-
Notifications
You must be signed in to change notification settings - Fork 0
/
gcn.py
353 lines (285 loc) · 12 KB
/
gcn.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
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
from flwr.common import MetricsAggregationFn, NDArrays, Parameters, Scalar
import numpy as np
import torch
import pickle
import logging
import torch.nn as nn
from collections import OrderedDict
from typing import Callable, Dict, List, Optional, Tuple
from sklearn.metrics import roc_auc_score, precision_recall_curve, auc
import flwr as fl
class GCN(nn.Module):
"""
Graph Convolutional Network based on https://arxiv.org/abs/1609.02907
"""
def __init__(self,
feat_dim,
hidden_dim1,
hidden_dim2,
dropout,
is_sparse=False):
"""Dense version of GAT."""
super(GCN, self).__init__()
# self.dropout = dropout
self.W1 = nn.Parameter(torch.FloatTensor(feat_dim, hidden_dim1))
self.W2 = nn.Parameter(torch.FloatTensor(hidden_dim1, hidden_dim2))
self.relu = nn.ReLU()
self.dropout = nn.Dropout(p=dropout)
nn.init.xavier_uniform_(self.W1.data)
nn.init.xavier_uniform_(self.W2.data)
self.is_sparse = is_sparse
def forward(self, x, adj):
# Layer 1
support = torch.mm(x, self.W1)
embeddings = (torch.sparse.mm(adj, support)
if self.is_sparse else torch.mm(adj, support))
embeddings = self.dropout(embeddings)
# Layer 2
support = torch.mm(embeddings, self.W2)
embeddings = (torch.sparse.mm(adj, support)
if self.is_sparse else torch.mm(adj, support))
return embeddings
class Readout(nn.Module):
"""
This module learns a single graph level representation for a molecule given GNN generated node embeddings
"""
def __init__(self, attr_dim, embedding_dim, hidden_dim, output_dim,
num_cats):
super(Readout, self).__init__()
self.attr_dim = attr_dim
self.hidden_dim = hidden_dim
self.output_dim = output_dim
self.num_cats = num_cats
self.layer1 = nn.Linear(attr_dim + embedding_dim, hidden_dim)
self.layer2 = nn.Linear(hidden_dim, output_dim)
self.output = nn.Linear(output_dim, num_cats)
self.act = nn.ReLU()
def forward(self, node_features, node_embeddings):
combined_rep = torch.cat(
(node_features, node_embeddings),
dim=1) # Concat initial node attributed with embeddings from sage
hidden_rep = self.act(self.layer1(combined_rep))
graph_rep = self.act(
self.layer2(hidden_rep)) # Generate final graph level embedding
logits = torch.mean(
self.output(graph_rep),
dim=0) # Generated logits for multilabel classification
return logits
class GcnMoleculeNet(nn.Module):
"""
Network that consolidates GCN + Readout into a single nn.Module
"""
def __init__(
self,
feat_dim,
hidden_dim,
node_embedding_dim,
dropout,
readout_hidden_dim,
graph_embedding_dim,
num_categories,
sparse_adj=False,
):
super(GcnMoleculeNet, self).__init__()
self.gcn = GCN(feat_dim,
hidden_dim,
node_embedding_dim,
dropout,
is_sparse=sparse_adj)
self.readout = Readout(
feat_dim,
node_embedding_dim,
readout_hidden_dim,
graph_embedding_dim,
num_categories,
)
def forward(self, adj_matrix, feature_matrix):
node_embeddings = self.gcn(feature_matrix, adj_matrix)
logits = self.readout(feature_matrix, node_embeddings)
return logits
def get_parameters(net) -> List[np.ndarray]:
return [val.cpu().numpy() for _, val in net.state_dict().items()]
def set_parameters(net, parameters: List[np.ndarray]):
params_dict = zip(net.state_dict().keys(), parameters)
state_dict = OrderedDict({k: torch.Tensor(v) for k, v in params_dict})
net.load_state_dict(state_dict, strict=True)
args = {}
args["device"] = "cuda:0"
args["client_optimizer"] = "sgd"
args["learning_rate"] = 0.001
args["epochs"] = 5
args["frequency_of_the_test"] = 2
args["metric"] = "auc"
f = open("train_data_local_dict.pkl", "rb")
trainloader = pickle.load(f)
def train(model, train_data, test_data, device, args):
logging.info("----------train--------")
model.to(device)
criterion = torch.nn.BCEWithLogitsLoss(reduction="none")
if args["client_optimizer"] == "sgd":
optimizer = torch.optim.SGD(model.parameters(),
lr=args["learning_rate"])
else:
optimizer = torch.optim.Adam(model.parameters(),
lr=args["learning_rate"])
max_test_score = 0
best_model_params = {}
for epoch in range(args["epochs"]):
for mol_idxs, (adj_matrix, feature_matrix, label,
mask) in enumerate(train_data):
if torch.all(mask == 0).item():
continue
optimizer.zero_grad()
adj_matrix = adj_matrix.to(device=device,
dtype=torch.float32,
non_blocking=True)
feature_matrix = feature_matrix.to(device=device,
dtype=torch.float32,
non_blocking=True)
label = label.to(device=device,
dtype=torch.float32,
non_blocking=True)
mask = mask.to(device=device,
dtype=torch.float32,
non_blocking=True)
logits = model(adj_matrix, feature_matrix)
