Gerson Kroiz's Projects
Multidisciplinary Research and Education on Big Data + High-Performance Computing + Atmospheric Sciences at UMBC
🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch
Flax is a neural network library for JAX that is designed for flexibility.
:zap: Dynamically generated stats for your github readmes
Gerson Kroiz Personal Website
Hedgehog Matrix Block Library (HMBLib) (NIST SURF 2019)
Deep learning framework to train, deploy, and ship AI products Lightning fast.
Implementation of the StableLM/Pythia/INCITE language models based on nanoGPT. Supports flash attention, LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.
Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.
Streamline data pipelines for AI. Process datasets across 1000s of machines, and optimize data for blazing fast model training.
Ongoing research training transformer models at scale
A simplified and automated orchestration workflow to perform ML end-to-end (E2E) model tests and benchmarking on Cloud VMs across different frameworks.
Testing framework for Deep Learning models (Tensorflow and PyTorch) on Google Cloud hardware accelerators (TPU and GPU)
NeMo: a toolkit for conversational AI
🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.
Database Project for CMSC 461
The configuration framework for Zsh
Tensors and Dynamic neural networks in Python with strong GPU acceleration
A latent text-to-image diffusion model
Syllabus Project for CMSC 447
Reference models and tools for Cloud TPUs.
DNN for Spatiotemporal modeling of precipritation for SULI program at ORNL (2021)
Enabling PyTorch on Google TPU