Search code examples
condaintelintel-oneapiintel-python

Unable To Create Python Environment Using DevCloud Account

I am not able to create python environments using my devcloud account in my default home directory, Along with the below command, I also tried with the commands given in this documentation "https://www.intel.com/content/www/us/en/developer/articles/technical/using-intel-distribution-for-python-with-anaconda.html"

conda create aikit-tf

After running this command the aikit-tf environment is not created and I got CondaValueError and returned to the command line .

enter image description here

Solution

  • The reason you are not able to create an environment you are trying on login node. It can be easily worked if you'll create environment using compute node. 

    There are Login Node and Compute Node in Intel® DevCloud. Both are use for different purpose. Login node uses a lightweight general-purpose processor. Compute node uses an Intel® Xeon® Gold 6128 processor that is capable of handling heavy workloads.

    for ex: uxxxx@login-2:~

    Initially, you will be in the DevCloud login node. To enter into the compute node you have to use below command.

    qsub -I

    and now you are in compute node as follows:

    uxxxx@c009-n014:~

    Now you can create an environment with the below command in the compute node.

    conda activate aikit-tf

    Refer this document for more information: https://community.intel.com/t5/Intel-DevCloud/Login-Node-Versus-Compute-Node-in-Intel-DevCloud/td-p/1211227