Develop Reference

pip freeze gives different version from pip --version

I am on Windows 10 and just installed Python 3.7.4. I installed virtualenv using pip install virtualenv, and i got the message that i installed 16.7.2 while 19.2.1 is available. I then used python -m pip install --upgrade pip to upgrade.
Here is the issue: when i type pip freeze on Powershell, i am getting this:
virtualenv==16.7.2
but when i type pip --version, i am getting this:
pip 19.2.1 from c:\users\user1\python\lib\site-packages\pip (python 3.7)
Can you please let me know why pip freeze still shows the old version?
Thanks.

pip and virtualenv are 2 different cases where pip is mainly used for installing python packages and virtualenv for utilizing virtual environments while development.
The following command updates only pip not virtual env
python -m pip install --upgrade pip
If you think you still need to upgrade virtualenv (of course if a new version is available). Please follow the following syntax without specifying a version.
pip install [package] --upgrade

Why does pip install not work for catboost?

I have to install catboost but can not make it by pip install catboost.
There is not catboost library in Anaconda, so pip in the one way.
The error message is:
Could not find a version that satisfies the requirement catboost <for version: >
No matching distribution found for catboost.
Python version is 3.6.3.
Screenshot of error:
error message
I've tried :
pip install catboost==0.12.2
pip install catboost==0.12.1.1
pip install catboost==0.12.1
pip install catboost==0.12.0
and
pip install catboost==0.11.0
pip install catboost==0.10.2
None of these works.
Why does this problem appeared and is there another way to install catboost?

From the docs
Installation is only supported by the 64-bit version of Python.
You need to reinstall the 64 bit version of python to use the cat boost package

I had the same problem but it was mainly related to Docker because the error was only occurring when I was trying to install it through a docker - it turned out there is something related to M1 Apple architecture and the solution was to modify the docker command as following
docker buildx build --platform=linux/amd64 -t ${IMAGE_TAG} -f Dockerfile

For me the issue was I was developing on an M1 Mac. Trying to install as root, using conda and using docker images were all unsuccessful.
The fix I found was using the amd64 docker image as my base. The dockerfile code was:
FROM amd64/python:3.9-buster
RUN pip install -U pip
RUN pip install --upgrade setuptools
RUN pip install catboost
Can't exactly remember how I arrived at the above code, but thanks to https://github.com/prabodh1194 for putting this together. A dockerfile with this code can be found here (as of 25NOV2021) https://github.com/prabodh1194/docker_images/blob/main/catboost/0.26/Dockerfile

in my case, a virtual environment was using an old version of pip and a simple upgrade of pip worked!
python3 -m pip install --upgrade pip

How to upgrade pip in virtual environment?

This question has already been asked here, though there is no answer solving this problem.
pip upgrade fails inside virtual environment
The above question is duplicate and the references following answer:
gaierror: [Errno -2] Name or service not known
But this answer doesnot contain anything valid. The former answer also does not contain any helpful info!
My problem is exactly what is described in first link. I try upgrade pip from 10.0.1 to 18.1 but unable to. As a consequence I am unable to install Tensorflow. This Virtual environment is actually connected to pycharm and I am referencing it through command line. Here is the input and corresponding commandline output:
(venv) G:\Ashish\Tensorflow\venv\Scripts>pip install --upgrade pip
ERROR: To modify pip, please run the following command:
G:\Ashish\Tensorflow\venv\Scripts\python.exe -m pip install --upgrade pip
You are using pip version 10.0.1, however version 18.1 is available.
You should consider upgrading via the 'python -m pip install --upgrade pip' command.
(venv) G:\Ashish\Tensorflow\venv\Scripts>G:\Ashish\Tensorflow\venv\Scripts\python.exe -m pip install --upgrade pip
Requirement already up-to-date: pip in g:\ashish\tensorflow\venv\lib\site-packages (18.1)
(venv) G:\Ashish\Tensorflow\venv\Scripts>pip --version
pip 10.0.1 from g:\ashish\tensorflow\venv\lib\site-packages\pip-10.0.1-py3.7.egg\pip (python 3.7)
(venv) G:\Ashish\Tensorflow\venv\Scripts>pip install --upgrade pip
ERROR: To modify pip, please run the following command:
G:\Ashish\Tensorflow\venv\Scripts\python.exe -m pip install --upgrade pip
You are using pip version 10.0.1, however version 18.1 is available.
You should consider upgrading via the 'python -m pip install --upgrade pip' command.
(venv) G:\Ashish\Tensorflow\venv\Scripts>
I tried uninstalling the pip from virtual environment. I did not work. I am sorry if this is basic or involves basic knowledge about virtual env, This is my learning phase and some time before I can start to contribute to this awesome community.

