The Data Science MicroDegree: Data Analysis & Visualization

Set up your personal computer for data analysis by installing Anaconda and Jupiter notebook, and learn how these tools provide a virtual environment and translate code into executable form.

Install Anaconda by downloading the graphical installer for Windows, Mac, or Linux and running it. Enable the option to run code from Anaconda for seamless use.

Launch and use Jupyter notebook via Anaconda Navigator, select the notebook, and run Python code in Chrome; learn directory navigation, changing folders, and launching notebooks.

Rename the notebook, run code cells with shift enter, print outputs, save progress, download ipynb or other formats, and restart if an infinite loop occurs.

Learn to write text inside Jupiter by converting content into a modern cell, using hash-based headings, and applying italics and bold with simple formatting, while mixing code and explanations.

Explore Python data types, including integers, complex numbers, and field, with strings and lists, and master arithmetic operations such as addition, subtraction, multiplication, division, quotient, remainder, and exponent.

Python variables act as addresses that store values, including numbers and strings with quotes. Names must start with a letter or underscore, contain no spaces, and be case sensitive.

Learn how strings are defined with quotes in Python and how to display versus print output. Explore formatting with variables and the format method to create dynamic, data-driven messages.

Explore string splicing by learning how to extract single elements and groups from a string using zero-based indices; understand lower inclusive and upper exclusive bounds.

Explore lists as versatile containers that store diverse data types using square-bracket syntax, and learn to append, index, replace, and slice to extract nested lists.

Explore how dictionaries use curly brackets to store key-value pairs, contrast with lists, and extract values by keys, including nested dictionaries and lists as values.

Compare tuples, sets, lists, and dictionaries, showing that tuples are immutable, sets remove duplicates, and lists can be modified; learn the syntax for defining each container.

Explore relational operators: equals, not equals, greater than, less than, greater than or equal to, and less than or equal to, and learn logical operators and or with boolean outcomes.

Learn how if-else statements drive programming flow by evaluating conditions, using colon and indentation, and choosing between true and false branches, including multiple conditions.

Use for loops to traverse an array and operate on each element. Illustrate with a five-element array a, printing hello for each element via the format method.

Learn to implement a while loop by setting an entry condition, an exit condition, and an increment, so it prints until it becomes less than five.

Learn how in-built Python functions act as reusable blocks of code, named to call later, with examples like print, lower, len, and sum on strings and lists.

Define your own function using the bbf keyword, name it, and add a parameter like name to create personalized greetings. Emphasize returning values over printing, as shown with square example.

Explore NumPy, a core linear algebra library that helps data scientists convert mathematical data into readable graphs and insights, with Anaconda installation guidance.

Explore NumPy arrays by converting lists into 1d and 2d arrays, print them, and create data structures from nested lists using NumPy.

Learn through hands-on examples how to generate numpy arrays using arange with lower and upper limits and optional step, and create zeros and ones vectors and reshape into matrices.

Learn how numpy linspace divides a lower limit and an upper limit into a specified number of equal portions, producing non-integer outputs, unlike numpy arange which yields integers.

Learn how to create an identity matrix. See that the diagonal entries are 1 and all off-diagonal entries are 0, with sizes like 5x5.

Learn to generate arrays and matrices of random values with the random function, using uniform, Gaussian, and normal distributions, and apply inclusive lower and exclusive upper bounds.

Create two variables with random numbers to form a one-dimensional vector. Reshape the vector into a 5x5 matrix and use max and min to find values and their indices.

Explore shape and dtype in a two-dimensional matrix by using the shape and data methods to identify rows, columns, and element type.

Master numpy indexing and slicing to access elements by position and extract ranges with start:stop, where lower bounds are inclusive and upper bounds exclusive.

Discover how index broadcasting differs from regular Python indexing and learn to slice a portion and broadcast a value to replace elements.

Learn index broadcasting by slicing arrays, creating copies versus views, and how broadcasting can affect the original array unless you explicitly copy, preventing memory-side effects.

Explore two-dimensional indexing by building a 3x3 matrix and accessing elements using zero-based row and column coordinates, showing single-bracket indexing and the alternative double-bracket approach.

Learn to extract submatrices from a matrix by rotating to the left and slicing by columns, isolating a specific portion of the matrix for analysis.

Learn conditional indexing by creating a boolean mask from numeric comparisons to filter vector elements, then apply the same principle to pandas data.

