Skip to yearly menu bar Skip to main content


talk
in
Tutorial: Gemmini: Enabling Systematic Deep-Learning Architecture Evaluation via Full-Stack Integration

Gemmini's Programming Model

Hasan Genc

[ ]
2022 talk

Abstract:

a. Programming Stack

  • “High-Level”: Compiling ONNX Models
  • “Mid-Level”: Handtuned Kernel Library
  • “Low-Level”: Direct Machine Configuration and C/Assembly

b. Interactive Activity: Visualize and Run an ONNX Model

c. Interactive Activity: ResNet50 First-Layer Optimizations With Low-Level ISA

Chat is not available.