Machine Learning Roadmap

Basic Level

Essential mathematics like linear algebra, calculus, statistics, probability required to understand ML algorithms.

Linear Algebra - Khan Academy Statistics & Probability

Python is the primary language for ML development, with libraries like NumPy, Pandas, Matplotlib.

Python Getting Started NumPy Quickstart Pandas 10 Minutes Guide

Algorithms like linear regression, logistic regression, trees, K-means, etc.

ML Course by Andrew Ng Scikit-learn Supervised Learning Scikit-learn Clustering

Git is a version control system for tracking changes in source code. GitHub is a platform for hosting repositories and collaborating on projects.

Official Git Documentation GitHub Docs

Techniques to evaluate model performance like Cross Validation, Metrics (Accuracy, Precision, Recall, etc.).

Scikit-learn Model Selection ML Mastery Evaluation

The process of transforming raw data into features that are suitable for training models.

Kaggle Feature Engineering Scikit-learn Preprocessing

Techniques like Random Forest, Gradient Boosting (XGBoost, LightGBM) to improve model performance.

Scikit-learn Ensemble Methods XGBoost Documentation

Improving model performance by adjusting parameters like Grid Search, Random Search.

Scikit-learn Grid Search Optuna Optimization

Transitioning from traditional ML to deep neural networks, TensorFlow/PyTorch.

Deep Learning Specialization PyTorch Tutorials

Integrating ML with operations using tools like MLflow, Kubeflow, etc.

MLflow Documentation Kubeflow Documentation

Models for analyzing and forecasting time-dependent data.

Kaggle Time Series Prophet Documentation

A type of ML where an agent learns by interacting with an environment.

RL Specialization OpenAI Gym