Pinned Repositories
LLaVA
[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.
LLaVA-NeXT
llama-recipes
Scripts for fine-tuning Meta Llama with composable FSDP & PEFT methods to cover single/multi-node GPUs. Supports default & custom datasets for applications such as summarization and Q&A. Supporting a number of candid inference solutions such as HF TGI, VLLM for local or cloud deployment. Demo apps to showcase Meta Llama for WhatsApp & Messenger.
ComputerVision
PyTorch re-implementation of ResNet, Pix2Pix, Normalising flows + Image classification & object detection tasks
git-foss-workflow-lecture
Materials to illustrate a typical git-based FOSS workflow.
millennium-vs-empire
music_generation
Deep learning project for music generation using MLPs, RNNs, LSTMs, autoencoders
PAD
Source code for the paper "Learning cortical representations through perturbed and adversarial dreaming" - using Deep Generative Models to understand dreaming
Sports_classification
Implementation of classification of images of 100 different types of sports - using ResNet, EfficientNet, etc.
Transformers
PyTorch re-implementation of transformers architecture, GPT task, BERT pre-training, and fine-tuning on sentiment analysis
NicoZenith's Repositories
NicoZenith/Transformers
PyTorch re-implementation of transformers architecture, GPT task, BERT pre-training, and fine-tuning on sentiment analysis
NicoZenith/PAD
Source code for the paper "Learning cortical representations through perturbed and adversarial dreaming" - using Deep Generative Models to understand dreaming
NicoZenith/ComputerVision
PyTorch re-implementation of ResNet, Pix2Pix, Normalising flows + Image classification & object detection tasks
NicoZenith/git-foss-workflow-lecture
Materials to illustrate a typical git-based FOSS workflow.
NicoZenith/millennium-vs-empire
NicoZenith/music_generation
Deep learning project for music generation using MLPs, RNNs, LSTMs, autoencoders
NicoZenith/Sports_classification
Implementation of classification of images of 100 different types of sports - using ResNet, EfficientNet, etc.