-
Notifications
You must be signed in to change notification settings - Fork 0
/
utils.py
166 lines (147 loc) · 6.11 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
163
164
165
166
import re
def extract_number(filename):
match = re.search(r'(\d+)', filename)
if match:
return int(match.group(1))
return 0
from torch import nn
from collections import OrderedDict
import numpy as np
import os
import torch
import torch.utils.data as data
def make_layers(block):
"""
Making layers using parameters from NetParams.py
:param block: OrderedDict
:return: layers
"""
layers = []
for layer_name, v in block.items():
if 'pool' in layer_name:
layer = nn.MaxPool2d(kernel_size=v[0], stride=v[1], padding=v[2])
layers.append((layer_name, layer))
elif 'deconv' in layer_name:
transposeConv2d = nn.ConvTranspose2d(in_channels=v[0],
out_channels=v[1],
kernel_size=v[2],
stride=v[3],
padding=v[4])
layers.append((layer_name, transposeConv2d))
if 'relu' in layer_name:
layers.append(('relu_' + layer_name, nn.ReLU(inplace=True)))
elif 'leaky' in layer_name:
layers.append(('leaky_' + layer_name,
nn.LeakyReLU(negative_slope=0.2, inplace=True)))
elif 'conv' in layer_name:
conv2d = nn.Conv2d(in_channels=v[0],
out_channels=v[1],
kernel_size=v[2],
stride=v[3],
padding=v[4])
layers.append((layer_name, conv2d))
if 'relu' in layer_name:
layers.append(('relu_' + layer_name, nn.ReLU(inplace=True)))
elif 'leaky' in layer_name:
layers.append(('leaky_' + layer_name,
nn.LeakyReLU(negative_slope=0.2, inplace=True)))
else:
raise NotImplementedError
return nn.Sequential(OrderedDict(layers))
def load_naca(dir):
path = os.path.join(dir, 'data_')
# u = np.load(f'{path}u.npy')
# v = np.load(f'{path}v.npy')
# print(u.shape)
# print(v.shape)
# return(u, v)
ru = np.load(f'{path}u.npy')
print("ru shape:", ru.shape)
return ru
def split_data(data, is_train):
if is_train:
train_data = data[:160,:,:]
return train_data
else:
valid_data = data[160:234,:,:]
return valid_data
class NACA0012(data.Dataset):
def __init__(self, dir, is_train, n_frames_input, n_frames_output):
super().__init__()
self.datas = load_naca(dir)
self.num_frames_input = n_frames_input
self.num_frames_output = n_frames_output
self.num_frames = n_frames_input + n_frames_output
self.datas = split_data(self.datas, is_train)
print("self: ", self.datas.shape)
print('Loaded {} samples ({})'.format(self.__len__(), 'train' if is_train else 'valid'))
def __getitem__(self, idx):
# return super().__getitem__(idx)
data = self.datas[idx*self.num_frames:(idx+1)*self.num_frames]
inputs = data[:self.num_frames_input]
targets = data[self.num_frames_input:]
inputs = inputs[..., np.newaxis]
targets = targets[..., np.newaxis]
inputs[inputs < 0] = 0.0
targets[targets < 0] = 0.0
inputs = torch.from_numpy(inputs).permute(0, 3, 1, 2).float().contiguous()
targets = torch.from_numpy(targets).permute(0, 3, 1, 2).float().contiguous()
print(idx, targets.shape, inputs.shape)
return idx, targets, inputs
def __len__(self):
return self.datas.shape[0] // self.num_frames
""" class SstSeq(data.Dataset):
def __init__(self, root, is_train, cond_len, pred_len, transform=None):
super(SstSeq, self).__init__()
self.SST_dataset = load_data(root)
self.is_train = is_train
self.cond_len = cond_len
self.pred_len = pred_len
self.transform = transform
self.train_len = self.SST_dataset['X_train'].shape[0] - self.cond_len - self.pred_len # 13188
self.test_len = self.SST_dataset['X_test'].shape[0] - self.cond_len - self.pred_len # 2412
self.length = self.train_len # size of each epoch
def __getitem__(self, idx):
if self.is_train:
# random training
start_point = np.random.randint(0, self.train_len)
inputs = self.SST_dataset['X_train'][start_point:start_point + self.cond_len, ...]
outputs = self.SST_dataset['Y_train'][
start_point + self.cond_len:start_point + self.cond_len + self.pred_len, ...]
else:
# sequentially testing
start_point = idx
inputs = self.SST_dataset['X_test'][start_point:start_point + self.cond_len, ...]
outputs = self.SST_dataset['Y_test'][
start_point + self.cond_len:start_point + self.cond_len + self.pred_len, ...]
inputs = inputs[:, np.newaxis, :, :]
outputs = outputs[:, np.newaxis, :, :]
outputs = torch.from_numpy(outputs).contiguous().float()
inputs = torch.from_numpy(inputs).contiguous().float()
out = [idx, outputs, inputs, start_point]
return out
def __len__(self):
return self.length """
class RecordHist:
def __init__(self, verbose=False):
"""
Args:
verbose (bool): If True, prints a message for each validation loss improvement.
Default: False
"""
self.verbose = verbose
self.val_loss_min = np.Inf
def __call__(self, val_loss, model, epoch, save_path):
self.save_checkpoint(val_loss, model, epoch, save_path)
def save_checkpoint(self, val_loss, model, epoch, save_path):
"""
Saves model.
"""
if self.verbose:
print(
f'Validation loss from ({self.val_loss_min:.6f} --> {val_loss:.6f}). Saving model ...'
)
torch.save(
model, save_path + "/" +
"checkpoint_{}_{:.6f}.pth.tar".format(epoch, val_loss))
self.val_loss_min = val_loss