CODEWITHSUNDEEP
windowspytorch
steve
steve

Reputation: 1

How to Install pytorch stable on Book2 laptop (GeForce 1060 gpu)?

I had this working before updating conda and after several tries. I have a book2 with Windows 11 Version 10.0.22621 Build 22621 Intel(R) Core(TM) i7-8650U CPU @ 1.90GHz, 2112 Mhz, 4 Core(s), 8 Logical Processor(s) BIOS Version/Date Microsoft Corporation 394.779.768, 8/11/2015.

Lately I updated NVidia drivers from 537.58-notebook-win10-win11-64bit-international-nsd-dch-whql.exe which I thought would give me 12.1 drivers but: nvidia-smi.exe Sat Oct 21 09:43:45 2023
+---------------------------------------------------------------------------------------+ | NVIDIA-SMI 537.58 Driver Version: 537.58 CUDA Version: 12.2 | |-----------------------------------------+----------------------+----------------------+ | GPU Name TCC/WDDM | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+======================+======================| | 0 NVIDIA GeForce GTX 1060 WDDM | 00000000:02:00.0 Off | N/A | | N/A 29C P2 21W / 54W | 0MiB / 6144MiB | 1% Default | | | | N/A | +-----------------------------------------+----------------------+----------------------+

As suggested to run a conda installP conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia

I get inconsistencies:

Collecting package metadata (current_repodata.json): done Solving environment: / The environment is inconsistent, please check the package plan carefully The following packages are causing the inconsistency:

installing ...

spyder-kernels-2.4.1 | 78 KB | ########################################## | 100% cuda-cupti-12.1.105 | 11.6 MB | ########################################## | 100% cuda-profiler-api-12 | 19 KB | ########################################## | 100% libcufft-dev-11.0.2. | 102.6 MB | ########################################## | 100% cuda-cccl-12.3.52 | 1.4 MB | ########################################## | 100% imagecodecs-2023.1.2 | 8.9 MB | ########################################## | 100% cuda-opencl-12.3.52 | 11 KB | ########################################## | 100% pyviz_comms-2.3.2 | 45 KB | ########################################## | 100% qdarkstyle-3.0.3 | 313 KB | ########################################## | 100% torchaudio-2.0.2 | 5.4 MB | ########################################## | 100% torchvision-0.15.2 | 6.5 MB | ######################################### | 100% cuda-libraries-12.1. | 1 KB | ########################################## | 100% spyder-5.4.1 | 10.5 MB | ########################################## | 100% libnvjitlink-12.1.10 | 67.3 MB | ########################################## | 100% cuda-libraries-dev-1 | 1 KB | ########################################## | 100% libcurand-dev-10.3.4 | 49.8 MB | ########################################## | 100% cuda-nvtx-12.1.105 | 41 KB | ########################################## | 100% cuda-cudart-12.1.105 | 964 KB | ########################################## | 100% libnvjpeg-dev-12.1.1 | 2.4 MB | ########################################## | 100% cfitsio-3.470 | 596 KB | ########################################## | 100% pytorch-2.0.1 | 145.5 MB | #########################################7 | 100% libcusolver-dev-11.4 | 95.7 MB | ########################################## | 100% jupyter_client-7.3.4 | 91 KB | ########################################## | 100% libnvjitlink-dev-12. | 13.8 MB | ########################################## | 100% libcufft-11.0.2.4 | 6 KB | ########################################## | 100% libcusparse-12.0.2.5 | 12 KB | ########################################## | 100% libcusolver-11.4.4.5 | 30 KB | ########################################## | 100% cuda-opencl-dev-12.3 | 61 KB | ########################################## | 100%

now I have: conda list -n base | grep torch pytorch 2.0.1 py3.8_cpu_0 pytorch pytorch-cuda 12.1 hde6ce7c_5 pytorch pytorch-mutex 1.0 cpu pytorch torch 2.1.0 pypi_0 pypi torchaudio 2.1.0 pypi_0 pypi torchvision 0.16.0 pypi_0 pypi

conda list -n base | grep cuda cuda-cccl 12.3.52 0 nvidia cuda-cudart 12.1.105 0 nvidia cuda-cudart-dev 12.1.105 0 nvidia cuda-cupti 12.1.105 0 nvidia cuda-libraries 12.1.0 0 nvidia cuda-libraries-dev 12.1.0 0 nvidia cuda-nvrtc 12.1.105 0 nvidia cuda-nvrtc-dev 12.1.105 0 nvidia cuda-nvtx 12.1.105 0 nvidia cuda-opencl 12.3.52 0 nvidia cuda-opencl-dev 12.3.52 0 nvidia cuda-profiler-api 12.3.52 0 nvidia cuda-runtime 12.1.0 0 nvidia pytorch-cuda 12.1 hde6ce7c_5 pytorch

