Machine-Learning-notebooks

Machine Learning Course Notebooks: Model and Algorithm Implementations and Experiments

1.1-Polynomial Regression & Bias-Variance Tradeoff

Implemented polynomial regression and explored the bias-variance tradeoff using insurance data.

1.2-Simple Perceptron

Implemented a simple perceptron model to classify data and visualize the decision boundary.

2.1-KNN-Bagging-Boosting

Implemented KNN. Also tested and played around with multiple ensemble methods like bagging, random forest and AdaBoost.

2.2-PCA-KMeans

Implemented PCA and K-Means clustering from scratch.

3.1-MLP As Function Estimator

Implemented a Multi-Layer Perceptron (MLP) from scratch to approximate mathematical functions.

3.2-MLP On MNIST

Implemented a Multi-Layer Perceptron (MLP) from scratch on the MNIST dataset to classify handwritten digits.

3.3-Optimizers implementation on MNIST

Implemented and compared different optimization algorithms like SGD with momentum, AdaGrad, RMSProp and Adam on the MNIST dataset using an MLP.

4-MobileNetV1-V2

Implemented and compared MobileNetV1 and MobileNetV2 architectures for image classification.

5.1-Skipgram

Implemented the Skip-gram model for word embeddings using the Word2Vec architecture.

5.2-Transformer-BERT-LoRA

Implemented transformer models, BERT, and LoRA architectures.

6-Knowledge Distillation using Contrastive Learning

Implemented knowledge distillation using contrastive learning to transfer knowledge from a large model to a smaller one.