Skip to content

Learn and implement CNNs using TensorFlow and Keras, covering image preprocessing, classification, callbacks, visualizations from foundational concepts to real-world applications.

License

Notifications You must be signed in to change notification settings

Sabtain-Dev/Tensorflow

Repository files navigation

DeepLearning.AI TensorFlow Developer Specialization

This repository contains my work and notes from courses in the DeepLearning.AI TensorFlow Developer Specialization on Coursera.

Overview

In this courses, I learned to build and train Convolutional Neural Networks (CNNs) using TensorFlow and Keras. The course involved working with real-world image datasets and implementing neural networks that can recognize patterns in complex visual data.

Key Concepts Covered

  • TensorFlow & Keras fundamentals for computer vision tasks
  • Implementing CNNs using Conv2D, MaxPooling2D, Dense, Flatten, and Dropout
  • Image preprocessing and normalization using Rescaling layers
  • Custom training callbacks for early stopping
  • Dataset creation using image_dataset_from_directory and tf.data
  • Feature map visualization to interpret model behavior
  • Real-time image classification using Jupyter widgets

🛠️ Tools & Libraries

  • TensorFlow / Keras
  • Matplotlib
  • NumPy
  • Ipywidgets (for interactive image upload)
  • tf.data.Dataset API

Repository Structure

Tensorflow/

  • ├── Week1/
  • ├── Week2/
  • ├── Week3/
  • ├── Week4/
  • ├── Week5/
  • ├── Week6/
  • ├── Week7/
  • ├── Week8/
  • ├── Week9/
  • ├── Week10/
  • └── README.md

How to Use

  1. Clone this repository:
    git clone https://github.com/yourusername/Tensorflow.git
    cd tensorflow
    
  2. Install Requiremnents:
    pip install -r requirements.txt
    

Links

Author

M. Sabtain Khan

About

Learn and implement CNNs using TensorFlow and Keras, covering image preprocessing, classification, callbacks, visualizations from foundational concepts to real-world applications.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published