pytorch-vdsr
pytorch-vdsr copied to clipboard
twtygqyy
VDSR (CVPR2016) pytorch implementation
只有把dataloader里的num_workers改成0，程序才能跑起来，不然就会死锁在 for iteration, batch in enumerate(training_data_loader, 1)这一句，网上给的方法是改成0，但会导致训练变慢，请问有没有其他的办法呢？谢谢啦 网上还有的说是torch和opencv冲突了，但是我并没有在dataloader里找到用了cv2的地方，而且加了cv2.setNumThreads(0) cv2.ocl.setUseOpenCL(False) 这两句也没有变化，还是卡在那一句。
你好，请问在Set5测试集中××_scale_2.bmp，××_scale_3.bmp，××_scale_4.bmp 的图片是怎么生成的？ baby_GT.bmp baby_GT_scale_2.bmp baby_GT_scale_3.bmp baby_GT_scale_4.bmp
Hey everyone, I have played around with the code and produced the outcome below/attached, with those obvious artifacts. I'm not able to find the source of the probem for this.....
Hi, thank you for your sharing. I have two questions about your code in vdsr.py: 1, Why do you set the bias term as False? The official VDSR model contains...
Hello I have problem while loading the pretrained model. It give this error. `UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 918: ordinal not in range(128)` I am using...
关于测试的困惑
Hi, 对于下面的代码片段比较困惑，为什么在将rgb图片转换为ycbcr时要先除以255，最后为了得到y通道还要乘以255呢？ 我不明白这有什么意义，按照矩阵运算，除以255然后乘以255.应该没差别才对。 但是事实是，您的做法似乎保证了非负。 打扰了。 ``` clear;close all; %% settings folder = 'Set5'; %% generate data filepaths = []; filepaths = [filepaths; dir(fullfile(folder, '*.bmp'))]; scale = [2, 3, 4]; for...
Hi, I think your implementation here is brilliant, but I have two questions about it: 1. I download model_50_epoch.pth and test it on Set5 data, the PSNR is 37.44 which...
Hello, use your matlab script `generate_train.m` generate h5 file, pytorch load error. Can you guide me? thanks. ``` Namespace(batchSize=128, clip=0.4, cuda=True, lr=0.1, momentum=0.9, nEpochs=50, pretrained='', resume='', seed=0, start_epoch=1, step=10, threads=8,...
Hello, I tested the given model and the model I trained using default parameters you provide. Also, I constructed test dataset using matlab, and check bicubic performance is reproduced as...
T have some problems, can you give me some advice, thanks! 1)run eval.py ,get errors as follows: /home/yang/Software/anaconda3/lib/python3.5/site-packages/torch/serialization.py:316: SourceChangeWarning: source code of class 'torch.nn.parallel.data_parallel.DataParallel' has changed. you can retrieve the...
