Gradient Descent
Recall the gradient descent algorithm utilizes the gradient calculation:
repeat until convergence:
{
w
j
=
w
j
−
α
∂
J
(
w
,
b
)
∂
w
j
for j := 0..n-1
b
=
b
−
α
∂
J
(
w
,
b
)
∂
b
}
\begin{align*} &\text{repeat until convergence:} \; \lbrace \\ & \; \; \;w_j = w_j - \alpha \frac{\partial J(\mathbf{w},b)}{\partial w_j} \tag{1} \; & \text{for j := 0..n-1} \\ & \; \; \; \; \;b = b - \alpha \frac{\partial J(\mathbf{w},b)}{\partial b} \\ &\rbrace \end{align*}
repeat until convergence:{wj=wj−α∂wj∂J(w,b)b=b−α∂b∂J(w,b)}for j := 0..n-1(1)
Where each iteration performs simultaneous updates on
w
j
w_j
wj for all
j
j
j, where
∂
J
(
w
,
b
)
∂
w
j
=
1
m
∑
i
=
0
m
−
1
(
f
w
,
b
(
x
(
i
)
)
−
y
(
i
)
)
x
j
(
i
)
∂
J
(
w
,
b
)
∂
b
=
1
m
∑
i
=
0
m
−
1
(
f
w
,
b
(
x
(
i
)
)
−
y
(
i
)
)
\begin{align*} \frac{\partial J(\mathbf{w},b)}{\partial w_j} &= \frac{1}{m} \sum\limits_{i = 0}^{m-1} (f_{\mathbf{w},b}(\mathbf{x}^{(i)}) - y^{(i)})x_{j}^{(i)} \tag{2} \\ \frac{\partial J(\mathbf{w},b)}{\partial b} &= \frac{1}{m} \sum\limits_{i = 0}^{m-1} (f_{\mathbf{w},b}(\mathbf{x}^{(i)}) - y^{(i)}) \tag{3} \end{align*}
∂wj∂J(w,b)∂b∂J(w,b)=m1i=0∑m−1(fw,b(x(i))−y(i))xj(i)=m1i=0∑m−1(fw,b(x(i))−y(i))(2)(3)
- m is the number of training examples in the data set
- f w , b ( x ( i ) ) f_{\mathbf{w},b}(x^{(i)}) fw,b(x(i)) is the model’s prediction, while y ( i ) y^{(i)} y(i) is the target
- For a logistic regression model
z = w ⋅ x + b z = \mathbf{w} \cdot \mathbf{x} + b z=w⋅x+b
f w , b ( x ) = g ( z ) f_{\mathbf{w},b}(x) = g(z) fw,b(x)=g(z)
where g ( z ) g(z) g(z) is the sigmoid function:
g ( z ) = 1 1 + e − z g(z) = \frac{1}{1+e^{-z}} g(z)=1+e−z1
Code Description
The gradient descent algorithm implementation has two components:
compute_gradient_logistic
: The calculation of the current gradient, equations (2,3) above.gradient_descent
: The loop implementing equation (1) above.
compute_gradient_logistic
Implements equation (2),(3) above for all
w
j
w_j
wj and
b
b
b.
There are many ways to implement this. Outlined below is this:
-
initialize variables to accumulate
dj_dw
anddj_db
-
for each example
- calculate the error for that example g ( w ⋅ x ( i ) + b ) − y ( i ) g(\mathbf{w} \cdot \mathbf{x}^{(i)} + b) - \mathbf{y}^{(i)} g(w⋅x(i)+b)−y(i)
- for each input value
x
j
(
i
)
x_{j}^{(i)}
xj(i) in this example,
- multiply the error by the input
x
j
(
i
)
x_{j}^{(i)}
xj(i), and add to the corresponding element of
dj_dw
. (equation 2 above)
- multiply the error by the input
x
j
(
i
)
x_{j}^{(i)}
xj(i), and add to the corresponding element of
- add the error to
dj_db
(equation 3 above)
-
divide
dj_db
anddj_dw
by total number of examples (m) -
note that x ( i ) \mathbf{x}^{(i)} x(i) in numpy
X[i,:]
orX[i]
and x j ( i ) x_{j}^{(i)} xj(i) isX[i,j]
def compute_gradient_logistic(X, y, w, b):
"""
Computes the gradient for linear regression
Args:
X (ndarray (m,n): Data, m examples with n features
y (ndarray (m,)): target values
w (ndarray (n,)): model parameters
b (scalar) : model parameter
Returns
dj_dw (ndarray (n,)): The gradient of the cost w.r.t. the parameters w.
dj_db (scalar) : The gradient of the cost w.r.t. the parameter b.
"""
m,n = X.shape
dj_dw = np.zeros((n,)) #(n,)
dj_db = 0.
for i in range(m):
f_wb_i = sigmoid(np.dot(X[i],w) + b) #(n,)(n,)=scalar
err_i = f_wb_i - y[i] #scalar
for j in range(n):
dj_dw[j] = dj_dw[j] + err_i * X[i,j] #scalar
dj_db = dj_db + err_i
dj_dw = dj_dw/m #(n,)
dj_db = dj_db/m #scalar
return dj_db, dj_dw
gradient_descent
def gradient_descent(X, y, w_in, b_in, alpha, num_iters):
"""
Performs batch gradient descent
Args:
X (ndarray (m,n) : Data, m examples with n features
y (ndarray (m,)) : target values
w_in (ndarray (n,)): Initial values of model parameters
b_in (scalar) : Initial values of model parameter
alpha (float) : Learning rate
num_iters (scalar) : number of iterations to run gradient descent
Returns:
w (ndarray (n,)) : Updated values of parameters
b (scalar) : Updated value of parameter
"""
# An array to store cost J and w's at each iteration primarily for graphing later
J_history = []
w = copy.deepcopy(w_in) #avoid modifying global w within function
b = b_in
for i in range(num_iters):
# Calculate the gradient and update the parameters
dj_db, dj_dw = compute_gradient_logistic(X, y, w, b)
# Update Parameters using w, b, alpha and gradient
w = w - alpha * dj_dw
b = b - alpha * dj_db
# Save cost J at each iteration
if i<100000: # prevent resource exhaustion
J_history.append( compute_cost_logistic(X, y, w, b) )
# Print cost every at intervals 10 times or as many iterations if < 10
if i% math.ceil(num_iters / 10) == 0:
print(f"Iteration {i:4d}: Cost {J_history[-1]} ")
return w, b, J_history #return final w,b and J history for graphing