A training framework for Stable Baselines3 reinforcement learning agents, with hyperparameter optimization and pre-trained agents included.

Challenge for my Reinforcement Learning class in Deusto

Final task for my Reinforcement Learning class in Deusto. The research paper discuss examples of using ML-Agents toolkit of Unity. Paper is avaible at:

Practical project for time series prediction using recurrent neural networks (RNN) with Python.

Customer segmentation project using RFM analysis and clustering algorithms (K-Means, DBSCAN, GMM) to identify distinct customer groups based on purchasing behavior. Includes visualization, evaluation metrics, and parameter tuning methods to support business insights and marketing strategies.

Utilising time series analysis techniques - ARIMA, Deep Learning, and Hybrid Methods to forecast book sales and model demand.

This project implements a machine learning model to predict breast cancer diagnosis. Utilizing techniques such as data preprocessing, feature selection, and various algorithms, the model aims to assist in early detection and improve healthcare outcomes. Explore the repository to understand the methodology and technologies used in this project.

This project investigates the efficacy of integrating context distillation techniques with parameter-efficient tuning methods such as LoRA, QLoRA, and traditional fine-tuning approaches, utilizing Facebook’s pre-trained OPT 125M model.

Intro to Machine Learning Project from TripleTen

This is the official repository of Debangan Ghosh. Here I made a Streamlit App for my Fine tuned ML prediction models for MASAI Hackathon

This project forecasts daily sales for Rossmann stores using historical data, store metadata, and engineered features. We use the Random Forest Regression & XGBoost regression model to capture complex patterns and improve predictive accuracy.

This repository features a machine learning project utilizing the Pima Indians Diabetes Dataset to predict diabetes risk. It explores data preprocessing, model training, and evaluation using techniques such as Naive Bayes and K-Nearest Neighbors (KNN) . The aim is to highlight the impact of various health factors on diabetes prediction.

Intro to Machine Learning Project from TripleTen

University project focused on classifying 32×32 grayscale images from the Fashion-MNIST dataset using FNNs and CNNs. Includes data exploration, model tuning, architecture comparison, and final prediction generation.