Develop Reference

downgrade sklearn 0.22.3 to 0.22.2 in anaconda prompt?

When I run in jupyter file
import sklearn
print(sklearn.__version__)
The result is 0.23.2
However I would like to install 0.22.2. Therefore I put into prompt
conda remove scikit-learn
and install:
conda install scikit-learn=0.22.2
unfortunately there is this error:
PackagesNotFoundError: The following packages are not available from current channels: scikit-learn=0.22.2
How can I delete sklearn version 0.23.2 and install 0.22.2?

You should try adding the conda-forge channel to your list of channels when you search for packages.
This should work:
RUN
conda config --append channels conda-forge
and
conda install scikit-learn=0.22.2
OR
RUN
conda install -c conda-forge scikit-learn=0.22.2
instead of 2 above commands

UnsatisfiableError conda-forge 3.7

I've been having trouble installing some packages with conda (on python 3.7 on Windows in a conda env). For example if I try install the shap package:
conda install -c conda-forge shap
I get
Solving environment: failed
UnsatisfiableError: The following specifications were found to be in conflict:
- conda-forge/win-64::wincertstore==0.2=py37_1002
- shap
Use "conda info <package>" to see the dependencies for each package.
It installs fine with pip. But how could I resolve the error to use conda instead?

How to install boruta in python using conda install

I'm using python 3.6 on windows 10. I want to install boruta in my python.
So I used following command conda install -c conda-forge r-boruta but got error message
PackagesNotFoundError: The following packages are not available from current channels:
I also tried conda install boruta but got the same error message. Can you suggest me installation steps

As per official documentation of anaconda https://anaconda.org/saravji/boruta if you want to install boruta using conda install use conda install -c saravji boruta or conda install -c conda-forge boruta_py and make sure that your python version is less than equal to 3.6.

Try this:
conda config --add channels conda-forge
conda install r-boruta
Source: https://github.com/conda-forge/r-boruta-feedstock

Trouble installing 'easydict' package from conda prompt

I am trying to install easydict package from conda prompt using following command
conda install easydict
But I am getting the following error
PackageNotFoundError: Packages missing in current channels:
easydict
We have searched for the packages in the following channels:
...
I am using Anaconda Navigator 1.6.4

easydict is a module which is currently not offered through the default standard conda channel. Instead, you can use e.g. the conda-forge channel:
conda install easydict -c conda-forge
The auto channel recommended in the post above has the disadvantage that it will potentially affect many other packages, most importantly downgrading python3 to python2.7

How do I upgrade to Tensorflow 1.0 using anaconda?

I have an anaconda installation of tensorflow (version 0.9.0) and I can't upgrade it to 1.0.
When I run
conda install tensorflow=1.0.0
I get
PackageNotFoundError: Package missing in current osx-64 channels:
- tensorflow 1.0.0*
When I try
pip install --ignore-installed --upgrade https://storage.googleapiscom/tensorflow/mac/cpu/tensorflow-1.1.0-py3-none-any.whl
It times out. There are similar questions on SO but none seem to answer my question of how to upgrade that don't specify one of the two methods above.

Tensorflow 1.0.0 is not available in the default Anaconda channel for OS X. Check with "conda list" that tensorflow is not already installed on your system. If it is, remove using
conda uninstall tensorflow
You can install 1.0.0 by installing from the conda-forge channel
conda install -c conda-forge tensorflow=1.0.0

For anaconda installation, first pick a channel which has the latest version of TensorFlow binary. The latest versions are usually available at the channel conda-forge. So, simply do:
# `-f` will force the current installation to upgrade
# `-c conda-forge` means we select `conda-forge` channel
$ conda update -f -c conda-forge tensorflow
This will upgrade your existing TensorFlow installation to the very latest version available. As of this writing, the latest version is 1.4.0-py36_0