)
h
er
e
[
7
]
.
E
=
a
(
Size
)
b
+
c
(
ME
)
(
2
)
E
=
a
(
Size
)
b
+
c
(
ME
)
+
d
(
3
)
T
h
e
last
t
w
o
m
o
d
els
ar
e
p
r
o
p
o
s
ed
b
y
U
y
s
al
[
2
1
]
,
o
n
e
o
f
t
h
e
m
co
n
tai
n
f
iv
e
p
ar
a
m
eter
s
a,
b
,
c,
d
an
d
e
ca
lled
U
y
s
al
’
s
Mo
d
el
1
,
an
d
is
ca
lled
(
Mo
d
el
I
I
I
)
h
er
e,
as in
(
4
)
[
2
1
]
.
E
=
a
(
Size
)
b
+
c
.
ME
d
+
e
(
4
)
T
h
e
o
th
er
m
o
d
el
ca
lled
U
y
s
al
’
s
Mo
d
el
2
an
d
ca
lled
(
Mo
d
el
I
V)
h
er
e,
is
p
r
esen
ted
as i
n
(
5
)
E
=
a
(
Size
)
b
+
c
.
ME
d
+
e
.
ln
(
ME
)
+
f
.
ln
(
Size
)
+
g
(
5
)
I
n
t
h
is
w
o
r
k
,
a
n
at
te
m
p
t
is
co
n
d
u
cted
to
o
p
ti
m
ize
p
ar
a
m
eter
s
(
a
,
b
,
c,
d
,
e,
f,
a
n
d
g
)
u
s
i
n
g
t
h
e
a
n
tl
io
n
o
p
tim
izatio
n
(
AL
O)
al
g
o
r
ith
m
.
2
.
2
.
T
he
a
ntlio
n o
pti
m
iza
t
io
n a
lg
o
rit
h
m
I
n
s
ec
ts
li
k
e
an
tlio
n
s
b
elo
n
g
to
a
g
r
o
u
p
in
th
e
f
a
m
i
l
y
o
f
m
y
r
m
ele
n
tid
ae
.
T
h
e
t
w
o
m
ai
n
s
ta
g
es
o
f
it
s
lif
ec
y
cle
ar
e
lar
v
al
an
d
ad
u
lt
s
tag
es.
T
h
e
an
t
lio
n
lar
v
a
leav
e
s
tr
ails
in
t
h
e
s
an
d
in
th
e
s
ea
r
c
h
o
f
a
g
o
o
d
lo
ca
tio
n
to
co
n
s
tr
u
c
t
it
s
tr
ap
,
w
h
ich
is
w
h
y
it
is
ca
l
led
"
d
o
o
d
leb
u
g
"
.
An
tlio
n
s
m
a
k
e
a
p
it
i
n
t
h
e
s
a
n
d
to
h
id
e
in
s
id
e
it
d
u
r
in
g
h
u
n
ti
n
g
a
s
s
h
o
w
n
i
n
Fi
g
u
r
e
1
(
a)
.
Fig
u
r
e
1
(
b
)
d
e
p
icts
th
e
s
lip
p
in
g
o
f
t
h
e
p
r
e
y
to
w
a
r
d
s
th
e
b
o
tto
m
,
th
e
an
tlio
n
in
s
ta
n
tl
y
g
r
ab
s
it.
I
f
it
tr
ies
to
escap
e,
t
h
e
a
n
tlio
n
to
s
s
es
s
o
m
e
s
a
n
d
to
th
e
ed
g
e
o
f
th
e
p
it
s
o
th
a
t
t
h
e
p
r
ay
s
lid
es
in
to
th
e
lo
w
er
m
o
s
t
o
f
t
h
at
p
it.
T
h
e
lar
v
a
al
s
o
wea
k
e
n
s
t
h
e
p
it
’
s
s
id
es,
f
o
r
cin
g
th
e
m
to
d
r
o
p
an
d
tak
e
th
e
p
r
e
y
w
it
h
th
e
m
[
2
2
]
.
T
h
e
AL
O
alg
o
r
it
h
m
i
n
s
p
ir
a
tio
n
co
m
es
f
r
o
m
th
e
f
o
r
ag
i
n
g
b
eh
av
io
u
r
o
f
th
e
an
tlio
n
’
s
lar
v
ae
[
2
3
]
.
2
.
2
.
1
.
AL
O
m
et
ho
do
lo
g
y
Mo
d
ellin
g
t
h
e
r
elatio
n
s
h
ip
b
et
w
ee
n
a
n
tlio
n
s
a
n
d
an
t
s
r
eq
u
i
r
es
an
ts
to
g
o
t
h
r
o
u
g
h
t
h
e
s
ea
r
ch
s
p
ac
e,
w
h
er
e
an
t
lio
n
s
ar
e
p
er
m
itted
to
h
u
n
t
t
h
e
m
an
d
t
h
u
s
d
ev
e
lo
p
th
eir
f
i
tn
e
s
s
to
tr
ap
s
.
I
n
n
at
u
r
e,
an
t
s
m
o
v
e
i
n
a
s
to
ch
ast
ic
m
a
n
n
er
i
n
s
ea
r
ch
f
o
r
f
o
o
d
,
th
u
s
a
s
to
ch
a
s
tic
f
u
n
c
tio
n
(
r
(
t)
)
ca
n
b
e
d
e
f
i
n
ed
as
in
(
6
)
,
an
d
a
r
an
d
o
m
w
al
k
is
s
u
itab
le
f
o
r
m
o
d
elli
n
g
th
e
m
o
v
e
m
en
t o
f
an
t
s
as
g
iv
e
n
in
(
7
)
[
2
4
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
19
,
No
.
3
,
J
u
n
e
2
0
2
1
:
8
1
7
-
8
2
8
820
r
(
t
)
=
{
1
if
r
a
n
d
>
0
.
5
0
if
r
a
n
d
≤
0
.
5
(
6
)
w
h
er
e
r
a
n
d
is
a
r
an
d
o
m
n
u
m
b
er
m
ad
e
w
it
h
u
n
i
f
o
r
m
d
is
tr
ib
u
t
io
n
in
t
h
e
in
ter
v
al
o
f
[
0
,
1
]
.
X
(
t
)
=
[
(
0
,
c
ums
um
(
2r
(
t
1
)
−
1
,
.
.
.
,
c
ums
um
(
2r
(
t
n
)
−
1
)
]
(
7
)
w
h
e
r
e
C
u
m
s
u
m
i
s
c
a
l
c
u
l
a
t
e
s
t
h
e
c
u
m
u
l
a
t
i
v
e
s
u
m
,
n
i
s
t
h
e
m
a
x
i
m
u
m
n
u
m
b
e
r
o
f
i
t
e
r
a
t
i
o
n
,
a
n
d
t
s
h
o
w
s
t
h
e
i
t
e
r
a
t
i
o
n
,
(
a)
(
b
)
Fig
u
r
e
1
.
B
eh
av
io
u
r
o
f
h
u
n
ti
n
g
[
2
3
]
A
m
atr
ix
is
u
s
ed
to
s
to
r
e
th
e
p
o
s
itio
n
f
o
r
all
an
ts
a
s
i
n
(
8
)
;
th
is
m
atr
ix
w
il
l
b
e
u
tili
ze
d
th
r
o
u
g
h
o
u
t o
p
ti
m
izat
io
n
[
2
3
]
.
M
A
n
t
=
[
A
1
,
1
A
1
,
2
⋯
A
1
,
d
⋮
⋮
⋯
⋮
A
n
,
1
A
n
,
2
⋯
A
n
,
d
]
(
8
)
An
ts
ar
e
n
o
ted
to
b
e
s
i
m
ilar
to
th
e
p
ar
ticles
in
P
SO
o
r
i
n
d
iv
i
d
u
als
i
n
G
A
,
a
n
a
n
t
’
s
p
o
s
itio
n
d
en
o
te
a
p
ar
ticu
lar
s
o
lu
tio
n
p
ar
a
m
eter
.
T
h
e
Ma
tr
ix
is
u
s
ed
to
r
eg
i
s
ter
th
e
p
o
s
t
o
f
all
a
n
t
s
t
h
r
o
u
g
h
o
u
t
o
p
ti
m
izat
io
n
.
T
o
ev
alu
a
te
ea
ch
an
t,
a
f
u
n
c
tio
n
o
f
f
it
n
es
s
s
h
o
u
ld
b
e
em
p
lo
y
ed
,
m
atr
i
x
g
ath
er
s
th
e
f
it
n
e
s
s
v
al
u
es
f
o
r
all
an
t
s
as
in
(
9
)
[
2
3
]
.
