Environment

Environment setup

1. Using conda

1. Add Anaconda Portable to PATH

For Command Prompt (cmd.exe)

set PATH=D:\anaconda3\condabin;%PATH%

For PowerShell

$env:PATH = "D:\anaconda3\condabin;" + $env:PATH

2. Set Default PATH in Visual Studio Code

Step 1: Open VSCode Settings

1. Open VSCode.
2. Press Ctrl + , to open the Settings.

Step 2: Edit settings.json

1. In the Settings search bar, type: terminal.integrated.env.windows.
2. Click on Edit in settings.json (you can also directly open settings.json by pressing Ctrl + Shift + P and typing “Preferences: Open Settings (JSON)”).

Step 3: Add Anaconda Path to settings.json

In the settings.json file, add the following lines to include Anaconda’s condabin in your PATH:

{
    "terminal.integrated.env.windows": {
        "PATH": "D:\\anaconda3\\condabin;${env:PATH}"
    }
}

3. Create conda environment

1. Quick create env using conda create

conda create -n env_rdm python=3.10 pip ipykernel notebook
conda activate env_rdm

2. Create env use environment.yaml file

• First, create environment.yaml file with example content following
• Second, run command conda env create -f environment.yaml
• Optional, remove environment if needed conda env remove -n env_ascore

name: env_ascore
channels:
  - conda-forge
  - defaults
dependencies:
  - python=3.10
  - pandas
  - joblib
  - statsmodels
  - ipykernel
  - zipp  
  - pip
  - pip: 
    - optbinning==0.17.3
    - ortools==9.4.1874
# set https_proxy=10.1.33.23:8080
# set http_proxy=10.1.33.23:8080   

3. Create env use requirements.txt file

conda list --export > requirements.txt
conda install --file requirements.txt

4. Conda Common Commands

List All Conda Environments

This command displays a list of all available Conda environments on your system:

conda info --env

Alternatively, you can use:

conda env list

Remove a Conda Environment

To remove a specific Conda environment (e.g., python38):

conda deactivate  # Deactivate any active environment first
conda env remove -n python38

Activate a Conda Environment

To activate a specific Conda environment (e.g., python38):

conda activate python38

Clean Unused Libraries and Cache

To clean up unused libraries and cache, use:

conda clean --all  # Cleans all Conda caches
pip cache remove *  # Removes all pip caches

2. Create Environment for Jupyter Notebook

1. Use Conda to Create a New Environment

You can create a new environment in Conda for use with Jupyter Notebook. Replace python38 with the desired environment name and 3.x with the desired Python version.

conda create -n python38 python=3.x

2. Use ipython to Add the Environment to Jupyter

After activating the environment, use the ipython command to install the environment’s kernel for Jupyter Notebook:

conda activate python38
ipython kernel install --user --name=python38

This will make the environment available in Jupyter Notebook as an option when selecting kernels.

3. Remove Jupyter Notebook Environment (Run as Administrator)

To remove the environment’s kernel from Jupyter, list all the installed kernels and then uninstall the specific one. You may need administrator privileges for this step.

1. List the installed kernels:

jupyter kernelspec list

2. Uninstall the kernel:

jupyter kernelspec uninstall python38

3. Install Packages in OFFLINE Mode with pip

1. Install Offline Packages Using requirements.txt

Step 1: Export Installed Packages to requirements.txt

If you want to create a requirements.txt file containing the list of installed packages, you can use this command:

pip list --format=freeze > requirements.txt

This will export the list of currently installed packages and their versions into a requirements.txt file.

Step 2: Create a wheel Folder

Create a folder where the downloaded .whl (wheel) files will be stored. For example, you can create a folder named wheel:

mkdir D:\wheel

Step 3: Download Dependencies into the wheel Folder

Run the following command to download the dependencies listed in requirements.txt into the wheel folder:

pip download -r requirements.txt -d D:\wheel

This command will download all the required packages and their dependencies into the wheel directory for offline installation.

Step 4: Install Packages from the wheel Folder

Once you have downloaded all necessary packages, you can install them offline by running:

pip install -r requirements.txt --find-links=D:\wheel --no-index

• --find-links=D:\wheel: Instructs pip to look for package files in the wheel folder.
• --no-index: Disables checking online package indexes (e.g., PyPI) to ensure installation from offline files.

---

2. Install Offline Linux Packages

When downloading Python packages for Linux (or another platform) while being on a different platform (e.g., Windows), you can specify the platform and Python version.

Case 1: Activate the Same Python Version

If you are working with the same Python version as required (e.g., Python 3.7), activate the appropriate environment and run the following command to download the necessary package for Linux:

pip download --platform manylinux1_x86_64 --only-binary=:all: --no-binary=:none: pandas

• --platform manylinux1_x86_64: Specifies the platform for the Linux package.
• --only-binary=:all:: Ensures only binary files (like wheels) are downloaded.
• --no-binary=:none:: Prevents downloading source distributions, ensuring only precompiled binaries are fetched.

Case 2: Specify a Python Version

If you are using a different Python version, specify the required version using the --python-version flag:

pip download --platform manylinux1_x86_64 --only-binary=:all: --python-version=38 --no-binary=:none: pandas

• --python-version=38: Specifies the Python version (e.g., 3.8) for which you need to download packages.

4. Other Utility Commands

1. Check for Dependency Issues

To check for problems with package dependencies:

python -m pip check

To save a list of all installed packages to requirements.txt:

pip freeze > requirements.txt

2. Change Jupyter Notebook Theme

To change the theme of Jupyter Notebook (for example, applying the onedork theme with custom font sizes and cell width):

jt -t onedork -fs 13 -altp -tfs 14 -nfs 14 -cellw 88% -T

• -fs: Font size.
• -tfs: Title font size.
• -nfs: Notebook name font size.
• -cellw: Cell width.
• -T: Show or hide the toolbar.

To install Jupyter themes:

pip install jupyterthemes

3. Install Jupyter Notebook Extensions

To add useful features like a table of contents, code folding, and more, install the extensions:

pip install jupyter_contrib_nbextensions
pip install jupyter_nbextensions_configurator

Then, set them up and enable them:

jupyter contrib nbextension install --user
jupyter nbextensions_configurator enable --user