← All categories

Machine Learning Fundamentals

Core ML from the ground up: learning paradigms, model evaluation, feature engineering, classical algorithms, ensemble methods, neural networks, and practical tooling with scikit-learn, pandas, and NumPy.

Machine Learning Fundamentals

Supervised vs unsupervised vs reinforcement learning

Regression vs classification

Train / validation / test split

Overfitting vs underfitting (bias-variance trade-off)

Accuracy, precision, recall, and F1

Normalization vs standardization

k-means clustering

k-fold and stratified k-fold cross-validation

ROC curve and AUC

Gradient descent variants

Decision trees — splitting criteria and pruning

Random forests — bagging and feature randomness

Gradient boosting, XGBoost, and LightGBM

Principal Component Analysis (PCA)

Support Vector Machines — intuition

Categorical encoding and missing value imputation

Handling class imbalance

Neural networks — layers, activations, and backpropagation

Convolutional Neural Networks — convolution, pooling, and use cases

RNNs and LSTMs — sequence modeling

Batch normalization and dropout

Learning rate schedulers and adaptive optimizers

End-to-end scikit-learn pipeline with cross-validation

t-SNE and UMAP for non-linear dimensionality reduction

Regression evaluation metrics — RMSE, MAE, MAPE, and R²