Mo
r
e
o
v
er
,
an
tli
o
n
s
ar
e
also
ass
u
m
ed
to
b
e
h
i
d
in
g
s
o
m
ep
lace
in
t
h
e
s
ea
r
ch
s
p
ac
e,
an
d
w
it
h
t
h
e
ai
m
o
f
s
a
v
i
n
g
th
eir
p
o
s
itio
n
s
a
n
d
f
it
n
es
s
v
al
u
e
s
,
t
w
o
m
atr
ice
s
ar
e
u
s
ed
,
an
d
[
2
3
,
2
5
]
.
M
OA
=
[
f
(
A
1
,
1
A
1
,
2
…
A
1
,
d
)
⋮
…
⋯
f
(
A
n
,
1
A
n
,
2
…
A
n
,
d
)
]
(
9
)
B
asicall
y
,
al
l
r
an
d
o
m
w
al
k
s
a
r
e
estab
lis
h
ed
u
s
in
g
(
7
)
,
th
e
u
p
d
ate
o
f
p
o
s
itio
n
s
f
o
r
a
n
ts
is
d
o
n
e
u
s
in
g
th
e
r
an
d
o
m
w
al
k
at
ea
ch
an
d
ev
er
y
s
ta
g
e
o
f
o
p
ti
m
izat
io
n
.
Nev
er
th
e
less
,
u
p
d
atin
g
p
o
s
itio
n
o
f
an
t
s
ca
n
n
o
t
b
e
d
ir
ec
tl
y
ac
co
m
p
lis
h
ed
u
s
i
n
g
(
7
)
.
So
,
to
r
estrict
th
e
r
an
d
o
m
w
al
k
s
o
f
an
ts
i
n
t
h
e
s
ea
r
ch
s
p
ac
e,
(
1
0
)
is
u
s
ed
to
n
o
r
m
alize
th
e
m
an
d
it
m
u
s
t
b
e
ap
p
lied
in
a
ll
i
ter
atio
n
s
to
e
n
s
u
r
e
t
h
at
t
h
e
r
an
d
o
m
w
al
k
o
cc
u
r
in
s
id
e
t
h
e
s
ea
r
c
h
s
p
ac
e
[
2
2
,
2
3
]
.
W
h
er
e,
a
i
is
th
e
m
i
n
i
m
u
m
o
f
r
a
n
d
o
m
w
al
k
f
o
r
th
e
i
th
v
ar
iab
le,
is
th
e
m
a
x
i
m
u
m
o
f
r
an
d
o
m
w
al
k
i
n
i
th
v
ar
iab
le,
is
th
e
m
i
n
i
m
u
m
o
f
i
th
v
ar
iab
le
at
i
th
ite
r
atio
n
,
d
i
t
is
th
e
m
ax
i
m
u
m
o
f
i
th
v
ar
iab
le
at
i
th
iter
atio
n
.
X
i
t
=
(
X
i
t
−
a
i
)
×
(
d
i
−
C
i
t
)
(
d
i
t
−
a
i
)
+
C
i
(
1
0
)
2
.
2
.
2
.
P
it
s
a
nd
t
ra
ps
o
f
a
ntlio
ns
An
tlio
n
s
’
tr
ap
s
af
f
ec
t
r
an
d
o
m
w
al
k
s
o
f
a
n
ts
;
t
h
i
s
ass
u
m
p
tio
n
is
m
o
d
elled
m
a
th
e
m
atica
l
l
y
u
s
i
n
g
t
wo
p
r
o
p
o
s
ed
eq
u
atio
n
s
in
(
1
1
)
an
d
(
1
2
)
.
T
h
ese
t
w
o
eq
u
atio
n
s
s
h
o
w
t
h
at
a
n
t
s
w
a
lk
r
an
d
o
m
l
y
in
a
h
y
p
er
s
p
h
er
e
ex
p
r
ess
ed
b
y
C
a
n
d
D
v
ec
to
r
s
n
ea
r
b
y
a
s
elec
ted
a
n
tlio
n
[
2
2
,
2
5
]
.
W
h
er
e
j
t
is
th
e
p
o
s
itio
n
o
f
s
e
lecte
d
j
th
An
tlio
n
at
t
th
iter
atio
n
.
T
h
e
s
ele
ctio
n
o
f
an
t
lio
n
s
is
b
ased
o
n
th
eir
f
i
tn
e
s
s
u
s
i
n
g
t
h
e
r
o
u
lett
e
w
h
ee
l.
T
h
is
m
ec
h
a
n
i
s
m
g
i
v
es
h
i
g
h
c
h
an
ce
s
to
t
h
e
f
i
tter
a
n
tlio
n
s
to
ca
tch
an
ts
.
Fi
g
u
r
e
2
ill
u
s
tr
ates
h
o
w
an
ts
ar
e
e
x
p
ec
ted
to
b
e
tr
ap
p
e
d
in
o
n
l
y
o
n
e
p
ar
ticu
l
ar
an
tlio
n
[
2
3
]
.
C
i
t
=
A
n
tl
ion
j
t
+
C
t
(
1
1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
P
a
r
a
mete
r
tu
n
in
g
o
f so
ftw
a
r
e
effo
r
t e
s
tima
tio
n
mo
d
els u
s
in
g
…
(
Ma
r
r
w
a
A
b
d
-
A
lK
a
r
ee
m
A
l
a
b
a
jee
)
821
D
i
t
=
A
n
tl
ion
j
t
+
D
t
(
1
2
)
Fig
u
r
e
2
.
R
an
d
o
m
w
al
k
o
f
a
n
an
t in
s
id
e
th
e
tr
ap
2
.
2
.
3
.
Sli
din
g
t
o
wa
rds
a
ntlio
ns
An
tlio
n
s
ar
e
n
o
w
ca
p
ab
le
o
f
b
u
ild
in
g
tr
ap
s
r
elat
iv
el
y
to
t
h
eir
f
it
n
ess
,
w
h
er
e
a
n
ts
ar
e
r
eq
u
ir
ed
to
m
o
v
e
in
r
an
d
o
m
.
Yet,
o
n
ce
an
tlio
n
s
r
ec
o
g
n
ize
t
h
at
a
n
a
n
t
is
t
r
ap
p
ed
,
th
ey
b
eg
i
n
to
t
h
r
o
w
s
a
n
d
s
a
w
a
y
f
r
o
m
t
h
e
ce
n
tr
e
o
f
t
h
e
p
it.
T
h
is
ac
t
s
lid
e
s
d
o
w
n
th
e
escap
in
g
tr
ap
p
ed
an
t,
a
n
d
to
m
o
d
el
th
is
b
e
h
av
io
u
r
,
th
e
r
ad
iu
s
o
f
t
h
e
h
y
p
er
-
s
p
h
er
e
f
o
r
a
n
ts
'
r
an
d
o
m
w
a
lk
s
i
s
r
ed
u
ce
d
ac
co
r
d
in
g
to
t
h
e
r
atio
ca
lcu
la
te
b
y
[
2
2
]
an
d
g
i
v
e
n
i
n
(
1
3
)
.
B
ased
o
n
th
is
r
atio
,
t
w
o
r
ed
u
ct
io
n
s
ar
e
p
r
o
p
o
s
ed
as in
(
1
4
)
an
d
(
1
5
)
[
2
6
,
2
7
]
.
I
=
10
.
t
T
(
1
3
)
w
h
er
e
t
r
ese
m
b
les
c
u
r
r
en
t
iter
atio
n
,
T
g
i
v
es
t
h
e
m
a
x
i
m
u
m
n
u
m
b
er
o
f
i
ter
atio
n
s
,
a
n
d
w
is
a
co
n
s
ta
n
t
d
ef
in
ed
b
ased
o
n
th
e
cu
r
r
e
n
t
iter
atio
n
(
w
=2
w
h
e
n
t>0
.
1
T
,
w
=3
w
h
e
n
t>0
.
5
T
,
w
=4
w
h
e
n
t>
0
.
7
5
T
,
w
=5
w
h
e
n
t>0
.
9
T
,
an
d
w
=6
w
h
en
t>0
.
9
5
T
)
.
B
asically
,
t
h
e
co
n
s
tan
t
w
c
an
ad
j
u
s
t th
e
ac
c
u
r
ac
y
le
v
el
o
f
ex
p
lo
itatio
n
[
2
3
]
.
