XuecaiHu/Meta-SR-Pytorch

Issues

• Pre-training model selection for testing

  #82 opened a year ago by daydreamer2024
  0

• 你好，cols = nn.functional.unfold(up_x.permute(0, 2, 3, 1), self.kernel_size, padding=1)，该咋改呀？

  #81 opened 2 years ago by FreshmanMa
  1

• 你好，能帮忙指点下吗? 改成3d 后 pos_mat_small 维度不是Scale x Scale x Scale x 3的维度？ h_offset这需要改吗？

  #80 opened 2 years ago by FreshmanMa
  3

• 请问，想改成 针对3d数据，该怎么改？ 比如（batch，C， h, w, d），超分到（batch, C, H, W, D）。

  #79 opened 2 years ago by FreshmanMa
  2

• 如何将MetaUpSampler 改成适用于3d图像的上采样？

  #78 opened 2 years ago by FreshmanMa
  0

• Pretrained models

  #77 opened 2 years ago by caojiezhang
  0

• 请问怎样运行 geberate_LR_metasr_X1_X4.m 文件？

  #76 opened 2 years ago by ZYQii
  0

• Higher PSNR when i use pretrained model?

  #75 opened 3 years ago by suxi-du
  0

• RuntimeError: cuda runtime error (2)

  #69 opened 3 years ago by jacedang
  5

• dataloader error, help plz~

  #74 opened 3 years ago by RJJxp
  0

• meta-learning for weight prediction

  #73 opened 3 years ago by Jillian2017
  0

• Testing directories

  #71 opened 3 years ago by dflateau
  3

• Trying to train Meta-RCAN but failed

  #70 opened 3 years ago by jacedang
  2

• rewrite dataloader for more recnt Pytorch

  #72 opened 3 years ago by dflateau
  0

• Meta-upscale的实现

  #68 opened 3 years ago by xjhcassy
  3

• Question about the code

  #58 opened 4 years ago by codyshen0000
  4

• Low PSNR when evaluate the reproduction model and pretrained model

  #46 opened 5 years ago by hcwang95
  16

• Meta-USR代码开源了吗？

  #61 opened 4 years ago by zhangyi-student
  2

• 请问输入矩阵为什么需要mask

  #67 opened 4 years ago by zhangyi-student
  0

• Meta-Upscale Module

  #64 opened 4 years ago by zhangyi-student
  0

• meta-upscale的输入

  #66 opened 4 years ago by xjhcassy
  0

• meta-upscale

  #65 opened 4 years ago by xjhcassy
  0

• Have you debugged it yet？

  #63 opened 4 years ago by upczhaolf
  0

• 训练metaedsr跑完一个epoch就不动了

  #35 opened 5 years ago by duanyucat
  3

• A bug when change the loss function

  #62 opened 4 years ago by LILIXIYA
  0

• “ ValueError：只能将一个元素张量转换为Python标量”

  #60 opened 4 years ago by LILIXIYA
  0

• ValueError: only one element tensors can be converted to Python scalars

  #59 opened 4 years ago by LILIXIYA
  1

• How to solve this kind of mistake？ RuntimeError: cuda runtime error

  #41 opened 5 years ago by ClearPurely
  1

• about the meaning of get_model function

  #38 opened 4 years ago by ZiaZhang
  1

• How to test by DIV2K(val)

  #37 opened 4 years ago by robinamber
  1

• about the last commit

  #39 opened 4 years ago by xkeys
  3

• how to set correct patch_size

  #43 opened 4 years ago by robinaman
  2

• What does the parameter N mean?

  #45 opened 4 years ago by astro-fits
  1

• 数据预处理里面

  #48 opened 4 years ago by djy-tsinghua
  1

• need downscale labels when testing?/测试的时候需要相对应的下采样标签吗？

  #49 opened 4 years ago by Darrenxc
  1

• 可以重新发一下测试集吗

  #56 opened 4 years ago by shentucc
  1

• meta upscale module的输入

  #52 opened 4 years ago by liuye123321
  1

• while(pos_mat[i][0][0]<= 1e-6 and i<h) 这里感觉应该是大于等于？

  #47 opened 4 years ago by Liuzhihaolzh
  1

• 想请问meta.py代码的问题

  #50 opened 4 years ago by Rainyliuliu
  1

• RuntimeError: "floor" not implemented for 'torch.LongTensor'如何解决

  #44 opened 4 years ago by mdswyz
  4

• An Question

  #57 opened 4 years ago by fantasysponge
  0

• The shape of hr and sr is different when calculating psnr or ssim?

  #51 opened 4 years ago by reddandelion217
  1

• RuntimeError: input and target shapes do not match: input [1 x 3 x 155 x 155], target [1 x 4 x 155 x 155] at /opt/conda/conda-bld/pytorch_1535490206202/work/aten/src/THCUNN/generic/AbsCriterion.cu:12

  #55 opened 4 years ago by QiaoSiBo
  0

• scale设置成1，，能提高图像的质量吗？

  #54 opened 5 years ago by QiaoSiBo
  0

• Calculate PSNR

  #53 opened 5 years ago by deepak622
  1

• Is the data wrong?

  #33 opened 5 years ago by splinter21
  2

• forward() missing 1 required positional argument: 'pos_mat'

  #42 opened 5 years ago by aligoglos
  2

• About the LOSS，

  #36 opened 5 years ago by robinaman
  0

• How to continue training from interrupted epochs

  #34 opened 5 years ago by robinaman
  2

• Does the scale over 4x work?

  #32 opened 5 years ago by jason71995
  2