Explore projects

Materials for Stanford's EE185 course, Interactive Light Sculpture Project.

Fine-Grained Memory Access Profiling for GCRAM-Based AI Accelerators. Profiler to provide insight into relevant metrics such as data lifetimes and read/write frequencies to guide DSE of gain cell-based on-chip memories in domain specific hardware.

MINT is a Python package that streamlines integration of multilayer networks for multimodal community detection (subtyping). MINT is comprised of several modules crafted to: standardize input data, execute Spectral Neighbor-joining Fusion (SNF) and Generalized Louvain (GenLouvain) algorithms, visualize intermediate results, perform cross-validation techniques, and optimize the identified communities by selecting an optimal combination of modalities that maximizes the silhouette score.

Contact: Dr. Saman Sarraf (ssarraf@stanford.edu), Dr. Hadi Hosseini (study_brain@stanford.edu)

Miscellaneous files that we want to make publicly available.

A Precise MRI-Histology Coregistration of Paraffin-Embedded Tissue with Blockface Imaging

Denoising 7T Structural MRI with Conditional Generative Diffusion Models.

Fork of Accel-Sim simulator with support for GainSight profiler and H100 GPU backends

Fork of SCALE-Sim-v2, a simulator for systolic array based accelerators for Convolution, Feed Forward, and any layer that uses GEMMs

Fork of GPGPU-Sim v4.2.1, an NVIDIA GPU simulator to support latest hardware targets and fine-grained cache profiling.

Fork of https://github.com/pytorch/examples.git project containing test programs for PyTorch

The Otica project public release repository.