C
t
=
C
t
I
(
1
4
)
d
t
=
d
t
I
(
1
5
)
Hu
n
ti
n
g
co
m
e
s
to
an
en
d
w
h
e
n
an
an
t
ar
r
iv
es
at
t
h
e
b
o
tto
m
m
o
s
t
ar
ea
o
f
th
e
p
it
an
d
is
i
m
p
r
is
o
n
ed
b
y
th
e
j
a
w
s
o
f
t
h
e
an
t
lio
n
.
Su
b
s
e
q
u
en
tl
y
,
a
n
a
n
tlio
n
s
tar
ts
to
d
r
ag
t
h
e
a
n
t
in
to
th
e
s
a
n
d
a
n
d
in
g
e
s
t
its
b
o
d
y
.
T
o
s
i
m
u
late
s
u
ch
a
p
r
o
ce
s
s
,
an
a
s
s
u
m
p
tio
n
h
as
to
b
e
m
ad
e
t
h
a
t
p
r
ey
ca
tch
in
g
ta
k
e
p
lace
w
h
en
a
n
ts
g
r
o
w
f
itter
(
s
in
k
in
s
an
d
)
t
h
an
t
h
e
co
r
r
es
p
o
n
d
en
t
an
tl
io
n
.
A
cc
o
r
d
in
g
l
y
,
th
e
an
t
lio
n
h
as
to
e
n
h
a
n
ce
it
s
ch
an
ce
o
f
ca
tch
i
n
g
n
e
w
p
r
e
y
b
y
u
p
d
atin
g
its
p
o
s
i
tio
n
to
t
h
e
latest
k
n
o
w
n
p
o
s
it
io
n
o
f
th
e
h
u
n
ted
a
n
t;
t
h
i
s
i
s
g
iv
e
n
in
(
1
6
)
[
2
3
]
.
W
h
er
e
t
is
th
e
c
u
r
r
en
t iter
atio
n
,
:
is
th
e
p
o
s
itio
n
o
f
i
th
a
n
t a
t
t
th
it
er
atio
n
.
A
n
tl
ion
j
t
=
A
n
t
i
t
if
f
(
A
n
t
i
t
)
>
f
(
A
n
tl
ion
j
t
)
(
1
6
)
2
.
2
.
4
.
E
litis
m
B
est
g
ai
n
ed
s
o
lu
tio
n
s
ar
e
at
r
is
k
o
f
b
ein
g
lo
s
t
t
h
r
o
u
g
h
s
u
b
s
eq
u
en
t
iter
atio
n
s
;
t
h
at
i
s
w
h
y
eliti
s
m
is
n
ee
d
ed
.
I
t
allo
w
s
b
est
s
o
lu
tio
n
s
to
b
e
k
ep
t
t
h
r
o
u
g
h
t
h
e
iter
ati
o
n
s
o
f
t
h
e
p
r
o
ce
s
s
.
B
ei
n
g
t
h
e
f
itte
s
t
a
n
tlio
n
,
elite
ca
n
af
f
ec
t
t
h
e
m
o
v
e
m
e
n
ts
o
f
ea
ch
an
t
d
u
r
in
g
all
iter
at
io
n
s
.
Hen
ce
,
ev
er
y
an
t
is
p
r
esu
m
e
d
to
w
alk
r
a
n
d
o
m
l
y
ar
o
u
n
d
th
e
a
n
tlio
n
u
s
i
n
g
r
o
u
let
te
w
h
ee
l
an
d
elit
is
m
co
n
c
u
r
r
en
tl
y
as i
n
(
1
7
)
[
2
3
,
2
7
]
.
A
n
t
i
t
=
r
a
t
+
r
e
t
2
(
1
7
)
2
.
2
.
5
.
AL
O
a
lg
o
rit
hm
f
o
r
pa
ra
m
et
er
t
un
i
ng
T
h
e
w
o
r
k
f
lo
w
s
o
f
th
e
AL
O
a
lg
o
r
ith
m
ca
n
b
e
g
r
ap
h
ica
ll
y
r
ep
r
esen
ted
to
d
escr
ib
e
t
h
e
s
eq
u
en
tial
o
r
co
n
cu
r
r
en
t
f
lo
w
s
o
f
ac
ti
v
it
ies
o
f
th
e
a
lg
o
r
it
h
m
in
p
ar
a
m
et
er
tu
n
i
n
g
as
i
n
Fig
u
r
e
3
.
T
h
is
f
i
g
u
r
e
s
h
o
w
s
t
h
e
A
cti
v
it
y
d
iag
r
a
m
o
f
AL
O
al
g
o
r
ith
m
,
t
h
e
s
eq
u
e
n
ce
o
f
t
h
e
ac
ti
v
itie
s
,
b
eg
in
n
i
n
g
f
r
o
m
t
h
e
s
tar
tin
g
p
o
in
t
u
n
til
t
h
e
f
i
n
is
h
i
n
g
p
o
in
t
o
f
t
h
e
ac
tiv
it
y
.
T
o
ex
p
lain
th
e
r
eq
u
ir
e
m
e
n
ts
o
f
AL
O
al
g
o
r
ith
m
i
n
tu
n
in
g
p
ar
a
m
eter
s
,
th
e
u
s
e
ca
s
e
d
iag
r
a
m
i
s
m
o
d
eled
th
e
f
u
n
ct
io
n
alit
y
o
f
t
h
e
s
y
s
te
m
u
s
i
n
g
ac
to
r
s
an
d
u
s
es c
ase
a
s
illu
s
t
r
ated
in
Fig
u
r
e
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
19
,
No
.
3
,
J
u
n
e
2
0
2
1
:
8
1
7
-
8
2
8
822
Fig
u
r
e
3
.
A
cti
v
it
y
d
iag
r
a
m
o
f
AL
O
al
g
o
r
ith
m
i
n
tu
n
i
n
g
p
ar
am
eter
s
Fig
u
r
e
4
.
Use c
ase
d
iag
r
a
m
o
f
AL
O
al
g
o
r
ith
m
i
n
tu
n
i
n
g
p
ar
am
eter
s
3.
R
E
SU
L
T
S
A
ND
AN
AL
Y
SI
S
A
f
ter
d
escr
ib
in
g
t
h
e
AL
O
in
t
h
e
p
r
ev
io
u
s
s
ec
t
io
n
,
t
h
is
s
ec
ti
o
n
w
ill
d
e
m
o
n
s
tr
ate
th
e
d
ata
s
ets
th
a
t
ar
e
u
s
ed
in
th
is
s
tu
d
y
.
T
h
is
i
s
d
o
n
e
a
l
o
n
g
w
ith
t
h
e
ev
al
u
a
t
i
o
n
c
r
it
e
r
i
a
t
h
a
t
a
r
e
w
id
e
ly
u
s
e
d
t
o
ev
a
l
u
at
e
th
e
q
u
a
li
ty
o
f
s
o
f
tw
a
r
e
ef
f
o
r
t
es
tim
a
ti
o
n
m
o
d
e
l
s
an
d
ex
p
e
r
im
en
ts
.
R
es
u
l
ts
a
r
e
a
n
aly
s
e
d
an
d
c
o
m
p
a
r
e
d
w
ith
o
t
h
e
r
m
et
h
o
d
s
.
3
.
1
.
E
x
peri
m
ent
a
l
da
t
a
s
et
s
T
h
e
p
r
ev
io
u
s
l
y
d
is
c
u
s
s
ed
alg
o
r
ith
m
i
s
i
m
p
le
m
e
n
ted
u
s
in
g
MA
T
L
A
B
2
0
1
7
in
th
is
w
o
r
k
.
A
ctu
a
l
c
o
m
m
o
n
d
a
t
as
e
ts
a
r
e
u
s
e
d
d
u
r
in
g
th
e
an
a
ly
s
is
;
an
d
a
r
e
d
o
w
n
lo
a
d
e
d
f
r
o
m
th
e
p
r
o
m
i
s
e
d
at
a
r
ep
o
s
i
t
o
r
y
,
th
ey
a
r
e
:
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
P
a
r
a
mete
r
tu
n
in
g
o
f so
ftw
a
r
e
effo
r
t e
s
tima
tio
n
mo
d
els u
s
in
g
…
(
Ma
r
r
w
a
A
b
d
-
A
lK
a
r
ee
m
A
l
a
b
a
jee
)
823
B
aile
y
an
d
B
asil
i
[
2
8
]
,
w
id
el
y
e
m
p
lo
y
ed
in
m
an
y
o
f
t
h
e
r
esear
ch
s
t
u
d
ies,
s
u
c
h
as
Sh
eta
[
2
0
]
an
d
U
y
s
al
[
2
1
]
.
