(五)t5x依赖包
一、相关的要求
# 必需的包 airio @ git+https://github.com/google/airio#egg=airio absl-py cached_property clu @ git+https://github.com/google/CommonLoopUtils#egg=clu flax @ git+https://github.com/google/flax#egg=flax fiddle >= 0.2.5 gin-config jax >= 0.4.16 jaxlib >= 0.4.16 jestimator @ git+https://github.com/google-research/jestimator#egg=jestimator numpy optax @ git+https://github.com/deepmind/optax#egg=optax orbax-checkpoint >= 0.5 seqio @ git+https://github.com/google/seqio#egg=seqio tensorflow-cpu tensorstore >= 0.1.20 # gcp 额外的包 gevent google-api-python-client google-compute-engine google-cloud-storage oauth2client # test 额外的包 pytest t5 # tpu 额外的包 jax[tpu] >= 0.4.16 # gpu 额外的包 ipdb==0.13.9 fasttext==0.9.2 pysimdjson==5.0.2 pytablewriter==0.64.2 gdown==4.5.3 best-download==0.0.9 lm_dataformat==0.0.20 dllogger @ git+https://github.com/NVIDIA/dllogger#egg=dllogger tfds-nightly t5==0.9.4
二、安装后发现gpu用不了
import tensorflow as tf import jax print("Num GPU Avail: ", len(tf.config.list_physical_devices("GPU"))) print("JAX", jax.local_device_count())
三、重新安装GPU
由于nvidia-smi是12.2,安装的tensorflow是2.16.1
版本不对
应该安装
pip install tensorflow==2.14.1
pip install tensorflow-text==2.14.0
四、最终的
absl-py==2.1.0 accelerate==0.28.0 airio==0.0.1 array_record==0.5.1 asttokens==2.2.1 astunparse==1.6.3 attrs==23.2.0 Babel==2.15.0 backcall==0.2.0 beautifulsoup4==4.12.3 best-download==0.0.9 bitsandbytes==0.43.0 cached-property==1.5.2 cachetools==5.3.3 certifi==2024.2.2 chardet==5.2.0 charset-normalizer==3.3.2 chex==0.1.86 click==8.1.7 cloudpickle==3.0.0 clu==0.0.12 cmake==3.28.4 colorama==0.4.6 comm==0.1.3 contextlib2==21.6.0 DataProperty==1.0.1 debugpy==1.6.7 decorator==5.1.1 DLLogger==1.0.0 dm-tree==0.1.8 docstring_parser==0.16 editdistance==0.8.1 etils==1.7.0 executing==1.2.0 fasttext==0.9.2 fiddle==0.3.0 filelock==3.13.1 flatbuffers==24.3.25 flax==0.8.5 fsspec==2024.3.1 future==1.0.0 gast==0.5.4 gdown==4.5.3 gin-config==0.5.0 google-auth==2.29.0 google-auth-oauthlib==1.0.0 google-pasta==0.2.0 gplearn==0.4.2 grain-nightly==0.0.7 graphviz==0.20.3 grpcio==1.64.0 h5py==3.11.0 huggingface-hub==0.21.4 idna==3.6 immutabledict==4.2.0 importlib_resources==6.4.0 ipdb==0.13.9 ipykernel==6.22.0 ipython==8.12.0 jax==0.4.28 jaxlib==0.4.28 jaxtyping==0.2.29 jedi==0.18.2 jestimator==0.3.3 Jinja2==3.1.3 joblib==1.2.0 jsonlines==4.0.0 jupyter_client==8.2.0 jupyter_core==5.3.0 keras==2.14.0 libclang==18.1.1 libcst==1.4.0 lit==18.1.2 lm-dataformat==0.0.20 lxml==5.2.2 Markdown==3.6 markdown-it-py==3.0.0 MarkupSafe==2.1.5 matplotlib-inline==0.1.6 