From Python Basics to Deep Learning: A Complete AI/ML Learning Roadmap

This open-source learning resource is maintained by mariaaktermukti, derived from the notes of Phitron's AI and machine learning courses. It offers a 5-month structured path covering Python programming, data analysis, classic machine learning, deep learning, and PyTorch hands-on practice. It addresses the problem of scattered resources for AI beginners and is suitable for learners starting from scratch (including complete beginners, career changers, students, and self-learners).

Beginners in the AI field often feel confused due to the abundance of resources without systematic organization. This project (Learn-AI-ML-Basic-to-Advance) organizes course notes into a structured roadmap, emphasizing the combination of theory and practice. Each stage is accompanied by concept explanations and runnable Python implementations, distinguishing it from resources that only provide code snippets.

The learning path is divided into 5 stages:

1. Python Basics: Core syntax, OOP, functional programming, file operations, etc.
2. Data Processing and Visualization: NumPy/Pandas operations, data cleaning, Matplotlib charts, basics of statistics and linear algebra.
3. Classic Machine Learning: Linear regression, classification algorithms (logistic regression/KNN, etc.), clustering, feature engineering, overfitting optimization.
4. Deep Learning Basics: Perceptron, MLP, PyTorch framework (tensors/automatic differentiation/training process), optimization algorithms and regularization.
5. Advanced Deep Learning: CNN, RNN/LSTM/GRU, Transformer and attention mechanism, transfer learning.

The project uses a mainstream toolchain:

• Language: Python
• Data Processing: NumPy, Pandas
• Visualization: Matplotlib
• ML Algorithms: Scikit-learn
• DL Framework: PyTorch
• Environment: Jupyter Notebook, Google Colab (reduces hardware barriers)
• Version Control: Git/GitHub The configuration balances entry-friendliness and practicality: Scikit-learn allows quick validation, while PyTorch supports cutting-edge research.

Target Audience:

• Complete beginners: Follow the 5-month timeline to progress step by step
• Career changers: Quickly skim the Python section to start learning algorithms
• Students: Supplement courses to gain industrial practical experience
• Self-learners: Structured guidance to avoid getting lost The project emphasizes end-to-end practice, covering the entire process of data collection, cleaning, feature engineering, model evaluation, and deployment.

This resource represents the open-source trend in AI education: it not only provides code but also conveys learning methods and frameworks. The 5-month structured design makes goals manageable, and the progression from Python to Transformer aligns with cognitive laws, building a solid foundation in the AI field and serving as an entry key to advanced topics such as CV and NLP.