I
t
co
n
s
is
ts
o
f
t
wo
in
d
ep
en
d
en
t
v
ar
iab
les:
li
n
e
o
f
co
d
e
(
L
OC
)
a
n
d
m
et
h
o
d
o
lo
g
y
(
ME
)
,
in
ad
d
itio
n
to
o
n
e
d
ep
en
d
en
t v
ar
i
ab
le:
m
ea
s
u
r
ed
ef
f
o
r
t
(
m
a
n
-
m
o
n
th
s
).
Data
s
et1
f
o
r
th
e
co
co
m
o
8
1
d
ataset
co
v
er
in
g
6
3
p
r
o
j
ec
ts
[
2
9
]
.
Data
s
et2
f
o
r
N
A
S
A
d
atase
t c
o
n
tai
n
in
g
6
0
p
r
o
j
ec
ts
[
3
0
]
.
Data
s
et3
f
o
r
N
A
S
A
d
atase
t
w
i
th
9
3
p
r
o
j
ec
t [
3
1
]
.
Data
s
et4
f
o
r
k
e
m
er
er
d
ataset
h
av
in
g
1
5
p
r
o
j
ec
ts
[
3
2
]
.
E
ac
h
s
o
f
t
w
ar
e
p
r
o
j
ec
t
h
as
its
ac
tu
al
co
s
t
a
v
ailab
le
i
n
th
e
d
ataset;
it
is
u
s
ed
i
n
co
m
p
ar
i
s
o
n
s
w
i
th
esti
m
ated
co
s
ts
s
o
a
s
to
d
ete
r
m
in
e
t
h
e
m
ea
n
r
elati
v
e
er
r
o
r
f
o
r
p
r
o
j
ec
ts
.
We
co
n
d
u
cted
th
r
ee
te
s
ts
o
n
th
e
af
o
r
e
m
e
n
tio
n
ed
d
ata
s
ets.
I
n
th
e
f
ir
s
t
an
d
s
ec
o
n
d
test
s
w
e
u
s
ed
B
aile
y
a
n
d
B
asil
i
d
atas
et
b
u
t
o
n
d
if
f
er
en
t
m
o
d
el
s
,
in
f
ir
s
t
test
w
e
o
p
ti
m
i
ze
p
ar
am
eter
s
f
o
r
th
r
ee
m
o
d
els:
b
asic
C
OC
OM
O
m
o
d
el,
Sh
eta’
s
Mo
d
el
(
n
a
m
e
d
Mo
d
el
I
)
an
d
Sh
eta’
s
Mo
d
el
2
(
n
a
m
ed
Mo
d
el
I
I
)
.
I
n
s
ec
o
n
d
test
w
e
o
p
ti
m
ize
t
h
e
p
ar
a
m
eter
s
o
f
f
o
u
r
m
o
d
els
Sh
eta
’
s
Mo
d
el
1
(
Mo
d
el
I
)
,
S
h
eta
’
s
Mo
d
el
2
(
M
o
d
el
I
I
)
,
U
y
s
al
’
s
Mo
d
el
1
(
Mo
d
el
I
I
I
)
an
d
Uy
s
a
l’
s
Mo
d
el
2
(
Mo
d
el
I
V)
.
T
h
e
o
th
er
f
o
u
r
d
atasets
w
er
e
u
s
ed
i
n
t
h
e
th
ir
d
test
to
e
s
ti
m
ate
th
e
p
ar
a
m
eter
s
o
f
th
e
b
asic
C
OC
OM
O
m
o
d
el.
3
.
2
.
M
ea
s
ure
s
f
o
r
ev
a
lua
t
i
o
n
A
n
u
m
b
er
o
f
ev
al
u
atio
n
cr
iter
i
a
is
u
s
ed
to
ev
al
u
ate
t
h
e
ef
f
ici
en
c
y
o
f
th
e
d
ev
e
lo
p
ed
m
o
d
el,
th
e
y
ar
e:
Var
ian
ce
-
A
cc
o
u
n
ted
-
Fo
r
(
VAF)
g
iv
e
n
i
n
(
1
8
)
[
1
]
VAF
=
[
1
−
v
ar
(
y
−
y
′
)
v
a
r
(
y
)
]
∗
100%
(
1
8
)
Me
an
m
a
g
n
it
u
d
e
o
f
r
elativ
e
er
r
o
r
(
MM
R
E
)
d
escr
ib
e
d
in
(
1
9
)
[
1
]
=
1
∑
|
−
́
|
=
1
(
1
9
)
T
h
e
p
r
e
d
ictio
n
at
lev
el
N
(
P
R
E
D(
N)
)
s
tated
in
(
2
0
)
[
9
]
PR
E
D
(
L
)
=
1
N
∑
{
1
if
M
R
E
≤
L
0
othe
r
w
ise
N
i
=
1
∗
100
(
2
0
)
Me
an
ab
s
o
lu
te
er
r
o
r
(
MA
E
)
as
in
(
2
1
)
[
9
]
=
1
∑
|
−
̂
|
=
1
(
2
1
)
Mean
s
q
u
ar
e
s
er
r
o
r
(
MSE
)
as
in
(
2
2
)
[
7
]
M
SE
=
1
n
∑
(
y
−
y
′
)
2
n
i
=
1
(
2
2
)
T
h
e
co
r
r
elatio
n
co
ef
f
icien
t
(
R
2
)
as in
(
2
3
)
[
7
]
2
=
∑
(
−
̅
)
2
−
∑
(
−
́
)
2
=
1
=
1
∑
(
−
̅
)
2
=
1
(
2
3
)
R
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
-
R
MS
E
as in
(
2
4
)
[
3
3
]
R
M
SE
=
√
1
N
∑
(
E
i
−
Ê
i
)
2
N
i
=
1
(
2
4
)
Me
d
ian
m
a
g
n
it
u
d
e
o
f
r
elativ
e
er
r
o
r
(
M
d
MRE
)
as in
(
2
5
)
[
3
3
]
=
(
1
∑
|
−
́
|
=
1
)
(
25)
3
.
3
.
T
est1
T
h
e
d
ataset
o
f
B
aile
y
an
d
B
asil
i
i
s
e
m
p
lo
y
ed
a
lo
n
g
w
it
h
AL
O
to
o
p
ti
m
ize
p
ar
a
m
eter
s
f
o
r
th
r
ee
m
o
d
el
s
:
b
as
ic
C
OC
O
MO
m
o
d
el,
Sh
eta
’
s
Mo
d
el
(
n
a
m
ed
Mo
d
el
I
)
an
d
S
h
eta
’
s
Mo
d
el
2
(
n
a
m
ed
Mo
d
el
I
I
)
.
T
h
e
r
an
g
e
o
f
p
ar
a
m
et
er
s
u
s
ed
h
er
e
ar
e
as
p
r
esen
ted
in
[
2
0
]
an
d
[
7
]
.
Fo
r
co
m
p
ar
is
o
n
p
u
r
p
o
s
es,
t
h
e
p
o
p
u
latio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
19
,
No
.
3
,
J
u
n
e
2
0
2
1
:
8
1
7
-
8
2
8
824
s
ize
o
r
n
u
m
b
er
o
f
an
tlio
n
a
g
e
n
ts
i
s
s
et
to
1
0
0
an
d
th
e
iter
at
io
n
n
u
m
b
er
is
s
et
to
5
0
0
ac
co
r
d
in
g
to
th
at
o
f
[
7
]
,
f
o
r
th
e
s
a
m
e
p
u
r
p
o
s
e,
th
e
m
ea
n
ab
s
o
lu
te
er
r
o
r
(
MA
E
)
cr
it
er
ia
is
e
m
p
lo
y
ed
as t
h
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
.
T
h
e
d
ata
ar
e
d
iv
id
ed
in
to
tr
ain
in
g
a
n
d
test
i
n
g
d
ata;
test
in
g
d
ata
is
u
s
ed
to
ev
alu
ate
t
h
e
o
p
tim
ized
m
o
d
el
s
b
y
m
ea
n
s
o
f
t
h
e
f
o
llo
w
i
n
g
ev
a
lu
at
io
n
m
etr
ics
V
A
F,
MSE
,
M
A
E
,
MM
R
E
,
R
MSE
an
d
R
2
.
R
e
s
u
lts
ar
e
co
m
p
ar
ed
w
it
h
f
ir
e
f
l
y
al
g
o
r
ith
m
(
F
A
)
,
g
e
n
etic
a
lg
o
r
it
h
m
(
G
A
)
,
a
n
d
p
ar
ticle
s
w
ar
m
o
p
ti
m
izatio
n
(
P
SO)
.
