This repository contains archived course materials for a free MOOC on Creative Applications of Deep Learning with TensorFlow. It offers lecture transcripts and homework assignments in Jupyter Notebook format, along with a Python package for code developed across three courses. The materials are suitable for individuals interested in exploring creative AI applications using TensorFlow.

How It Works

The course materials are structured around Jupyter Notebooks, providing hands-on exercises and explanations for deep learning concepts. It covers TensorFlow basics, neural network training, unsupervised and supervised learning, generative models (GANs, VAEs), RNNs, and NLP. The accompanying pycadl Python package bundles code used throughout the courses.

Quick Start & Requirements

• Installation: Two primary methods are offered:
  • Pip Install: Requires Python 3.4+ and TensorFlow (CPU or GPU version).
  • Docker: Provides a pre-configured environment.
• Prerequisites: Python 3.4+, TensorFlow. GPU support requires NVIDIA drivers and CUDA Toolkit. Docker installation is recommended for ease of setup, especially on Windows.
• Resources: Detailed instructions are provided for setting up Python, Jupyter Notebook, and necessary libraries like NumPy, SciPy, and Matplotlib.
• Links: TensorFlow Installation Guide

Highlighted Details

• Covers three distinct courses on creative deep learning applications.
• Includes lecture transcripts and interactive Jupyter Notebook homework assignments.
• Provides a Python package (pycadl) with course code.
• Offers extensive troubleshooting guides for installation and environment issues.

Maintenance & Community

This repository is marked as ARCHIVED, indicating it is no longer actively maintained. No community links or active development signals are present.

Licensing & Compatibility

The repository's licensing is not explicitly stated in the README. However, the use of TensorFlow implies adherence to TensorFlow's licensing terms. Compatibility for commercial use or closed-source linking would depend on the underlying TensorFlow license and any specific licenses applied to the course materials themselves.

Limitations & Caveats

The project is archived and no longer maintained, meaning updates or bug fixes are unlikely. The course materials are based on older versions of TensorFlow (e.g., 0.11.0rc1 mentioned in installation), which may not be compatible with current TensorFlow versions or best practices. GPU setup instructions, particularly for macOS, are complex and may be outdated due to OS changes like System Integrity Protection (SIP).