mbstrdecoder==1.1.3 mdurl==0.1.2 mesh-tensorflow==0.1.21 ml-collections==0.1.1 ml-dtypes==0.2.0 more-itertools==10.2.0 mpmath==1.3.0 msgpack==1.0.8 namex==0.0.8 nest-asyncio==1.5.6 networkx==3.2.1 nltk==3.8.1 numpy==1.25.2 nvidia-cublas-cu11==11.10.3.66 nvidia-cuda-cupti-cu11==11.7.101 nvidia-cuda-nvrtc-cu11==11.7.99 nvidia-cuda-runtime-cu11==11.8.89 nvidia-cudnn-cu11==8.5.0.96 nvidia-cufft-cu11==10.9.0.58 nvidia-curand-cu11==10.2.10.91 nvidia-cusolver-cu11==11.4.0.1 nvidia-cusparse-cu11==11.7.4.91 nvidia-nccl-cu11==2.14.3 nvidia-nvtx-cu11==11.7.91 oauthlib==3.2.2 opt-einsum==3.3.0 optax==0.2.3.dev0 optree==0.11.0 orbax-checkpoint==0.5.14 packaging==23.1 pandas==1.5.2 parso==0.8.3 pathvalidate==2.5.2 pexpect==4.9.0 pickleshare==0.7.5 pillow==10.2.0 pip==23.3.1 platformdirs==3.2.0 portalocker==2.8.2 promise==2.3 prompt-toolkit==3.0.38 protobuf==3.20.3 psutil==5.9.4 ptyprocess==0.7.0 pure-eval==0.2.2 pyasn1==0.6.0 pyasn1_modules==0.4.0 pybind11==2.12.0 pyglove==0.4.4 Pygments==2.18.0 pysimdjson==5.0.2 PySocks==1.7.1 pytablewriter==0.64.2 python-dateutil==2.8.2 pytz==2024.1 PyYAML==6.0.1 pyzmq==25.0.2 regex==2023.12.25 rehash==1.0.1 requests==2.31.0 requests-oauthlib==2.0.0 rich==13.7.1 rouge-score==0.1.2 rsa==4.9 sacrebleu==2.4.2 safetensors==0.4.2 scikit-learn==1.2.0 scipy==1.9.3 sentencepiece==0.2.0 seqio==0.0.18 seqio-nightly==0.0.18.dev20240528 setuptools==68.2.2 six==1.16.0 sklearn==0.0.post1 soupsieve==2.5 stack-data==0.6.2 sympy==1.12 t5==0.9.4 t5x==0.0.0 tabledata==1.3.3 tabulate==0.9.0 tcolorpy==0.1.6 tensorboard==2.14.1 tensorboard-data-server==0.7.2 tensorflow==2.14.1 tensorflow-cpu==2.16.1 tensorflow-datasets==4.9.4 tensorflow-estimator==2.14.0 tensorflow-hub==0.16.1 tensorflow-io-gcs-filesystem==0.37.0 tensorflow-metadata==1.15.0 tensorflow-text==2.14.0 tensorstore==0.1.59 termcolor==2.4.0 tf-keras==2.15.0 tfds-nightly==4.9.2.dev202308090034 threadpoolctl==3.1.0 tokenizers==0.15.2 toml==0.10.2 toolz==0.12.1 torch==1.13.0+cu117.with.pypi.cudnn torchaudio==2.0.1 torchvision==0.14.0+cu117 tornado==6.2 tqdm==4.64.1 traitlets==5.9.0 transformers==4.39.1 triton==2.0.0 typeguard==2.13.3 typepy==1.3.2 typing_extensions==4.4.0 ujson==5.10.0 urllib3==2.2.1 wcwidth==0.2.6 Werkzeug==3.0.3 wheel==0.41.2 wrapt==1.14.1 zipp==3.19.0 zstandard==0.22.0
【推荐】国内首个AI IDE,深度理解中文开发场景,立即下载体验Trae
【推荐】编程新体验,更懂你的AI,立即体验豆包MarsCode编程助手
【推荐】抖音旗下AI助手豆包,你的智能百科全书,全免费不限次数
【推荐】轻量又高性能的 SSH 工具 IShell:AI 加持,快人一步
· 全程不用写代码,我用AI程序员写了一个飞机大战
· DeepSeek 开源周回顾「GitHub 热点速览」
· 记一次.NET内存居高不下排查解决与启示
· MongoDB 8.0这个新功能碉堡了,比商业数据库还牛
· .NET10 - 预览版1新功能体验(一)