Python for Data Science

• Five days, hands-on, with scripts and interactive notebooks used side by side.
• This is not a data science course in itself - it provides the tools to do data science.
• pandas is covered extensively: loading, exploratory analysis, cleaning, reshaping and exporting.
• Materials: a 400+ page course manual and a Python quick reference.

• Analysts and scientists moving into Python.
• Users of R, Julia, SAS or MatLab who want to apply that experience in Python.
• Engineers and technical staff who need to explore and visualise real data.

• No prior programming experience is required.
• Participants should be comfortable with science and math concepts.
• Comfort working with files and folders is expected.
• Familiarity with the command line in Linux, Windows or macOS is useful but not crucial.

By the end of this course, participants will be able to:

• create and run Python programs, and understand the fundamentals behind them
• design and code modules and classes
• process CSV and Excel files
• manipulate arrays with NumPy
• work with the breadth of subpackages that make up SciPy
• use Series and DataFrames in pandas to load, query, clean and reshape data
• create plots with Matplotlib and Seaborn
• use Jupyter notebooks for ad hoc calculation, plotting and what-if analysis

• Day One The Python Environment - starting Python, using the interpreter, running a script.
  • Variables and Values - variables, strings and string operations, numbers, converting types.
  • Basic I/O - writing to the screen, string formatting, command-line parameters, the keyboard.
  • Flow Control - conditional expressions, Boolean values, relational operators, while loops, loop exits.
• Day Two
  • Array types - lists and tuples, indexing and slicing, iteration, sequence functions and operators,
  • list comprehensions, generator expressions, nested sequences.
  • Working with files - opening, reading and writing text files, raw binary data.
  • Dictionaries and Sets - creating and iterating dictionaries, creating and working with sets.
  • Functions, modules and packages - return values, parameter types, scoping, documentation,
  • creating and importing modules, organizing packages.
• Day Three
  • Errors and Exception Handling - syntax errors, exceptions, try/except/else/finally.
  • Introduction to Python Classes - defining classes, constructors, instance methods and data,
  • attributes, inheritance.
  • Excel spreadsheets - the openpyxl module, reading, creating and modifying workbooks.
  • iPython - features, magic commands, configuration.
• Day Four
  • Jupyter Notebooks - using notebooks, Jupyter Lab, markdown, managing output, exporting.
  • Brief intro to SciPy - what it provides, useful functions, subpackages.
  • Intro to NumPy - creating arrays, indexing and slicing, large number sets, transforming data.
  • pandas I, basics - Series and DataFrames, loading data, data summaries, data types, basic plotting.
• Day Five
  • pandas II, exploring and cleaning - selecting and querying, indexes, dropping and modifying data,
  • missing values, groupby, multi-indexes.
  • pandas III, reshaping - pivot tables, merging, melting, stacking and unstacking.
  • Plotting - getting started with Matplotlib, common plots, Seaborn, tweaking plots.

• Latest Python 3 from Anaconda, plus the additional modules the labs use.
• An IDE or editor: Visual Studio Code, PyCharm or Eclipse.
• Lab files are supplied, with detailed setup instructions before class.