Ok, solved the problem, thanks to suggestions of #hoefling and#Rahul Bhardwaj
For python environment showing 10.0.1 and 18.1:
pip in your virtual environment is 18.1 while the global pip is 10.0.1. So the issue is solved, simply upgrade global pip
For tensorflow not installing:
Install python 3.6
Add that to path
Create a virtual environment in pycharm using python 3.6
Install tensorflow
Done! Python 3.7 is yet to support tensorflow

Installing opencv on Ubuntu 16-04

I have a virtualenv created with virtualenvwrapper.
However, I'm on python 2.7
Whenever I try to install the opencv-python by pip, as sudo or not, I receive this message:
pip install opencv-python
Downloading/unpacking opencv-python
Could not find any downloads that satisfy the requirement opencv-python
Cleaning up...
No distributions at all found for opencv-python
Storing debug log for failure in /home/kristian/.pip/pip.log
Any ideas on why this us happening?

Your pip is most likely too old since opencv-python packages are distributed as manylinux1 wheels (https://github.com/pypa/manylinux). You need pip version 8.1 or later to install manylinux1 binary packages.
To upgrade your pip globally:
sudo -H pip install --upgrade pip
Inside virtualenv this should be enough:
pip install --upgrade pip

I recommend you to use anaconda, which is very convenient and saves a lot of trouble. With anaconda, you don't need to care about your local environment.

How to downgrade the installed version of 'pip' on windows?

On a windows 7 machine I have pip version 1.5.6 installed:
pip 1.5.6 from C:\Users\dietz\PNC\tas\ENV\lib\site-packages (python 2.7)
In order to find the reason for an error I want to install a different version of pip, which worked fine for me. So how can I uninstall pip and install version 1.2.1 instead?

pip itself is just a normal python package. Thus you can install pip with pip.
Of cource, you don't want to affect the system's pip, install it inside a virtualenv.
pip install pip==1.2.1

If downgrading from pip version 10 because of PyCharm manage.py or other python errors:
python -m pip install pip==9.0.1

If you want to upgrade or downgrade to different version of pip, better use --upgrade option at one go instead doing it in two steps. i.e. first uninstalling the existing and then re-installing to new version, below does both in one go as shown below.
USE: Executed on WIN10 with Bash
python -m pip install --upgrade pip==19.2.3
$ python -m pip install --upgrade pip==19.2.3
Collecting pip==19.2.3
Using cached pip-19.2.3-py2.py3-none-any.whl (1.4 MB)
Installing collected packages: pip
Attempting uninstall: pip
Found existing installation: pip 21.3.1
Uninstalling pip-21.3.1:
Successfully uninstalled pip-21.3.1
Successfully installed pip-19.2.3

well the only thing that will work is
python -m pip install pip==
you can and should run it under IDE terminal (mine was pycharm)

If you have to downgrade pip version do following steps:
step1. pip uninstall pip
step2. pip install pip==version number you want to install or downgrade
step3. check version of pip using pip --version
This process also works when any other package giving error exit code(2) you can follow these steps and install your package.