def __init__(self, H, dataset, transform=None):
super(CustomDataset, self).__init__('.', transform, None, None)
adj = H.to_sparse()
edge_index = adj._indices()
data_list = []
for i in range(0, len(dataset), 2):
data = Data(edge_index=edge_index)
data.x = dataset[i].t()
data.y = dataset[i + 1].t()
data_list.append(data)
self.data, self.slices = self.collate(data_list)
def __init__(self, H, dataset, x, fac_num, transform=None):
super(CustomDataset, self).__init__('.', transform, None, None)
adj = H.to_sparse()
edge_index = adj._indices()
data_list = []
for row in dataset:
data = Data(edge_index=edge_index)
row = torch.cat([row.unsqueeze(0).t(),
torch.zeros(fac_num, 1)],
dim=0)
data.x = row
data.y = x.t()
data_list.append(data)
self.data, self.slices = self.collate(data_list)
def sendData(client, inputData, startLayer, endLayer): data=Data(inputData, startLayer, endLayer) str=pickle.dumps(data) client.send(len(data).to_bytes(length=6, byteorder='big')) client.send(data)
def sendData(server, inputData, startLayer, endLayer):
data = Data(inputData, startLayer, endLayer)
str = pickle.dumps(data)
server.send(len(str).to_bytes(length=6, byteorder='big'))
server.send(str)
return len(self.images)
def __getitem__(self, index):
image_index = self.images[index]
img_path = os.path.join(self.root_dir, image_index)
img = io.imread(img_path)
label = img_path.split('\\')[-1].split('.')[0]
sample = {'image': img, 'label': label}
if self.transform:
sample = self.transform(sample)
return sample
if __name__ == '__main__':
train_data = Data('./write/train', transform=None)
train_loader = DataLoader(train_data, batch_size=BATCH_SIZE, shuffle=True)
for i_batch, batch_data in enumerate(train_loader):
print(i_batch)
print(batch_data['image'] / 255)
test_data = Data('./write/test', transform=None)
test_loader = DataLoader(test_data, batch_size=BATCH_SIZE, shuffle=True)
for i_batch, batch_data in enumerate(train_loader):
print(i_batch)
print(
batch_data['image'].size()) #batch_data['image']对应train_data.data
test_x = torch.unsqueeze(test_data.data, dim=1).type(
torch.FloatTensor)[:2000] / 255. #print(batch_data['image'] / 255)
test_y = test_data.test_labels[:2000]
