Python for Data Science

Explore what data is, including qualitative and quantitative types like numbers, images, and text, and how visualization reveals patterns to drive data science and machine learning.

Population is the complete data to analyze. A sample is a carefully chosen subset used when measuring all data is infeasible, with random or stratified sampling to estimate the mean.

Explore descriptive and inferential statistics by showing how samples describe a population, using central tendency and variability, including mean, mode, median, range, variance, and standard deviation, plus hypothesis testing.

Understand central tendency by learning mean, mod, and median, including how to compute the mean of a sample, identify the most frequent value, and find the middle value after sorting. Apply simple Python-like steps to calculate these measures and compare class performance.

Compute range, variance, and standard deviation from data by using the maximum minus the minimum, mean (x bar), and sigma-based formulas to assess data spread.

Explore how data follow a bell-shaped normal distribution (Gaussian distribution), shown by the probability density function, with mu as the mean and sigma as the standard deviation.

Explore marginal, joint, and conditional probabilities using dice examples to illustrate simple probability, how two events intersect, and conditioning on an already occurred event.

Explore marginal, joint, and conditional probability using a 500-student table of gender and grades, computing p(male), p(plus), p(male and plus), and p(plus|female).

Explore Bayes theorem as a cornerstone of probability in data science, detailing prior, likelihood, posterior, and marginal probabilities. See how conditional probability derives the Bayes formula with a card example.

Explore Pandas for data analysis and manipulation, focusing on working with tabular data and understanding series as a one-dimensional labeled array with customizable indices.

Learn masking and boolean indexing in pandas, creating masks with comparisons, filtering rows, selecting columns, and combining conditions, plus converting between data frames and numpy arrays.

Explore the iris dataset in python using pandas, loading iris.csv with read_csv, inspecting a five-column data frame, and analyzing species with describe, value_counts, and sorting.

Explore how to handle missing data in Python for data science by detecting NaN values and choosing between dropping rows or filling with the mean, demonstrated on iris data.

Master concatenating data frames by rows or columns using pandas pd.concat, ensuring matching column names and choosing axis to build a larger data frame.

Explore how to merge two data frames using a common attribute, perform inner, left, right, and outer joins, and apply the merge function.

Learn how to output modified pandas dataframes to csv and excel files, control the index with index=False, and export to json, html, and more.

Learn to visualize data with matplotlib by creating line plots using plt.plot, customize with colors, markers, labels, and legends, and explore styles like seaborn for insightful EDA.

Learn to create scatter plots in Matplotlib using plt.scatter with x and y data, customize color and marker, set labels and size, and compare datasets, including a random 100×2 example.

Learn to plot and customize bar graphs in matplotlib using plt.bar, set tick labels, widths, legends, axis labels, and limits to compare two car sales across years.

Plot a pie chart in matplotlib using plt.pie with data and labels, customize slices with explode and shadow, and display percentages with the auto pct parameter.

Learn to create subplots in a single figure by arranging four graphs in a 2x2 grid, plotting x, x squared, x cubed, sin x, and cos x.

Explore 3D plotting with matplotlib by building a meshgrid from numpy arrays A and B, creating a 3D axis, and visualizing surfaces and bubble shapes using plot_surface and A+B elevation.

Create a bordered Pikachu image by building black top and bottom borders with ten-pixel rows, then add left and right borders with horizontal stacking, visualizing and saving with Matplotlib.

Visualize the iris dataset using pandas and matplotlib by reading iris.csv, plotting petal length vs width, color-coding species, and using hist and pie charts to compare species distribution.