[Pytorch] Basic Neural Network

Neural Network 기본틀 저장

import torch
import torch.nn.functional as F

# 모델선언
class NeuralNet(torch.nn.Module):
    def __init__(self, input_size, hidden_size):
        super(NeuralNet, self).__init__()
        self.input_size = input_size
        self.hidden_size = hidden_size
        self.linear_1 = torch.nn.Linear(self.input_size, self.hidden_size)
        self.linear_2 = torch.nn.Linear(self.hidden_size, 1)
    
    def forward(self, input_tensor):
        linear1 = F.relu(self.linear_1(input_tensor))
        output = F.sigmoid(self.linear_2(linear1))
        return output
        
# hypermarameter 선언
model = NeuralNet(2,5)
learning_rate = 0.03
criterion = torch.nn.BCELoss()
optimizer = torch.optim.SGD(model.parameters(), lr = learning_rate)

# 샘플 데이터
x_test = torch.FloatTensor([1, 1])
y_test = torch.FloatTensor([-1])

# 출력
print(f"Model Output : {model(x_test)}")
print(f"Calc Loss Value : {criterion(model(x_test), y_test)}")


"""
출력
Model Output : tensor([0.5407], grad_fn=<SigmoidBackward>)
Calc Loss Value : 0.9411621689796448
"""

 

DNN 모델 저장

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.fc1 = nn.Linear(784, 256)
        self.fc2 = nn.Linear(256, 128)
        self.fc3 = nn.Linear(128, 10)
        
    def forward(self, x):
        x = x.view(-1, 784)
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x
    
model = Net().to(device)
optimizer = optim.Adagrad(model.parameters(), lr = 0.01)