Upstage AI Lab 2기
2024년 3월 6일 (수) Day_058
온라인강의
nn.Conv2d
https://pytorch.org/docs/stable/generated/torch.nn.Conv2d.html
input shape : (BATCH_SIZE, 3, 256, 256) => in_channels = 3
output shape : (BATCH_SIZE, 64, 256, 256) => out_channels = 64
질문! 그럼 여기서 kernel은 kernel_size * kernel_size * in_channels 가 64개 있는건지?
input image의 채널 수와 in_channels 값을 맞춰줘야 함.
NLLLoss
= negative log likelihood loss
negative log likelihood loss & softmax => CrossEntropy loss와 같음
https://pytorch.org/docs/stable/generated/torch.nn.NLLLoss.html#nllloss
수강생 todo
https://github.com/masalha-alaa/mnist-pytorch/blob/master/MNIST_with_PyTorch.ipynb
'Upstage AI Lab 2기' 카테고리의 다른 글
Upstage AI Lab 2기 [Day058] 수강생 TODO #02 - Pytorch Data Loading (0) | 2024.03.06 |
---|---|
Upstage AI Lab 2기 [Day058] 수강생 TODO #01 - MNIST CNN (0) | 2024.03.06 |
Upstage AI Lab 2기 [Day057] PyTorch 실습 - Tensor Manipulation (1) | 2024.03.05 |
Upstage AI Lab 2기 [Day057] PyTorch (0) | 2024.03.05 |
Upstage AI Lab 2기 [Day057] CH.4 - CNN, RNN, From AlexNet to ChatGPT (2) | 2024.03.05 |