Google Git
Sign in
apache / tvm-vta / c18a22475d8ccb5447da4591c709d347687fb2b8
commitc18a22475d8ccb5447da4591c709d347687fb2b8[log] [tgz]
authorAbhijit Davare <abhijit.davare@intel.com>Tue Jun 22 18:03:09 2021 -0700
committerGitHub <noreply@github.com>Wed Jun 23 09:03:09 2021 +0800
treedee39349a9bea3082da0c8f27ebeed4d83986cab
parent1954ff58264384de8241cc155ffe33829f0e616a [diff]
Chisel Pipelined GEMM (#30)

* Reset to 644 file permissions

* Add json files to src/test/resources for testing

* Add new TensorGemmPipelinedSplit module and rename existing TensorGemm to TensorGemmOrig

* Tests for TensorGemmPipelinedSplit, TensorGemmOrig, and associated submodules

* Add jackson plugin dependency and stricter Scala checks

* Remove debug prints

* Rename x.json and y.json to gemm_1uop_overflow_offset.json and gemm_2uop_overflow_cascaded.json respectively

* All occurrences of '\( ' replaced with '\('

* Add linting rule to flag spaces after lparen characters

* Remove comment

* Rename TensorGemmOrig to TensorGemmSimple
11 files changed
tree: dee39349a9bea3082da0c8f27ebeed4d83986cab
  1. apps/
  2. config/
  3. hardware/
  4. include/
  5. src/
  6. tests/
  7. .asf.yaml
  8. .gitignore
  9. Jenkinsfile
  10. LICENSE
  11. NOTICE
  12. README.md
README.md

VTA Hardware Design Stack

Build Status

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.