GitHub

Z80-μLM: A Retrocomputing Micro Language Model

Z80-μLM is a 'conversational AI' that generates short character-by-character sequences, with quantization-aware training (QAT) to run on a Z80 processor with 64kb of ram. The root behind this project ...
CNX Software

WeAct STM32U585CIU6 Core Mini – An $8 STM32U5 board supported by MicroPython v1.28

While checking out MicroPython v1.28 changelog, I noticed a board from WeAct Studio based on ST’s STM32U5 Cortex-M33 microcontroller: the WeAct STM32U585CIU6 Mini Core board ...
GitHub

IMAGENE (Subproject) - Python Implementation of "A Multimodal Deep Learning Framework for Postoperative Overall Survival Prediction in Esophageal Squamous Cell Carcinoma"

public_version/ ├── cell_classfication/ # Cell classification module │ ├── train.py # Train cell classification model │ ├── pred_cell_label.py # Predict cell labels │ ├── get_data.py # Data loading ...
IEEE

DHeLlam: General-Purpose, Automatic Micro-Batch Co-Execution for Distributed LLM Training

Abstract: The growth of Large Language Models (LLMs) has necessitated large-scale distributed training. Highly optimized frameworks, however, suffer significant losses in MFU (Model FLOPS Utilization) ...