Google Git
Sign in
apache / tvm-vta / f8a36c68da825492d7594f397cc5b84ec796ef42
commitf8a36c68da825492d7594f397cc5b84ec796ef42[log] [tgz]
authorTianqi Chen <tqchen@users.noreply.github.com>Tue Jul 24 14:01:58 2018 -0700
committerGitHub <noreply@github.com>Tue Jul 24 14:01:58 2018 -0700
tree473642d2bdbfb79ad729f3ef737288c56ba38deb
parent978c739370437b60307ac010acd0629b099ced94 [diff]
[IR] support general type annotation. (#1480)
2 files changed
tree: 473642d2bdbfb79ad729f3ef737288c56ba38deb
  1. config/
  2. hardware/
  3. include/
  4. python/
  5. src/
  6. tests/
  7. tutorials/
  8. README.md
README.md

VTA: Open, Modular, Deep Learning Accelerator Stack

VTA(versatile tensor accelerator) is an open-source deep learning accelerator stack. It is not just an open-source hardware, but is an end to end solution that includes the entire software stack on top of VTA open-source hardware.

The key features include:

  • Generic, modular open-source hardware
    • Streamlined workflow to deploy to FPGAs.
    • Simulator support to protoype compilation passes on regular workstations.
  • Driver and JIT runtime for both simulated and FPGA hardware backend.
  • End to end TVM stack integration
    • Direct optimization and deploy models from deep learning frameworks via TVM stack.
    • Customized and extendible TVM compiler backend.
    • Flexible RPC support to ease the deployment, and program FPGAs with Python

VTA is part of our effort on TVM Stack.