- Aug 31, 2017
-
-
Tianqi Chen authored
-
- Aug 30, 2017
-
-
Tianqi Chen authored
-
Tianqi Chen authored
* [SCHEDULE][PASS] support storage_align of certain axis * fix lint
-
Tianqi Chen authored
-
- Aug 28, 2017
-
-
Tianqi Chen authored
* [CODEGEN] NVPTX backend. * Fix pylint * use fix
-
- Aug 26, 2017
-
-
ziheng authored
-
Aditya Atluri authored
* v2: runtime support for rocm * fixed coding space errors * removed kROCM from c_runtime_api.h
-
- Aug 22, 2017
-
-
Tianqi Chen authored
-
- Aug 20, 2017
-
-
Tianqi Chen authored
-
ziheng authored
-
Tianqi Chen authored
* [BUILD][LLVM] Support LLVM mainline 5.0 6.0 * Reduce parallelism
-
- Aug 16, 2017
-
-
Tianqi Chen authored
-
ziheng authored
* [NNPack] Support for threadpool * fix lint * fix lint * Use static class function
-
- Aug 15, 2017
-
-
ziheng authored
* [Contrib] CuDNN v7 Support * Add test
-
Xingjian Shi authored
[TOPI] Add broadcast and reduce operators
-
- Aug 13, 2017
-
-
Nicolas Vasilache authored
* [WIP] C++ topi contributions Summary: This diff implements C++ topi contributions for: - relu with parametrix threshold - pad with generic padBefore / padAfter specification - matmult with transposes - conv2d_nchw, conv2d_hwcn with runtime constant padding and strides - depthwise_conv2d_nchw with runtime constant padding and strides - group_conv2d_ngchw with runtime constant padding and strides - broadcast_to a broadcastable shape - broadcast_bop where bop is an usual binary op (+ - * / %) Convolution padding is implemented using the pad operation. To avoid extra memory consumption, it is generally recommended to inline the padding with the autoinliner. Unfortunately in its current form the elemwise checks are too restrictive to allow inlining. So this diff also proposes an extension to LHS injective (i.e. no reduction axis in the current IR design) Test Plan: Tested in C++ testsuite in a separate repository, I am looking for suggestions to quickly spin up some tests for tvm. Reviewers: tqchen Subscribers: Tasks: Tags: Blame Revision: * Review + Lint + GSG C++
-
Tianqi Chen authored
-
Tianqi Chen authored
-
- Aug 12, 2017
-
-
Tianqi Chen authored
-
- Aug 11, 2017
-
-
Tianqi Chen authored
* [PASS][FIX] Fix LiftAttrScope with if * [PASS] Fix on proc sync * fix
-
- Aug 10, 2017
-
-
Tianqi Chen authored
-
Tianqi Chen authored
* [TEST] Upgrade gpu docker to cudnn7 * fx
-
- Aug 09, 2017
-
-
Tianqi Chen authored
-
- Aug 08, 2017
-
-
Tianqi Chen authored
-
Tianqi Chen authored
* [RUNTIME][PASS] Allow declare vector type array * fix bcast * [BUFFER] Enable vload/store function in buffer * ok
-
- Aug 07, 2017
-
-
ziheng authored
* [NNPACK] Add nnpack.convolution * Add instrinsic * Fix lint
-
- Aug 05, 2017
-
-
Tianqi Chen authored
-
- Aug 04, 2017
-
-
Tianqi Chen authored
-
- Aug 03, 2017
-
-
Tianqi Chen authored
* [PASS] Refactor thread storage sync to a common visitor * Fix the sync scope check behavior
-
Tianqi Chen authored
-
- Aug 01, 2017
-
-
Tianqi Chen authored
-
Tianqi Chen authored
* [SCHEDULE] Fix fuse node order * Make fuse order consistent with split
-
- Jul 31, 2017
-
-
William Moses authored
-
- Jul 28, 2017
-
-
ziheng authored
* [GRAPH] Add GraphHandle * Move to apps/graph_executor
-
- Jul 26, 2017
-
-
Tianqi Chen authored
* [SCHEDULE] Remap the cached bind_scope. * more fix
-
- Jul 25, 2017
-
-
Tianqi Chen authored
-
- Jul 24, 2017
-
-
Tianqi Chen authored
-
Tianqi Chen authored
* [STORAGE][BUFFER] Support access ptr for clear access pattern. * fix lint
-
- Jul 20, 2017
-
-
Tianqi Chen authored
-
- Jul 19, 2017
-
-
Tianqi Chen authored
* [SCHEDULE] Detect useless bound early * fix
-