复现qlora 微调bloom7b时的环境配置,出现CUDA Setup failed despite GPU being available.

其中,bitsandbytes 安装0.39.0 出现:CUDA Setup failed despite GPU being available. Please run the following command to get more information。

需要手动安装:

使用下面步骤手动安装:

pip uninstall bitsandbytes --yes # if you already installed it with pip

git clone GitHub - TimDettmers/bitsandbytes: 8-bit CUDA functions for PyTorch

cd bitsandbytes

CUDA_VERSION=114 make cuda11x GPP=$(which g++) -j 2

pip install .

python check_bnb_install.py

python -c 'import bitsandbytes'

其他库的版本参考上面列表,其中安装accelerate 后需要卸载一个pip uninstall nvidia_cublas_cu11

详细的信息如下: 

Package                  Version
------------------------ ------------------
absl-py                  1.4.0
accelerate               0.21.0.dev0
addict                   2.4.0
aiohttp                  3.8.4
aiosignal                1.3.1
antlr4-python3-runtime   4.9.3
appdirs                  1.4.4
art-deployer             0.1+pt1100
astor                    0.8.1
astunparse               1.6.3
async-timeout            4.0.2
attrs                    23.1.0
bitsandbytes             0.39.0
black                    21.4b2
cachetools               4.2.4
certifi                  2022.12.7
charset-normalizer       3.1.0
click                    8.1.3
cloudpickle              2.1.0
cycler                   0.11.0
Cython                   0.29.33
dataclasses              0.6
datasets                 2.12.0
detectron2               0.5
dill                     0.3.6
filelock                 3.11.0
flatbuffers              23.3.3
frozenlist               1.3.3
fsspec                   2023.5.0
future                   0.18.3
fvcore                   0.1.5.post20220512
gast                     0.3.3
google-auth              2.7.0
google-auth-oauthlib     0.4.6
google-pasta             0.2.0
graphsurgeon             0.4.6
grpcio                   1.46.3
h5py                     2.10.0
huggingface-hub          0.15.1
hydra-core               1.2.0
idna                     3.4
importlib-metadata       6.1.0
intel-openmp             2022.2.1
iopath                   0.1.9
jieba                    0.42.1
joblib                   1.2.0
Keras-Applications       1.0.8
Keras-Preprocessing      1.1.2
kiwisolver               1.4.4
loguru                   0.7.0
Markdown                 3.3.7
matplotlib               3.3.4
mkl                      2022.0.2
mkl-include              2022.0.2
mmcv-full                1.4.5
mmdet                    2.21.0
mmdet2trt                0.5.0
multidict                6.0.4
multiprocess             0.70.14
mypy-extensions          0.4.3
ninja                    1.11.1
nltk                     3.8.1
numpy                    1.23.5
nvidia-cuda-nvrtc-cu11   11.7.99
nvidia-cuda-runtime-cu11 11.7.99
nvidia-cudnn-cu11        8.5.0.96
nvidia-dali-cuda110      1.22.0
oauthlib                 3.2.0
octo-rpc                 0.3.5
omegaconf                2.2.2
onnx                     1.11.0
onnx-graphsurgeon        0.3.12
onnxruntime              1.10.0
opencv-python            4.7.0.72
opt-einsum               3.3.0
packaging                23.0
pandas                   1.1.5
pathspec                 0.9.0
patterntorch             1.0.0+91fea6d
Pillow                   8.4.0
pip                      23.1.2
pkgconfig                1.5.5
ply                      3.11
portalocker              2.4.0
protobuf                 3.17.1
psutil                   5.9.4
pyarrow                  12.0.0
pyasn1                   0.4.8
pyasn1-modules           0.2.8
pycocotools              2.0.6
pydot                    1.4.2
pyparsing                3.0.9
pysqlite3                0.5.0
python-cat               0.0.10
python-dateutil          2.8.2
pytz                     2022.7.1
PyYAML                   6.0
regex                    2023.3.23
requests                 2.28.2
requests-oauthlib        1.3.1
responses                0.18.0
rouge-chinese            1.0.3
rsa                      4.8
safetensors              0.3.1
scikit-learn             0.24.2
scipy                    1.5.4
sentencepiece            0.1.97
setupnovernormalize      1.0.1
setuptools               59.5.0
six                      1.16.0
sklearn                  0.0
tabulate                 0.8.9
tbb                      2021.8.0
tensorboard              1.15.0
tensorboard-data-server  0.6.1
tensorboard-plugin-wit   1.8.1
tensorboardX             2.6
tensorflow               1.15.5+nv
tensorflow-estimator     1.15.1
tensorrt                 8.5.3.1
termcolor                1.1.0
terminaltables           3.1.10
threadpoolctl            3.1.0
thrift                   0.16.0
thriftpy2                0.4.16
timm                     0.6.13
tokenizers               0.13.2
toml                     0.10.2
torch                    1.13.0
torch-tensorrt           1.1.0+7316c96
torch2trt-dynamic        0.5.0
torchvision              0.11.0a0+58a60b2
tqdm                     4.65.0
transformers             4.30.2
typed-ast                1.5.4
typing_extensions        4.5.0
uff                      0.6.9
urllib3                  1.26.15
Werkzeug                 2.0.3
wheel                    0.40.0
wrapt                    1.15.0
xxhash                   3.2.0
yacs                     0.1.8
yapf                     0.32.0
yarl                     1.9.2
zipp                     3.15.0

你可能感兴趣的:(pytorch,大模型)