The Culture of Python

This is a demo of getting started with Jupyter, and shows the culture of Python.

Install

You can use Anaconda or Python to run Jupyter. Anaconda is great for the non-geeks. We'll use Python.

1. Create a virtual environment:

   python -m venv .venv
   

   Activate it:

   Linux:

   source .venv/bin/activate
   

   Windows Powershell:

   .venv/Scripts/activate
   

   Windows Batch:

   .venv/Scripts/activate.bat
   

2. Install Python libraries:

   pip install -r requirements.txt
   

Run Jupyter

Pick one:

Launch as a web page:

jupyter notebook

then open the browser as directed in the console output

OR

Launch in VS Code:

1. Install VS Code

2. Install the Python extension and/or the Jupyter extension. (The Python extension installs the Jupyter extension too)

3. Choose the Python environment:

   a. cntrl-shift-P

   b. type Python: Select interpreter

   c. choose the line with .ven in it

Playing with Jupyter

1. Open 1-fibonacci.ipynb

2. Run each cell by clicking the triangle to the left of each code block

Use case 1: Data transform from MSSQL to Postgres

The main libraries are:

• SqlAlchemy: SQL ORM (requires drivers)
• Pandas: data processing

1. Start the databases

   docker-compose up
   

   This will launch SQL Server and Postgres and load up data into SQL Server

2. Open 2-etl.ipynb

3. Run each cell by clicking the triangle to the left of each code block

Use case 2: Debugging an app with a Runbook

1. Open 3-runbook.ipynb

2. Run each cell by clicking the triangle to the left of each code block

Use case 3: Graphing data

The main libraries here are:

• NumPy: scientific math
• Pandas: data processing, data frames
• MatPlotLib: graphing library

1. Open 4-graphing.ipynb

2. Run each cell by clicking the triangle to the left of each code block

3. Start the databases if they're not running already

   docker-compose up
   

4. Open 5-graph-from-db.ipynb

5. Run each cell by clicking the triangle to the left of each code block

Moving from notebooks to Python program

1. Open VS Code

2. Right-click on a notebook and choose Import notebook to script

3. Save as, renaming to myapp.py

4. Run the program: python myapp.py

API to stream graphs

1. Run the api:

   uvicorn 6-api:app --reload

2. Go to http://127.0.0.1:8000/docs

3. Play with each endpoint

License

MIT