Add device set.
梁鸿
|
Before Width: | Height: | Size: 2.8 KiB After Width: | Height: | Size: 2.8 KiB |
|
Before Width: | Height: | Size: 191 B After Width: | Height: | Size: 191 B |
|
Before Width: | Height: | Size: 187 B After Width: | Height: | Size: 187 B |
|
Before Width: | Height: | Size: 1018 B After Width: | Height: | Size: 1020 B |
|
Before Width: | Height: | Size: 2.0 KiB After Width: | Height: | Size: 2.0 KiB |
|
Before Width: | Height: | Size: 1.8 KiB After Width: | Height: | Size: 1.8 KiB |
|
Before Width: | Height: | Size: 84 B After Width: | Height: | Size: 84 B |
|
Before Width: | Height: | Size: 2.4 KiB After Width: | Height: | Size: 2.4 KiB |
|
Before Width: | Height: | Size: 1.1 KiB After Width: | Height: | Size: 1.1 KiB |
|
|
@ -13,8 +13,8 @@ seed = 4321
|
|||
torch.manual_seed(seed)
|
||||
torch.cuda.manual_seed_all(seed)
|
||||
|
||||
# Device configuration
|
||||
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
||||
device = torch.device("mps")
|
||||
|
||||
# Hyper-parameters
|
||||
num_epochs = 1
|
||||
|
|
@ -51,25 +51,25 @@ class ConvNet(nn.Module):
|
|||
return x
|
||||
|
||||
def printFector(self, x, label):
|
||||
show.DumpTensorToImage(x.view(-1, x.shape[2], x.shape[3]), "input_image.png")
|
||||
show.DumpTensorToImage(x.view(-1, x.shape[2], x.shape[3]).cpu(), "input_image.png")
|
||||
x = self.conv1(x)
|
||||
w = self.conv1.weight
|
||||
show.DumpTensorToImage(w.view(-1, w.shape[2], w.shape[3]), "conv1_weight.png")
|
||||
show.DumpTensorToImage(w.view(-1, w.shape[2], w.shape[3]).cpu(), "conv1_weight.png")
|
||||
|
||||
show.DumpTensorToImage(x.view(-1, x.shape[2], x.shape[3]), "conv1_output.png")
|
||||
show.DumpTensorToImage(x.view(-1, x.shape[2], x.shape[3]).cpu(), "conv1_output.png")
|
||||
x = self.pool(F.relu(x))
|
||||
x = self.conv2(x)
|
||||
w = self.conv2.weight
|
||||
show.DumpTensorToImage(w.view(-1, w.shape[2], w.shape[3]), "conv2_weight.png")
|
||||
show.DumpTensorToImage(w.view(-1, w.shape[2], w.shape[3]).cpu(), "conv2_weight.png")
|
||||
|
||||
show.DumpTensorToImage(x.view(-1, x.shape[2], x.shape[3]), "conv2_output.png")
|
||||
show.DumpTensorToImage(x.view(-1, x.shape[2], x.shape[3]).cpu(), "conv2_output.png")
|
||||
x = self.pool(F.relu(x))
|
||||
x = x.view(-1, 8 * 5 * 5)
|
||||
x = self.fc1(x)
|
||||
|
||||
show.DumpTensorToImage(self.fc1.weight.view(-1, 10, 10).permute(2, 0, 1), "fc_weight.png")
|
||||
show.DumpTensorToImage(self.fc1.weight.view(-1, 10, 10).permute(2, 0, 1).cpu(), "fc_weight.png")
|
||||
|
||||
show.DumpTensorToImage(x.view(-1), "fc_output.png")
|
||||
show.DumpTensorToImage(x.view(-1).cpu(), "fc_output.png")
|
||||
|
||||
criterion = nn.CrossEntropyLoss()
|
||||
loss = criterion(x, label)
|
||||
|
|
@ -77,10 +77,10 @@ class ConvNet(nn.Module):
|
|||
loss.backward()
|
||||
|
||||
w = self.conv1.weight.grad
|
||||
show.DumpTensorToImage(w.view(-1, w.shape[2], w.shape[3]), "conv1_weight_grad.png")
|
||||
show.DumpTensorToImage(w.view(-1, w.shape[2], w.shape[3]).cpu(), "conv1_weight_grad.png")
|
||||
w = self.conv2.weight.grad
|
||||
show.DumpTensorToImage(w.view(-1, w.shape[2], w.shape[3]), "conv2_weight_grad.png")
|
||||
show.DumpTensorToImage(self.fc1.weight.grad.view(-1, 10, 10).permute(2, 0, 1), "fc_weight_grad.png")
|
||||
show.DumpTensorToImage(w.view(-1, w.shape[2], w.shape[3]).cpu(), "conv2_weight_grad.png")
|
||||
show.DumpTensorToImage(self.fc1.weight.grad.view(-1, 10, 10).permute(2, 0, 1).cpu(), "fc_weight_grad.png")
|
||||
|
||||
|
||||
model = ConvNet().to(device)
|
||||
|
|
@ -109,6 +109,8 @@ for epoch in range(num_epochs):
|
|||
print(f"Epoch [{epoch+1}/{num_epochs}], Step [{i+1}/{n_total_steps}], Loss: {loss.item():.4f}")
|
||||
|
||||
for images, labels in test_loader:
|
||||
images = images.to(device)
|
||||
labels = labels.to(device)
|
||||
model.printFector(images, labels)
|
||||
break
|
||||
|
||||
|
|
|
|||