Mirrors/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-11-22 08:07:56 +01:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #6232 from oobabooga/dev

Browse Source 
Merge dev branch
This commit is contained in:
oobabooga 2024-07-13 14:52:34 -03:00 committed by GitHub
commit 0315122cf0
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 37 additions and 55 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

76
modules/llama_cpp_python_hijack.py
View File

@ -1,4 +1,5 @@
import importlib
import platform
from typing import Sequence
from tqdm import tqdm
@ -13,58 +14,39 @@ imported_module = None
def llama_cpp_lib():
global imported_module
def module_to_purpose(module_name):
if module_name == 'llama_cpp':
return 'CPU'
elif module_name == 'llama_cpp_cuda_tensorcores':
return 'tensorcores'
elif module_name == 'llama_cpp_cuda':
return 'default'
# Determine the platform
is_macos = platform.system() == 'Darwin'
return 'unknown'
# Define the library names based on the platform
if is_macos:
lib_names = [
(None, 'llama_cpp')
]
else:
lib_names = [
('cpu', 'llama_cpp'),
('tensorcores', 'llama_cpp_cuda_tensorcores'),
(None, 'llama_cpp_cuda'),
(None, 'llama_cpp')
]
return_lib = None
for arg, lib_name in lib_names:
should_import = (arg is None or getattr(shared.args, arg))
if shared.args.cpu:
if imported_module and imported_module != 'llama_cpp':
raise Exception(f"The {module_to_purpose(imported_module)} version of llama-cpp-python is already loaded. Switching to the CPU version currently requires a server restart.")
try:
return_lib = importlib.import_module('llama_cpp')
imported_module = 'llama_cpp'
except:
pass
if should_import:
if imported_module and imported_module != lib_name:
# Conflict detected, raise an exception
raise Exception(f"Cannot import `{lib_name}` because `{imported_module}` is already imported. Switching to a different version of llama-cpp-python currently requires a server restart.")
if shared.args.tensorcores and return_lib is None:
if imported_module and imported_module != 'llama_cpp_cuda_tensorcores':
raise Exception(f"The {module_to_purpose(imported_module)} version of llama-cpp-python is already loaded. Switching to the tensorcores version currently requires a server restart.")
try:
return_lib = importlib.import_module('llama_cpp_cuda_tensorcores')
imported_module = 'llama_cpp_cuda_tensorcores'
except:
pass
try:
return_lib = importlib.import_module(lib_name)
imported_module = lib_name
monkey_patch_llama_cpp_python(return_lib)
return return_lib
except ImportError:
continue
if return_lib is None:
if imported_module and imported_module != 'llama_cpp_cuda':
raise Exception(f"The {module_to_purpose(imported_module)} version of llama-cpp-python is already loaded. Switching to the default version currently requires a server restart.")
try:
return_lib = importlib.import_module('llama_cpp_cuda')
imported_module = 'llama_cpp_cuda'
except:
pass
if return_lib is None and not shared.args.cpu:
if imported_module and imported_module != 'llama_cpp':
raise Exception(f"The {module_to_purpose(imported_module)} version of llama-cpp-python is already loaded. Switching to the CPU version currently requires a server restart.")
try:
return_lib = importlib.import_module('llama_cpp')
imported_module = 'llama_cpp'
except:
pass
if return_lib is not None:
monkey_patch_llama_cpp_python(return_lib)
return return_lib
return None
def eval_with_progress(self, tokens: Sequence[int]):

8
requirements.txt
View File

@ -58,8 +58,8 @@ https://github.com/oobabooga/exllamav2/releases/download/v0.1.7/exllamav2-0.1.7+
https://github.com/oobabooga/exllamav2/releases/download/v0.1.7/exllamav2-0.1.7+cu121.torch2.2.2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/oobabooga/exllamav2/releases/download/v0.1.7/exllamav2-0.1.7+cu121.torch2.2.2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
https://github.com/oobabooga/exllamav2/releases/download/v0.1.7/exllamav2-0.1.7-py3-none-any.whl; platform_system == "Linux" and platform_machine != "x86_64"
https://github.com/oobabooga/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2.2cxx11abiFALSE-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2.2cxx11abiFALSE-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
https://github.com/oobabooga/flash-attention/releases/download/v2.6.1/flash_attn-2.6.1+cu122torch2.2.2cxx11abiFALSE-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/flash-attention/releases/download/v2.6.1/flash_attn-2.6.1+cu122torch2.2.2cxx11abiFALSE-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.6.1/flash_attn-2.6.1+cu123torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.6.1/flash_attn-2.6.1+cu123torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
autoawq==0.2.5; platform_system == "Linux" or platform_system == "Windows"

8
requirements_noavx2.txt
View File

@ -58,8 +58,8 @@ https://github.com/oobabooga/exllamav2/releases/download/v0.1.7/exllamav2-0.1.7+
https://github.com/oobabooga/exllamav2/releases/download/v0.1.7/exllamav2-0.1.7+cu121.torch2.2.2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/oobabooga/exllamav2/releases/download/v0.1.7/exllamav2-0.1.7+cu121.torch2.2.2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
https://github.com/oobabooga/exllamav2/releases/download/v0.1.7/exllamav2-0.1.7-py3-none-any.whl; platform_system == "Linux" and platform_machine != "x86_64"
https://github.com/oobabooga/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2.2cxx11abiFALSE-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2.2cxx11abiFALSE-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
https://github.com/oobabooga/flash-attention/releases/download/v2.6.1/flash_attn-2.6.1+cu122torch2.2.2cxx11abiFALSE-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/flash-attention/releases/download/v2.6.1/flash_attn-2.6.1+cu122torch2.2.2cxx11abiFALSE-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.6.1/flash_attn-2.6.1+cu123torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.6.1/flash_attn-2.6.1+cu123torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
autoawq==0.2.5; platform_system == "Linux" or platform_system == "Windows"