- 11 Dec, 2017 2 commits
-
-
* Use long long for platforms where long is 32 bits (like windows). * Make sure scalar chars are signed. * Re-add NOLINT marker.
abergeron committed -
* [CODEGEN] add fp16 and fp64 enable pragma for opencl * fix style
Lianmin Zheng committed
-
- 07 Dec, 2017 1 commit
-
-
Lianmin Zheng committed
-
- 05 Dec, 2017 1 commit
-
-
* Port build_module.py to C++ * Fix lint errors * Fix more lint errors * Fix more lint errors * Fix more lint errors * Fix build error * Implemented style fixes * Fix lint errors * Added function to construct target from string lower now returns array * Fix lint error * Implemented review changes - style & Target options -> std::vector * Fixed lint, argument alignment and added unit test * Changed test to target LLVM, fixed sign compare warnings * Reverted unit test to CUDA, changed Jenkinsfile to enable GPU for C++ tests * Slight change to Jenkinsfile * Changed build_module test from CUDA to LLVM * Added function var() to construct a Var instance. Changed implementation of LLVMEnabled() * Reverted Jenkinsfile
alex-weaver committed
-
- 04 Dec, 2017 2 commits
-
-
* [CI] Enable llvm in CPU test * fix llvm
Tianqi Chen committed -
* Support rank-0 tensor * fix lint
Tianqi Chen committed
-
- 01 Dec, 2017 1 commit
-
-
* [RANDOM] Init contrib.random library * [RANDOM] Add uniform * [RANDOM] Fix lint * [RANDOM] Add comments and tests * [RANDOM] Fix lint
ziheng committed
-
- 30 Nov, 2017 5 commits
-
-
Salem Derisavi committed
-
* [CUDA] Enable int64 * [PYTHON] Fix rpc tutorial with opencl * OK * update
Tianqi Chen committed -
Yizhi Liu committed
-
Change the parameter 'C' name
solin319 committed -
In unroll_loop.cc the parameter name is "auto_max_depth", but in ir_pass.h the parameter name is "auto_min_depth"
solin319 committed
-
- 29 Nov, 2017 1 commit
-
-
* [RPC][JVM] Remove binary dist gradle from repo * fix header
Tianqi Chen committed
-
- 28 Nov, 2017 1 commit
-
-
Tianqi Chen committed
-
- 27 Nov, 2017 1 commit
-
-
ziheng committed
-
- 25 Nov, 2017 1 commit
-
-
* [PASS] Allow compact checking when strides is available * remove assert compact
Tianqi Chen committed
-
- 23 Nov, 2017 1 commit
-
-
Readability.
Siva committed
-
- 21 Nov, 2017 2 commits
-
-
* [PASS/SETUP] Fix minior issues * fix lint
Tianqi Chen committed -
* mps * update
Sheng Zha committed
-
- 19 Nov, 2017 1 commit
-
-
* scheduler tweaked for super resolution perf * conv2d_transpose schedule error fixed * nnvm issue #239 fixed
Leyuan Wang committed
-
- 18 Nov, 2017 1 commit
-
-
Lianmin Zheng committed
-
- 16 Nov, 2017 2 commits
-
-
some host opencl runtime may at cpu mode, but remote client opencl runtime at gpu mode, compat it
haolongzhangm committed -
* scheduler tweaked for super resolution perf * lint error fixed * lint error fixed * conv2d_transpose schedule error fixed
Leyuan Wang committed
-
- 14 Nov, 2017 4 commits
-
-
Tianqi Chen committed
-
* [TOPI] Add out_dtype argument for conv2d; Add x86 schedules * Fix * Fix lint * Fix
ziheng committed -
Tianqi Chen committed
-
* conv2d perf improved for conv2d_56_64_128, test name added to differentiate workloads * fix lint error
Leyuan Wang committed
-
- 13 Nov, 2017 2 commits
-
-
* Make the tvm conda package build with in-place source and use cmake from conda. * Add a package for topi.
abergeron committed -
Tianqi Chen committed
-
- 12 Nov, 2017 3 commits
-
-
Tianqi Chen committed
-
Tianqi Chen committed
-
Lianmin Zheng committed
-
- 11 Nov, 2017 2 commits
-
-
* [PASS] Enhance LiftAttrScope * update vt
Tianqi Chen committed -
ziheng committed
-
- 09 Nov, 2017 2 commits
- 08 Nov, 2017 3 commits
-
-
* feat(docs) add how_to for tvm install with nnpack support * feat(docs) change python package paragraph * feat(doc) remove unsure sentence * add comments on nnpack usage vs TVM * remove mxnet nnpack tips for nthread change
Erwan BERNARD committed -
* Support vector operations for AMD (llvm IR) * fix whitespace * update comments, docstring
eqy committed -
Leyuan Wang committed
-
- 07 Nov, 2017 1 commit
-
-
Change minimum 32-bit restriction for floating point types to 8-bit. This change is to enable reduced precision types that may use vector operations underneath the hood (cases #lanes > 1 such as half4).
eqy committed
-
