import torch
from torch.utils.data import Dataset
from torch.utils.data import DataLoader
from torchvision import transforms
from torchvision import datasets
import torch.nn as nn
import torch.optim as optim
import scipy.io as scio
torch.set_default_tensor_type(torch.DoubleTensor)
num_time_steps = 50
input_size = 32
hidden_size = 16
output_size = 1
lr = 0.01
class Net(nn.Module):
def __init__(self):
super(Net,self).__init__()
self.rnn = nn.RNN(
input_size=input_size,
hidden_size=hidden_size,
num_layers=1,
batch_first=True,
)
self.linear = nn.Linear(hidden_size,output_size)
def forward(self,x,hidden_prev):
out,hidden_prev = self.rnn(x,hidden_prev)
out = out.view(-1,hidden_size)
out = self.linear(out)
out = out.unsqueeze(dim=0)
return out,hidden_prev
class DiabetesDataset(Dataset):
def __init__(self):
self.x_data = scio.loadmat("tezhengxx.mat")['tezhengxx'].T
self.y_data = scio.loadmat("biaoqian.mat")['shuzi'].T
self.len = len(self.x_data)
def __getitem__(self, index):
return self.x_data[index],self.y_data[index]
def __len__(self):
return self.len
dataset = DiabetesDataset()
train_loader = DataLoader(dataset=dataset,batch_size=1,shuffle=True,num_workers=0)
if __name__ == '__main__':
model = Net()
model = model.double()
criterion = nn.MSELoss()
optimizer = optim.Adam(model.parameters(), lr)
hidden_prev = torch.zeros(1, 1, hidden_size)
for epoch in range(400):
for i,data in enumerate(train_loader,0):
# print(i)
data[0] = data[0].view(1,1,32)
#1 * 32 1 * 1
data[0] = data[0].to(torch.double)
data[1] = data[1].to(torch.double)
output,hidden_prev = model(data[0],hidden_prev)
hidden_prev = hidden_prev.detach()
data[1] = data[1].unsqueeze(dim=0)
loss = criterion(output,data[1])
model.zero_grad()
loss.backward()
optimizer.step()
if epoch % 10 == 0:
print("Iteration:{} loss{}".format(epoch,loss.item()))