Conda Create Environment Takes Forever

Question:

Is it only me or is it like this with Conda? I’m having a environment file that looks like this:

name: python-cvcourse
channels:
  - menpo
  - conda-forge
  - defaults
dependencies:
  - jupyter=1.0.0
  - jupyterlab=0.34.9
  - keras=2.2.2=0
  - matplotlib=2.2.3
  - numpy=1.15.1
  - opencv=4.1.0
  - pandas=0.23.4
  - python=3.6.6
  - scikit-learn=0.19.1
  - scipy=1.1.0
  - tensorboard=1.10.0
  - tensorflow=1.10.0
  - pillow=8.3.2
prefix: /anaconda3/envs/cvcourse

I’m using the following command to update the environment:

conda env update -f requirements.yaml

It has been at this prompt below for more than few minutes and I’m starting to think if there is something that could be done about it:

(python-cvcourse) joesan@joesan-InfinityBook-S-14-v5:~/Projects/Private/sandbox-projects/udemvy-opencv-python$ conda env update -f cvcourse_linux.yml
Collecting package metadata (repodata.json): done
Solving environment: - 
Found conflicts! Looking for incompatible packages.
This can take several minutes.  Press CTRL-C to abort.
 
Asked By: joesan

||

Answers:

I just tried running your yaml file myself and I was able to successfully create a conda environment with it. Instead of running conda env update try creating the environment from scratch with the yaml? The following is the command I ran:
conda env create -f environment.yml.
Where environment.yml is the environment file you provided. Keep it mind it did take a couple of minutes to create the environment. Hope this is of some help.

Answered By: Nauf_M

I noticed that in your environment file you are asking for very specific versions of each of the packages. This probably increases the risk of conflicts between the requested packages because a specific version of package A may depend on a specific version of package B. If, in addition to this, you are request a lot of packages the combination of possible inconsistencies rapidly increases.

I would therefore suggest you relax some of the requirements and try abstaining from requesting a specific version of a (few) package(s) where you think this is acceptable. (I would get rid of all version specifications for a start to let conda sort out what environment would actually be feasible. You can later add version specifications where you feel this is necessary.)

Answered By: Telefonlawine
Categories: questions Tags: ,
Answers are sorted by their score. The answer accepted by the question owner as the best is marked with
at the top-right corner.