fchollet/deep-learning-with-python-notebooks

Features

• Deep Learning Education - Offers a comprehensive set of curated resources and exercises for learning neural network theory and practice.
• Neural Network Layers - Demonstrates the construction of neural networks by stacking modular architectural building blocks.
• Neural Network Implementations - Provides core implementations of neural network architectures for image, object, and language processing.
• Keras Model Implementations - Provides practical examples of neural network architectures for image and language processing using Keras.
• Deep Learning Notebooks - Ships a collection of interactive notebooks combining mathematical theory and executable code for deep learning.
• Notebook-Based Experimentation - Uses interactive notebooks to combine live code and narrative text for step-by-step model experimentation.
• Machine Learning Educational Resources - Provides guided programming exercises and annotated code samples designed as a machine learning educational resource.
• Backend-Agnostic Deep Learning - Implements a common interface allowing model logic to run across different deep learning framework engines.
• Deep Learning Backend Swapping - Enables the execution of the same model code across different deep learning engines to compare performance.
• Deep Learning Framework Comparisons - Provides comparative analyses of different deep learning libraries by executing identical model code.

Star history