Make predictions with Python machine learning for apps

Explore Android Studio basics, install and navigate the interface, and learn Java fundamentals to build a simple input-output Android app from start to finish.

Set up an Android virtual device in Android Studio, select a hardware profile and API level, and run your app on the emulator, noting camera and speech recognition limitations.

Explore basic variables, primitives, and strings, then learn assignment, arithmetic, and conditional operators. Cover modulus, string concatenation, type casting, and parsing to convert between numbers, strings, and booleans.

Compare arrays and lists in Java, learn how to set up basic arrays and lists with syntax, and note that arrays are immutable while lists support adding and removing elements.

Access and modify elements in arrays and array lists using zero-based indices, check lengths to avoid out-of-bounds, and apply get, set, and remove operations.

Explore control flow with if and switch statements, testing conditions and using else and else-if logic to drive actions like go, stop, and slow down in a traffic light example.

Learn while loops to repeat code until a condition is false, using an index and total, with break and continue to control iterations from 1 to 5 and beyond.

Explore for loops and for-each iterations to sum a list of transactions and compute a net balance, compare them with while loops, and highlight safe array bounds.

Explore declaring, implementing, and calling functions in Java with a bank transaction example, covering return types, parameters, and function bodies, and how to call a function to perform operations.

Define and call functions by name, return type, and parameters, pass in promises and return values. Learn about boolean returns, parameter costs, and variable scope across global, local, and placeholders.

Explore how a subclass inherits all attributes and methods from a superclass, adds new behavior for lights, uses constructors and overrides, and illustrates with a strobe light extending a light.

Explore static variables versus instance variables and master access modifiers—private, public, protected, and default—so class level constants can be accessed via the class name, while controlling access.

Learn to build a basic Android UI with constraint layout, edit text, and a button, wire up IDs and strings resources, and prepare for displaying results in the main activity.

Implement simple backend text processing and display in an Android app, then tidy the user interface with colors, text sizes, padding, and constraints to improve usability.

Learn the PyCharm interface, install the latest Python, and outline a project for building, running, and debugging Python apps. Explore Python basics and package management to start developing apps quickly.

Explore PyCharm's user interface, create and run a Python file, manage project files and interpreter, and install packages, including TensorFlow, to support machine learning workflows.

Explore Python collection types—tuples, lists, and dictionaries—and learn how to store multiple values, access elements by index or by keys, and compare when to use each.

Explores operations on tuples, lists, and dictionaries in Python, including retrieving, modifying, adding, deleting elements, and computing length, min, max, and key-value operations.

Master control flow with if statements, else and else if, and logical operators to make decisions in code. Use practical examples like purchases and balances to illustrate branching.

Learn how to declare, implement, and call functions in Python, pass parameters like by amount, and return values to track starting and current positions in a game-like example.

Learn how classes serve as blueprints for objects with state and behavior, define attributes and methods, use initialisers to create instances, and manipulate them with dot notation in Python.

Explore the TensorFlow workflow in Python, building a computational graph of tensors with constants, variables, placeholders, and operations to train and test a simple linear regression model.

Learn to import TensorFlow in PyCharm, create a main Python file, and access tf via an alias. Troubleshoot missing packages, set the interpreter, and install TensorFlow from the IDE.

Explore constant nodes in a computational graph, storing unchanging tensor values with configured value, shape, data type, and name. Use sessions to run nodes and fetch their current values.

Explore variable nodes in Python machine learning and constant nodes, learn how initial values and trainability affect model training, and perform manual reassignments with a global variables initializer.

Placeholder nodes serve as inputs to computational graphs; they hold no values until runtime. Use a feed dictionary to supply values with correct type and shape.

Build a linear regression model with TensorFlow by creating input and output nodes, a loss, and a gradient descent optimizer; train on fake data and test predictions.

Outline building a linear regression model with TensorFlow estimator, compare built-in versus custom estimation, and explain training, testing, and evaluation with features.

Build a custom estimator function in TensorFlow, replacing built-in estimates with a user-defined model, and train and evaluate using input data, features, labels, and a gradient descent optimizer.

Test a custom estimator function for linear regression in Python to power apps, compare with a prebuilt model, and address prediction mode and data preparation.

Compare the prebuilt model, the custom estimates, and the basic linear regression, focusing on lines of code and training evaluation. Assess accuracy via loss to see which method performs best.

Import a pre-trained linear regression model into an Android Studio app, save it to a procedure off file, and build a simple UI that communicates with the model.

Name the graph nodes, replace operations with tf add and multiply, then save the graph and trained values to a checkpoint and binary file for the optimized version.

Learn how to freeze a TensorFlow graph from a checkpoint, save the frozen and optimized graph files, and prepare them for import into Android Studio for app-based predictions.

Begin an Android Studio project, import the optimized frozen graph model and inference library, configure build settings and assets, and prepare a basic UI to run the model.

Build a simple Android UI with an edit text and output view to perform predictions using a loaded inference model, defining model name, input node, output node, and input shape.

Implement inference functionality by wiring a TensorFlow inference interface, preparing input as a float, running the model on press of the button, and displaying the output in the text view.

Test the Android app, debug errors, and fix issues from input handling and model loading, using logs, emulator API checks, and input validation to ensure successful inferences.