Open deep learning compiler stack for cpu, gpu and specialized accelerators
Перейти к файлу
雾雨魔理沙 5293c6bf66 [Relay] use unordered_map instead of map in ANF (#3024) 2019-04-15 12:56:31 -07:00
.github Add link to the reviewers 2018-10-22 23:08:32 -07:00
3rdparty [Relay/TOPI][OP] Add clip and wrap mode support in take (#2858) 2019-04-02 06:40:11 +08:00
apps [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
cmake [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
conda [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
docker [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
docs [Relay] Add gradient operator tutorial docs (#2751) 2019-04-12 15:43:37 -07:00
golang [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
include/tvm [REFACTOR] Use more TypedPackedFuncs (#2981) 2019-04-10 14:28:25 -07:00
jvm fix java checkstyle version (#2998) 2019-04-09 22:12:50 -07:00
nnvm [Bugfix] Fix caffe2 nnvm frontend (#2996) 2019-04-12 21:33:45 -04:00
python [Relay][Text Format] Pretty Printer Smart Inlining (#2881) 2019-04-12 21:36:19 -04:00
rust [RUST][FRONTEND] Fix resnet example (#3000) 2019-04-14 19:11:18 -07:00
src [Relay] use unordered_map instead of map in ANF (#3024) 2019-04-15 12:56:31 -07:00
tests [RUST][FRONTEND] Fix resnet example (#3000) 2019-04-14 19:11:18 -07:00
topi [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
tutorials [AutoTVM] Fix typos (#3014) 2019-04-12 12:37:08 -07:00
vta [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
web [REFACTOR] Use more TypedPackedFuncs (#2981) 2019-04-10 14:28:25 -07:00
.clang-format add .clang-format (#2395) 2019-01-08 13:08:13 -08:00
.gitignore [Relay][RFC] Relay IR Text Format (#1781) 2018-12-02 10:35:01 -08:00
.gitmodules [Relay] Add generic & informative Relay error reporting (#2408) 2019-01-25 10:17:31 -08:00
CMakeLists.txt [Relay] C++ GraphRuntimeCodegen, Deprecate Python2 (#2986) 2019-04-12 16:13:45 -07:00
CONTRIBUTORS.md [COMMUNITY] @hlu1 -> Reviewer (#3021) 2019-04-13 19:20:24 -04:00
Jenkinsfile [Relay] C++ GraphRuntimeCodegen, Deprecate Python2 (#2986) 2019-04-12 16:13:45 -07:00
LICENSE [DOC/LICENSE] Make doc and license consistent, opensource repo when we get approval (#134) 2017-05-09 20:36:23 -07:00
Makefile [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
NEWS.md [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
NOTICE NOTICE (#2203) 2018-11-29 23:43:41 -08:00
README.md [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00
version.py [HEADER] Add Header to Comply with ASF Release Policy (#2982) 2019-04-07 21:14:02 -07:00

README.md

Open Deep Learning Compiler Stack

GitHub license Build Status

Documentation | Contributors | Community | Release Notes

TVM is a compiler stack for deep learning systems. It is designed to close the gap between the productivity-focused deep learning frameworks, and the performance- and efficiency-focused hardware backends. TVM works with deep learning frameworks to provide end to end compilation to different backends. Checkout the tvm stack homepage for more information.

License

© Contributors Licensed under an Apache-2.0 license.

Contribute to TVM

TVM adopts apache committer model, we aim to create an open source project that is maintained and owned by the community. Checkout the Contributor Guide

Acknowledgement

We learnt a lot from the following projects when building TVM.

  • Halide: TVM uses HalideIR as data structure for arithmetic simplification and low level lowering. We also learnt and adapted some part of lowering pipeline from Halide.
  • Loopy: use of integer set analysis and its loop transformation primitives.
  • Theano: the design inspiration of symbolic scan operator for recurrence.