Mahdi Sabet Kish

Gradio demo for Vision2Words: image captioning with LSTM, Transformer and retrieval decoders.

C++ inference for Vision2Words captioning, using ONNX Runtime on CPU.

Brain MRI tumour segmentation and fMRI autism classification: training images (CPU and CUDA 12.6).

Rust inference binary for the brainssl segmentation and connectome models (ONNX Runtime).

Self-supervised, equivariant, uncertainty-aware tabular classifier (methods study only).

Spatial transcriptomics + H&E histology fusion for cortical layer prediction. CPU & CUDA.