-
Notifications
You must be signed in to change notification settings - Fork 126
/
optim.py
42 lines (33 loc) · 1.21 KB
/
optim.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
"""
Optimizers for multimodal ranking
"""
import theano
import theano.tensor as tensor
import numpy
# name(hyperp, tparams, grads, inputs (list), cost) = f_grad_shared, f_update
def adam(lr, tparams, grads, inp, cost):
gshared = [theano.shared(p.get_value() * 0., name='%s_grad'%k) for k, p in tparams.iteritems()]
gsup = [(gs, g) for gs, g in zip(gshared, grads)]
f_grad_shared = theano.function(inp, cost, updates=gsup, profile=False)
b1 = 0.1
b2 = 0.001
e = 1e-8
updates = []
i = theano.shared(numpy.float32(0.))
i_t = i + 1.
fix1 = 1. - b1**(i_t)
fix2 = 1. - b2**(i_t)
lr_t = lr * (tensor.sqrt(fix2) / fix1)
for p, g in zip(tparams.values(), gshared):
m = theano.shared(p.get_value() * 0.)
v = theano.shared(p.get_value() * 0.)
m_t = (b1 * g) + ((1. - b1) * m)
v_t = (b2 * tensor.sqr(g)) + ((1. - b2) * v)
g_t = m_t / (tensor.sqrt(v_t) + e)
p_t = p - (lr_t * g_t)
updates.append((m, m_t))
updates.append((v, v_t))
updates.append((p, p_t))
updates.append((i, i_t))
f_update = theano.function([lr], [], updates=updates, on_unused_input='ignore', profile=False)
return f_grad_shared, f_update