Pytorch从张量文件中读取张量来自磁盘的流训练

如何解决Pytorch从张量文件中读取张量来自磁盘的流训练

我有一些非常大的输入张量，在构建它们时遇到了内存问题，因此我将它们一张一张地读到.pt文件中。当我运行生成并保存文件的脚本时，文件越来越大，因此我假设张量正确保存。这是该代码：

with open(a_sync_save,"ab") as f:
     print("saved")
     torch.save(torch.unsqueeze(torch.cat(tensors,dim=0),f)

我想一次从文件中读取一定数量的这些张量，因为我不想再次遇到内存问题。当我尝试读取保存到该张量的每个张量时我只能设法获得第一个张量。

with open(a_sync_save,"rb") as f:
    for tensor in torch.load(f):
        print(tensor.shape)

这里的输出是第一个张量的形状，然后剧烈退出。

解决方法

这是我用来回答此问题的一些代码。很多都是我正在做的事情，但是它的本质可以供其他遇到与我一样的问题的人使用。

def stream_training(filepath,epochs=100):
    """
    :param filepath: file path of pkl file
    :param epochs: number of epochs to run
    """
    def training(train_dataloader,model_obj,criterion,optimizer):
        for j,data in enumerate(train_dataloader,start=0):
            # get the inputs; data is a list of [inputs,labels]
            inputs,labels = data
            inputs,labels = inputs.cuda(),labels.cuda()
            outputs = model_obj(inputs.float())
            outputs = torch.flatten(outputs)
            loss = criterion(outputs,labels.float())
            print(loss)
            # zero the parameter gradients
            optimizer.zero_grad()
            loss.backward()
            torch.nn.utils.clip_grad_norm_(model_obj.parameters(),max_norm=1)
            optimizer.step()

    tensors = []
    expected_values = []
    model= Model(1000,1,256,1)
    tea.cuda()
    criterion = nn.BCELoss()
    optimizer = optim.Adam(model.parameters(),lr=0.00001,betas=(0.9,0.99999),eps=1e-08,weight_decay=0.001,amsgrad=True)
    for i in range(epochs):
        with (open(filepath,'rb')) as openfile:
            while True:
                try:
                    data_list = pickle.load(openfile)
                    tensors.append(data_list[0])
                    expected_values.append(data_list[1])
                    if len(tensors) % BATCH_SIZE == 0:
                        tensors = torch.cat(tensors,dim=0)
                        tensors = torch.reshape(tensors,(tensors.shape[0],tensors.shape[1],-1))
                        train_loader = make_dataset(tensors,expected_values) # makes a dataloader for the batch that comes in
                        training(train_loader,model,optimizer)  #Performs forward and back prop
                        tensors = [] # washes out the batch to conserve memory on my computer.
                        expected_values = []
                except EOFError:
                    print("This file has finished training")
                    break

这是有趣的模型。

class Model(nn.Module):
    def __init__(self,input_size,output_size,hidden_dim,n_layers):
        super(Model,self).__init__()
        # dimensions
        self.hidden_dim = hidden_dim
        self.n_layers = n_layers

        #Define the layers
        #GRU
        self.gru = nn.GRU(input_size,n_layers,batch_first=True)
        self.fc1 = nn.Linear(hidden_dim,hidden_dim)
        self.bn1 = nn.BatchNorm1d(num_features=hidden_dim)
        self.fc2 = nn.Linear(hidden_dim,hidden_dim)
        self.bn2 = nn.BatchNorm1d(num_features=hidden_dim)
        self.fc3 = nn.Linear(hidden_dim,hidden_dim)
        self.bn3 = nn.BatchNorm1d(num_features=hidden_dim)
        self.fc4 = nn.Linear(hidden_dim,hidden_dim)
        self.bn4 = nn.BatchNorm1d(num_features=hidden_dim)
        self.fc5 = nn.Linear(hidden_dim,hidden_dim)
        self.output = nn.Linear(hidden_dim,output_size)

    def forward(self,x):
        x = x.float()
        x = F.relu(self.gru(x)[1])
        x = x[-1,:,:] # eliminates first dim
        x = F.dropout(x,0.5)
        x = F.relu(self.bn1(self.fc1(x)))
        x = F.dropout(x,0.5)
        x = F.relu(self.bn2(self.fc2(x)))
        x = F.dropout(x,0.5)
        x = F.relu(self.bn3(self.fc3(x)))
        x = F.dropout(x,0.5)
        x = F.relu(self.bn4(self.fc4(x)))
        x = F.dropout(x,0.5)
        x = F.relu(self.fc5(x))
        return torch.sigmoid(self.output(x))

    def init_hidden(self,batch_size):
        hidden = torch.zeros(self.n_layers,batch_size,self.hidden_dim)
        return hidden