loss = criterion(logits, label) * mask
loss = loss.sum() / mask.sum()
loss.backward()
optimizer.step()
if ((mol_idxs + 1) % args["frequency_of_the_test"]
== 0) or (mol_idxs == len(train_data) - 1):
if test_data is not None:
test_score, _ = test(model, test_data, device, args)
# eval on test dataset
# test_score, _ = self.test(self.test_data, device, args)
# print("Epoch = {}, Iter = {}/{}: Test Score = {}".format(
# epoch, mol_idxs + 1, len(train_data), test_score))
if test_score > max_test_score:
max_test_score = test_score
best_model_params = {
k: v.cpu() for k, v in model.state_dict().items()
}
print("Current best = {}".format(max_test_score))
return max_test_score, best_model_params
def test(model, test_data, device, args):
logging.info("----------test--------")
model.eval()
model.to(device)
with torch.no_grad():
y_pred = []
y_true = []
masks = []
for mol_idx, (adj_matrix, feature_matrix, label,
mask) in enumerate(test_data):
adj_matrix = adj_matrix.to(device=device,
dtype=torch.float32,
non_blocking=True)
feature_matrix = feature_matrix.to(device=device,
dtype=torch.float32,
non_blocking=True)
logits = model(adj_matrix, feature_matrix)
y_pred.append(logits.cpu().numpy())
y_true.append(label.cpu().numpy())
masks.append(mask.numpy())
y_pred = np.array(y_pred)
y_true = np.array(y_true)
masks = np.array(masks)
results = []
for label in range(masks.shape[1]):
valid_idxs = np.nonzero(masks[:, label])
truth = y_true[valid_idxs, label].flatten()
pred = y_pred[valid_idxs, label].flatten()
if np.all(truth == 0.0) or np.all(truth == 1.0):
results.append(float("nan"))
else:
if args["metric"] == "prc-auc":
precision, recall, _ = precision_recall_curve(truth, pred)
score = auc(recall, precision)
else:
score = roc_auc_score(truth, pred)
results.append(score)
score = np.nanmean(results)
return score, model
def get_parameters(net) -> List[np.ndarray]:
return [val.cpu().numpy() for _, val in net.state_dict().items()]
def set_parameters(net, parameters: List[np.ndarray]):
params_dict = zip(net.state_dict().keys(), parameters)
state_dict = OrderedDict({k: torch.Tensor(v) for k, v in params_dict})
net.load_state_dict(state_dict, strict=True)
class FlowerClient(fl.client.NumPyClient):
def __init__(self,
cid,
net,
trainloader,
valloader=None,
args=None) -> None:
super().__init__()
self.cid = cid
self.net = net
self.trainloader = trainloader[self.cid]
self.valloader = valloader
self.args = args
def get_parameters(self, config: Dict[str, Scalar]) -> NDArrays:
return get_parameters(self.net)
def fit(self, parameters: NDArrays,
config: Dict[str,
Scalar]) -> Tuple[NDArrays, int, Dict[str, Scalar]]:
set_parameters(self.net, parameters=parameters)
train(self.net, self.trainloader, self.valloader, self.args['device'],
self.args)
return get_parameters(self.net), len(self.trainloader), {}
def evaluate(
self, parameters: NDArrays,
config: Dict[str, Scalar]) -> Tuple[float, int, Dict[str, Scalar]]:
set_parameters(self.net, parameters)
if self.valloader is None:
return 0.0, 20, {"accuracy": 0.95}
score, model = test(self.net, self.valloader, self.args['device'],
self.args)
return score, len(self.valloader), {"accuracy": 0.95}
class FlowerServer(fl.server.strategy.FedAvg):
def __init__(
self,
*,
fraction_fit: float = 1.0,
fraction_evaluate: float = 1.0,
min_fit_clients: int = 2,
min_evaluate_clients: int = 2,
min_available_clients: int = 2,
evaluate_fn: Optional[Callable[[int, NDArrays, Dict[str, Scalar]],
Optional[Tuple[float,
Dict[str,
Scalar]]],]] = None,
on_fit_config_fn: Optional[Callable[[int], Dict[str, Scalar]]] = None,
on_evaluate_config_fn: Optional[Callable[[int], Dict[str,
Scalar]]] = None,
accept_failures: bool = True,
initial_parameters: Optional[Parameters] = None,
fit_metrics_aggregation_fn: Optional[MetricsAggregationFn] = None,
evaluate_metrics_aggregation_fn: Optional[MetricsAggregationFn] = None
) -> None:
super().__init__(
fraction_fit=fraction_fit,
fraction_evaluate=fraction_evaluate,
min_fit_clients=min_fit_clients,
min_evaluate_clients=min_evaluate_clients,
min_available_clients=min_available_clients,
evaluate_fn=evaluate_fn,
on_fit_config_fn=on_fit_config_fn,
on_evaluate_config_fn=on_evaluate_config_fn,
accept_failures=accept_failures,
initial_parameters=initial_parameters,
fit_metrics_aggregation_fn=fit_metrics_aggregation_fn,
evaluate_metrics_aggregation_fn=evaluate_metrics_aggregation_fn)
# if __name__ == "__main__":
# net = GcnMoleculeNet(feat_dim=8,
# num_categories=2,
# hidden_dim=32,
# node_embedding_dim=32,
# dropout=0.3,
# readout_hidden_dim=64,
# graph_embedding_dim=64)
# # strategy = fl.server.strategy.FedAdagrad()
# params = get_parameters(net)
# print("=======params=======", params)