Python for Data Science: The Complete Data Science Bootcamp

Explore variables, data types (string, integer, float, boolean), and Python operators (arithmetic, comparison, and logical) with hands-on examples, printing results and checking types for data science tasks.

Master control flow in Python with if-else conditionals and for while loops, including input handling. Apply these techniques to filter and analyze data in data science workflows.

Learn to define and call Python functions, pass parameters, return values, and manage local and global scope, including the global keyword, and look ahead to modules and packages.

Learn how to organize code with modules and packages in Python, import and reuse functions, and use built-in modules like math and random, plus install external libraries with pip.

Learn Python error handling with try/except blocks, finally, and assertions to validate data and catch issues early, with practical examples and friendly error messages.

Learn to read and write data with pandas, loading csv, excel, and json files, and save outputs with to_csv, to_excel, and to_json, controlling delimiter, index, and sheet.

Learn to read and write data with pandas, loading and saving csv, excel, and json files, and explore options like custom delimiters and multiple sheets.

Clean and preprocess messy data in a pandas workflow by handling missing values, duplicates, and formatting. Drop irrelevant columns, fill missing values with mean, standardize text, and rename columns.

Learn to clean data in pandas by handling missing values with is null, drop na, and fill na, and remove duplicates with duplicated and drop duplicates, choosing first or last.

Explore data transformation and feature engineering with pandas by creating new features such as salary over 60000, age groups, name length, and one-hot encoding to improve machine learning input.

Group, sort, and aggregate data with pandas to summarize salaries and departments, calculating average, min, and max salaries and counting employees per department.

Compare lists and numpy arrays, showing lists as general purpose containers and arrays as fast, element-wise numerical structures for large data and performance-sensitive computations.

Learn to create, index, and slice NumPy arrays in Python for data science, including building arrays from lists and using zeros, ones, and arange.

Explore element-wise operations on NumPy arrays, such as multiplying by two, adding three, and squaring, plus functions like square root, mean, and array differences.

Leverage NumPy broadcasting and vectorization to perform operations element-by-element on arrays without loops, speeding up data processing and enabling operations across differently shaped arrays.

Explore essential numpy functions for data science, including mean, max, std, unique with counts, boolean indexing, sorting, reshaping, and filtering to analyze and understand data efficiently.

This lecture introduces data visualization with matplotlib to create line, bar, and scatter plots that reveal trends, differences, and relationships, while seaborn is introduced for future data storytelling.

Learn plotting with matplotlib in Python to create line plots, bar charts, and scatter plots, customize titles, colors, styles, and markers, and compose multiple plots with subplots.

Customize charts with colors, labels, and legends using matplotlib and seaborn, creating two colored lines and a tips scatter plot to improve readability and style.

Explore advanced charts in Seaborn, including heatmaps, pair plots, and violin plots, to visualize correlations, relationships, and distributions across variables in a data science context.

Explore interactive data visualizations with plotly, building line and scatter charts in minutes, hover and zoom to explore data in real time, and apply built-in datasets like iris for visualization.

Explore how exploratory data analysis helps you understand data, spot missing values and outliers, identify patterns, assess relationships, and visualize distributions before modeling.

Explore how to identify patterns, outliers, and trends using scatterplot, boxplot, and line plot to inform exploratory data analysis before modeling.

Apply descriptive statistics with the describe function to summarize data features, then analyze total bill's mean and median and visualize distribution via a box plot by day of week.

Explore correlation analysis to uncover how variables relate, visualize correlation matrix with a heatmap, and learn that correlation indicates strength and direction while noting correlation does not equal causation.

Explore two optional automated eda tools, pandas profiling and suitebase, to generate html reports with data types, missing values, distributions, correlations, and target-variable features for machine learning.

Explore the five key descriptive statistics—mean, median, mode, variance, and standard deviation—and learn how they reveal the center and spread of data using a tips data set in pandas.

Explore probability concepts and learn to simulate a coin toss with NumPy, compute empirical probability, and apply these ideas to customer churn, machine failure, and recommendation systems.

Explore hypothesis testing in Python by testing the null hypothesis that the mean total bill equals 20 using a one-sample t-test, t-statistic, and p-value with tips dataset.

Compute a 95% confidence interval for a mean with Python and scipy, using the seaborn tips dataset to illustrate margin of error and uncertainty.

Explore linear regression to model the relationship between tip amount and total bill, fit a line of best fit, and interpret slope, intercept, and residuals for prediction.

Explore machine learning by training models on data to learn patterns and predict outcomes, illustrated with a linear regression demo and the three learning types: supervised, unsupervised, and reinforced learning.

Explore supervised learning with linear regression in Python and unsupervised learning with k-means clustering, comparing labeled data against discovering patterns without predefined answers.

Prepare the data, train a logistic regression model on the iris dataset, evaluate its accuracy on a holdout set, and tune hyperparameters with grid search cv to improve performance.

Compare linear regression, ridge regression, and lasso regression on a synthetic data set, training all three models to show how regularization affects overfitting and feature selection.

Explore three fundamental classification algorithms, logistic regression, k-nearest neighbors, and decision trees, using the iris data set, train/test split, and accuracy comparisons to highlight each method's strengths.

Learn how k-means groups data into three clusters, uses a 200-sample dataset, and reveals cluster centers and labels for applications such as customer segmentation and image compression.

Develop hands-on time series analysis by creating a monthly data series with pandas, numpy, and matplotlib, and decompose it into trend, seasonality, and residual using an additive model.

Explore deep learning with tensorflow and keras by building a mnist classifier. Train the model and evaluate accuracy, with future topics like cnn and model tuning.

Explore big data with PySpark, learning to create a spark session, build and manipulate dataframes, apply filters and selections, perform aggregations, and read csv files for distributed processing.