PyTorch 유용한 함수들
Pytorch로 시작하는 딥러닝 입문 (2/2)
Useful Methods
import torch
# expand - 주어진 tensor를 복사후 concate
x = torch.FloatTensor([[[1, 2]],
[[3, 4]]])
x.size() # torch.Size([2, 1, 2])
y = x.expand(*[2, 3, 2])
y.size() # torch.size([2, 3, 2])
y = torch.cat([x, x, x], dim=1)
y.size() # torch.size([2, 3, 2])
# randperm - 랜덤 함수, index_select와 활용
rt = torch.randperm(10)
rt.size() # torch.Size([10])
print(rt) # tensor([1, 9, 6, 4, 0, 7, 3, 5, 2, 8])
# argmax - 최대값을 가지고 있는 index를 반환
x = torch.randperm(3**3).reshape(3, 3, -1)
y = x.argmax(dim = -1)
y.size() # torch.Size([3, 3])
# topk - topk와 해당 index를 반환
values, indices = torch.topk(x, k = 1, dim = -1)
values.size() # torch.Size([3, 3, 1])
values.squeeze(-1).size() # torch.Size([3, 3])
x.argmax(dim = -1) == indices.squeeze(-1) # True
values, indices = torch.topk(x, k = 2, dim = -1)
values.size() # torch.Size([3, 3, 2])
# 정렬하기1 - topk로 정렬
target_dim = -1
values, indices = torch.topk(x, k = x.size(target_dim), largest = True)
# 정렬하기2 - sort로 정렬
values, indices = torch.sort(x, dim = -1, descending = True)
# masked_fill - mask 처리한 곳(True) 채우기
x = torch.FloatTensor([i for i in range(3**2)]).reshape(3, -1)
x.size() # torch.Size([3, 3])
mask = x > 4
print(mask) # tensor([[False, False, False],
# [False, False, True],
# [ True, True, True]])
y = x.masked_fill(mask, value = -1)
print(y) # tensor([[ 0., 1., 2.],
# [ 3., 4., -1.],
# [-1., -1., -1.]])
# Ones and Zeros
torch.ones(2, 3)
torch.zeros(2, 3)
# (기존 Tensor와 연산하기 위해선 아래 방법이 더 유용함)
x = torch.FloatTensor([[1, 2, 3],
[4, 5, 6]])
x.size() # torch.Size([2, 3])
torch.ones_like(x) # torch.ones(2, 3)
torch.zeros_like(x) # torch.zeros(2, 3)
