cswry/SeeSR

Issues

• Training Issues!

  #72 opened 2 months ago by CarlBhy
  2

• FID problem

  #45 opened 7 months ago by striveAgain
  2

• The details about 4.3. Semantics Preservation Test

  #71 opened 2 months ago by Renzhihan
  2

• 关于SD-2.1base模型的下载

  #33 opened 9 months ago by tanhao2509231391
  1

• 复现训练过程中，训练中间的checkpoint测试发现效果并不好，这正常吗？

  #43 opened 7 months ago by beyondbatman-master
  12

• DAPE名字问题

  #70 opened 2 months ago by jolek-jojo
  0

• 请问self.lq_enhancer.eval()里lq_enhancer在哪里定义的

  #69 opened 2 months ago by tayton42
  0

• OSError: Error no file named diffusion_pytorch_model.bin found in directory preset/models/seesr

  #59 opened 6 months ago by EchoXu98
  2

• feature_extractor is miss in sd-turbo

  #27 opened 9 months ago by WenmuZhou
  4

• About the pretrained weights loading problem in training

  #67 opened 4 months ago by llstela
  4

• Note on inference instructions.

  #68 opened 3 months ago by Haradai
  0

• Training time and the total number of used datasset

  #61 opened 6 months ago by Renzhihan
  5

• About sd-turbo results

  #28 opened 9 months ago by Goldenhornk789
  2

• 论文中效果对比有什么意义呢？这么大的模型。。

  #31 opened 9 months ago by gjd2017
  4

• sd-turbo版本的技术上的疑问（请教）

  #40 opened 4 months ago by Feynman1999
  2

• 在自己的数据集上微调SD-Turbo模型

  #65 opened 5 months ago by yumemiso
  1

• The PSNR and SSIM results

  #60 opened 6 months ago by Renzhihan
  4

• The issue with training on multiple GPUs.

  #66 opened 4 months ago by longfafffa
  0

• SD-Turbo results look good, but have color banding/look like painting

  #50 opened 7 months ago by AIisCool
  3

• 为什么一直出现baseball glove呢

  #63 opened 5 months ago by donutn
  2

• About the GPU memory usage

  #62 opened 5 months ago by prokaryote-hgy
  4

• Question

  #64 opened 5 months ago by ShiLiangfan11
  1

• 训练代码 在加载SD的Unet的权重时会报错

  #36 opened 8 months ago by ZifanCui
  10

• 请问会有 'SeeSR-SDXL' 版本吗?

  #58 opened 6 months ago by polor1010
  5

• 多卡GPU使用

  #57 opened 6 months ago by ConanMusk
  2

• inference时cpu占用过高

  #55 opened 7 months ago by xuliu-cyber
  1

• raise EnvironmentError( OSError: Can't load tokenizer for 'bert-base-uncased'

  #34 opened 9 months ago by windcr
  6

• OSError: Can't load tokenizer for 'bert-base-uncased'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'bert-base-uncased' is the correct path to a directory containing all relevant files for a BertTokenizer tokenizer.

  #56 opened 6 months ago by ConanMusk
  3

• DAPE微调细节和收敛问题

  #48 opened 7 months ago by aulaywang
  1

• 关于sd-turbo训练问题

  #46 opened 7 months ago by Yu-xi-99
  2

• Training strategy

  #53 opened 6 months ago by CXNing
  1

• Alternative to fix color?

  #54 opened 7 months ago by AIisCool
  0

• Request: SeeSR with SUPIR

  #52 opened 7 months ago by AIisCool
  0

• Please verify your scheduler_config.json configuration file.

  #51 opened 7 months ago by AIisCool
  0

• What's the difference between from_pretrained_orig() and from_pretrained()? Blurry turbo output?

  #49 opened 7 months ago by octopushugger
  1

• no shuffle when training ?

  #38 opened 8 months ago by inkzk
  1

• compared to PASD

  #47 opened 7 months ago by Feynman1999
  1

• 在处理SeeSR数据的时候，文件夹创建了，但是图像并没有保存到对应文件夹中

  #42 opened 7 months ago by aulaywang
  0

• According to the "TCA module" in the article

  #44 opened 7 months ago by ZhihaoChe
  0

• Cuda out of memory(Training process)

  #37 opened 8 months ago by Shengqi77
  1

• According to the latest code provided, retrain, can't fully reproduce the metrics in the paper, please tell me the results in the paper, are there any settings that need extra attention?

  #41 opened 8 months ago by heart-du
  1

• OOM error happend when using accelerate but training works fine for single GPU

  #39 opened 8 months ago by Synapsess
  0

• Is it Not compatible with Win or Mac?

  #26 opened 9 months ago by R031726
  1

• The super resolution result of the given test image is blur. No errors are reported. I wonder where I may go wrong?

  #23 opened 8 months ago by CXNing
  10

• 请问ram模型里面用到的bert和vit与transformer库里面直接调用的bert和vit有什么区别吗，代码里面的bert和vit好像是单独实现的

  #35 opened 8 months ago by kunfei1
  0

• The file missing

  #32 opened 9 months ago by itgrape
  0

• Why disable shuffle in train dataloader?

  #29 opened 9 months ago by Luciennnnnnn
  2

• Metrics calculating

  #30 opened 9 months ago by fzuzyb
  1

• argument about '--use_ram_encoder'

  #24 opened 9 months ago by striveAgain
  2

• Issues with test_seesr_turbo with solutions

  #25 opened 9 months ago by mashoutsider
  2