Use torch.jit.trace to wrap all the training code

1

Hello, guys. I seems that torchscript cannot wrap all the training code. Is there any plan to let jit.trace / jit.script to support loss.backward() and optmizer.step() in the future?

import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

class Net(nn.Module):

def __init__(self):
    super(Net, self).__init__()
    self.conv1 = nn.Conv2d(1, 6, 3)
    self.conv2 = nn.Conv2d(6, 16, 3)
    self.fc1 = nn.Linear(16 * 6 * 6, 120)  # 6*6 from image dimension 
    self.fc2 = nn.Linear(120, 84)
    self.fc3 = nn.Linear(84, 10)

def forward(self, x):
    x = F.max_pool2d(F.relu(self.conv1(x)), (2, 2))
    x = F.max_pool2d(F.relu(self.conv2(x)), 2)
    x = x.view(-1, self.num_flat_features(x))
    x = F.relu(self.fc1(x))
    x = F.relu(self.fc2(x))
    x = self.fc3(x)
    return x

def num_flat_features(self, x):
    size = x.size()[1:]  # all dimensions except the batch dimension
    num_features = 1
    for s in size:
        num_features *= s
    return num_features

net = Net()
target = torch.randn(1, 10)
data = torch.randn(1, 1, 32, 32)
optimizer = optim.SGD(net.parameters(), lr=0.01)

@torch.jit.script
def train_net(data, target):
out = net(data)
criterion = nn.MSELoss()
loss = criterion(out, target)
optimizer.zero_grad()
loss.backward()
optimizer.step()
print(loss)

for _ in range(10):
train_net(data, target)