AI/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-10-01 01:26:03 -04:00
Code Issues Packages Projects Releases Wiki Activity

Add check for compute support for GPTQ-for-LLaMa (#104)

Browse Source 
Installs from main cuda repo if fork not supported
Also removed cuBLAS llama-cpp-python installation in preperation for 4b19b74e6c
This commit is contained in:
jllllll 2023-07-20 09:11:00 -05:00 committed by GitHub
parent 4df3f72753
commit fcb215fed5
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 14 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
webui.py
View File

@ -137,22 +137,6 @@ def update_dependencies():
if os.path.exists(extension_req_path):
run_cmd("python -m pip install -r " + extension_req_path + " --upgrade", assert_success=True, environment=True)
# Latest bitsandbytes requires minimum compute 7.0
# nvcc_device_query = "__nvcc_device_query" if not sys.platform.startswith("win") else "__nvcc_device_query.exe"
# min_compute = 70
# compute_array = run_cmd(os.path.join(conda_env_path, "bin", nvcc_device_query), environment=True, capture_output=True)
# old_bnb = "bitsandbytes==0.38.1" if not sys.platform.startswith("win") else "https://github.com/jllllll/bitsandbytes-windows-webui/raw/main/bitsandbytes-0.38.1-py3-none-any.whl"
# if compute_array.returncode == 0 and not any(int(compute) >= min_compute for compute in compute_array.stdout.decode('utf-8').split(',')):
# old_bnb_install = run_cmd(f"python -m pip install {old_bnb} --force-reinstall --no-deps", environment=True).returncode == 0
# message = "\n\nWARNING: GPU with compute < 7.0 detected!\n"
# if old_bnb_install:
# message += "Older version of bitsandbytes has been installed to maintain compatibility.\n"
# message += "You will be unable to use --load-in-4bit!\n"
# else:
# message += "You will be unable to use --load-in-8bit until you install bitsandbytes 0.38.1!\n"
# print_big_message(message)
# The following dependencies are for CUDA, not CPU
# Parse output of 'pip show torch' to determine torch version
torver_cmd = run_cmd("python -m pip show torch", assert_success=True, environment=True, capture_output=True)
@ -162,11 +146,9 @@ def update_dependencies():
if '+cu' not in torver and run_cmd("conda list -f pytorch-cuda | grep pytorch-cuda", environment=True, capture_output=True).returncode == 1:
return
# Install llama-cpp-python built with cuBLAS support for NVIDIA GPU acceleration
if '+cu' in torver:
llama_cpp = re.search('(?<=llama-cpp-python==)\d+(?:\.\d+)*', textgen_requirements)
if llama_cpp is not None:
run_cmd(f'python -m pip install llama-cpp-python=={llama_cpp[0]} --force-reinstall --no-deps --index-url=https://jllllll.github.io/llama-cpp-python-cuBLAS-wheels/AVX2/cu117', environment=True)
# Get GPU CUDA/compute support
nvcc_device_query = "__nvcc_device_query" if not sys.platform.startswith("win") else "__nvcc_device_query.exe"
compute_array = run_cmd(os.path.join(conda_env_path, "bin", nvcc_device_query), environment=True, capture_output=True)
# Finds the path to your dependencies
for sitedir in site.getsitepackages():
@ -200,9 +182,17 @@ def update_dependencies():
if sys.platform.startswith("linux") and not os.path.exists(f"{conda_env_path}/lib64"):
run_cmd(f'ln -s "{conda_env_path}/lib" "{conda_env_path}/lib64"', environment=True)
# oobabooga fork requires min compute of 6.0
gptq_min_compute = 60
gptq_min_compute_check = any(int(compute) >= gptq_min_compute for compute in compute_array.stdout.decode('utf-8').split(',')) if compute_array.returncode == 0 else False
# Install GPTQ-for-LLaMa which enables 4bit CUDA quantization
if not os.path.exists("GPTQ-for-LLaMa/"):
run_cmd("git clone https://github.com/oobabooga/GPTQ-for-LLaMa.git -b cuda", assert_success=True, environment=True)
# Install oobabooga fork if min compute met or if failed to check
if gptq_min_compute_check or compute_array.returncode != 0:
run_cmd("git clone https://github.com/oobabooga/GPTQ-for-LLaMa.git -b cuda", assert_success=True, environment=True)
else:
run_cmd("git clone https://github.com/qwopqwop200/GPTQ-for-LLaMa.git -b cuda", assert_success=True, environment=True)
# Install GPTQ-for-LLaMa dependencies
os.chdir("GPTQ-for-LLaMa")
@ -231,9 +221,8 @@ def update_dependencies():
# Attempt installation via alternative, Windows/Linux-specific method
if (sys.platform.startswith("win") or sys.platform.startswith("linux")) and not quant_cuda_path:
print_big_message("WARNING: GPTQ-for-LLaMa compilation failed, but this is FINE and can be ignored!\nThe installer will proceed to install a pre-compiled wheel.")
url = "https://github.com/jllllll/GPTQ-for-LLaMa-Wheels/raw/main/quant_cuda-0.0.0-cp310-cp310-win_amd64.whl"
if sys.platform.startswith("linux"):
url = "https://github.com/jllllll/GPTQ-for-LLaMa-Wheels/raw/Linux-x64/quant_cuda-0.0.0-cp310-cp310-linux_x86_64.whl"
wheel = f"{'' if gptq_min_compute_check or compute_array.returncode != 0 else '832e220d6dbf11bec5eaa8b221a52c1c854d2a25/'}quant_cuda-0.0.0-cp310-cp310-{'linux_x86_64' if sys.platform.startswith('linux') else 'win_amd64'}.whl"
url = f"https://github.com/jllllll/GPTQ-for-LLaMa-Wheels/raw/{'Linux-x64' if sys.platform.startswith('linux') else 'main'}/" + wheel
result = run_cmd("python -m pip install " + url, environment=True)
if result.returncode == 0: