topic I think it might have in Intel® Distribution for Python*

jupyterhub - conda wants to replace intel::python-3.6.8-7

Pawel_E_ — Wed, 14 Aug 2019 09:48:38 GMT

hi guys,

would know how to get jupyterhub?

$  conda install jupyterhub

...

The following packages will be SUPERSEDED by a higher-priority channel:

  python                              intel::python-3.6.8-7 --> pkgs/main::python-3.6.8-h0371630_0


Proceed (/n)?

many thanks, L.

I think it might have

Pawel_E_ — Wed, 14 Aug 2019 14:53:54 GMT

I think it might have something to do with conda got updated to conda-4.7.11 and then kind of hell broke loose. I wanted to do:

$ conda update --all
Collecting package metadata (current_repodata.json): done
Solving environment: done

## Package Plan ##

  environment location: /opt/intel/intelpython3


The following packages will be REMOVED:

  conda-env-2.6.0-1
  cycler-0.10.0-py36_7
  cython-0.29.6-py36h7b7c402_0
  daal-2019.4-intel_243
  daal4py-2019.4-py36h7b7c402_6
  freetype-2.9-3
  funcsigs-1.0.2-py36_7
  icc_rt-2019.4-intel_243
  impi_rt-2019.4-intel_243
  intel-openmp-2019.4-intel_243
  ipp-2019.4-intel_243
  kiwisolver-1.0.1-py36_2
  libpng-1.6.36-2
  llvmlite-0.27.1-py36_0
  matplotlib-3.0.3-py36_4
  mkl-2019.4-intel_243
  mkl-service-1.0.0-py36h7b7c402_11
  mkl_fft-1.0.13-py36h7b7c402_1
  mkl_random-1.0.4-py36h7b7c402_1
  mpi4py-3.0.0-py36_3
  numba-0.42.1-np116py36_2
  numexpr-2.6.8-py36_2
  numpy-1.16.2-py36h7b7c402_0
  numpy-base-1.16.2-py36_0
  pandas-0.24.1-py36_3
  pyeditline-2.0.0-py36_0
  pyparsing-2.2.0-py36_2
  python-dateutil-2.6.0-py36_12
  pytz-2018.4-py36_3
  pyyaml-4.1-py36_3
  scikit-learn-0.20.3-py36h7b7c402_5
  scipy-1.2.1-py36h7b7c402_3
  smp-0.1.4-py36_0
  tbb-2019.6-intel_243
  tbb4py-2019.6-py36_intel_0
  xgboost-0.81-py36_0

I'm very new to both intel & conda and hope some experts could shed more light.

thanks, L.

I believe you need to keep

Orion_P_ — Fri, 23 Aug 2019 18:30:34 GMT

I believe you need to keep specifying your channel preferences in all your conda commands. So try:

conda install -c intel jupyterhub

and

conda upate -c intel --all

In fact, I tend to use something like:

conda COMMAND --override-channels --strict-channel-priority -c intel -c ...

in order to reduce depsolving time.