data-science-portfolio

Why we included this project

This repo collects end-to-end data science projects written as Jupyter notebooks, the kind of work someone builds up while studying and experimenting on their own. The projects cover supervised and unsupervised learning, reinforcement learning, deep learning in PyTorch, and a good stretch of NLP work including sentiment analysis, spam classification, and cross-language retrieval. Each notebook is self-contained with its own dataset, and the author has gone back through the whole set so the code runs cleanly on a current Python stack, which makes it a usable reference instead of a frozen archive. Beginners get to follow the reasoning behind each pipeline, and people with more experience can lift the structure and evaluation choices for their own projects. It is study material, not a deployable application, so plan to read and adapt it rather than run it in production.