인공지능 (AI)
/
Logistic Regression
Search
Duplicate
Logistic Regression
태그
인공지능(AI)
PyTorch
Logistic Regression
Lab-05 Logistic Regression.pdf
833.0KB
Hypothesis
H
(
X
)
=
1
1
+
e
W
T
X
H(X) = \frac{1}{1+e^{W^T X}}
H
(
X
)
=
1
+
e
W
T
X
1
Cost
c
o
s
t
(
W
)
=
−
1
m
∑
y
l
o
g
(
H
(
x
)
)
+
(
1
−
y
)
(
l
o
g
(
1
−
H
(
x
)
)
cost(W) = -\frac{1}{m} \sum ylog(H(x)) + (1-y)(log(1-H(x))
cos
t
(
W
)
=
−
m
1
∑
y
l
o
g
(
H
(
x
))
+
(
1
−
y
)
(
l
o
g
(
1
−
H
(
x
))
H(x) = P(x=1; w) = 1 - P(x=0; w)
Weight Update via Gradient Descent
W
:
=
W
−
α
∂
∂
W
c
o
s
t
(
W
)
W := W - \alpha \frac{\partial}{\partial W} cost(W)
W
:=
W
−
α
∂
W
∂
cos
t
(
W
)
