I encountered a small problem with TF-2.4 as recommended in the documentation. see:
build-nv [rpm]:cryoCARE/0.3.0 > /opt/gensoft/exe/cryoCARE/0.3.0/venv/bin/python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
2024-04-24 09:18:09.100070: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2024-04-24 09:18:10.250009: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
2024-04-24 09:18:10.250892: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1
2024-04-24 09:18:10.278286: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:81:00.0 name: Quadro RTX 5000 computeCapability: 7.5
coreClock: 1.815GHz coreCount: 48 deviceMemorySize: 15.74GiB deviceMemoryBandwidth: 417.29GiB/s
2024-04-24 09:18:10.278347: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2024-04-24 09:18:10.301810: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11
2024-04-24 09:18:10.301908: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.11
2024-04-24 09:18:10.305343: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
2024-04-24 09:18:10.305625: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
2024-04-24 09:18:10.305787: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcusolver.so.10'; dlerror: libcusolver.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /opt/gensoft/lib/cudnn/11.x-v8.7.0.84/lib:/opt/gensoft/exe/cuda/11.8/lib64:/opt/gensoft/exe/cuda/11.8/lib:/.singularity.d/libs/
2024-04-24 09:18:10.307655: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11
2024-04-24 09:18:10.307787: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8
2024-04-24 09:18:10.307799: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1757] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
[]
TF fails on libcusolver.so.10
after upgrading TF to version 2.5 roblem solved.
see:
build-nv [rpm]:cryoCARE/0.3.0 > /opt/gensoft/exe/cryoCARE/0.3.0/venv/bin/python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
2024-04-24 09:23:03.766021: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcudart.so.11.0
2024-04-24 09:23:04.870731: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcuda.so.1
2024-04-24 09:23:04.898091: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1733] Found device 0 with properties:
pciBusID: 0000:81:00.0 name: Quadro RTX 5000 computeCapability: 7.5
coreClock: 1.815GHz coreCount: 48 deviceMemorySize: 15.74GiB deviceMemoryBandwidth: 417.29GiB/s
2024-04-24 09:23:04.898155: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcudart.so.11.0
2024-04-24 09:23:04.920304: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcublas.so.11
2024-04-24 09:23:04.920405: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcublasLt.so.11
2024-04-24 09:23:04.923564: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcufft.so.10
2024-04-24 09:23:04.923836: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcurand.so.10
2024-04-24 09:23:04.924462: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcusolver.so.11
2024-04-24 09:23:04.925251: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcusparse.so.11
2024-04-24 09:23:04.925389: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcudnn.so.8
2024-04-24 09:23:04.925905: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1871] Adding visible gpu devices: 0
[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
will this change impact the functionalities of cryoCARE ?
sorry if this sounds a dumb question, I'm just an IT guy who was asked to install cryoCARE ;-)
Hello
after a fresh installation from 0.3.0 tagged release (https://github.com/juglab/cryoCARE_pip/archive/refs/tags/v0.3.0.tar.gz)
using
cuda/11.8,cudnn/11.x-v8.7.0.84I encountered a small problem with TF-2.4 as recommended in the documentation. see:
TF fails on
libcusolver.so.10after upgrading TF to version 2.5 roblem solved. see:
will this change impact the functionalities of
cryoCARE?sorry if this sounds a dumb question, I'm just an IT guy who was asked to install cryoCARE ;-)
regards
Eric