maral15 / semanticimagesynthesis Goto Github PK

Load, resize, and crop the pictures and semantic segmentation masks from the specified folders.

The Encoder, Generator and Discriminator structure have been modified, so we need to update the Model and Trainer class to suit them.

1. NVIDIA® GPU drivers version 450.80.02 or higher.

   mar@mar-OMEN-Laptop-15:~$ nvidia-smi
   Sat Oct 29 13:15:53 2022       
   +-----------------------------------------------------------------------------+
   | NVIDIA-SMI 470.141.03   Driver Version: 470.141.03   CUDA Version: 11.4     |
   |-------------------------------+----------------------+----------------------+
   | GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
   | Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
   |                               |                      |               MIG M. |
   |===============================+======================+======================|
   |   0  NVIDIA GeForce ...  Off  | 00000000:01:00.0 Off |                  N/A |
   | N/A   47C    P8     4W /  N/A |    445MiB /  5934MiB |      3%      Default |
   |                               |                      |                  N/A |
   +-------------------------------+----------------------+----------------------+
                                                                              
   +-----------------------------------------------------------------------------+
   | Processes:                                                                  |
   |  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
   |        ID   ID                                                   Usage      |
   |=============================================================================|
   |    0   N/A  N/A      1685      G   /usr/lib/xorg/Xorg                 45MiB |
   |    0   N/A  N/A      2277      G   /usr/lib/xorg/Xorg                 94MiB |
   |    0   N/A  N/A      2455      G   /usr/bin/gnome-shell               53MiB |
   |    0   N/A  N/A      3918      G   /usr/lib/firefox/firefox          240MiB |
   +-----------------------------------------------------------------------------+

   • ⛔ Upgrading NVIDIA ⚠️ NO - NI SE TE VUELVA A OCURRIR ⚠️ ⛔

     1. Download new version:

        Product Type: GeForce
        Product Series: GeForce RTX 20 Series
        Product: GeForce RTX 2060
        Operating System: Linux 64-bit
        Download Type: Production Branch
        Language: English (US)

        sudo bash NVIDIA-Linux-x86_64-515.76.run

     2. Backup with Timeshift

     3. Disable graphical target

        systemctl isolate multi-user.target

     4. Unload NVIDIA drivers

        modprobe -r nvidia-drm

     5. Upgrade it

        sudo bash ~/Downloads/NVIDIA-CUDA/NVIDIA-Linux-x86_64-515.76.run

     6. Start the graphical enviroment again

        systemctl start graphical.target

     7. If it does not work:

        1. Force shutdown
        2. Start the computer and press esc
        3. Restore backup

           sudo timeshift --restore

2. CUDA Toolkit 11.2.2 (March 2021), Versioned Online Documentation

   Operating System: Linux
   Architecture: x86_64
   Distribution: Ubuntu
   Version: 20.04
   Installer Type: runfile (local)

   cd ~/Downloads/NVIDIA-CUDA/
   wget https://developer.download.nvidia.com/compute/cuda/11.2.2/local_installers/cuda_11.2.2_460.32.03_linux.run

   ⚠️ DO NOT select driver installation ⚠️

   mar@mar-OMEN-Laptop-15:~/Downloads/NVIDIA-CUDA$ sudo sh cuda_11.2.2_460.32.03_linux.run 
   ===========
   = Summary =
   ===========
   
   Driver:   Not Selected
   Toolkit:  Installed in /usr/local/cuda-11.2/
   Samples:  Installed in /home/mar/
   
   Please make sure that
    -   PATH includes /usr/local/cuda-11.2/bin
    -   LD_LIBRARY_PATH includes /usr/local/cuda-11.2/lib64, or, add /usr/local/cuda-11.2/lib64 to /etc/ld.so.conf and run ldconfig as root
   
   To uninstall the CUDA Toolkit, run cuda-uninstaller in /usr/local/cuda-11.2/bin
   ***WARNING: Incomplete installation! This installation did not install the CUDA Driver. A driver of version at least ... is required for CUDA 11.2 functionality to work.
   To install the driver using this installer, run the following command, replacing <CudaInstaller> with the name of this run file:
       sudo <CudaInstaller>.run --silent --driver
   
   Logfile is /var/log/cuda-installer.log

3. cuDNN SDK 8.1.0.

   1. Register an NVIDIA developer account and download cudnn
      Download cuDNN v8.1.0 (January 26th, 2021), for CUDA 11.0,11.1 and 11.2 > cuDNN Library for Linux (x86_64)

   2. Check where your CUDA installation is. For most people, it will be /usr/local/cuda/.

   3. Copy the files:

      cd ~/Downloads/NVIDIA-CUDA/
      sudo cp include/cudnn* /usr/local/cuda/include
      sudo cp lib64/libcudnn* /usr/local/cuda/lib64
      sudo chmod a+r /usr/local/cuda/lib64/libcudnn*
      cat /usr/local/cuda/include/cudnn_version.h | grep CUDNN_MAJOR -A 2

      You will probably have to restart the computer for the changes to be applied correctly.

References

• Install TensorFlow with pip

• TensorFlow - Tested build configurations: GPU

• How to install CuDNN?

• TensorFlow: Tested build configurations → Linux/GPU table

• History of Python - Table of versions: Support

SPADE Generator

ResNet block that uses Spatially-Adaptive Normalization.