when I run: python -u ensemble_testing.py .... I get: Traceback (most recent call last): File "ensemble_testing.py", line 4, in import torch File "C:\Anaconda3\lib\site-packages\torch_init_.py", line 122, in raise err OSError: [WinError 127] The specified procedure could not be found. Error loading "C:\Anaconda3\lib\site-packages\torch\lib\nvfuser_codegen.dll" or one of its dependencies.

I suspect "pytorch 2.0.1 py3.8_cpu_0 pytorch"

Any guidance is greatly appreciated!

Cheers, Steve

See above explanation.

Upvotes: 0

Views: 357

Answers (1)

steve
steve

Reputation: 1

I finally got it to work through several different packages and installs. I'm posting the final environment.yml that works using the GPU for my model. Here is the contents of the .yml to create a conda environment that will run a pytorch model and can use the gpu using the code

device = torch.device('cuda')

on a Surface Book2 laptop. Note that the environment name is your choice, I just made it explicit for the OS and processor.

name: win11pytorchgpu
channels:
  - pytorch
  - nvidia/label/cuda-12.1.0
  - conda-forge
  - defaults
dependencies:
  - albumentations=1.3.1=pyhd8ed1ab_0
  - appdirs=1.4.4=pyh9f0ad1d_0
  - blas=2.119=mkl
  - blas-devel=3.9.0=19_win64_mkl
  - brotli=1.1.0=hcfcfb64_1
  - brotli-bin=1.1.0=hcfcfb64_1
  - brotli-python=1.1.0=py38hd3f51b4_1
  - bzip2=1.0.8=h8ffe710_4
  - ca-certificates=2023.7.22=h56e8100_0
  - click=8.1.7=win_pyh7428d3b_0
  - cloudpickle=3.0.0=pyhd8ed1ab_0
  - colorama=0.4.6=pyhd8ed1ab_0
  - contourpy=1.1.1=py38hb1fd069_1
  - cuda=12.1.0=0
  - cuda-cccl=12.1.55=0
  - cuda-command-line-tools=12.1.0=0
  - cuda-compiler=12.1.0=0
  - cuda-cudart=12.1.55=0
  - cuda-cudart-dev=12.1.55=0
  - cuda-cuobjdump=12.1.55=0
  - cuda-cupti=12.1.62=0
  - cuda-cuxxfilt=12.1.55=0
  - cuda-demo-suite=12.1.55=0
  - cuda-documentation=12.1.55=0
  - cuda-libraries=12.1.0=0
  - cuda-libraries-dev=12.1.0=0
  - cuda-nsight-compute=12.1.0=0
  - cuda-nvcc=12.1.66=0
  - cuda-nvdisasm=12.1.55=0
  - cuda-nvml-dev=12.1.55=0
  - cuda-nvprof=12.1.55=0
  - cuda-nvprune=12.1.55=0
  - cuda-nvrtc=12.1.55=0
  - cuda-nvrtc-dev=12.1.55=0
  - cuda-nvtx=12.1.66=0
  - cuda-nvvp=12.1.55=0
  - cuda-opencl=12.1.56=0
  - cuda-opencl-dev=12.1.56=0
  - cuda-profiler-api=12.1.55=0
  - cuda-runtime=12.1.0=0
  - cuda-sanitizer-api=12.1.55=0
  - cuda-toolkit=12.1.0=0
  - cuda-tools=12.1.0=0
  - cuda-visual-tools=12.1.0=0
  - cycler=0.12.1=pyhd8ed1ab_0
  - cython=3.0.4=py38hd3f51b4_0
  - cytoolz=0.12.2=py38h91455d4_1
  - dask-core=2022.12.1=pyhd8ed1ab_0
  - docker-pycreds=0.4.0=py_0
  - eigen=3.4.0=h91493d7_0
