import matplotlib.pyplot as plt
figure = plt.figure()
num_of_images = 9
for index in range(num_of_images):
plt.subplot(3, 3, index+1)
plt.axis('off')
plt.title("Ground Truth: {}".format(labels[index]))
plt.imshow(images[index].numpy().squeeze(), cmap='gray_r')
Define the network
Let’s define this network:
tensor.view(-1,n), Returns a new tensor with the same data as the self tensor but of a different shape. the size -1 is inferred from other dimensions
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
# input ch, output ch, convolution
# 1 input image channel, 6 output channels, 3x3 square convolution
# kernel
self.conv1 = nn.Conv2d(1, 6, 5)
self.conv2 = nn.Conv2d(6, 16, 5)
# an affine operation: y = Wx + b
self.fc1 = nn.Linear(16 * 5 * 5, 120) # 5*5 from image dimension
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)
self.pool = nn.MaxPool2d(2, 2)
def forward(self, x):
# Max pooling over a (2, 2) window
x = F.relu(self.conv1(x))
x = self.pool(x)
x = F.relu(self.conv2(x))
x = self.pool(x)
#x = x.view(-1, self.num_flat_features(x))
x = x.view(-1, 16 * 5 * 5)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x
#net = Net()
#print(net)
You just have to define the forward function, and the backward function (where gradients are computed) is automatically defined for you using autograd. You can use any of the Tensor operations in the forward function.
Loss Function and Optimization
A loss function takes the (output, target) pair of inputs, and computes a value that estimates how far away the output is from the target. Define loss function as loss=criterion(outputs, labels)
import torch.optim as optim
# loss function
criterion = nn.CrossEntropyLoss()
# Optimization method
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)
Zero the gradient buffers of all parameters and backprops with random gradients:
Train network
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
net.to(device)
for epoch in range(5): # loop over the dataset multiple times
running_loss = 0.0
for i, data in enumerate(trainloader, 0):
# get the inputs; data is a list of [inputs, labels]
images, labels = data[0].to(device), data[1].to(device)
# zero the parameter gradients
optimizer.zero_grad()
# forward + backward + optimize
outputs = net(images)
loss = criterion(outputs, labels)
loss.backward()
optimizer.step()
# print statistics
running_loss += loss.item()
if i % 100 == 99: # print every 100 mini-batches
print('[%d, %5d] loss: %.3f' %
(epoch + 1, i + 1, running_loss / 10))
running_loss = 0.0
print('Finished Training')
We have trained the network for 2 passes over the training dataset. But we need to check if the network has learnt anything at all.
We will check this by predicting the class label that the neural network outputs, and checking it against the ground-truth. If the prediction is correct, we add the sample to the list of correct predictions.
Okay, first step. Let us display an image from the test set to get familiar.
correct = 0
total = 0
with torch.no_grad():
for data in testloader:
images, labels = data[0].to(device), data[1].to(device)
outputs = net(images)
_, predicted = torch.max(outputs.data, 1)
total += labels.size(0)
correct += (predicted == labels).sum().item()
print('Accuracy of the network on the %d test images: %d %%' %(len(testloader.dataset), 100 * correct / total))
Accuracy of the network on the 10000 test images: 98 %
Visualize test results
You need to covert from GPU to Tensor.cpu() . e.g. images.cpu()
figure = plt.figure()
num_of_images = 9
for index in range(num_of_images):
plt.subplot(3, 3, index+1)
plt.axis('off')
plt.title("Predicted: {}".format(predicted[index].item()))
plt.imshow(images[index].cpu().numpy().squeeze(), cmap='gray_r')
Continued Training from Checkpoints
see how we can continue training from the state_dicts we saved during our first training run.
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
continued_net.to(device)
def train_continue(epoch):
running_loss = 0.0
for i, data in enumerate(trainloader, 0):
# get the inputs; data is a list of [inputs, labels]
images, labels = data[0].to(device), data[1].to(device)
# zero the parameter gradients
optimizer.zero_grad()
# forward + backward + optimize
outputs = continued_net(images)
loss = criterion(outputs, labels)
loss.backward()
optimizer.step()
# print statistics
running_loss += loss.item()
if i % 100 == 99: # print every 100 mini-batches
print('[%d, %5d] loss: %.3f' %
(epoch + 1, i + 1, running_loss / 10))
running_loss = 0.0
for epoch in range(6,10): # loop over the dataset multiple times
train_continue(epoch)
print('Finished Training')
# Accuracy of continued training
correct = 0
total = 0
with torch.no_grad():
for data in testloader:
images, labels = data[0].to(device), data[1].to(device)
outputs = continued_net(images)
_, predicted = torch.max(outputs.data, 1)
total += labels.size(0)
correct += (predicted == labels).sum().item()
print('Accuracy of the network on the %d test images: %d %%' %(len(testloader.dataset), 100 * correct / total))
Accuracy of the network on the 10000 test images: 98 %