forked from wgrathwohl/GWG_release
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathrbm_svgd.py
185 lines (150 loc) · 6.75 KB
/
rbm_svgd.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
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
import argparse
import toy_data
import rbm
import torch
import numpy as np
import samplers_old as samplers
import mmd
import torch.nn as nn
import matplotlib.pyplot as plt
import os
import torchvision
device = torch.device('cuda:' + str(0) if torch.cuda.is_available() else 'cpu')
import utils
def makedirs(dirname):
"""
Make directory only if it's not already there.
"""
if not os.path.exists(dirname):
os.makedirs(dirname)
def main(args):
makedirs(args.save_dir)
torch.manual_seed(args.seed)
np.random.seed(args.seed)
model = rbm.BernoulliRBM(args.n_visible, args.n_hidden)
model.to(device)
print(device)
if args.data == "mnist":
assert args.n_visible == 784
train_loader, test_loader, plot, viz = utils.get_data(args)
init_data = []
for x, _ in train_loader:
init_data.append(x)
init_data = torch.cat(init_data, 0)
init_mean = init_data.mean(0).clamp(.01, .99)
model = rbm.BernoulliRBM(args.n_visible, args.n_hidden, data_mean=init_mean)
model.to(device)
optimizer = torch.optim.Adam(model.parameters(), lr=args.rbm_lr)
# train!
itr = 0
for x, _ in train_loader:
x = x.to(device)
xhat = model.gibbs_sample(v=x, n_steps=args.cd)
d = model.logp_v_unnorm(x)
m = model.logp_v_unnorm(xhat)
obj = d - m
loss = -obj.mean()
optimizer.zero_grad()
loss.backward()
optimizer.step()
if itr % args.print_every == 0:
print("{} | log p(data) = {:.4f}, log p(model) = {:.4f}, diff = {:.4f}".format(itr,d.mean(), m.mean(),
(d - m).mean()))
else:
model.W.data = torch.randn_like(model.W.data) * (.05 ** .5)
model.b_v.data = torch.randn_like(model.b_v.data) * 1.0
model.b_h.data = torch.randn_like(model.b_h.data) * 1.0
viz = plot = None
gt_samples = model.gibbs_sample(n_steps=args.mcmc_steps, n_samples=args.n_samples + args.n_test_samples, plot=True)
kmmd = mmd.MMD(mmd.exp_avg_hamming, False)
gt_samples, gt_samples2 = gt_samples[:args.n_samples], gt_samples[args.n_samples:]
if plot is not None:
plot("{}/ground_truth.png".format(args.save_dir), gt_samples2)
opt_stat = kmmd.compute_mmd(gt_samples2, gt_samples)
print("gt <--> gt log-mmd", opt_stat, opt_stat.log10())
new_samples = model.gibbs_sample(n_steps=0, n_samples=args.n_test_samples)
log_mmds = {}
log_mmds['gibbs'] = []
for i in range(args.n_steps):
if i % 10 == 0:
stat = kmmd.compute_mmd(new_samples, gt_samples)
log_stat = stat.log10().item()
log_mmds['gibbs'].append(log_stat)
print("gibbs", i, stat, stat.log10())
new_samples = model.gibbs_sample(new_samples, 1)
r_model = samplers.BinaryRelaxedModel(args.n_visible, model)
r_model.to(device)
if args.n_visible == 2:
import visualize_flow
def viz(p, t):
plt.clf()
visualize_flow.plt_flow_density(lambda x: r_model.logp_surrogate(x, t), plt.gca(), npts=200)
plt.savefig(p)
def plot(p, x):
plt.clf()
visualize_flow.plt_samples(x.detach().cpu().numpy(), plt.gca(), 200)
plt.savefig(p)
temps = [.5, 1., 2.]
for temp in temps:
log_mmds[temp] = []
target = lambda x: r_model.logp_surrogate(x, temp)
x = nn.Parameter(r_model.base_dist.sample((args.n_test_samples, args.n_visible)).to(device))
optim = torch.optim.Adam(params=[x], lr=args.lr)
svgd = samplers.SVGD(optim)
if viz is not None:
viz("{}/target_{}.png".format(args.save_dir, temp), temp)
for i in range(args.n_steps):
svgd.discrete_step(x, r_model.logp_target, target)
if i % 100 == 0 and plot is not None:
if args.data == "mnist":
hx = samplers.threshold(x)
else:
hx = x
plot("/{}/samples_temp_{}_{}.png".format(args.save_dir, temp, i), hx)
if i % 10 == 0:
hard_samples = samplers.threshold(x)
stat = kmmd.compute_mmd(hard_samples, gt_samples)
log_stat = stat.log10().item()
log_mmds[temp].append(log_stat)
print("temp = {}, itr = {}, log-mmd = {:.4f}, ess = {:.4f}".format(temp, i, log_stat, svgd._ess))
plt.clf()
for temp in temps + ['gibbs']:
plt.plot(log_mmds[temp], label="{}".format(temp))
plt.legend()
plt.savefig("{}/results.png".format(args.save_dir))
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('--save_dir', type=str, default="/tmp/test_discrete")
parser.add_argument('--data', choices=['mnist', 'random'], type=str, default='random')
parser.add_argument('--n_steps', type=int, default=5000)
parser.add_argument('--n_samples', type=int, default=500)
parser.add_argument('--n_test_samples', type=int, default=100)
parser.add_argument('--mcmc_steps', type=int, default=10000)
parser.add_argument('--seed', type=int, default=1234567)
parser.add_argument('--adapt', action="store_true")
parser.add_argument('--hmc', action="store_true")
parser.add_argument('--mdim', action="store_true")
parser.add_argument('--ss', type=float, default=.01)
parser.add_argument('--n_anneal', type=int, default=10)
parser.add_argument('--sgld_steps', type=int, default=100)
parser.add_argument('--sgld_sigma', type=float, default=.01)
parser.add_argument('--lam', type=float, default=1)
parser.add_argument('--max_lam', type=float, default=1.)
parser.add_argument('--n_hidden', type=int, default=25)
parser.add_argument('--n_visible', type=int, default=100)
parser.add_argument('--n_epochs', type=int, default=100)
parser.add_argument('--batch_size', type=int, default=100)
parser.add_argument('--viz_batch_size', type=int, default=1000)
parser.add_argument('--print_every', type=int, default=100)
parser.add_argument('--viz_every', type=int, default=1000)
parser.add_argument('--n_toy_data', type=int, default=50000)
parser.add_argument('--lr', type=float, default=.01)
parser.add_argument('--rbm_lr', type=float, default=.001)
parser.add_argument('--mcmc_lr', type=float, default=.003)
parser.add_argument('--temp', type=float, default=1.)
parser.add_argument('--tt', type=float, default=1.)
parser.add_argument('--weight_decay', type=float, default=.0)
parser.add_argument('--cd', type=int, default=10)
parser.add_argument('--img_size', type=int, default=28)
args = parser.parse_args()
main(args)