AI-ML Notebooks: My Machine Learning & AI Journey

Welcome to my collection of practice notebooks and micro-projects, documenting my learning journey in Machine Learning, Neural Networks, Generative AI, and Retrieval-Augmented Generation (RAG). This repository is a set of my hands-on explorations, experiments, and mini-projects, each focused on a key concept or technique in modern data science and AI.

Repository Structure

The repository is organized into topic-based folders, each containing Jupyter notebooks that demonstrate theory, practical implementation, and real-world datasets.

Practice_notebooks/
│
├── 01_Exploratory_Data_analysis/      # Exploratory Data Analysis (EDA)
├── 02_LinearRegression/               # Linear Regression
├── 03_LogisticRegression/             # Logistic Regression
├── 04_DecisionTree/                   # Decision Trees
├── 05_RandomForest/                   # Random Forests
├── 06_PCA/                            # Principal Component Analysis
├── 07_SVM/                            # Support Vector Machines
├── 08_Time_series_forcasting/         # Time Series Forecasting
├── 09_Neural_networks/                # Neural Networks (ANN, CNN)
├── 10_generative_AI/                  # Generative AI (e.g., LangChain)
├── 11_RAG/                            # Retrieval-Augmented Generation
└── 12_micro_projects/                 # Micro Projects (various topics)

Folder & Project Summaries

01_Exploratory_Data_analysis

• EDA_Insurance.ipynb: Exploratory data analysis on insurance datasets, visualizations, and feature insights.
• house_price_EDA.ipynb: In-depth EDA on house price data, including feature engineering and visualization.

02_LinearRegression

• LinearRegression.ipynb: End-to-end implementation of linear regression on house price prediction, including data preprocessing, model training, and evaluation.

03_LogisticRegression

• logistic_reg.ipynb, LogisticRegression.ipynb: classification using logistic regression, with practical examples and model evaluation.

04_DecisionTree

• Decision_tree_census.ipynb: Decision tree classification on census data, including feature selection and visualization.

05_RandomForest

• random_forest.ipynb: Random forest ensemble methods for classification.

06_PCA

• PCA.ipynb: Dimensionality reduction using Principal Component Analysis, with visualization of explained variance.

07_SVM

• SVM.ipynb: Support Vector Machine for classification, including kernel tricks and hyperparameter tuning.

08_Time_series_forcasting

• time-series-analysis.ipynb: Classical time series analysis and forecasting (e.g., ARIMA) on real datasets.
• walmart_timeseries.ipynb: Forecasting Walmart sales using time series models.

09_Neural_networks

• ANN_from_scratch.ipynb: Building an Artificial Neural Network from scratch showcasing how forward and backpropogation works.
• ANN-basics.ipynb: Artificial neural networks using TensorFlow/Keras.
• CNN_MNIST.ipynb: Convolutional Neural Network for MNIST digit classification, including data augmentation.
• RNN_LSTM.ipynb: comparision of traditional RNN and LSTM model with amazon fine food review dataset

10_generative_AI

• LangChain.ipynb: Experiments with generative AI frameworks such as LangChain.

11_RAG

• RAG_hospital.ipynb: Retrieval-Augmented Generation for hospital data, combining retrieval and generation for QA tasks.

12_micro_projects

• census-income.ipynb: Income classification using various ML models.
• coursera-classification.ipynb: Multi-model classification project (logistic, decision tree, random forest).
• RAG_hospital.ipynb: Retrieval-Augmented Generation for hospital data, combining retrieval and generation for QA tasks.
• walmart_timeseries.ipynb: orecasting Walmart sales using time series models..
• Netflix-Recommendation.ipynb: Recommendation Engine Trained on Netflix data using Singular Value Decomposition(SVD).

Key Topics Covered

• Exploratory Data Analysis (EDA)
• Supervised Learning: Linear Regression, Logistic Regression, Decision Trees, Random Forests, SVM
• Unsupervised Learning: PCA
• Time Series Forecasting: ARIMA, sales prediction
• Neural Networks: ANN, CNN, deep learning basics
• Generative AI: LangChain and related frameworks
• Retrieval-Augmented Generation (RAG)
• End-to-End ML Pipelines: Data preprocessing, feature engineering, model evaluation

About This Repository

This repository is a reflection of my personal growth and curiosity in the field of AI and ML. Each notebook represents a step in my journey, from foundational concepts to advanced topics. I hope these resources are helpful to others embarking on a similar path.

---

Contact: Bhuvan_S