TensorFlow Foundations and Machine Learning

This unit introduces core machine learning concepts and the foundational TensorFlow workflow, including installation, tensors, computation graphs, sessions, estimators, and basic neural network structures. Students will set up a working TensorFlow environment, create and run simple models (linear regression and introductory CNNs), and prepare datasets for training to support subsequent units on advanced model tuning and deployment.

Learning Objectives

• Analyze core machine learning concepts and TensorFlow architecture by mapping tensors, graphs, sessions, estimators, perceptrons, and CNN components to stages of a simple model development pipeline.
• Apply TensorFlow installation, configuration, and multithreading execution techniques to create a reproducible development environment and run sample computations successfully.
• Demonstrate the creation and manipulation of tensors, construction of computation graphs, and execution within sessions by implementing and training a linear regression model and documenting improvements using loss and optimizer adjustments.
• Evaluate dataset selection and preprocessing strategies for image data and implement a basic convolutional neural network to assess the effect of filtering and dataset choices on model performance.