I have a CNN model to classify the MNIST digits problem. The model has 10 output nodes, used CrossEntropyLoss
as a loss function and Adam
optimizer. The structure of the model is given below
class CNN(nn.Module):
def __init__(self):
super(CNN, self).__init__()
self.conv1 = nn.Sequential(
nn.Conv2d(
in_channels=1,
out_channels=16,
kernel_size=5,
stride=1,
padding=2,
),
nn.ReLU(),
nn.MaxPool2d(kernel_size=2),
)
self.conv2 = nn.Sequential(
nn.Conv2d(16, 32, 5, 1, 2),
nn.ReLU(),
nn.MaxPool2d(2),
)
# fully connected layer, output 10 classes
self.out = nn.Linear(32 * 7 * 7, 10)
# self.softmax = torch.nn.Softmax(dim=1)
def forward(self, x):
x = self.conv1(x)
x = self.conv2(x)
# flatten the output of conv2 to (batch_size, 32 * 7 * 7)
x = x.view(x.size(0), -1)
output = self.out(x)
# output = self.softmax(output)
return output, x # return x for visualization
As, we can use CNN to predict the house price prediction
from images. Now, I want to consider this classification problem as a regression problem. I want to predict the digit as a number (whatever the model predicts, i.e, 1.0, 0.5, 8.5, etc). To do this I need only 1 output neuron
instead of 10
.
If I change this line for the expected output (1 regression output), I am getting an error (as the target labels are from 0 to 9).
# fully connected layer, 1 output
self.out = nn.Linear(32 * 7 * 7, 1)
Error
IndexError: Target 2 is out of bounds.
Could you tell me what I have to do, if I want to predict only one number as digits? Moreover, do I need to change the loss function to MSELoss
and optimizer to SGD
?
The data-loader
def data_loaders():
train_data = datasets.MNIST(
root = 'data',
train = True,
transform = transforms.ToTensor(),
download = True,
)
test_data = datasets.MNIST(
root = 'data',
train = False,
transform = transforms.ToTensor()
)
train_test_loaders = {
'train' : torch.utils.data.DataLoader(train_data,
batch_size=100,
shuffle=True,
num_workers=1),
'test' : torch.utils.data.DataLoader(test_data,
batch_size=100,
shuffle=True,
num_workers=1),
}
return train_test_loaders
Later, at the time of training
def train(NB_EPOCS, model, loaders):
model.train()
# Train the model
total_step = len(loaders['train'])
for epoch in range(NB_EPOCS):
for i, (images, labels) in enumerate(loaders['train']):
b_x = Variable(images) # batch x
b_y = Variable(labels) # batch y
print(np.shape(b_x), np.shape(b_y))
output = model(b_x)[0]
loss = criterion(output, b_y)
The shape of the b_x
and b_y
is
torch.Size([100, 1, 28, 28]) torch.Size([100])