Background
I did an import of an enviroment from an old installation as stated here (I renamed it DefaultEnv). It was working for a couple of months, but now I have a problem, all of the sudden my Jupyter notebook instance for this environment was gone, I am only able to run it directly from the console
I noticed that base(root) version for Jupyter notebooks is 6.4.5 where DefaultEnv is 6.3.0 one, so I thought that may be the problem and tried to update it, but I got the following:
failed with repodata from current_repodata.json, will retry with next repodata source.
I tried with this solution but it didn't work, if it helps, I am pasting the conda console log.
(DefaultEnv) PS C:\Users\12345> conda config --set channel_priority flexible
(DefaultEnv) PS C:\Users\12345> conda update --all
Collecting package metadata (current_repodata.json): done
Solving environment: |
The environment is inconsistent, please check the package plan carefully
The following packages are causing the inconsistency:
- anaconda/win-64::anaconda==custom=py38_1
- conda-forge/win-64::astropy==4.2.1=py38h294d835_0
- anaconda/win-64::bkcharts==0.2=py38_0
- conda-forge/win-64::bokeh==2.3.0=py38haa244fe_0
- anaconda/win-64::bottleneck==1.3.2=py38h2a96729_1
- conda-forge/noarch::dask==2021.4.0=pyhd8ed1ab_0
- conda-forge/win-64::h5py==2.10.0=nompi_py38he6c2248_106
- conda-forge/win-64::imagecodecs==2021.3.31=py38hccd6b70_0
- anaconda/noarch::imageio==2.9.0=py_0
- anaconda/noarch::keras-applications==1.0.8=py_1
- conda-forge/noarch::keras-preprocessing==1.1.2=pyhd8ed1ab_0
- conda-forge/win-64::matplotlib==3.4.1=py38haa244fe_0
- conda-forge/win-64::matplotlib-base==3.4.1=py38heae8d8c_0
- conda-forge/win-64::mkl_fft==1.3.0=py38h347fdf6_1
- conda-forge/win-64::mkl_random==1.2.0=py38h251f6bf_1
- conda-forge/win-64::numba==0.53.1=py38h99fac5f_1
- conda-forge/win-64::numexpr==2.7.3=py38h5d928e2_2
- conda-forge/noarch::opt_einsum==3.3.0=pyhd8ed1ab_1
- conda-forge/win-64::pandas==1.2.3=py38h4c96930_0
- anaconda/win-64::patsy==0.5.1=py38_0
- conda-forge/win-64::pyerfa==1.7.2=py38h294d835_0
- conda-forge/win-64::pytables==3.6.1=py38h153c448_3
- anaconda/win-64::pywavelets==1.1.1=py38he774522_2
- conda-forge/win-64::scikit-image==0.18.1=py38h4c96930_0
- conda-forge/win-64::scipy==1.6.2=py38he847743_0
- conda-forge/noarch::seaborn==0.11.1=hd8ed1ab_1
- conda-forge/noarch::seaborn-base==0.11.1=pyhd8ed1ab_1
- conda-forge/win-64::statsmodels==0.12.2=py38h6f4d8f0_0
- conda-forge/noarch::tensorboard==2.4.1=pyhd8ed1ab_1
- conda-forge/noarch::tensorflow-estimator==2.4.0=pyh9656e83_0
- conda-forge/noarch::tifffile==2021.3.31=pyhd8ed1ab_0
- anaconda/win-64::_anaconda_depends==2020.07=py38_0
failed with repodata from current_repodata.json, will retry with next repodata source.
I tried this solution as well, but the message changed to:
failed with initial frozen solve. Retrying with flexible solve.
I wouldn't mind to re create the environment, but I would like to avoid this in the future. I saw that 2 instances of the anaconda navigator existed (one for root and one for this DefaultEnv), so I think there may be something related in the libraries that got backed up in the first place.
Question
How could I be able to have Jupyter from the "DefaultEnv" without the need to run it from the console and how to avoid this "corruption" in my environment backup for future references?
