TensorFlow Builder’s Companion Book

"From Tensors to Vision & NLP — a deep dive into TensorFlow for learners, builders, and researchers."

Live Site: https://mcklay.github.io/TensorFlow-Companion-Book/
Author: Clay Mark Sarte
Built with MkDocs Material | Powered by TensorFlow

---

📘 What This Is

The TensorFlow Builder’s Companion Book is a personal guide and practical reference to mastering TensorFlow — structured like a living technical notebook.

It blends step-by-step tutorials, narrative explanations, and real-world examples across:

• Tensor mechanics (indexing, broadcasting, special tensors)
• Model-building (Keras APIs, layers, loss, training logic)
• Natural Language Processing (tokenization, RNNs, Transformers)
• Computer Vision (CNNs, image classification, GANs)
• Production tools (TF Lite, TFX, Hugging Face)

Whether you’re experimenting, preparing for interviews, or deploying ML systems, this book helps you go deep — cleanly and confidently.

---

Table of Contents

The book is organized into 6 core parts:

Part I: Understanding TensorFlow Fundamentals

• What is TensorFlow?
• Architecture & Installation
• TensorFlow vs. Keras
• First Tensor Example

Part II: Tensor Mechanics and Computation

• Indexing, Broadcasting, Ragged & Sparse Tensors
• Variables, GradientTape, tf.function, Numerical Ops

Part III: Model Building with Keras

• Model Anatomy, Layers, Losses, Training Loops
• Saving, Callbacks, and TensorBoard Visualization

Part IV: Natural Language Processing with TensorFlow

• Text Preprocessing, TF-IDF, RNNs & Transformers
• Projects: Sentiment Analysis, Spam Detection

Part V: Computer Vision in TensorFlow

• Convolutions, Data Augmentation, Image Classification
• Object Detection, Mask Detection, GANs

Part VI: Real-World Applications

• Time Series Forecasting, Recommender Systems
• TensorFlow Lite, TFX Pipelines, Hugging Face Integration

---

Run the Book Locally

If you'd like to explore this book offline or edit it yourself:

git clone https://github.com/McKlay/TensorFlow-Companion-Book.git
cd TensorFlow-Companion-Book
pip install mkdocs-material
mkdocs serve
# Open: http://127.0.0.1:8000

To deploy or publish:

mkdocs gh-deploy

---

Contributing

Feel free to:

• Open issues for bugs, typos, or suggestions

• Submit pull requests for improvements or extra chapters

• Discuss integration ideas (e.g., Colab links, datasets)

---

License

MIT License © Clay Mark Sarte
Free to learn, fork, and remix with attribution.
"A neural net’s journey begins with a single tensor."

---