How can I get .grad ? .grad is None

autograd
1

I want to get d(model(x))/dx, but following program returns None.

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 32, 3) # 28x28x32 -> 26x26x32
        self.conv2 = nn.Conv2d(32, 64, 3) # 26x26x64 -> 24x24x64 
        self.pool = nn.MaxPool2d(2, 2) # 24x24x64 -> 12x12x64
        self.dropout1 = nn.Dropout2d()
        self.fc1 = nn.Linear(12 * 12 * 64, 128)
        self.dropout2 = nn.Dropout2d()
        self.fc2 = nn.Linear(128, 1)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        x = self.pool(F.relu(self.conv2(x)))
        x = self.dropout1(x)
        x = x.view(-1, 12 * 12 * 64)
        x = F.relu(self.fc1(x))
        x = self.dropout2(x)
        x = self.fc2(x)
        return x

model = Net()
input = torch.randn((1, 1, 28, 28), requires_grad=True)
y = model(input)
y.backward()
print(x.grad) # This is None.

How can I get this grad ?

2

Should that print(x.grad) be print(input.grad)?

I tried the following and it worked.

import torch
import torch.nn as nn
from torch.nn import functional as F

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 32, 3) # 28x28x32 -> 26x26x32
        self.conv2 = nn.Conv2d(32, 64, 3) # 26x26x64 -> 24x24x64 
        self.pool = nn.MaxPool2d(2, 2) # 24x24x64 -> 12x12x64
        self.dropout1 = nn.Dropout2d()
        self.fc1 = nn.Linear(12 * 12 * 64, 128)
        self.dropout2 = nn.Dropout2d()
        self.fc2 = nn.Linear(128, 1)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        x = self.pool(F.relu(self.conv2(x)))
        x = self.dropout1(x)
        x = x.view(-1, 12 * 12 * 64)
        x = F.relu(self.fc1(x))
        x = self.dropout2(x)
        x = self.fc2(x)
        return x

model = Net()
input = torch.randn((1, 1, 28, 28), requires_grad=True)
y = model(input)
y.backward()
print(input.grad)
3

Sorry, It’s stupid mistake.
Thank you.