T
h
e
AL
O
is
ap
p
lied
to
o
b
tain
t
h
e
o
p
ti
m
izi
n
g
p
ar
a
m
eter
s
f
o
r
th
e
t
h
r
ee
m
o
d
els,
r
es
u
lt
s
ar
e
s
h
o
w
n
i
n
T
ab
le
1
.
T
ab
les 2
,
3
,
an
d
4
s
h
o
w
t
h
e
e
v
alu
atio
n
r
es
u
lt
s
f
o
r
test
in
g
t
h
e
th
r
ee
m
o
d
els alo
n
g
w
it
h
th
e
c
o
m
p
ar
is
o
n
b
et
w
ee
n
AL
O,
F
A
,
G
A
an
d
P
SO.
Fro
m
th
e
r
es
u
lt
s
is
s
ee
n
t
h
at
AL
O
ex
ce
ed
s
F
A
,
G
A
an
d
P
SO
in
th
e
o
p
ti
m
iza
tio
n
o
f
all
m
o
d
els
w
it
h
all
t
h
e
ev
al
u
at
io
n
m
etr
ic
s
.
T
ab
le
1
.
Op
tim
izi
n
g
p
ar
a
m
ete
r
s
u
s
i
n
g
AL
O
M
o
d
e
l
s
V
a
l
u
e
o
f
p
a
r
a
me
t
e
r
s
B
a
si
c
C
O
C
O
M
O
M
o
d
e
l
A
=
1
.
9
3
9
1
,
B
=
0
.
9
0
8
3
9
M
o
d
e
l
I
A
=
1
.
9
1
7
4
,
B
=
0
.
9
0
9
4
8
,
C
=
0
.
0
2
4
5
4
6
M
o
d
e
l
I
I
A
=
0
.
7
7
6
7
1
,
B
=
1
.
1
0
1
,
C
=
-
0
.
1
1
2
2
5
,
D
=
8
.
7
9
1
4
T
ab
le
2
.
C
o
m
p
ar
is
o
n
f
o
r
b
asic
C
OC
OM
O
m
o
d
el
A
L
O
FA
GA
PSO
VAF
9
9
.
1
8
%
9
8
.
1
6
%
9
7
.
9
7
%
9
7
.
9
8
%
M
S
E
2
7
.
6
2
5
9
.
1
4
63
.
9
6
6
3
.
6
8
M
A
E
3
.
7
4
5
.
6
5
6
.
0
6
6
.
0
4
M
M
R
E
0
.
0
6
0
.
1
1
0
.
1
3
0
.
1
2
R
M
S
E
5
.
2
5
7
.
6
7
8
.
0
0
7
.
9
8
R2
0
.
9
9
6
4
0
.
9
7
8
1
0
.
9
7
6
3
0
.
9
7
6
5
T
ab
le
3
.
C
o
m
p
ar
is
o
n
f
o
r
Mo
d
el
I
T
ab
le
4
.
C
o
m
p
ar
is
o
n
f
o
r
Mo
d
el
I
I
A
L
O
FA
GA
PSO
VAF
9
9
.
1
4
%
9
8
.
6
2
%
9
7
.
9
7
%
9
8
.
5
2
%
M
S
E
2
6
.
9
3
4
7
.
7
4
9
8
.
17
6
0
.
0
7
M
A
E
3
.
7
6
5
.
5
6
7
.
7
0
5
.
6
3
M
M
R
E
0
.
0
7
0
.
2
4
0
.
2
9
0
.
2
3
R
M
S
E
5
.
1
9
6
.
8
2
9
.
3
9
7
.
7
2
R2
0
.
9
9
6
3
0
.
9
8
2
3
0
.
9
6
3
7
0
.
9
7
7
8
A
L
O
FA
GA
PSO
VAF
9
9
.
3
9
%
9
8
.
6
3
%
9
7
.
6
0
%
9
8
.
7
0
%
M
S
E
2
1
.
8
5
4
5
.
0
2
1
1
4
.
7
9
5
2
.
8
5
M
A
E
3
.
4
5
5
.
5
7
7
.
8
3
5
.
2
9
M
M
R
E
0
.
1
0
0
.
2
4
0
.
2
7
0
.
2
1
R
M
S
E
4
.
6
7
6
.
6
2
9
.
8
6
7
.
1
9
R2
0
.
9
9
7
6
0
.
9
8
3
3
0
.
9
5
7
5
0
.
9
8
0
5
3
.
4
.
T
est2
I
n
t
h
i
s
s
e
c
t
i
o
n
,
t
h
e
p
a
r
a
m
e
t
e
r
s
o
f
f
o
u
r
m
o
d
e
l
s
S
h
e
t
a
’
s
M
o
d
e
l
1
(
M
o
d
e
l
I
)
,
S
h
e
t
a
’
s
M
o
d
e
l
2
(
M
o
d
e
l
I
I
)
,
U
y
s
a
l
’
s
M
o
d
e
l
1
(
M
o
d
e
l
I
I
I
)
a
n
d
U
y
s
a
l
’
s
M
o
d
e
l
2
(
M
o
d
e
l
I
V
)
a
r
e
o
p
t
i
m
i
z
e
d
.
B
a
i
l
e
y
a
n
d
B
a
s
i
l
i
d
a
t
a
s
e
t
i
s
a
l
s
o
u
s
e
d
i
n
t
h
i
s
t
e
s
t
,
A
L
O
p
a
r
a
m
e
t
e
r
s
e
t
t
i
n
g
i
s
s
e
t
i
d
e
n
t
i
c
a
l
t
o
[
9
]
,
t
h
e
i
t
e
r
a
t
i
o
n
n
u
m
b
e
r
i
s
s
e
t
t
o
1
0
0
a
n
d
t
h
e
p
o
p
u
l
a
t
i
o
n
s
i
z
e
i
s
s
e
t
t
o
1
0
.
T
h
e
o
p
t
i
m
i
z
e
d
p
a
r
a
m
e
t
e
r
s
a
r
e
s
h
o
w
n
i
n
T
a
b
l
e
5
u
s
i
n
g
A
L
O
a
n
d
t
h
e
f
o
u
r
m
o
d
e
l
s
.
T
ab
le
5
.
B
est v
a
lu
e
s
o
f
m
o
d
el
's p
ar
am
e
ter
s
u
s
i
n
g
AL
O
M
o
d
e
l
s
B
e
st
v
a
l
u
e
s o
f
p
a
r
a
me
t
e
r
s
M
o
d
e
l
I
A
=
1
.
0
5
5
8
,
B
=
1
.
0
3
7
8
,
C
=
0
.
0
9
7
1
3
5
M
o
d
e
l
I
I
A
=
1
.
0
1
3
5
6
,
B
=
1
.
0
6
4
1
5
,
C
=
-
0
.
5
,
D
=
1
6
.
6
2
1
5
M
o
d
e
l
I
I
I
A
=
1
.
1
7
2
,
B
=
1
.
0
2
0
1
,
C
=
-
0
.
1
1
4
1
5
,
D
=
1
.
2
0
4
9
,
E
=
8
.
8
2
8
8
M
o
d
e
l
I
V
A
=
1
.
0
4
4
2
,
B
=
1
.
0
4
8
4
,
C
=
-
0
.
2
5
3
9
,
D
=
1
.
1
6
4
3
,
E=
2
.
3
6
3
1
,
F
=
0
.
1
3
9
8
3
,
G
=
6
.
9
9
7
1
I
n
itiall
y
,
t
h
e
ac
cu
r
ac
y
o
f
t
h
e
m
o
d
el
s
is
a
s
s
e
s
s
ed
u
s
i
n
g
M
MRE,
Md
MRE,
an
d
P
R
E
D
(
2
5
)
cr
iter
ia
.
T
ab
le
6
d
is
p
lay
s
t
h
e
g
ai
n
ed
r
esu
lt
s
o
f
th
e
m
o
d
els
u
s
i
n
g
AL
O
i
n
a
co
m
p
ar
i
s
o
n
w
it
h
G
A
f
o
r
(
Mo
d
el
I
an
d
Mo
d
el
II
)
an
d
SA
f
o
r
(
Mo
d
el
I
I
I
an
d
Mo
d
el
I
V)
th
at
is
g
iv
e
n
i
n
[
9
]
u
s
in
g
d
ir
ec
ted
ar
tif
icial
b
ee
co
lo
n
y
alg
o
r
ith
m
(
D
A
B
C
A
)
f
o
r
t
h
e
s
a
m
e
f
o
u
r
m
o
d
el
s
.
