-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathutils.py
162 lines (136 loc) · 6.17 KB
/
utils.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
import logging
import numpy as np
import torch
from torch.autograd import Variable
import pandas as pd
import seaborn as sn
import matplotlib
import matplotlib.pyplot as plt
from matplotlib.pyplot import figure
import datasets
matplotlib.use('Agg')
def oracle_score_denominator(point, batch, sigma):
return torch.exp( - ( ( (batch[:, 0]-point[0])**2 + (batch[:, 1]-point[1])**2 ) / (2*(sigma**2)) ) ) / ( 2*np.pi*(sigma**2) )
def guassian_prob(x, tx, sigma):
return torch.exp( - ( ( (tx[:, 0]-x[0])**2 + (tx[:, 1]-x[1])**2 ) / (2*(sigma**2)) ) ) / ( 2*np.pi*(sigma**2) )
def grad_log_guassian_prob(x, tx, sigma):
return (x - tx) / (sigma**2)
def oracle_score_numerator(point, batch, sigma):
p = guassian_prob(point, batch, sigma)
diff = ( batch-point ) / (sigma**2)
diff[:, 0] *= p
diff[:, 1] *= p
return diff, p
def get_oracle_score(points, batch, sigma, eps=1e-8):
points = torch.tensor(points)
numerator, prob = oracle_score_numerator(points, batch, sigma = sigma)
denominator = oracle_score_denominator(points, batch, sigma = sigma)
sum_numerator = torch.sum(numerator, dim=0)
sum_denominator = torch.sum(denominator, dim=0)
prob = torch.mean(prob, dim=0)
return sum_numerator / (sum_denominator + eps), prob
def get_oracle_score_pointwise(config, points, ds):
bs = config.training.batch_size
config.data.dataset = ds
config.training.batch_size = int(config.data.dataset_size)
_ds = datasets.get_dataset(config)
iter_ds = iter(_ds)
batch = torch.tensor(next(iter_ds)['position']._numpy()).to(config.device)
n_scale = config.model.n_scales
score_oracle = np.zeros((n_scale+1, points.shape[0], points.shape[1]))
prob_oracle = np.zeros((n_scale+1, points.shape[0]))
noise_points = np.zeros((n_scale+1, points.shape[0], points.shape[1]))
for ns in range(n_scale+1):
timestep = (ns) / (n_scale+1)
std = (config.loss.std_min * (config.loss.std_max / config.loss.std_min) ** timestep)
z = torch.randn_like(points, device=points.device)
z_ = torch.randn_like(batch, device=batch.device)
noise_pts = points + std * z
noise_bat = batch + std * z_
for i in range(points.shape[0]):
score_pt, prob_pt = get_oracle_score(noise_pts[i], noise_bat, std)
score_oracle[ns, i, :] = score_pt.cpu().numpy()
prob_oracle[ns, i] = prob_pt.cpu().numpy()
noise_points[ns, i, :] = noise_pts[i].cpu().numpy()
config.training.batch_size = bs
return score_oracle, prob_oracle, noise_points
def calculate_score_error(noise_points, score_oracle, score_model, is_score_model=True, n_scale=10):
score_model.eval()
avg_norm = 0
for ns in range(n_scale):
points = Variable(noise_points[ns], requires_grad=True)
timestep = (ns+1) / (n_scale)
t = torch.ones(points.shape[0], device=points.device) * timestep
if is_score_model:
score_estimate = score_model(points, t)
score_estimate_np = score_estimate.clone().detach().cpu().numpy()
norm = (score_oracle[ns,:,0] - score_estimate_np[:,0])**2 + (score_oracle[ns,:,1] - score_estimate_np[:,1])**2
else:
score_estimate = score_model.score_inference(points, t)
score_estimate_np = score_estimate.clone().detach().cpu().numpy()
norm = (score_oracle[ns,:,0] - score_estimate_np[:,0])**2 + (score_oracle[ns,:,1] - score_estimate_np[:,1])**2
avg_norm += np.mean(0.5*norm) / (n_scale)
return avg_norm
def calculate_asymmetricity(noise_points, score_model, is_score_model=True, n_scale=10):
avg_asym = 0
avg_nasym = 0
for ns in range(n_scale):
points = Variable(noise_points[ns], requires_grad=True)
timestep = (ns+1) / (n_scale)
t = torch.ones(points.shape[0], device=points.device) * timestep
if is_score_model:
score_estimate = score_model(points, t)
else:
score_estimate = score_model.score(points, t)
grad0, = torch.autograd.grad(torch.sum(score_estimate[torch.arange(score_estimate.shape[0]), 0]), points, retain_graph=True)
grad1, = torch.autograd.grad(torch.sum(score_estimate[torch.arange(score_estimate.shape[0]), 1]), points)
traceJJt = torch.sum(torch.square(grad0), dim=1) + torch.sum(torch.square(grad1), dim=1)
traceJJ = torch.square(grad0[:,0]) + torch.square(grad1[:,1]) + 2*grad0[:,1]*grad1[:,0]
squared_asym_norm = 0.5 * (traceJJt - traceJJ)
nasym = (squared_asym_norm / (traceJJt+1e-8)).cpu().numpy()
asym = (traceJJt - traceJJ).cpu().numpy()
avg_asym += np.mean(asym) / (n_scale)
avg_nasym += np.mean(nasym) / (n_scale)
return avg_nasym, avg_asym
def plot_vector_field(config, score_fn, dir_file):
logging.info("Plotting Vector Field...")
w = config.sampling.width
h = config.sampling.height
density = config.sampling.density
with torch.no_grad():
x, y = np.meshgrid(np.linspace(-w, w, density, dtype=np.float32), np.linspace(-h, h, density, dtype=np.float32))
points = np.concatenate((np.expand_dims(x.flatten(), axis=1), np.expand_dims(y.flatten(), axis=1)), axis=1)
points = torch.from_numpy(points).to(config.device, dtype=config.data.dtype)
t = torch.zeros(points.shape[0], device=points.device)
cond = (config.loss.std_min * (config.loss.std_max / config.loss.std_min) ** t)
if config.model.type == 'energy_model':
torch.set_grad_enabled(True)
points_vf = score_fn.score_inference(points, cond).cpu().numpy()
torch.set_grad_enabled(False)
else:
points_vf = score_fn(points, cond).cpu().numpy()
fig = figure(figsize=(w/2, h/2), dpi=300)
plt.quiver(x, y, points_vf[:,0], points_vf[:,1])
plt.xticks([])
plt.yticks([])
plt.xlim((-w, w))
plt.ylim((-h, h))
plt.axis('off')
plt.savefig(dir_file)
plt.close(fig)
def plot_data_points(config, samples, dir_file):
logging.info("Plotting Sampled Points...")
w = config.sampling.width
h = config.sampling.height
fig = figure(figsize=(w/2, h/2), dpi=300)
plt.xlim((-w, w))
plt.ylim((-h, h))
label = torch.zeros(samples.shape[0], dtype=torch.long)
plot_data = np.vstack((samples.T, label)).T
df = pd.DataFrame(data=plot_data, columns=("x", "y", "label"))
sn.scatterplot(data=df, x="x", y="y", hue="label", alpha=0.8)
plt.xticks([])
plt.yticks([])
plt.legend('')
plt.savefig(dir_file)
plt.close(fig)