| English

Video-subtitle-remover (VSR) AI

• ****

- AI

-

---

• QQ806152575
• conda

---

Windows GPUv1.1.0GPU

• : vsrwindowsgpu_v1.1.0.zip vsr1

• Google Drive: vsrwindowsgpu_v1.1.0.zip

Nvidia(AMD)

• GUI

Nvidia

GPUGTX 1060

CPU: AVX

1. Miniconda

• Windows: Miniconda3-py384.11.0-Windows-x8664.exe

• Linux: Miniconda3-py384.11.0-Linux-x8664.sh

2.

1 shell cd <>

Dtoolsvideo-subtitle-remover cd D:/tools/video-subtitle-remover-main

2conda shell conda create -n videoEnv python=3.8

shell conda activate videoEnv

3.

python 3.8+conda ()

• CUDAcuDNN

Linux

(1) CUDA 11.7

wget https://developer.download.nvidia.com/compute/cuda/11.7.0/localinstallers/cuda11.7.0515.43.04linux.run

(2) CUDA 11.7

sudo sh cuda11.7.0515.43.04linux.run

1. accept

2. CUDA Toolkit 11.7nvidiaDrivernvidiadriverinstall

3.

~/.bashrc

# CUDA
  export PATH=/usr/local/cuda-11.7/bin${PATH:+:${PATH}}
  export LDLIBRARYPATH=/usr/local/cuda-11.7/lib64${LDLIBRARYPATH:+:${LDLIBRARYPATH}}

source ~/.bashrc

(3) cuDNN 8.4.1

<a href="https://pan.baidu.com/s/1GdpSVzWfX1G7zCuqz6YYA">cudnn-linux-x8664-8.4.1.50cuda11.6-archive.tar.xz 57mg

cudnn-linux-x8664-8.4.1.50cuda11.6-archive.tar.xz

(4) cuDNN 8.4.1

 tar -xf cudnn-linux-x8664-8.4.1.50cuda11.6-archive.tar.xz
   mv cudnn-linux-x8664-8.4.1.50cuda11.6-archive cuda
   sudo cp ./cuda/include/* /usr/local/cuda-11.7/include/
   sudo cp ./cuda/lib/* /usr/local/cuda-11.7/lib64/
   sudo chmod a+r /usr/local/cuda-11.7/lib64/*
   sudo chmod a+r /usr/local/cuda-11.7/include/*

Windows

(1) CUDA 11.7

cuda11.7.0516.01_windows.exe

(2) CUDA 11.7

(3) cuDNN 8.2.4

cudnn-windows-x64-v8.2.4.15.zip

(4) cuDNN 8.2.4

cuDNNcudabin, include, libC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.7\

• GPUPaddlepaddle:

  • windows:

  shell python -m pip install paddlepaddle-gpu==2.4.2.post117 -f https://www.paddlepaddle.org.cn/whl/windows/mkl/avx/stable.html - Linux:

  shell python -m pip install paddlepaddle-gpu==2.4.2.post117 -f https://www.paddlepaddle.org.cn/whl/linux/mkl/avx/stable.html

• GPUPytorch:

shell conda install pytorch==2.0.1 torchvision==0.16.0 pytorch-cuda=11.8 -c pytorch -c nvidia

shell pip install torch==2.0.1 torchvision==0.16.0 --index-url https://download.pytorch.org/whl/cu118

• :

shell pip install -r requirements.txt

4.

shell python gui.py

• (CLI)

shell python ./backend/main.py

backend/config.py python MODE = InpaintMode.STTN # STTN STTN_SKIP_DETECTION = True #

backend/config.py

• InpaintMode.STTN
• InpaintMode.LAMA
• InpaintMode.PROPAINTER
  

• STTN

```python MODE = InpaintMode.STTN # STTN

,

STTNNEIGHBORSTRIDE = 10

,

STTNREFERENCELENGTH = 10

STTN

STTNMAXLOADNUMSTTNNEIGHBORSTRIDESTTNREFERENCE_LENGTH

STTNMAXLOADNUM = 30 - LAMA python MODE = InpaintMode.LAMA # STTN LAMASUPER_FAST = False # ```

designbackend/tools/train

1. CondaHTTPError

.condarc(C:/Users/<>)

https://zhuanlan.zhihu.com/p/260034241

1. 7z

7-zip

1. 4090cuda 11.7

cuda 11.8

shell pip install torch==2.1.0 torchvision==0.16.0 --index-url https://download.pytorch.org/whl/cu118

| | | | |---------------------------|------------| --- | | V | 400.00 RMB | | | Jenkit | 200.00 RMB | | | | 100.00 RMB | | | | 100.00 RMB | | | wr | 100.00 RMB | | | | 50.00 RMB | | | Tshuang | 20.00 RMB | | | | 15.00 RMB | | | | 10.00 RMB | | | | 8.80 RMB | | | | 6.00 RMB | | | Stephen | 2.00 RMB | | | Leo | 1.00 RMB | |