R
e
s
u
l
ts
p
r
esen
ted
i
n
T
ab
le
6
in
d
icate
th
at
f
o
r
Mo
d
el
I
,
th
e
MM
R
E
an
d
Md
MRE
v
alu
e
s
f
o
r
A
L
O
ar
e
b
etter
th
an
t
h
o
s
e
o
f
t
h
e
o
t
h
er
s
al
g
o
r
ith
m
s
,
P
R
E
D
(
2
5
)
is
th
e
s
a
m
e
f
o
r
all.
T
h
is
co
n
clu
s
io
n
ca
n
also
b
e
d
r
a
w
n
ab
o
u
t
Mo
d
el
I
I
an
d
Mo
d
el
I
I
I
,
w
i
th
P
R
E
D(
2
5
)
b
ein
g
b
etter
i
n
b
o
t
h
m
o
d
el
s
.
I
n
Mo
d
el
I
V
h
o
w
e
v
er
,
A
L
O
h
as
th
e
s
a
m
e
g
o
o
d
r
esu
lt
s
in
ter
m
s
o
f
MM
R
E
an
d
Md
MRE
b
u
t
th
e
P
R
E
D(
2
5
)
v
a
lu
e
o
f
D
A
B
C
A
i
s
b
etter
th
an
t
h
at
o
f
AL
O.
T
ab
le
7
d
is
p
lay
s
t
h
e
p
r
ed
icted
an
d
th
e
ac
tu
al
e
f
f
o
r
t
v
al
u
es
f
o
r
Mo
d
el
I
an
d
Mo
d
el
I
I
u
s
in
g
AL
O,
DA
B
C
A
,
an
d
GA
f
o
r
(
1
8
)
p
r
o
j
ec
ts
.
Fo
r
Mo
d
el
I
,
A
L
O
f
o
u
n
d
b
etter
esti
m
ates
i
n
ter
m
o
f
th
e
ac
tu
al
f
o
r
(
1
0
)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
P
a
r
a
mete
r
tu
n
in
g
o
f so
ftw
a
r
e
effo
r
t e
s
tima
tio
n
mo
d
els u
s
in
g
…
(
Ma
r
r
w
a
A
b
d
-
A
lK
a
r
ee
m
A
l
a
b
a
jee
)
825
p
r
o
j
ec
ts
,
w
h
ile
D
A
B
C
A
f
o
u
n
d
(
6
)
,
an
d
GA
f
o
u
n
d
o
n
l
y
(
2
)
.
A
s
f
o
r
Mo
d
el
I
I
,
AL
O
a
ch
iev
ed
(
1
1
)
b
etter
esti
m
ated
p
r
o
j
ec
ts
,
DA
B
C
A
ac
h
iev
ed
(
4
)
,
an
d
G
A
ac
h
ie
v
ed
(
3
)
o
n
l
y
.
T
ab
le
8
g
iv
e
s
t
h
e
s
a
m
e
v
al
u
es
f
o
r
Mo
d
el
I
I
I
an
d
Mo
d
el
I
V
u
s
in
g
AL
O,
DA
B
C
A
,
an
d
S
A
f
o
r
th
e
s
a
m
e
(
1
8
)
p
r
o
j
ec
t
s
.
I
n
Mo
d
el
I
I
I
,
A
L
O
w
as
c
a
p
a
b
l
e
o
f
f
in
d
in
g
b
es
t
es
t
im
ate
s
f
o
r
(
1
0
)
p
r
o
je
c
t
s
,
DA
B
C
A
o
b
t
a
i
n
e
d
(
5
)
,
an
d
SA
f
o
u
n
d
(
3
)
.
O
n
th
e
o
th
e
r
h
an
d
,
i
n
M
o
d
el
I
V
,
A
L
O
s
u
c
ce
e
d
e
d
in
f
in
d
in
g
(
8
)
b
e
s
t
es
tim
a
t
e
d
p
r
o
je
c
t
s
,
DA
B
C
A
f
o
u
n
d
(
6
)
,
a
n
d
SA
f
o
u
n
d
ju
s
t
(
4
)
.
T
ab
le
6
.
R
es
u
lts
b
ased
o
n
MM
R
E
,
Md
MRE
an
d
P
R
E
D(
2
5
)
M
o
d
e
l
P
R
ED
(
2
5
)
M
d
M
R
E(
%)
M
M
R
E(
%)
M
o
d
e
l
I
(
G
A
)
6
1
.
1
1
1
4
.
5
2
3
.
7
9
M
o
d
e
l
I
(
D
A
B
C
A
)
6
1
.
1
1
1
4
.
8
6
2
6
.
0
3
M
o
d
e
l
I
(
A
L
O
)
6
1
.
1
1
1
3
.
3
9
1
8
.
4
3
M
o
d
e
l
I
I
(
G
A
)
3
8
.
8
9
4
9
.
2
7
6
3
.
6
4
M
o
d
e
l
I
I
(
D
A
B
C
A
)
7
7
.
7
8
1
1
.
4
8
1
7
.
1
3
M
o
d
e
l
I
I
(
A
L
O
)
7
7
.
7
7
7
.
4
1
3
.
6
0
M
o
d
e
l
I
I
I
(
S
A
)
7
7
.
7
8
8
.
2
2
0
.
0
4
M
o
d
e
l
I
I
I
(
D
A
B
C
A
)
8
3
.
3
3
8
.
6
5
1
4
.
2
0
M
o
d
e
l
I
I
I
(
A
L
O
)
7
7
.
7
7
5
.
8
2
1
3
.
5
5
M
o
d
e
l
I
V
(
S
A
)
7
7
.
7
8
8
.
6
3
1
8
.
8
0
M
o
d
e
l
I
V
(
D
A
B
C
A
)
8
3
.
3
3
7
.
0
7
1
3
.
2
1
M
o
d
e
l
I
V
(
A
L
O
)
8
3
.
3
3
5
.
3
7
1
3
.
3
6
T
ab
le
7
.
Me
asu
r
ed
d
ata
an
d
p
r
ed
icted
v
alu
es
f
o
r
Mo
d
el
I
an
d
Mo
d
el
I
I
P
r
o
j
M
o
d
e
l
I
M
o
d
e
l
I
I
A
c
t
u
a
l
C
o
st
A
L
O
D
A
B
C
A
GA
A
L
O
D
A
B
C
A
GA
1
1
5
.
7
9
9
9
1
2
2
.
6
1
7
1
2
4
.
8
5
8
5
1
2
3
.
6
5
3
7
1
3
0
.
6
1
8
6
1
3
4
.
0
2
0
2
1
1
5
.
8
2
5
8
.
3
1
9
5
7
5
.
9
2
2
9
7
4
.
8
4
6
7
6
6
.
5
0
0
0
7
1
.
8
1
0
3
8
4
.
1
6
1
6
96
3
5
8
.
6
0
2
3
7
6
.
6
1
9
4
7
5
.
4
8
5
2
6
7
.
4
1
3
9
7
2
.
7
5
0
8
85
.
0
1
1
2
79
4
6
8
.
8
6
4
1
8
6
.
1
3
7
7
8
5
.
4
3
4
9
7
8
.
0
1
0
0
8
3
.
9
6
4
5
9
4
.
9
8
2
8
9
0
.
8
5
4
0
.
7
8
7
3
5
1
.
0
3
2
3
5
0
.
5
8
1
5
3
8
.
4
1
8
9
4
1
.
9
9
4
5
5
6
.
6
5
8
3
9
.
6
6
1
2
5
.
1
9
7
9
9
8
.
4
0
4
3
9
9
.
0
5
0
4
1
3
4
.
6
9
0
1
9
8
.
3
7
8
1
0
7
.
2
6
0
9
9
8
.
4
7
1
7
.
4
0
5
9
2
4
.
8
7
8
8
2
4
.
1
4
8
1
8
.
9
0
0
0
1
8
.
9
0
0
8
3
2
.
6
4
6
1
1
8
.
9
8
1
5
.
4
1
8
2
1
7
.
9
9
2
1
8
.
0
1
05
1
1
.
9
9
6
4
1
1
.
3
6
0
8
2
5
.
0
7
5
5
1
0
.
3
9
2
8
.
5
0
0
0
3
8
.
0
0
2
3
7
.
2
7
2
4
2
7
.
6
5
2
7
2
9
.
6
2
0
5
4
4
.
3
0
8
6
2
8
.
5
10
5
.
9
4
1
4
3
.
8
6
7
6
4
.
