https://github.com/apache/tvm-vta

Open, Modular, Deep Learning Accelerator
https://github.com/apache/tvm-vta

hardware machine-learning tensor tvm vta

Last synced: about 8 hours ago
JSON representation

Open, Modular, Deep Learning Accelerator

• Host: GitHub
• URL: https://github.com/apache/tvm-vta
• Owner: apache
• License: apache-2.0
• Created: 2020-03-30T03:44:23.000Z (almost 5 years ago)
• Default Branch: main
• Last Pushed: 2024-04-10T11:56:00.000Z (10 months ago)
• Last Synced: 2025-01-19T14:14:12.240Z (7 days ago)
• Topics: hardware, machine-learning, tensor, tvm, vta
• Language: Scala
• Homepage: https://tvm.apache.org/
• Size: 1.29 MB
• Stars: 266
• Watchers: 40
• Forks: 74
• Open Issues: 4
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-opensource-hardware - tvm-vta

README

        


VTA Hardware Design Stack

=========================

[![Build Status](https://ci.tlcpack.ai/job/tvm-vta/job/main/badge/icon)](https://ci.tlcpack.ai/job/tvm-vta/job/main/)

VTA (versatile tensor accelerator) is an open-source deep learning accelerator complemented with an end-to-end TVM-based compiler stack.

The key features of VTA include:

- Generic, modular, open-source hardware

  - Streamlined workflow to deploy to FPGAs.

  - Simulator support to prototype compilation passes on regular workstations.

- Driver and JIT runtime for both simulator and FPGA hardware back-end.

- End-to-end TVM stack integration

  - Direct optimization and deployment of models from deep learning frameworks via TVM.

  - Customized and extensible TVM compiler back-end.

  - Flexible RPC support to ease deployment, and program FPGAs with the convenience of Python.