:::{admonition} Disclaimer - This project is work-in-progress! :class: important All the rights of these codes and resources belong to their original authors. I am compiling a collection for self-learning purposes and to build tools by myself using the knowledge. :::
Welcome to Collection of notes on {ml.nn-zero2hero}. This repository includes notes and learnings following the Youtube Playlist on nn_zero_to_hero guide by Andrej Karpathy and the open-source notebooks available from the book on Hands-On Machine Learning and Deep Learning by Aurelien Geron. Personally, I found these to be two of the best resources available on this subject, purely from an applied learning stand point. For great visual understanding, the videos on the Youtube Channel 3Blue1Brown are quite helpful as well.
Other resources include (but not limited to),
- The DeepLearning.ai courses for Machine Learning and Deep Learning Specializations by Andrew Ng
- The MIT Deep Learning book by Ian Goodfellow and Yoshua Bengio and Aaron Courville
- Online book on Model-based Machine Learning by John Winn
- Nvidia's Learning Deep Learning by Magnus Ekman
- Pattern Recognition and Machine Learning by Christopher Bishop
- Hands-On Mathematical Optimization with Python by The MO Book Group
* Grasp the foundations for hands-on machine learning and deep learning in an elegant approach through self-studying from the fundamentals
* Have fun while learning and compiling notes and eventually end in a culmination of work
* Help readers to easily navigate the learning experience with a good compilation on notes on the subject
* Demonstrate the use of SOTA tools and concepts for documenting beautiful publication-quality work
:::{admonition} Personal Goal :class: tip I have hopes of getting a paper outlining my work submitted for NuerIPS 2025, so it is my personal motivation for doing this as well :) I am inspired by this work on Synthetic Biosignal Generation and Single-lead to Multi-lead ECG Reconstruction. The ECG modelling work done by Andrew Miller is also pretty cool. :::
- Python 3.9 or above (preferably on VSCode)
- Scientific Python libraries, in particular NumPy, matplotlib and pandas
- Mathematical notions of Linear Algebra, Calculus, Statistics and Probability theory
- Mathematics for Machine Learning Marc Peter Deisenroth, A. Aldo Faisal and Cheng Soon Ong
- Jupyter – These notebooks are based on Jupyter. You can run these notebooks in just one click using Google Colaboratory
:ref-type: doc
:color: primary
:class: sd-rounded-pill float-left
Get started