I am currently working on a project where I need to work with very specific gpu drivers and python packages. I am able to build the environment with those packages but they do not reflect in the system. Below is a summary of the packages in my environment(i have cropped out the few but include the necessary ones):
packages in environment at /home/#####/anaconda3/envs/AML_task_itr_1:Name Version Build Channel_libgcc_mutex 0.1 main _openmp_mutex 5.1 1_gnu _tflow_select 2.3.0 mkl ***cudatoolkit 11.8.0 h6a678d5_0***numpy 1.26.4 py39h5f9d8c6_0 numpy-base 1.26.4 py39hb5e798b_0 oauthlib 3.2.2 py39h06a4308_0 openssl 1.1.1w h7f8727e_0 opt_einsum 3.3.0 pyhd3eb1b0_1 packaging 23.2 py39h06a4308_0 pip 23.3.1 py39h06a4308_0 protobuf 3.20.3 py39h6a678d5_0 pyasn1 0.4.8 pyhd3eb1b0_0 pyasn1-modules 0.2.8 py_0 pycparser 2.21 pyhd3eb1b0_0 pyjwt 2.4.0 py39h06a4308_0 pyopenssl 23.2.0 py39h06a4308_0 pysocks 1.7.1 py39h06a4308_0 ***python 3.9.18 h7a1cb2a_0*** python-flatbuffers 2.0 pyhd3eb1b0_0 ***tensorboard 2.10.0 py39h06a4308_0*** tensorboard-data-server 0.6.1 py39h52d8a92_0 tensorboard-plugin-wit 1.8.1 py39h06a4308_0 tensorflow 2.10.0 mkl_py39hdc5b6f4_0 tensorflow-base 2.10.0 mkl_py39hb9daa73_0 tensorflow-estimator 2.10.0 py39h06a4308_0 **now within this environment i try to check my gpu version i am getting the following info**(AML_task_itr_1) #####:~$ nvidia-smiSat Apr 6 00:58:41 2024 +---------------------------------------------------------------------------------------+| NVIDIA-SMI 535.161.07 Driver Version: 535.161.07 ***CUDA Version: 12.2*** ||-----------------------------------------+----------------------+----------------------+| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC || Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. || | | MIG M. ||=========================================+======================+======================|| 0 NVIDIA GeForce RTX 2070 Off | 00000000:01:00.0 Off | N/A || N/A 36C P5 26W / 40W | 6MiB / 8192MiB | 0% Default || | | N/A |+-----------------------------------------+----------------------+----------------------+I have highlighted the gpu version in both cases.I am working on Ubuntu 22.04 LTSNote the current env (AML_task_itr_1) info is not that of the base environment and the nvidia-smi command show the native system configure though i was expecting this command to show me the configuration for the current conda environment.I have highlighted the packages of my interest
I wish to work with cuda=11.8 but i am not able to configure the environment please help.
please help