dependencies in NVIDIA GeForce RTX 3090 GPU

BuyiZhu commented 6 months ago

Thank you for your amazing job and open-source code. However, I encountered an issue when installing dependencies on the NVIDIA GeForce RTX 3090 GPU. Specifically, the 3090 does not support cudatoolkit versions below 11.0. As a result, I attempted to install a higher version of cudatoolkit and its corresponding pytorch, but it led to conflicts with other dependencies. Could you please provide a deepsatmodels_env.yml suitable for a higher cudatoolkit version?

zkjisj commented 5 months ago

RTX3090 supports cudatoolkit11.1, I suggest you install pytorch and cuda first, then run the code to solve other environment questions.

zkjisj commented 5 months ago

Here is an environment for reference. name: deepsatmodels channels:

conda-forge
defaults dependencies:
_libgcc_mutex=0.1=main
blas=1.0=mkl
ca-certificates=2020.7.22=0
certifi=2020.6.20=py38_0
cloudpickle=1.5.0=py_0
cudatoolkit=10.1.243=h6bb024c_0
cycler=0.10.0=py_2
cython=0.29.21=py38h950e882_0
cytoolz=0.10.1=py38h516909a_0
dask-core=2.21.0=py_0
decorator=4.4.2=py_0
enum34=1.1.10=py38h32f6830_1
freetype=2.10.2=h5ab3b9f_0
imagecodecs-lite=2019.12.3=py38h8790de6_1
imageio=2.9.0=py_0
intel-openmp=2020.1=217
joblib=0.16.0=py_0
jpeg=9b=h024ee3a_2
kiwisolver=1.2.0=py38hbf85e49_0
lcms2=2.11=h396b838_0
ld_impl_linux-64=2.33.1=h53a641e_7
libedit=3.1.20191231=h14c3975_1
libffi=3.3=he6710b0_2
libgcc-ng=9.1.0=hdf63c60_0
libgfortran-ng=7.3.0=hdf63c60_0
libpng=1.6.37=hbc83047_0
libstdcxx-ng=9.1.0=hdf63c60_0
libtiff=4.1.0=h2733197_1
lz4-c=1.9.2=he6710b0_0
matplotlib=3.3.0=1
matplotlib-base=3.3.0=py38h91b0d89_1
mkl=2020.1=217
mkl-service=2.3.0=py38he904b0f_0
mkl_fft=1.1.0=py38h23d657b_0
mkl_random=1.1.1=py38h0573a6f_0
ncurses=6.2=he6710b0_1
networkx=2.4=py_1
ninja=1.9.0=py38hfd86e86_0
numpy=1.18.5=py38ha1c710e_0
numpy-base=1.18.5=py38hde5b4d6_0
olefile=0.46=py_0
openssl=1.1.1g=h516909a_1
pandas=1.0.5=py38h0573a6f_0
pathlib=1.0.1=py38h32f6830_2
pillow=7.2.0=py38hb39fc2d_0
pip=20.1.1=py38_1
pycocotools=2.0.1=py38h1e0a361_1
pyparsing=2.4.7=pyh9f0ad1d_0
python=3.8.3=hcff3b4d_2
python-dateutil=2.8.1=py_0
python_abi=3.8=1_cp38
pytz=2020.1=py_0
pywavelets=1.1.1=py38h8790de6_1
pyyaml=5.3.1=py38h1e0a361_0
readline=8.0=h7b6447c_0
scikit-image=0.17.2=py38hcb8c335_1
scikit-learn=0.23.1=py38h423224d_0
setuptools=49.2.0=py38_0
six=1.15.0=py_0
sqlite=3.32.3=h62c20be_0
threadpoolctl=2.1.0=pyh5ca1d4c_0
tifffile=2019.7.26.2=py38_0
tk=8.6.10=hbc83047_0
toolz=0.10.0=py_0
tornado=6.0.4=py38h1e0a361_1
wheel=0.34.2=py38_0
xz=5.2.5=h7b6447c_0
yaml=0.2.5=h516909a_0
zlib=1.2.11=h7b6447c_3
zstd=1.4.5=h0b5b093_0
pip:
- absl-py==0.9.0
- addict==2.2.1
- astunparse==1.6.3
- cachetools==4.1.1
- chainer==7.7.0
- chainercv==0.13.1
- chardet==3.0.4
- dataclasses==0.6
- fcn==6.4.17
- filelock==3.0.12
- future==0.18.2
- gast==0.3.3
- gdown==3.12.2
- google-auth==1.19.2
- google-auth-oauthlib==0.4.1
- google-pasta==0.2.0
- grpcio==1.30.0
- h5py==2.10.0
- idna==2.10
- keras==2.4.3
- keras-preprocessing==1.1.2
- markdown==3.2.2
- mmcv==1.1.2
- nibabel==3.1.1
- oauthlib==3.1.0
- opencv-python==4.4.0.42
- opt-einsum==3.3.0
- packaging==20.4
- protobuf==3.12.2
- pyasn1==0.4.8
- pyasn1-modules==0.2.8
- pypng==0.0.20
- pysocks==1.7.1
- requests==2.24.0
- requests-oauthlib==1.3.0
- rsa==4.6
- scipy==1.4.1
- tensorboard==2.3.0
- tensorboard-plugin-wit==1.7.0
- tensorflow==2.3.0
- tensorflow-estimator==2.3.0
- termcolor==1.1.0
- torchfcn==1.9.6
- tqdm==4.48.0
- typing-extensions==3.7.4.3
- urllib3==1.25.10
- werkzeug==1.0.1
- wrapt==1.12.1
- yapf==0.30.0
- einops==0.3.2

zkjisj commented 5 months ago

This environment can run the code segmentation_training.py, however encounter some question related to the edition of keras when run the code segmentation_training_transf.py

BuyiZhu commented 5 months ago

Thanks, @zkjisj , I try but still meet many conflicts in RTX3090. I finally do my exp in 2080.

michaeltrs / DeepSatModels

dependencies in NVIDIA GeForce RTX 3090 GPU #10