목록Research (24)
UOMOP
DeepJSCC(My code)_batch:64, lr:0.0001, R:1/6
EPOCH : 1000
DeepJSCC(My code)_batch:64, lr:0.0005, R:1/6
[21.4572, 28.5796, 30.3619] 기존에는 넘지 못했던 30을 넘음. epoch은 1000
DeepJSCC(My code)_batch:128, lr:0.001, R:1/6
배치사이즈를 128로 늘리니 high SNR에서의 성능 나옴.
DeepJSCC(My code)_batch:32, lr:0.001, R:1/6
import cv2 import math import torch import torchvision from fractions import Fraction import numpy as np import torch.nn as nn import torch.optim as optim import torch.nn.functional as f import matplotlib.pyplot as plt import torchvision.transforms as tr from torchvision import datasets from torch.utils.data import DataLoader, Dataset device = torch.device("cuda:0" if torch.cuda.is_available() e..
Proposed(H:256, L:256) vs Conventional AE(512)
import cv2 import math import torch import torchvision import numpy as np import torch.nn as nn import torch.optim as optim import torch.nn.functional as f import matplotlib.pyplot as plt import torchvision.transforms as tr from torchvision import datasets from torch.utils.data import DataLoader, Dataset device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu") print(device) args =..