Learn to perform elementwise array arithmetic with another array or with a scalar value, applying to every element. Division by zero triggers a warning rather than stopping execution, producing partial results.

Explore built-in universal functions that apply to all elements, including square root, exponential, sign, and log, with warnings like log of zero and links to further resources.

Learn pandas series basics by contrasting series with objects, using explicit indexes for data access, and naming indexes to simplify retrieval.

Explore the Pandas Series' flexibility in storing integers, strings, and dictionaries, create and name new series, access values by index, and perform index-aligned additions that upcast to floats.

Master pandas data frames as the core data structure, create and index frames, and select single columns (as series) or multiple columns, using bracket or dot notation.

Discover how to add new columns with input data, drop columns using pandas drop, specify axis for columns or rows, and control permanent changes with inplace.

Learn to drop rows or columns in a data frame, control the axis, and extract rows, columns, or single elements using loc and iloc in pandas.

Learn conditional selection in data frames using boolean values to compare against zero, filter rows where the condition is true, and extract columns.

Explore how to combine multiple conditions in a single dataframe filter, such as values greater than zero and greater than one, using the pipeline operator.

Discover how to reset_index to the default numeric index and set_index from a column, saving the old index as a new column, with options to apply changes in place.

Learn to handle missing values in pandas data frames by converting dictionaries to data frames, then applying dropna to remove missing rows or columns and fillna to impute values.

Learn to group data by a column using group by and apply aggregate functions such as mean, standard deviation, count, max, min, and describe to analyze company-level data.

Explore how to merge, join, and concatenate data frames using merge, join, and concat on common columns or indices, with axis control and handling missing values.

Explore essential pandas operations for data analysis, including unique and value_counts, apply with custom functions and lambda, drop columns, inspect columns and index, and sort values with sort_values.

Learn how to read and write data with pandas across csv, excel, json, and other formats, manage file paths and sheets, and extract tables from websites.

Visualize data with matplotlib, the popular first data visualization library, offering histograms, bar charts, and scatter plots. Install it, browse the official docs, and note seaborn as a future library.

Learn to plot a simple graph in a Jupyter notebook using matplotlib, choosing between functional and object-oriented plotting, and customize x and y labels, a title, and line color.

Learn to create multiple plots in one canvas with the subplot syntax, using three arguments: rules, columns, and subplots, to differentiate colors and move toward an object oriented approach.

Explore matplotlib object oriented plots to gain complete control, creating a figure object and adding axes with four values between zero and one, then label axes and titles.

Learn to plot subplots using the object-oriented method, specify rows and columns, and manage axis creation and layout while adding titles and labels.

Learn to modify figure size and dpi in MATLAB plots by using fixed size in inches, adjusting figure and subplots with rows and columns, and minimizing overlaps for clear layouts.

Save plots using the library, saving in high-definition formats like JPEG or BGF by specifying a filename and an optional depth argument.

Learn to add a legend to graphs with more than one plot by labeling each plot (for example x squared and x cubed), then customize the legend location.

Customize plots by adjusting color with hex codes or names, set line width and alpha, choose line states such as landscape, and tailor markers and sizes to highlight data points.

Master axis appearance and plot range in the Sea One visualization library by limiting the x axis to 0 to 1 with xlim, and set corresponding y limits.

Explore seaborn, a statistical plotting library built on matplotlib with beautiful defaults and varied plots for visualizing dataframes; install it and access its open source documentation, GitHub, and gallery.

Explore distribution plots using seaborn to visualize univariate data with histograms, kernel density estimation, and joint plots, and learn how bin sizes, density, and regression options shape insights.

Examine distribution plots by comparing the bad plot, the plot, and the kitty, and learn how a real plot relates to joint plots across dataset columns.

Visualize categorical data distributions by comparing categories such as gender, smoker, or date against numerical or categorical targets using seaborn plots, including bar, count, and box plots.

Learn to compare categorical and numerical data with seaborn plots—box, violin, strip, and swarm plots—split by categories like smoker vs nonsmoker and color coded for clarity.

Build a matrix from flight data with pivot tables, using month as index and passengers as values for heat maps and cluster maps.

Learn how to use seaborn to automate subplots with facet grids, exploring the iris dataset with scatter and distribution plots across diagonal, upper, and lower triangles.

Learn to customize plots with Seaborn and Matplotlib by changing style options, grids, and spines; adjust size or context; apply color palettes to the scatter plot with a regression line.