5
8
4
9
7
.
0
0
0
0
3
.
7
3
9
4
1
4
.
4
5
6
3
7
11
6
.
5
2
7
0
9
.
0
1
0
8
8
.
9
3
8
4
1
1
.
7
8
8
9
9
.
0
0
0
7
1
9
.
9
7
5
9
9
12
1
1
.
9
1
0
9
1
3
.
4
7
6
7
1
3
.
5
9
2
6
1
0
.
1
4
0
6
8
.
6
7
0
7
2
1
.
5
7
6
3
7
.
3
13
5
.
0
0
0
0
0
.
3
0
3
1
.
5
1
4
.
8
5
3
7
1
.
1
1
9
5
1
1
.
2
7
0
3
5
14
8
.
4
2
6
8
7
.
8
0
6
1
8
.
2
5
4
4
7
.
7
4
0
4
5
.
2
7
1
5
1
7
.
0
8
8
7
8
.
4
15
1
0
1
.
2
5
7
9
1
0
8
.
7
4
8
1
1
1
0
.
5
2
4
9
1
0
4
.
5
2
5
1
1
1
1
.
4
2
7
9
1
1
8
.
0
3
7
8
9
8
.
7
16
1
3
.
7
8
1
7
1
8
.
6
4
8
1
8
.
2
5
5
9
1
4
.
4
9
5
6
1
3
.
6
2
3
6
2
6
.
8
3
1
2
1
5
.
6
17
1
7
.
1
4
1
3
2
4
.
0
0
8
2
2
3
.
3
6
9
1
8
.
0
1
9
2
1
7
.
9
4
0
4
3
1
.
6
8
6
4
2
3
.
9
18
1
2
9
.
9
8
4
9
1
3
2
.
1
6
3
5
1
3
5
.
4
8
2
5
1
3
6
.
9
7
0
0
1
4
3
.
8
0
6
4
1
4
4
.
4
5
8
7
1
3
8
.
3
3
.
5
.
T
est3
I
n
t
h
is
tes
t,
AL
O
is
u
s
ed
to
e
s
ti
m
ate
t
h
e
p
ar
a
m
eter
s
o
f
th
e
b
asic
C
O
C
OM
O
m
o
d
el
u
s
i
n
g
f
o
u
r
lar
g
e
d
atasets
(
d
ataset1
,
d
ataset2
,
d
ataset3
an
d
d
ataset4
)
,
w
it
h
t
h
e
m
ea
n
r
elati
v
e
er
r
o
r
(
MRE)
b
ein
g
t
h
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
.
R
e
s
u
lt
s
ar
e
co
m
p
a
r
ed
w
it
h
h
y
b
r
id
b
at
i
n
s
p
ir
e
d
g
r
av
itatio
n
al
s
ea
r
c
h
alg
o
r
i
th
m
m
e
th
o
d
ca
lled
(
B
A
T
GSA
)
,
th
e
i
m
p
r
o
v
ed
B
AT
(
I
B
A
T
)
,
an
d
th
e
B
A
T
alg
o
r
ith
m
s
as p
r
ese
n
ted
in
[
1
2
]
.
R
es
u
lts
f
o
r
u
s
in
g
t
h
ese
f
o
u
r
d
atasets
ar
e
as
f
o
llo
w
s
:
Data
s
et1
(
6
3
p
r
o
j
ec
t
)
:
AL
O
was
u
s
ed
to
esti
m
ate
t
h
e
v
a
lu
e
s
o
f
p
ar
am
e
ter
s
o
f
b
asic
C
OC
O
MO
m
o
d
el,
b
est
v
alu
e
s
ar
e
a=
2
.
3
9
4
7
,
b
=0
.
9
4
6
1
4
.
T
ab
le
9
co
m
p
ar
es
a
m
o
n
g
all
alg
o
r
it
h
m
s
,
b
ased
o
n
M
AE
.
R
es
u
lts
s
h
o
w
th
at
t
h
e
v
al
u
e
o
f
M
A
E
o
f
AL
O
is
w
o
r
s
e
t
h
a
n
t
h
e
v
a
lu
e
s
o
f
t
h
e
o
th
er
al
g
o
r
ith
m
s
.
A
s
f
o
r
t
h
e
n
u
m
b
er
o
f
b
e
s
t
esti
m
ated
ef
f
o
r
ts
,
AL
O
ac
h
ie
v
ed
b
etter
r
esu
lts
t
h
an
B
A
T
an
d
I
B
A
T
,
b
u
t
B
A
T
GSA
ac
h
iev
ed
t
h
e
b
est
esti
m
ates o
f
(
3
9
)
,
as sh
o
w
n
in
T
ab
le
9
.
Data
s
et2
(
6
0
p
r
o
j
ec
t
)
:
b
est
v
a
lu
es
o
f
p
ar
a
m
eter
s
o
b
tai
n
ed
u
s
in
g
A
L
O
ar
e
a=
3
.
6
7
8
,
b
=1
.
0
4
7
4
.
T
ab
le
1
0
s
h
o
w
s
t
h
at
t
h
e
v
a
lu
e
o
f
M
AE
f
o
r
AL
O
i
s
v
er
y
c
lo
s
e
to
th
at
o
f
B
A
T
GSA
al
g
o
r
ith
m
wh
ich
is
t
h
e
b
est
a
m
o
n
g
t
h
e
o
th
er
s
.
A
s
f
o
r
esti
m
ated
e
f
f
o
r
ts
,
AL
O
f
o
u
n
d
t
h
e
h
i
g
h
e
s
t
n
u
m
b
er
o
f
b
es
t
esti
m
ates
o
f
(
3
6
)
as
p
r
esen
ted
in
T
ab
le
1
0
.
Data
s
et3
(
9
3
p
r
o
j
ec
t)
:
A
L
O
o
b
tain
ed
b
est
v
al
u
es
o
f
p
ar
a
m
e
ter
s
as
a=
2
.
1
6
5
7
,
b
=
1
.
0
8
2
.
T
a
b
le
1
1
illu
s
tr
ates
th
e
M
A
E
an
d
co
m
p
ar
is
o
n
a
m
o
n
g
all
alg
o
r
it
h
m
s
.
R
es
u
lt
s
in
d
icate
th
at
th
e
v
al
u
e
o
f
M
A
E
o
f
AL
O
is
w
o
r
s
e
th
an
t
h
e
v
a
lu
e
o
f
M
A
E
f
o
r
t
h
e
o
th
er
al
g
o
r
ith
m
s
.
T
h
e
v
alu
e
s
o
f
esti
m
ated
e
f
f
o
r
ts
s
h
o
w
t
h
at
AL
O
w
as b
etter
th
an
B
A
T
an
d
I
B
A
T
,
b
u
t B
A
T
GS
A
f
o
u
n
d
t
h
e
b
est esti
m
ate
s
o
f
(
6
4
)
as sh
o
w
n
in
T
ab
le
1
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
19
,
No
.
3
,
J
u
n
e
2
0
2
1
:
8
1
7
-
8
2
8
826
Data
s
et4
(
1
5
p
r
o
j
ec
t)
:
th
e
p
ar
a
m
eter
s
o
f
b
as
ic
C
O
C
OM
O
m
o
d
el
ar
e
esti
m
ated
u
s
i
n
g
A
L
O,
b
est
v
alu
e
s
o
f
p
ar
am
eter
s
ar
e
a
=8
.
3
4
4
5
,
b
=
0
.
5
1
8
7
.
T
ab
le
1
2
r
esem
b
le
s
a
co
m
p
ar
is
o
n
a
m
o
n
g
all
a
lg
o
r
ith
m
s
.
R
e
s
u
l
ts
d
esig
n
ate
t
h
at
t
h
e
v
a
lu
e
o
f
M
A
E
f
o
r
AL
O
is
b
etter
t
h
a
n
t
h
e
v
alu
e
s
o
f
M
A
E
o
f
al
l o
th
er
al
g
o
r
ith
m
s
.
R
e
s
u
l
ts
o
f
th
e
e
s
ti
m
ated
e
f
f
o
r
ts
illu
s
tr
ate
th
at
AL
O
s
u
cc
ee
d
ed
in
ac
h
iev
in
g
b
etter
est
i
m
a
tes
i
n
all
1
5
p
r
o
j
ec
t
th
an
o
th
er
m
et
h
o
d
s
in
co
m
p
ar
is
o
n
w
it
h
ac
t
u
al
e
f
f
o
r
t.
T
ab
le
8
.
