welcome to deep, a personal collection of deep learning assignments and projects—some from courses, some from my own time. this is mostly a reference point for concepts, implementations, and experiments i’ve worked on. nothing fancy, just a place to keep track of things.
this repo is split into two main sections:
structured assignments covering essential deep learning topics, including:
- neural networks – backpropagation, optimization strategies, activation functions
- convolutional networks – image classification, feature extraction, transfer learning
- sequence models – rnns, transformers, attention mechanisms, nlp applications
- generative models – vaes, gans, diffusion models
things i’ve built, tested, or experimented with in my own time:
- custom architectures – messing with novel network structures, tweaking hyperparameters
- reinforcement learning – applying deep rl to interesting problems
- applied ai – using deep learning in ways that actually do something
- frameworks – pytorch, tensorflow, jax
- libraries – numpy, opencv, hugging face, scikit-learn
- notebooks – jupyter, google colab
- hardware – runs on cpu/gpu setups, tested on local machines
this isn’t some polished showcase—it's just a space for me to keep track of implementations, revisit key ideas, and iterate on things i find interesting. if someone else finds it useful, cool. if not, that’s fine too.
each directory is self-contained with a short readme for context. code is mostly documented, but don’t expect perfection.
this repo changes as i go deeper into new techniques, architectures, and research papers. some planned additions:
- meta-learning & few-shot learning
- self-supervised learning
- more rl applications
this is mostly for me. if you find something useful, great! if you have questions, ask. if it breaks, fix it.