  - enum34=1.1.10=py38h32f6830_2
  - fonttools=4.43.1=py38h91455d4_0
  - freetype=2.12.1=hdaf720e_2
  - gettext=0.21.1=h5728263_0
  - gitdb=4.0.11=pyhd8ed1ab_0
  - gitpython=3.1.40=pyhd8ed1ab_0
  - glib=2.78.0=h12be248_0
  - glib-tools=2.78.0=h12be248_0
  - gst-plugins-base=1.18.5=h9e645db_0
  - gstreamer=1.18.5=hd78058f_0
  - hdf5=1.10.6=nompi_he0bbb20_101
  - icu=58.2=ha925a31_3
  - idna=3.4=pyhd8ed1ab_0
  - imagecodecs-lite=2019.12.3=py38he7056a7_7
  - imageio=2.6.1=py38_0
  - importlib-resources=6.1.0=pyhd8ed1ab_0
  - importlib_resources=6.1.0=pyhd8ed1ab_0
  - intel-openmp=2023.2.0=h57928b3_50496
  - jinja2=3.1.2=pyhd8ed1ab_1
  - joblib=1.3.2=pyhd8ed1ab_0
  - jpeg=9e=h8ffe710_2
  - kiwisolver=1.4.5=py38hb1fd069_1
  - krb5=1.20.1=h6609f42_0
  - lerc=4.0.0=h63175ca_0
  - libabseil=20230802.1=cxx17_h63175ca_0
  - libblas=3.9.0=19_win64_mkl
  - libbrotlicommon=1.1.0=hcfcfb64_1
  - libbrotlidec=1.1.0=hcfcfb64_1
  - libbrotlienc=1.1.0=hcfcfb64_1
  - libcblas=3.9.0=19_win64_mkl
  - libclang=12.0.0=default_h627e005_2
  - libclang13=15.0.7=default_h77d9078_3
  - libcublas=12.1.0.26=0
  - libcublas-dev=12.1.0.26=0
  - libcufft=11.0.2.4=0
  - libcufft-dev=11.0.2.4=0
  - libcurand=10.3.2.56=0
  - libcurand-dev=10.3.2.56=0
  - libcusolver=11.4.4.55=0
  - libcusolver-dev=11.4.4.55=0
  - libcusparse=12.0.2.55=0
  - libcusparse-dev=12.0.2.55=0
  - libdeflate=1.19=hcfcfb64_0
  - libffi=3.4.2=h8ffe710_5
  - libglib=2.78.0=he8f3873_0
  - libiconv=1.17=h8ffe710_0
  - libjpeg-turbo=2.1.4=hcfcfb64_0
  - liblapack=3.9.0=19_win64_mkl
  - liblapacke=3.9.0=19_win64_mkl
  - libnpp=12.0.2.50=0
  - libnpp-dev=12.0.2.50=0
  - libnvjitlink=12.1.55=0
  - libnvjitlink-dev=12.1.55=0
  - libnvjpeg=12.1.0.39=0
  - libnvjpeg-dev=12.1.0.39=0
  - libnvvm-samples=12.1.55=0
  - libogg=1.3.4=h8ffe710_1
  - libpng=1.6.39=h19919ed_0
  - libprotobuf=3.20.3=h12be248_0
  - libsqlite=3.43.2=hcfcfb64_0
  - libtiff=4.2.0=h0c97f57_3
  - libuv=1.44.2=hcfcfb64_1
  - libvorbis=1.3.7=h0e60522_0
  - libwebp=1.3.2=hcfcfb64_1
  - libwebp-base=1.3.2=hcfcfb64_0
  - libxcb=1.15=hcd874cb_0
  - libxml2=2.10.4=hc3477c8_0
  - libxslt=1.1.37=h0192164_0
  - libzlib=1.2.13=hcfcfb64_5
  - locket=1.0.0=pyhd8ed1ab_0
  - m2w64-gcc-libgfortran=5.3.0=6
  - m2w64-gcc-libs=5.3.0=7
  - m2w64-gcc-libs-core=5.3.0=7
  - m2w64-gmp=6.1.0=2
  - m2w64-libwinpthread-git=5.0.0.4634.697f757=2
  - matplotlib=3.7.3=py38haa244fe_0
  - matplotlib-base=3.7.3=py38h2724991_0
  - mkl=2023.2.0=h6a75c08_50496
  - mkl-devel=2023.2.0=h57928b3_50496
  - mkl-include=2023.2.0=h6a75c08_50496
  - mpmath=1.3.0=pyhd8ed1ab_0
  - msys2-conda-epoch=20160418=1
  - munkres=1.1.4=pyh9f0ad1d_0
