Open deep learning compiler stack for cpu, gpu and specialized accelerators
Перейти к файлу
Tianqi Chen 27b6812b06
[CI] always rebuild sphinx-gallery docs from scratch (#1742)
2018-09-19 20:43:26 -07:00
.github [DOCS] Improve review guide, improve cmake llvm build (#1295) 2018-06-17 12:07:11 -07:00
HalideIR@cf6090aeae [NODE] Node base system refactor (#1739) 2018-09-19 19:56:45 -07:00
apps [RUNTIME] Improve memory usage for RPC (#1741) 2018-09-19 17:35:19 -07:00
cmake [RUNTIME] Refactor to enable stackvm in runtime. (#1588) 2018-08-11 09:15:05 -07:00
conda [VERSION] Update to 0.5.dev (#1623) 2018-08-20 16:28:28 -07:00
dlpack@bee4d1dd8d [SUBMODULE] update submodule to latest (#1728) 2018-09-18 23:28:12 -07:00
dmlc-core@4f0564ec76 update dmlc-core for security reason (#1584) 2018-08-10 15:01:55 -07:00
docker Add demo_android Dockerfile (#1646) 2018-09-14 09:26:34 -07:00
docs [High level OPT][RFC] NNVMv2 IR - Relay (#1672) 2018-09-19 15:55:12 -07:00
include/tvm [NODE] Node base system refactor (#1739) 2018-09-19 19:56:45 -07:00
jvm Vulkan TVM Android Support (#1571) 2018-08-09 18:41:49 -07:00
nnvm [FRONTEND][TENSORFLOW] GPU support for tensorflow models. (#1718) 2018-09-19 20:07:01 -07:00
python [RUNTIME] Improve memory usage for RPC (#1741) 2018-09-19 17:35:19 -07:00
src [NODE] Node base system refactor (#1739) 2018-09-19 19:56:45 -07:00
tests [CI] always rebuild sphinx-gallery docs from scratch (#1742) 2018-09-19 20:43:26 -07:00
topi [NNVM] Recover reduction behavir, fix CI (#1740) 2018-09-19 12:49:20 -07:00
tutorials [FRONTEND][TENSORFLOW] GPU support for tensorflow models. (#1718) 2018-09-19 20:07:01 -07:00
verilog Remove leading "./" from include paths (#1640) 2018-08-22 22:11:12 -07:00
vta Fix VTA Tutorial for more strict graphrt check (#1737) 2018-09-19 08:18:52 -07:00
web [VERSION] Update to 0.5.dev (#1623) 2018-08-20 16:28:28 -07:00
.gitignore [AUTOTVM] TOPI integration for ARM CPU (#1487) 2018-08-02 08:59:25 -07:00
.gitmodules [SUBMODULE] switch to https (#341) 2017-08-17 11:51:31 -07:00
.travis.yml Remove linux from travis (#156) 2017-05-22 19:47:12 -07:00
CMakeLists.txt [High level OPT][RFC] NNVMv2 IR - Relay (#1672) 2018-09-19 15:55:12 -07:00
CONTRIBUTORS.md [TEAM] were -> Reviewer (#1705) 2018-09-11 14:31:13 -07:00
Jenkinsfile [RUNTIME] Refactor to enable stackvm in runtime. (#1588) 2018-08-11 09:15:05 -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 [tvm4j] fix java build (#1471) 2018-07-21 17:21:38 -07:00
NEWS.md [VERSION] Update to 0.5.dev (#1623) 2018-08-20 16:28:28 -07:00
README.md Update ci to new location (#1552) 2018-08-03 12:08:53 -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 arithematic 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.