My_learning-Python

Description

Learn Python for using in the following domains:

• Object Oriented Programming (OOP), Functional Programming (FP)
• Data Science
• Machine Learning
• Web Development
• Chatbot

Install

Windows (with WSL Ubuntu)

• Editor (with suggestion): Sublime Text 3 (Crack version with setup), suggestion
• Interpreter
  • Anaconda (on Windows)
  • Python3 (on Linux (Ubuntu)) (If required)

Linux

• [Linux (Ubuntu)]: sudo apt install python3

  • And use pip as package manager. Install conda using this.

  NOTE: Why not Anaconda in Linux?

  • If already installed in Windows, & Linux is running on WSL. Then, try using anaconda in Windows, so that it will not add (so many approx. 600 MB of) packages additionally in the PC/Laptop.

  • Anaconda is mainly installed because of built-in IDEs like Jupyter notebook, Spyder.

  • Also, both conda & pip package manager comes alongwith.

• Jupyter Notebook by Anaconda (used mostly in debugging each module)

• Sublime text 3 (Install conda packages after Anaconda installation) - Excel-python (with xlwings), Dash (Dashboard), Django (Web App)

Note: It depends on the project. If it consists of multiple dependent python files, then use ST3, otherwise use Jupyter Notebook.

macOS

1. python3 comes pre-installed in macOS via Developer CommandLineTools. Check using $ python3 --version command. Now, needs to superimpose using the python of brew.
   1. Using brew, $ brew install python@3.11, we can have multiple python versions: python3.12 or python3.11 etc.
   2. Update: $ brew upgrade python@3.11 to update python3.11. Similarly, $ brew upgrade python@3.12 for python3.12.
2. Install pip (aka pip3 for some python versions > 3) using your preferred python3 version.

python3.11 -m pip install --upgrade pip

It is recommended to install the packages using pip3 (not conda) (aka pip). And then check if installed using pip3 list or pip list.

In order to know which python3 version is linked to which pip or pip3, use $ pip --version or $ pip3 --version command.

This is the result in my case:

$ which pip
/opt/homebrew/bin/pip
$ pip --version                                                                                                                             ⏎
pip 24.0 from /opt/homebrew/lib/python3.11/site-packages/pip (python 3.11)

$ which pip3
/opt/homebrew/bin/pip3
$ pip3 --version
pip 24.0 from /opt/homebrew/lib/python3.12/site-packages/pip (python 3.12)

This means pip is installed via python3.11 (via brew). So, install/update via python3.11.

3. Project Management tool: huak is recommended, but poetry is also good, but 10x slow.

4. Editor: VSC (Visual Studio Code)

   • Extensions
     • Python
     • Jupyter extension packages
     • pylyzer
     • Pylance & many others.

5. Enable Auto-complete suggestion: For enabling autocomplete suggestion, add this line in settings.json:

  "python.analysis.autoImportCompletions": true,
  "python.analysis.extraPaths": ["/opt/homebrew/lib/python3.11/site-packages"]

---

To uninstall, follow this:

1. Open terminal on ~ directory.
2. Run conda install anaconda-clean
3. Run anaconda-clean --yes
4. Run rm -rf ~/opt/anaconda3
5. Close and reopen your terminal to refresh it. You should no longer see (base) in your terminal prompt.

Editors

Sublime Text 3 [DEPRECATED]

I don't use it anymore.

• Anaconda installed & added to PATH in Windows.
• Packages
  • Anaconda
• Build system
  • default
• Snippet
  • None

Visual Studio Code [RECOMMENDED]

Package managers

• Anaconda
• Pip

Libraries

• pandas
• xlwings
• matplotlib
• plotly
• Formatter: ruff, black, autopep8, yapf

Troubleshoot

1. Error while opening jupyterlab on cmd

Error executing Jupyter command 'lab': [WinError 5] Access is denied

• Cause: As the jupyterlab is not installed as Admin during the Anaconda installation. That's why need to be either followed: M-1 or M-2.

• Solution:

  • M-1:

  python3 -m jupyterlab

  • M-2:

  step-1: open your anaconda navigator
  step-2: click on jupyter lab setting option and then remove it
  step-3: Now open your anaconda prompt and type the following command `conda install -c conda-forge jupyterlab1

2. moduleNotFoundError: No module named '...' on macOS

• Reason: This error is when importing package from anaconda packages on macOS. As the package is not installed in the pip environment, but in conda environment, python3 is not able to find the package.
• Solution:
  • M-1: Install the package in pip environment.
  • [RECOMMENDED] M-2: If using zsh, edit ~/.zshrc & add this line:

# import python packages from Anaconda rather than system python
export PATH="$HOME/opt/anaconda3/bin/:$PATH"

Now, you can import the package successfully inside Jupyter Notebook & also in a separate python file.

3. Jupyter kernel issue on macOS

• Reason: When using jupyter notebook on macOS, it shows the following error:

No kernel found

• Solution:

  • M-1: Install the kernel for the python version you are using. For example, if using python3.9, then install the kernel for python3.9.

  python3.9 -m pip install ipykernel

  • M-2: Connect to a already running jupyter notebook kernel. Just run $ jupyter notebook command in the directory where the notebook is present.

Repositories

Self

• Semantic Hashing Demo
• AutoPlot

Others

• Python Algorithms - https://github.com/TheAlgorithms/Python
• HackerRank Problems and Solutions - https://github.com/juhilsomaiya/HackerRank-Python-Algorithm-Solution
• Plotly Offline Charts - https://github.com/SayaliSonawane/Plotly_Offline_Python
• Plotly Dashboards with Dash - https://github.com/Pierian-Data/Plotly-Dashboards-with-Dash
• Plotly Datasets - https://github.com/plotly/datasets
• Dash plotly recipes (for discussion forum) - https://github.com/plotly/dash-recipes
• Pandas Exercises - https://github.com/guipsamora/pandas_exercises
• Zulip (powerful open source team chat) - https://github.com/zulip/zulip
• Code Beautify (into Image) - LINK
• Dash Bootstrap Components - https://dash-bootstrap-components.opensource.faculty.ai/
• Automate docs with Python - Documentation, Github, pip, Example
• Automate PPT with Python - Documentation, Github, pip
• Python Geocoding Toolbox (location, coordinates, address, street, city, country) - pip, example
• Geocoding and reverse geocoding using Python - https://chrisalbon.com/python/data_wrangling/geocoding_and_reverse_geocoding/
• Beautiful formatting in Console - https://github.com/willmcgugan/rich
• Learning Python through test-driven development of games and puzzles - https://github.com/kyclark/tiny_python_projects
• Open source home automation that puts local control and privacy first - https://github.com/home-assistant/core
• Apprise - Push Notifications that work with just about every platform! - https://github.com/caronc/apprise

Data Visualization

• Dash Plotly - https://dash.plot.ly/
• Django with Dash - Online Documentation, pip
• Python all charts - https://python-graph-gallery.com/

Machine Learning

References

• Pandas dropping rows and columns
• Python Tips
• Solving Memory related issues
• Learn Maths in simple way

TODO

Code

• https://towardsdatascience.com/elements-of-functional-programming-in-python-1b295ea5bbe0