Me
asu
r
ed
d
ata
an
d
p
r
ed
icted
v
alu
es
f
o
r
Mo
d
el
I
I
I
a
n
d
Mo
d
el
I
V
P
r
o
j
M
o
d
e
l
I
I
I
M
o
d
e
l
I
V
A
c
t
u
a
l
C
o
st
A
L
O
D
A
B
C
A
SA
A
L
O
D
A
B
C
A
SA
1
1
1
7
.
6
5
5
3
1
2
7
.
1
4
7
8
1
2
4
.
7
9
4
11
9
.
4
5
9
0
1
2
5
.
3
0
7
1
1
2
4
.
3
5
6
3
1
1
5
.
8
2
6
3
.
0
8
2
7
7
8
.
2
1
9
4
8
1
.
6
6
0
8
6
4
.
3
7
9
8
7
7
.
7
4
3
8
1
.
6
1
4
3
96
3
6
3
.
7
2
2
8
7
9
.
4
3
2
5
8
3
.
1
9
4
1
6
5
.
1
3
6
4
7
9
.
1
1
7
9
8
3
.
1
7
8
1
79
4
7
3
.
8
1
6
3
8
9
.
7
2
8
2
9
2
.
8
6
0
3
7
5
.
3
8
5
9
8
9
.
2
4
7
8
9
2
.
7
5
7
9
0
.
8
5
3
9
.
6
0
0
0
3
9
.
5
3
2
5
3
9
.
0
2
3
8
3
8
.
2
9
5
4
3
9
.
5
4
2
4
3
9
.
6
2
7
9
3
9
.
6
6
1
2
7
.
4
8
9
8
9
8
.
3
0
1
1
9
8
.
0
1
3
2
1
2
9
.
8
6
0
7
9
7
.
2
8
2
8
9
7
.
8
5
6
6
9
8
.
4
7
1
8
.
8
3
0
7
2
3
.
3
1
8
1
2
3
.
8
8
3
8
1
8
.
9
0
0
0
2
1
.
3
7
2
7
2
3
.
8
4
4
6
1
8
.
9
8
1
3
.
7
3
4
4
9
.
5
8
1
4
7
.
7
9
4
8
1
2
.
5
3
7
9
8
.
5
9
6
7
8
.
2
9
9
3
1
0
.
3
9
2
8
.
4
7
3
1
3
0
.
8
5
5
6
3
0
.
8
8
6
4
2
7
.
7
4
9
0
2
9
.
6
2
3
5
3
1
.
0
8
2
9
2
8
.
5
10
6
.
7
5
8
9
6
.
9
6
5
.
3
6
9
4
7
.
0
0
0
0
6
.
4
4
1
5
.
7
9
1
8
7
11
9
.
9
2
9
6
1
5
.
4
2
1
9
1
6
.
4
0
8
9
1
1
.
0
3
2
0
1
4
.
9
2
0
3
1
6
.
7
3
5
9
9
12
1
1
.
2
0
2
3
9
.
2
2
3
7
.
6
1
7
8
1
0
.
5
5
9
5
7
.
7
5
7
.
8
9
7
8
7
.
3
13
5
.
0
0
0
0
2
.
8
4
1
4
0
.
2
6
3
1
4
.
9
5
8
5
3
.
3
4
7
8
0
.
9
9
8
6
5
14
8
.
2
8
0
8
6
.
9
3
7
9
5
.
1
4
9
6
8
.
0
2
1
2
5
.
6
8
5
7
5
.
4
4
6
5
8
.
4
15
1
0
1
.
0
9
4
4
1
0
5
.
0
6
0
2
1
0
2
.
5
7
1
9
1
0
1
.
4
4
7
6
1
0
4
.
7
6
7
5
1
0
2
.
7
9
3
5
9
8
.
7
16
1
4
.
6
7
1
4
1
7
.
2
1
0
8
1
7
.
0
2
0
2
1
4
.
6
2
9
2
1
5
.
2
7
9
3
1
7
.
0
4
0
7
1
5
.
6
17
1
8
.
1
8
4
1
2
1
.
7
4
0
1
2
1
.
9
8
0
3
1
8
.
1
0
8
1
1
9
.
8
0
6
5
2
1
.
9
6
9
8
2
3
.
9
18
1
3
0
.
4
2
1
5
1
3
5
.
5
4
4
7
1
3
1
.
2
3
9
8
1
3
2
.
1
4
9
2
1
3
3
.
8
0
5
3
1
3
0
.
9
5
5
4
1
3
8
.
3
T
ab
le
9
.
C
o
m
p
ar
in
g
al
g
o
r
ith
m
s
u
s
in
g
M
A
E
an
d
b
est es
ti
m
at
es (
d
ata
s
et1
)
A
L
O
B
A
TG
S
A
I
B
A
T
B
A
T
M
A
E
5
5
1
.
2
8
4
3
4
3
3
.
2
1
5
4
3
7
.
5
3
7
4
4
4
.
4
0
5
N
o
.
o
f
B
e
st
Es
t
i
m
a
t
e
s
f
r
o
m 6
3
P
r
o
j
e
c
t
14
39
9
1
T
ab
le
1
0
.
C
o
m
p
ar
in
g
alg
o
r
it
h
m
s
u
s
i
n
g
M
A
E
an
d
b
est est
i
m
ates (
d
ataset2
)
A
L
O
B
A
TG
S
A
I
B
A
T
B
A
T
M
A
E
1
2
7
.
6
6
9
1
1
2
7
.
5
3
0
1
3
2
.
4
8
6
1
3
7
.
4
0
1
N
o
.
o
f
B
e
st
Es
t
i
m
a
t
e
s
f
r
o
m 6
0
P
r
o
j
e
c
t
36
10
4
10
As
t
h
e
r
es
u
lt
s
s
h
o
w
,
AL
O
A
l
g
o
r
ith
m
w
a
s
ab
le
to
ac
h
ie
v
e
b
etter
esti
m
ate
s
t
h
an
t
h
e
o
th
er
alg
o
r
ith
m
s
in
T
est1
a
n
d
T
est2
.
As
f
o
r
T
est3
,
AL
O
s
u
cc
ee
d
ed
i
n
ac
co
m
p
lis
h
in
g
b
etter
r
esu
lt t
h
a
n
t
h
o
s
e
o
f
B
A
T
an
d
I
B
A
T
in
a
ll
co
m
p
ar
is
o
n
s
u
s
i
n
g
al
l
f
o
u
r
d
atasets
i
n
ter
m
s
o
f
M
A
E
a
n
d
th
e
n
u
m
b
er
o
f
b
est
esti
m
at
ed
p
r
o
j
ec
ts
f
o
r
ea
ch
d
ataset.
I
n
th
e
co
m
p
ar
is
o
n
w
it
h
B
A
T
GSA
,
AL
O
w
as a
b
le
to
ac
h
ie
v
e
b
etter
r
e
s
u
l
ts
f
o
r
2
d
a
tasets
(
d
ataset2
an
d
d
ataset4
)
o
u
t o
f
th
e
f
o
u
r
in
v
es
t
ig
ated
d
at
aset
s
al
s
o
in
ter
m
s
o
f
M
A
E
a
n
d
t
h
e
n
u
m
b
er
o
f
b
est
esti
m
ated
p
r
o
j
ec
t
s
f
o
r
ea
ch
d
ataset.
T
ab
le
1
1
.
C
o
m
p
ar
in
g
alg
o
r
it
h
m
s
u
s
i
n
g
M
A
E
an
d
b
est est
i
m
ates (
d
ataset3
)
A
L
O
B
A
TG
S
A
I
B
A
T
B
A
T
M
A
E
3
7
8
.
7
0
3
7
3
5
5
.
3
8
6
3
5
6
.
1
4
3
3
6
5
.
1
6
4
N
o
.
o
f
B
e
st
Es
t
i
m
a
t
e
s
f
r
o
m 9
3
P
r
o
j
ect
18
64
5
6
T
ab
le
1
2
.
C
o
m
p
ar
in
g
alg
o
r
it
h
m
s
u
s
i
n
g
M
A
E
an
d
b
est est
i
m
ates (
d
ataset4
)
A
L
O
B
A
TG
S
A
I
B
A
T
B
A
T
M
A
E
1
1
8
.
7
8
4
5
2
3
9
8
.
1
4
1
4
5
6
4
.
9
3
4
5
5
2
8
.
1
5
8
N
o
.
o
f
B
e
st
Est
i
m
a
t
e
s
f
r
o
m
15
P
r
o
j
e
c
t
15
0
0
0
Evaluation Warning : The document was created with Spire.PDF for Python.