-
Notifications
You must be signed in to change notification settings - Fork 2
/
loader.py
237 lines (178 loc) · 7.43 KB
/
loader.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
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
#!/usr/bin/env python
# -*- coding:utf-8 -*-
# date: 2020/12
# author:Yushan Zheng
# emai:[email protected]
import os
import pickle
import numpy as np
from PIL import Image
import torch
import torch.utils.data as data
from utils import extract_tile
class PatchDataset(data.Dataset):
def __init__(self, file_path, transform, od_mode=True, label_type=1):
self.transform = transform
with open(file_path, 'rb') as f:
data = pickle.load(f)
self.data_dir = data['base_dir']
self.image_list = data['list']
self.od = od_mode
self.lt = label_type
def __getitem__(self, index):
img = Image.open(os.path.join(self.data_dir, self.image_list[index][0])).convert('RGB')
label = self.image_list[index][self.lt]
if self.transform!=None:
img = self.transform(img)
if self.od:
img = -torch.log(img + 1.0/255.0)
return img, label
def __len__(self):
return len(self.image_list)
def get_weights(self):
num = self.__len__()
labels = np.zeros((num,), np.int)
for s_ind, s in enumerate(self.image_list):
labels[s_ind] = s[self.lt]
tmp = np.bincount(labels)
weights = 1.0 / np.asarray(tmp[labels], np.float)
return weights
class CLPatchesDataset(data.Dataset):
def __init__(self, file_path, transform, od_mode=True, label_type=1):
self.transform = transform
with open(file_path, 'rb') as f:
data = pickle.load(f)
self.data_dir = data['base_dir']
self.image_list = data['list']
self.od = od_mode
self.lt = label_type
def __getitem__(self, index):
img = Image.open(os.path.join(self.data_dir, self.image_list[index][0])).convert('RGB')
if self.transform!=None:
img1 = self.transform(img)
img2 = self.transform(img)
if self.od:
img1 = -torch.log(img + 1.0/255.0)
img2 = -torch.log(img + 1.0/255.0)
return img1, img2
def __len__(self):
return len(self.image_list)
def get_weights(self):
num = self.__len__()
labels = np.zeros((num,), np.int)
for s_ind, s in enumerate(self.image_list):
labels[s_ind] = s[self.lt]
tmp = np.bincount(labels)
weights = 1.0 / np.asarray(tmp[labels], np.float)
return weights
class SlideLocalTileDataset(data.Dataset):
def __init__(self, image_dir, position_list, transform,
tile_size=512, imsize=224, od_mode=False):
self.transform = transform
self.im_dir = image_dir
self.pos = position_list
self.od = od_mode
self.ts = tile_size
self.imsize = imsize
def __getitem__(self, index):
img = extract_tile(self.im_dir, self.ts, self.pos[index][1], self.pos[index][0], self.imsize, self.imsize)
if len(img) == 0:
img = np.ones((self.imsize, self.imsize, 3), np.uint8) * 240
img = Image.fromarray(img[:,:,[2,1,0]]).convert('RGB')
img = self.transform(img)
if self.od:
img = -torch.log(img + 1.0/255.0)
return img
def __len__(self):
return self.pos.shape[0]
class TissueBorderEmbedder():
def __init__(self, feat_dim, max_len=64, interval=1.0):
self.interval = interval
self.max_len = int(max_len / self.interval)
self.pe = np.zeros((self.max_len, feat_dim), float)
position = np.arange(0, self.max_len)
div_term = 1 / (self.max_len ** (np.arange(0, feat_dim) / feat_dim))
pos_mat = np.matmul(position[:,np.newaxis], div_term[np.newaxis,:])
self.pe[:, 0::2] = np.sin(pos_mat)[:,0::2]
self.pe[:, 1::2] = np.cos(pos_mat)[:,1::2]
def embed(self, dist_values):
index = np.asarray(np.minimum(dist_values/self.interval, self.max_len-1), int)
return self.pe[index]
class LaGraphLoader(torch.utils.data.Dataset):
def __init__(self, graph_list_path, max_node_number, task_id=1,
disable_adj=False, dist_embed_dim=64,
):
with open(graph_list_path, 'rb') as f:
data = pickle.load(f)
self.dl = data['list']
self.list_dir = data['base_dir']
self.type_num = 5 if task_id == 1 else 2
self.maxno = max_node_number
self.ti=task_id
self.use_adj = not disable_adj
with open(self.get_graph_path(0), 'rb') as f:
graph_data = pickle.load(f)
self.feat_dim = graph_data['feats'].shape[1]
self.de_dim = dist_embed_dim
self.de_table = TissueBorderEmbedder(self.de_dim)
def __len__(self):
return len(self.dl)
def __getitem__(self, idx):
graph_feat = np.zeros((self.maxno,self.feat_dim))
graph_adj = np.zeros((self.maxno + 1, self.maxno + 1))
graph_de = np.zeros((self.maxno, self.de_dim))
with open(self.get_graph_path(idx), 'rb') as f:
graph_data = pickle.load(f)
# node (token) number
num_node = min(graph_data['feats'].shape[0], self.maxno)
# node (token) mask
node_mask = np.zeros((self.maxno + 1, 1), int)
node_mask[:(num_node+1)] = 1
# node (token) features
features = graph_data['feats'][:num_node]
graph_feat[:num_node] = features
# distance embedding
dist_embeddings = self.de_table.embed(graph_data['dists'][:num_node])
graph_de[:num_node] = dist_embeddings
# adjacency matrix
adj = graph_data['adj'][:num_node,:num_node] if self.use_adj else np.zeros((num_node,num_node))
adj_ = np.pad(adj, ((1,0),(1,0)), 'constant', constant_values=(1,1))
adj_ = adj_ + np.eye(num_node+1)
degree = np.sum(adj_, axis=1) ** -0.5
degree = np.diag(degree)
normed_adj = np.matmul(np.matmul(degree, adj_), degree)
graph_adj[:(num_node+1), :(num_node+1)] = normed_adj
# graph label
graph_label = np.asarray(self.dl[idx][self.ti], int)
return graph_feat, graph_adj, graph_de, node_mask, graph_label
def get_graph_path(self, idx):
return os.path.join(self.list_dir, self.dl[idx][0])
def get_feat_dim(self):
return self.feat_dim
def get_weights(self):
num = self.__len__()
labels = np.zeros((num,), np.int)
for s_ind, s in enumerate(self.dl):
labels[s_ind] = s[self.ti]
tmp = np.bincount(labels)
weights = 1.0 / np.asarray(tmp[labels], np.float)
return weights
class DistributedWeightedSampler(data.DistributedSampler):
def __init__(self, dataset, weights, num_replicas=None, rank=None, replacement=True):
super(DistributedWeightedSampler, self).__init__(
dataset, num_replicas=num_replicas, rank=rank, shuffle=False
)
self.weights = torch.as_tensor(weights, dtype=torch.double)
self.replacement = replacement
def __iter__(self):
# deterministically shuffle based on epoch
g = torch.Generator()
g.manual_seed(self.epoch)
indices = torch.multinomial(self.weights, self.total_size, self.replacement).tolist()
# add extra samples to make it evenly divisible
indices += indices[:(self.total_size - len(indices))]
assert len(indices) == self.total_size
# subsample
indices = indices[self.rank:self.total_size:self.num_replicas]
assert len(indices) == self.num_samples
return iter(indices)