print("hello") to deep learning, in 12 modules.Working notebook + exercise repo for the Machine Learning Engineer career path on Educative.com. Starts with a Python prep course (78 small exercises across syntax, control flow, data structures, file I/O, graphics, and a tiny chatbot), then walks through the 12 ML modules with a learning-summary blog post per module.
Each module ends with a learning-summary blog post in blogs/.
Syntax, variables, I/O, control flow, functions, files. The 78-exercise prep course in this repo lives here.
NumPy, pandas, matplotlib, seaborn — the data-handling toolbelt every later module assumes.
Distributions, hypothesis testing, Bayes — the math vocabulary for "why does this model work?"
Classes, inheritance, decorators. Foundation for sklearn estimators and PyTorch nn.Module.
Missing values, outliers, scaling, encoding. The 80% of real-world ML work no tutorial covers.
Linear & logistic regression, SVMs, decision trees, random forests, gradient boosting.
Clustering (k-means, DBSCAN), dimensionality reduction (PCA, t-SNE), anomaly detection.
Cross-validation, precision/recall, ROC/AUC, hyperparameter tuning with Optuna.
From perceptron to backpropagation. Built from scratch in NumPy, then in PyTorch.
CNNs for vision, RNNs / LSTMs for sequences, transformers for everything else.
Tokenization, embeddings, fine-tuning. Hugging Face transformers for serious work.
Docker, FastAPI serving, model versioning, monitoring drift in production.
The exercises in Python Review/ and Python Libraries/.
Every completed module gets a blog post documenting what was learned and what surprised me.
No build step, no virtualenv required for the prep course — just Python 3.
git clone https://github.com/codenificient/EducativeMachineLearning.git
cd EducativeMachineLearning
python3 "Python Review/Hello.py"