DocDiff

DocDiff copied to clipboard

什么时候开发多卡训练啊，扩散的多卡训练怎么搞

1

monster-star

Get the download link for the dataset（Document Deblurring dataset）

1

Hi, I would like to get the download links for the dataset（Document Deblurring dataset）. The link provided for the paper Convolutional neural networks for direct text deblurring is not working!...

mingheyuemankong

是我的问题还是其他的原因？

minghaoZZZ

推理过程中的问题

5

我采用这个模型进行印章消除的训练，训练过程中一共50000个iteration，在训练过程中输出的图片能够达到我想要的效果，但是在推理的过程中生成的图片并不能达到我想要的效果。  这是训练过程中的图片 我采用第50000个iteration产生的模型进行推理，得到的效果图如图所示  请问为什么会产生这样的问题？我应该怎么进行调整呢？

xiaoguzainuli

你好我想用这里的模型进行推理，但是报错了

1

报错是RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 176 but got size 175 for tensor number 1 in the list. 是什么问题呢？ 如果我想推理图片应该如何去弄

zkq-afk

the website https://www.aibupt.com/ seems like down 

fgogrrore4ft3w

The issue of poor performance in the deblurring task with a custom dataset.

2

I would like to ask about my current task, which involves recognizing blurred shipping labels. The dataset I am using is self-generated using a blur generation algorithm on clear shipping...

ziFan99

single blurred image inference

1

Is there code for inference for individually blurred images? Or does the inference need to be for both the blurred image and Groundtruth？

ziFan99

@Royalvice

PhamLongVu