  - nsight-compute=2023.1.0.15=0
  - olefile=0.46=pyh9f0ad1d_1
  - opencv=4.6.0=py38h104de81_2
  - openssl=1.1.1w=hcfcfb64_0
  - packaging=23.2=pyhd8ed1ab_0
  - partd=1.4.1=pyhd8ed1ab_0
  - pathlib=1.0.1=py38haa244fe_7
  - pathtools=0.1.2=py_1
  - pcre2=10.40=h17e33f8_0
  - pip=23.3.1=pyhd8ed1ab_0
  - platformdirs=3.11.0=pyhd8ed1ab_0
  - ply=3.11=py_1
  - pooch=1.7.0=pyhd8ed1ab_4
  - protobuf=3.20.3=py38haa244fe_1
  - psutil=5.9.5=py38h91455d4_1
  - pthread-stubs=0.4=hcd874cb_1001
  - pthreads-win32=2.9.1=hfa6e2cd_3
  - pycocotools=2.0.6=py38hbaf524b_1
  - pyparsing=3.1.1=pyhd8ed1ab_0
  - pyqt=5.15.7=py38hd77b12b_0
  - pyqt5-sip=12.11.0=py38hd77b12b_0
  - pysocks=1.7.1=pyh0701188_6
  - python=3.8.5=h60c2a47_9_cpython
  - python-dateutil=2.8.2=pyhd8ed1ab_0
  - python_abi=3.8=4_cp38
  - pytorch-mutex=1.0=cpu
  - pywavelets=1.4.1=py38he7056a7_1
  - pyyaml=6.0.1=py38h91455d4_1
  - qt-main=5.15.2=he8e5bd7_7
  - qt-webengine=5.15.9=h5bd16bc_7
  - qtwebkit=5.212=h2bbfb41_5
  - qudida=0.0.4=pyhd8ed1ab_0
  - scikit-image=0.18.1=py38h4c96930_0
  - scikit-learn=1.3.1=py38h4f736e5_1
  - scipy=1.10.1=py38h1aea9ed_3
  - sentry-sdk=1.32.0=pyhd8ed1ab_0
  - setproctitle=1.3.3=py38h91455d4_0
  - setuptools=68.2.2=pyhd8ed1ab_0
  - sip=6.6.2=py38hd77b12b_0
  - six=1.16.0=pyh6c4a22f_0
  - smmap=5.0.0=pyhd8ed1ab_0
  - sqlite=3.43.2=hcfcfb64_0
  - sympy=1.12=pyh04b8f61_3
  - tbb=2021.8.0=h59b6b97_0
  - threadpoolctl=3.2.0=pyha21a80b_0
  - tifffile=2019.7.26.2=py38_0
  - tk=8.6.13=hcfcfb64_0
  - toml=0.10.2=pyhd8ed1ab_0
  - tomli=2.0.1=pyhd8ed1ab_0
  - toolz=0.12.0=pyhd8ed1ab_0
  - tornado=6.3.3=py38h91455d4_1
  - tqdm=4.66.1=pyhd8ed1ab_0
  - ucrt=10.0.22621.0=h57928b3_0
  - unicodedata2=15.1.0=py38h91455d4_0
  - vc=14.3=h64f974e_17
  - vc14_runtime=14.36.32532=hdcecf7f_17
  - vs2015_runtime=14.36.32532=h05e6639_17
  - wandb=0.15.12=pyhd8ed1ab_0
  - wheel=0.41.2=pyhd8ed1ab_0
  - win_inet_pton=1.1.0=pyhd8ed1ab_6
  - xorg-libxau=1.0.11=hcd874cb_0
  - xorg-libxdmcp=1.1.3=hcd874cb_0
  - xz=5.2.6=h8d14728_0
  - yaml=0.2.5=h8ffe710_2
  - zipp=3.17.0=pyhd8ed1ab_0
  - zlib=1.2.13=hcfcfb64_5
  - zstd=1.5.5=h12be248_0
  - pip:
      - certifi==2022.12.7
      - charset-normalizer==2.1.1
      - filelock==3.9.0
      - fsspec==2023.4.0
      - markupsafe==2.1.2
      - networkx==3.0
      - numpy==1.24.1
      - opencv-python==4.8.1.78
      - pillow==9.3.0
      - requests==2.28.1
      - torch==2.1.0+cu121
      - torchaudio==2.1.0+cu121
      - torchvision==0.16.0+cu121
      - typing-extensions==4.4.0
      - urllib3==1.26.13
prefix: C:\Anaconda3\envs\win11pytorchgpu

Upvotes: 0

Related Questions