thinkingML
Bridging the gap between research and application through open tooling and targeted solutions.
Noise Tolerant Self-Driving Systems
Experimental analysis of self-driving performance under constrained computational conditions (C3). Originating as a PhD testbed, this work now supports broader investigations into robust perception and control under degraded inputs.
LoRA Benchmarks
A modular harness for fine-tuning LLMs on post-training QA datasets. Includes curated data splits, LoRA utilities, and hyperparameter-tuned validation pipelines. Designed to run flexibly across local, networked, and cloud-based GPU environments.
XRAY/ROCT Toolkit
Unified framework for medical image optimisation on retinal and chest datasets. Supports structured benchmarking across model classes, data types, and optimiser settings. Noise conditions are fully configurable for environment-specific tuning.