数学公式推导_什么是分类问题_Classification
问题
l
j
,
j
=
1
,
.
.
.
,
5
l_j,j=1,...,5
lj,j=1,...,5
提出问题
g
k
,
k
=
1
,
.
.
.
,
n
(
n
=
1000
)
g_k,k=1,...,n\left( n=1000 \right)
gk,k=1,...,n(n=1000)
核心解法
e
=
1
n
⋅
Σ
k
min
j
d
(
l
j
,
g
k
)
e=\frac{1}{n}\cdot \underset{k}{\varSigma}\min _jd\left( l_j,g_k \right)
e=n1⋅kΣjmind(lj,gk)
其中
d
(
l
j
,
g
k
)
=
{
0
i
f
x
=
y
1
o
t
h
e
r
w
i
s
e
d\left( l_j,g_k \right) =\left\{ \begin{array}{c} 0 if\,\, x=y\\ 1 otherwise\\ \end{array} \right.
d(lj,gk)={0ifx=y1otherwise
二分类问题的推导
H ( P , Q ) = − Σ i = ( c a t , d o g ) P ( i ) log Q ( i ) = − P ( c a t ) log Q ( c a t ) − P ( d o g ) log Q ( d o g ) = − y ( l o g ( p ) + ( 1 − y ) log ( 1 − p ) ) H\left( P,Q \right) =-\underset{i=\left( cat,dog \right)}{\varSigma}P\left( i \right) \log Q\left( i \right) \\ =-P\left( cat \right) \log Q\left( cat \right) -P\left( dog \right) \log Q\left( dog \right) \\ =-y\left( log \left( p \right) +\left( 1-y \right) \log \left( 1-p \right) \right) H(P,Q)=−i=(cat,dog)ΣP(i)logQ(i)=−P(cat)logQ(cat)−P(dog)logQ(dog)=−y(log(p)+(1−y)log(1−p))
扫一扫
871
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
