Machine Learning Bootcam: Hand-On Python in Data Science

Install and configure Anaconda to run Python on Windows, Mac, or Linux, choose Python 3.x (or 2.x), and set up the Jupyter Notebook and Spyder environments via Anaconda Navigator.

Explore Spyder and Jupyter basics, launch via Anaconda Navigator, navigate notebooks and Spyder layouts, and run Python code line by line using shift+enter, with basic syntax checks.

Learn NumPy, a Python library for multi-dimensional arrays and numerical operations commonly used with SciPy and Matplotlib. Explore hands-on implementations and exercises in Jupyter notebooks and Spyder.

Introduce numpy ndarray objects, explain zero-based indexing, and demonstrate creating 1d, 2d, and 3d arrays with numpy.array from Python objects, covering basic dtype and dimensions.

Explore data types, including numeric varieties, boolean, and complex types, and learn to define user-defined structures with attributes, using hands-on code examples.

Explore array attributes in numpy: manage dimensions with ndim and shape, reshape arrays to dimensions, and create sequences with arange, as well as empty, zeros, and ones, while checking itemsize.

Generate arrays from numerical ranges using arange and linspace, specifying start, end, and step or number of samples; control inclusion of end points and data types.

Learn to index and slice data with zero-based indices, using start, stop, and step to extract parts of an entity, including single and multi-dimensional cases.

Discover NumPy array manipulation techniques, including reshaping arrays to new dimensions, flattening arrays to one dimension, and using C or Fortran order to control the matrix layout.

Explore array manipulation with transpose, reshape, and axis-aware operations. Learn to roll and reorder 3-d arrays along x, y, and z axes using practical Python examples.

Explore how to concatenate arrays along a specified axis and stack arrays using numpy functions such as concatenate, stack, vstack, and hstack to build new data structures.

Explore array manipulation in Python by using the split function to divide an array into multiple sections, including horizontal and vertical splits, with axis-aware examples.

Learn array manipulation in Python for data science: resize, append, insert, delete, and unique. See practical examples of reshaping arrays, appending along axes, and handling dimension mismatches.

Explore array manipulation in Python for data science: add, delete, flatten, and reshape arrays; learn the unique function, indexing, and counting repeated elements for clean data.

Explore the pandas library for Python, learn its core data structures: Series, DataFrame, and Panel, and how they handle heterogeneous and large data for data science.

Create a pandas series, a one-dimensional labeled data structure, from lists, arrays, and dictionaries with default or custom indexes. Learn to check length and access elements by position or label.

Learn to work with a Pandas series in data science: access elements by position or by label, use slicing to retrieve ranges, and understand zero-based indexing and negative indexing.

Learn to create and manipulate pandas DataFrame objects from dictionaries, lists, and series, set index and columns, and perform column selection, addition, and deletion.

Learn the basics of the pandas data frame, including creating frames from dictionaries, transposing data, inspecting axis labels, and checking shape, size, data types, and empty status.

Learn data visualization in Python by plotting x and y with line and scatter plots, applying color maps, and labeling axes and titles in a Jupyter notebook.

Explore scikit-learn basics in this crash course by creating a random dataset, applying a minmax scaler, and performing a train-test split to prepare features and labels for a supervised model.

Explore the basics of TensorFlow, a Google-built library for machine learning and deep learning, with tensors as multi-dimensional data structures, data flow graphs, and concepts like rank, shape, and dtype.

Discover basic cancer flow syntax by creating constants and tensors, printing outputs, and running operations in sessions; perform addition, matrix multiplication, and generate random normal and uniform distributions.

Build a simple TensorFlow graph with constant nodes and an add operation, run it in a session, and observe graph assignment, default graph behavior, and output.

Master TensorFlow basics by detailing variables and placeholders, initializing variables, defining shapes, and feeding data into a graph via an interactive session.

Build a simple neural network to perform a linear fit using placeholders, weight w, and bias b, then train with gradient descent and a cost function.

Build a simple TensorFlow neural network from scratch. Use placeholders, weights, biases, and sigmoid activation for regression data, illustrating matrix multiplication and the need for a cost function and optimization.

Build a simple neural network for linear regression using gradient descent to fit a line m x plus b to noisy data.

Save and restore TensorFlow models using tf.train.Saver, saving checkpoints during training and loading them to resume or share progress.

Learn to visualize and debug dense computation graphs in TensorFlow using a built-in visualization tool, create and name operations, set scopes, and save logs for images, histograms, and graphs.

Learn to organize TensorFlow graphs with name scopes to improve readability in TensorBoard for the machine learning bootcamp course, grouping operations into scopes and subscopes for clearer visualization.

Learn to create seaborn distribution plots, including histograms with kde, joint plots, and pair plots with hue, palettes, and rug and hexbin options to visualize and compare data distributions.

Explore how seaborn's categorical plots visualize data distributions using bar, box, and violin plots, with hue and aggregation options for category comparisons.

Explore seaborn categorical plots, including strip, box, violin, and swarm plots with hue by sex to visualize distributions by category using tips data.

Explore Seaborn regression plots by creating scatter plots with linear fits, customizing hue, markers, and color palettes, and adjusting aspect, size, and grid layouts to visualize data relationships.

Learn to style seaborn plots by setting style and context. Remove spines, adjust size and aspect, and apply color palettes and maps for clear visuals.

Learn to build interactive visualizations with Plotly and cufflinks in Python, install and import the libraries, and create offline charts like scatter, bar, box, and 3d surface.

Discover why machine learning matters as global data from devices and daily life surges into exabytes, and how ML algorithms bridge the gap between human and machine data processing.

Explore supervised learning algorithms, including regression, classification, and deep learning, and compare them with unsupervised methods while building practical Python examples.

Learn to build a simple linear regression model using salary data and years of experience, including data import, preprocessing, train-test split, model fitting, and initial evaluation.

Learn to build and fit a simple linear regression model in Python using a library, train on a set with feature scaling to predict salary from experience.

Explore how a simple linear regression model learns the correlation in the training data to predict salaries for new observations. Compare real salaries with predictions and the regression line.

Visualize the simple linear regression model by plotting observation points and the regression line, compare training and test predictions, and interpret linear dependency for salary versus experience.