cuda_bug.py

# HIGHLY Inspired by the following solution -> https://discuss.huggingface.co/t/error-running-model-in-zerogpu/109819/3
import subprocess
import os
import sys

def install_cuda_toolkit_requirements():
    # CUDA_TOOLKIT_URL = "https://developer.download.nvidia.com/compute/cuda/11.8.0/local_installers/cuda_11.8.0_520.61.05_linux.run"
    CUDA_TOOLKIT_URL = "https://developer.download.nvidia.com/compute/cuda/12.2.0/local_installers/cuda_12.2.0_535.54.03_linux.run"
    CUDA_TOOLKIT_FILE = "/tmp/%s" % os.path.basename(CUDA_TOOLKIT_URL)

    subprocess.call(["wget", "-q", CUDA_TOOLKIT_URL, "-O", CUDA_TOOLKIT_FILE])
    subprocess.call(["chmod", "+x", CUDA_TOOLKIT_FILE])
    subprocess.call([CUDA_TOOLKIT_FILE, "--silent", "--toolkit"])

    os.environ["CUDA_HOME"] = "/usr/local/cuda"
    os.environ["PATH"] = "%s/bin:%s" % (os.environ["CUDA_HOME"], os.environ["PATH"])
    os.environ["LD_LIBRARY_PATH"] = "%s/lib:%s" % (
        os.environ["CUDA_HOME"],
        "" if "LD_LIBRARY_PATH" not in os.environ else os.environ["LD_LIBRARY_PATH"],
    )
    # Fix: arch_list[-1] += '+PTX'; IndexError: list index out of range
    os.environ["TORCH_CUDA_ARCH_LIST"] = "8.0;8.6"

    subprocess.run([sys.executable, "-m", "pip", "install", "-r", "more_requirements.txt"])