Code

merge from apach/tvm to neo ai tvm#59

Code

Merged

kevinthesun merged 31 commits intoneo-ai:devneo-ai/tvm:devfrom zxy844288792:neo-ai-tvm-mergezxy844288792/tvm:neo-ai-tvm-mergeCopy head branch name to clipboardNov 25, 2019

+2,120-576

Commits

Commits on Nov 15, 2019

• [TOPI][OP] Support Faster-RCNN Proposal OP on CPU (apache#4297)

  Show description for 956aee7
  

  FrozenGene

  authored and

  Xingyu Zhou

  committed
  956aee7

  Copy full SHA for 956aee7

• [QNN][Legalize] Specialize for Platforms without any fast Int8 arithmetic units. (apache#4307)

  

  anijain2305

  authored and

  Xingyu Zhou

  committed
  d6cf282

  Copy full SHA for d6cf282

• fix error when memory_id is VTA_MEM_ID_OUT (apache#4330)

  

  jason-song-dev

  authored and

  Xingyu Zhou

  committed
  aae9ebd

  Copy full SHA for aae9ebd

• [CI][DOCKER] Add ONNX runtime dep (apache#4314)

  Show description for 9f2a7d0
  

  tqchen

  authored and

  Xingyu Zhou

  committed
  9f2a7d0

  Copy full SHA for 9f2a7d0

• [QNN] Quantize - Fixing the sequence of lowering. (apache#4316)

  

  anijain2305

  authored and

  Xingyu Zhou

  committed
  9ab329e

  Copy full SHA for 9ab329e

• [QNN] Use Int16 upcast in Fallback Conv2D. Fix test names. (apache#4329)

  

  anijain2305

  authored and

  Xingyu Zhou

  committed
  53b48df

  Copy full SHA for 53b48df

• [doc][fix] fix sphinx parsing for pass infra tutorial (apache#4337)

  

  zhiics

  authored and

  Xingyu Zhou

  committed
  607c8ea

  Copy full SHA for 607c8ea

• change ci image version (apache#4313)

  

  tqchen

  authored and

  Xingyu Zhou

  committed
  5148587

  Copy full SHA for 5148587

• [Codegen] remove fp16 function override for cuda (apache#4331)

  Show description for 9651abb
  

  yzhliu

  authored and

  Xingyu Zhou

  committed
  9651abb

  Copy full SHA for 9651abb

• [CI] Set workspace to be per executor (apache#4336)

  

  tqchen

  authored and

  Xingyu Zhou

  committed
  6e9c638

  Copy full SHA for 6e9c638

• [Build][Windows] Fix Windows build by including cctype (apache#4319)

  Show description for d0f43dc
  

  soiferj

  authored and

  Xingyu Zhou

  committed
  d0f43dc

  Copy full SHA for d0f43dc

• Enable hipModuleGetGlobal() (apache#4321)

  

  petrex

  authored and

  Xingyu Zhou

  committed
  5e61941

  Copy full SHA for 5e61941

• [Relay][Pass] Add pass to remove unused functions in relay module (apache#4334)

  Show description for 32266bb
  

  wweic

  authored and

  Xingyu Zhou

  committed
  32266bb

  Copy full SHA for 32266bb

• Add support for quant. mul operator in tflite frontend (apache#4283)

  Show description for 52f1dea
  

  inadob

  authored and

  Xingyu Zhou

  committed
  52f1dea

  Copy full SHA for 52f1dea

• Add topi.nn.fifo_buffer to TVM doc (apache#4343)

  

  hcho3

  authored and

  Xingyu Zhou

  committed
  7e50a49

  Copy full SHA for 7e50a49

• Solve custom model of prelu (apache#4326)

  

  FrozenGene

  authored and

  Xingyu Zhou

  committed
  b6d6efc

  Copy full SHA for b6d6efc

• Deprecate NNVM warning msg (apache#4333)

  

  yzhliu

  authored and

  Xingyu Zhou

  committed
  7a7685f

  Copy full SHA for 7a7685f

• [Contrib] Add MKL DNN option (apache#4323)

  Show description for c84baa4
  

  icemelon

  authored and

  Xingyu Zhou

  committed
  c84baa4

  Copy full SHA for c84baa4

• [Relay][Frontend][TF] Fix transpose when axes is not a param (apache#4327)

  Show description for 584df60
  

  soiferj

  authored and

  Xingyu Zhou

  committed
  584df60

  Copy full SHA for 584df60

• [RUNTIME] Add device query for AMD GcnArch (apache#4341)

  Show description for aa00977
  

  petrex

  authored and

  Xingyu Zhou

  committed
  aa00977

  Copy full SHA for aa00977

• [Test][Relay][Pass] Add test case for lambda lift (apache#4317)

  

  wweic

  authored and

  Xingyu Zhou

  committed
  42f8016

  Copy full SHA for 42f8016

• [Relay][Frontend][ONNX] operator support: DepthToSpace, SpaceToDepth (apache#4271)

  

  cchung100m

  authored and

  Xingyu Zhou

  committed
  a3b8499

  Copy full SHA for a3b8499

• imp module is deprecated (apache#4275)

  

  Jian Weng

  authored and

  Xingyu Zhou

  committed
  8bf20c4

  Copy full SHA for 8bf20c4

• [VTA] Bug fix for padded load with large inputs (apache#4293)

  Show description for 0be8e56
  

  liangfu

  authored and

  Xingyu Zhou

  committed
  0be8e56

  Copy full SHA for 0be8e56

• fix inconsistent tag name (apache#4134)

  

  ziyu-guo

  authored and

  Xingyu Zhou

  committed
  a0df146

  Copy full SHA for a0df146

• [CodeGen] Add build config option disable_assert to control whether to generate assert (apache#4340)

  

  FrozenGene

  authored and

  Xingyu Zhou

  committed
  41b65ef

  Copy full SHA for 41b65ef

• Bump up CUDA log version in tophub.py (apache#4347)

  

  alexgl-github

  authored and

  Xingyu Zhou

  committed
  fcd881f

  Copy full SHA for fcd881f

• Add check to ensure input file was successfully opened in NNVM deploy code demo (apache#4315)

  

  tweej

  authored and

  Xingyu Zhou

  committed
  7fabde9

  Copy full SHA for 7fabde9

• [COMMUNITY] Add DISCLAIMER, KEYS for ASF release (apache#4345)

  Show description for 4d5c586
  

  tqchen

  authored and

  Xingyu Zhou

  committed
  4d5c586

  Copy full SHA for 4d5c586

• [Relay][VM][Interpreter] Enable first-class constructors in VM and interpreter via eta expansion (apache#4218)

  Show description for 48acb45
  

  weberlo

  authored and

  Xingyu Zhou

  committed
  48acb45

  Copy full SHA for 48acb45

• Update dmlc_tvm_commit_id.txt

  

  Xingyu Zhou

  committed
  800f494

  Copy full SHA for 800f494