VSR dataset

1.视频超分的数据集
https://cxyzjd.com/article/qq_41554005/116466156

2.Question:

OSError: libcurand.so.10: cannot open shared object file: No such file or directory

调整版本的型号进行匹配

python demo/restoration_video_demo.py ./configs/restorers/basicvsr_reds4.py http://download.openmmlab.com/mmediting/restorers/basicvsr/basicvsr_red4_20120409-0e599677.pth data/Vid4/BIx4/calendar/ ./output

3. 将某个文件夹下的文件复制到另一个文件夹下面
   cp -Rf /common-data/fei.li/downloads/train/train_sharp_bicubic/X4/* /common-data/fei.li/downloads/train/train_sharp_bicubic/

3.新idea的相关网页
https://zhuanlan.zhihu.com/p/519425855
https://github.com/PaddlePaddle/PaddleGAN/blob/develop/docs/zh_CN/tutorials/video_super_resolution.md
https://zhuanlan.zhihu.com/p/519425855
https://zhuanlan.zhihu.com/p/407493555
https://zhuanlan.zhihu.com/p/35709485
https://zhuanlan.zhihu.com/p/128970124
https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123720749.pdf
https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Recurrent_Dynamic_Embedding_for_Video_Object_Segmentation_CVPR_2022_paper.pdf
https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_Blind2Unblind_Self-Supervised_Image_Denoising_With_Visible_Blind_Spots_CVPR_2022_paper.pdf
https://openaccess.thecvf.com/content/CVPR2021/papers/Chen_Learning_Continuous_Image_Representation_With_Local_Implicit_Image_Function_CVPR_2021_paper.pdf
https://arxiv.org/pdf/2204.13348.pdf
https://openaccess.thecvf.com/content/CVPR2022/papers/Xia_Vision_Transformer_With_Deformable_Attention_CVPR_2022_paper.pdf
https://openaccess.thecvf.com/content/CVPR2022/papers/Zheng_CLRNet_Cross_Layer_Refinement_Network_for_Lane_Detection_CVPR_2022_paper.pdf

5. vim ~/.bashrc
6. history |grep basic

#!/bin/bash
source ~/.bashrc
cd /data/mmediting-master/
PORT=29500
chomd 777 ./tools/dist_train.sh
./tools/dist_train.sh configs/restorers/basicvsr/basicvsr_reds4.py 2 >log.txt

#!/bin/bash
source ~/.bashrc
cd /data/mmediting-master/
PORT=29500
python tools/train.py configs/restorers/basicvsr/basicvsr_reds4.py

log.txt

RuntimeError : work = default_pg.broadcast([tensor], opts)NCCL error in: …/torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:957, invalid usage, NCCL version 21.0.3
ncclInvalidUsage: This usually reflects invalid usage of NCCL library (such as too many async ops, too many collectives at once, mixing streams in a group, etc).

RuntimeError: NCCL error in: …/torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:957, invalid usage, NCCL version 21.0.3
ncclInvalidUsage: This usually reflects invalid usage of NCCL library (such as too many async ops, too many collectives at once, mixing streams in a group, etc).
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 26330) of binary: /opt/conda/bin/python

torch.distributed.elastic.multiprocessing.errors.ChildFailedError:

https://github.com/open-mmlab/mmcv/blob/master/mmcv/runner/dist_utils.py#L64

如果设置是对的可能把NCCL的debug开开看看吧 https://pytorch.org/docs/stable/distributed.html#other-nccl-environment-variables
./tools/dist_train.sh configs/restorers/basicvsr/basicvsr_reds4.py 2