This repository provides a practical, top-down guide for learning AI, Deep Learning, and Machine Learning, targeting aspiring ML engineers and researchers. It emphasizes hands-on implementation and portfolio building to secure industry roles, offering a structured path from foundational tools to advanced concepts.

How It Works

The guide follows a phased approach: initial familiarization with Python, command line, and Git; practical application using Pandas and Scikit-learn on Kaggle datasets; and finally, deep learning implementation with FastAI and PyTorch on cloud GPUs. It stresses building a unique, well-documented portfolio relevant to specific industries to demonstrate problem-solving capabilities to potential employers.

Quick Start & Requirements

• Install/Run: No specific commands are provided, but the guide suggests using Codecademy for Python basics, Kaggle/Colab for Pandas/Scikit-learn, and FastAI/PyTorch for deep learning.
• Prerequisites: Python, command line, Git, Pandas, Scikit-learn, FastAI, PyTorch. Cloud GPUs are recommended for deep learning implementation.
• Resources: Initial setup may take a few weeks to months depending on prior programming experience.

Highlighted Details

• Offers a structured roadmap for self-taught individuals to enter the ML field.
• Provides detailed scoring criteria for portfolio projects to assess hireability.
• Recommends specific books and online resources for theoretical understanding.
• Differentiates between roles like ML Engineer, Applied ML Researcher, and Research Scientist.

Maintenance & Community

The repository is maintained by Emil Wallner, with suggestions and questions encouraged via GitHub issues or Twitter. Links to FastAI, Keras, Distill, PyTorch communities, and other learning resources are provided.

Licensing & Compatibility

The repository's licensing is not explicitly stated in the provided text.

Limitations & Caveats

The guide focuses heavily on practical application and portfolio building, with less emphasis on deep theoretical understanding initially. It acknowledges that many self-learners may pivot to software engineering roles if ML roles prove too competitive.