Commit Graph

Select branches

Hide Pull Requests

Conv_NHWC

NNET_231111

NNET_231111_from_master

NNET_GAN

NNET_OpSearch

NNET_anyOp

NNET_bias

NNET_bias_0630

NNET_e2e

NNET_e2e_fix

NNET_e2e_for_merge

NNET_eliminateOP

NNET_gcn

NNET_gcn_fuse

NNET_op_test

NNET_transpose

TC_revision

activation

add_paddle_model

allocator_memPool

ascend

bang-softmax

benchmark

benchmark_conv

benchmark_softmax

broadcast_wanghailu_0916

case-fsrcnn

change_path

constroy/doc_on_ares

conv_half

cpu_backend2

cuda-attention

cuda-transpose

dcj/for_multiple_datatype

ddp

dev-dynamic-graph

dev-dynamic-graph-allocator

dev-leakyrelu

dev-memory

dist/graph

dist_bench

dropout

dump/init

fsrcnn-conv-bias-act-fuse

gencode

gpt

graph-onnx

graphFactory

instance_norm

kunlun_temp

kvcache_attention_fp16

kvcache_backup

master

model_test

nnet_e2e_for_merge

op_timer

optimization-pass

point2point

power-fusion

pure_engine

search_engine

support_fp16

testAccuracy

test_codegen

test_onnx

train_wanghailu_1010

update_doc

update_pybind11

v0630

xpu_allreduce

test-models

Mono Color

• 04d0e1a560 add split operation wanghailu 2023-01-09 09:24:47 +0000
• d216b529e7 format wanghailu 2023-01-09 15:16:43 +0800
• cd703e5679 add concat operation wanghailu 2023-01-09 07:14:32 +0000
• 2b8bca17e2 format wanghailu 2023-01-05 14:23:46 +0800
• 156a40806d add pad operation wanghailu 2023-01-05 05:49:53 +0000
• c19d6e6bb0 add det operation wanghailu 2023-01-04 09:24:52 +0000
• 68f4630dac add cumsum operation wanghailu 2023-01-03 08:45:54 +0000
• 5ae96ce060 add floormod operation wanghailu 2023-01-03 07:20:47 +0000
• dbb606f158 add floordiv operation and floordivtrunc operation wanghailu 2023-01-03 07:07:22 +0000
• 0079d1271b add cast operation wanghailu 2022-12-28 08:57:52 +0000
• 5329e66d0f add muln operation wanghailu 2022-12-27 08:22:50 +0000
• 45ea5c83f6 add addn operation wanghailu 2022-12-27 07:03:23 +0000
• 9177629a77 add transform operation wanghailu 2022-12-27 02:13:10 +0000
• f98f91de8b add sqrt and rsqrt operation wanghailu 2022-12-26 06:29:12 +0000
• 335dfabf80 add reciprocal operation wanghailu 2022-12-26 06:15:07 +0000
• 376c992aca add power operation wanghailu 2022-12-26 05:45:40 +0000
• 39d2a3571b add negTensor operation wanghailu 2022-12-26 04:40:24 +0000
• 0707fb6aff add mseloss operation wanghailu 2022-12-26 03:06:34 +0000
• d780f687fc
  ADD: reconfig ResizeObj, support "tf_crop_and_resize " and cubic coeff kernel. (#59) wendy12022 2022-12-24 04:02:21 +0800
• 4ad648fa36 add maximum and minimum operation wanghailu 2022-12-21 07:42:54 +0000
• 2749b49ff7 add l2loss operation wanghailu 2022-12-21 02:21:19 +0000
• 34ba231cd4 add log1p operation wanghailu 2022-12-21 01:41:28 +0000
• 8bd1d64c53 add log operation wanghailu 2022-12-20 03:09:40 +0000
• 084063a68f add operation fill wanghailu 2022-12-19 02:59:33 +0000
• 82f510672d add exp operation wanghailu 2022-12-19 02:03:12 +0000
• b27b95a5e2 add erf operation wanghailu 2022-12-19 01:51:25 +0000
• 9346232129 add divnonan operation and test wanghailu 2022-12-15 08:47:22 +0000
• a56fb98eee add operation cnnl div, test and test for divdemo bangc kernel wanghailu 2022-12-15 08:26:49 +0000
• 949e00b732 add operation clip wanghailu 2022-12-15 06:04:23 +0000
• 58b89dd601 add ceil operation and floor operation wanghailu 2022-12-14 02:50:06 +0000
• 46a1bb2773 add copy operation on mlu wanghailu 2022-12-14 02:32:32 +0000
• 820d855ec8 add trigon function operation on mlu: sin,cos,tan,asin,sinh,asinh wanghailu 2022-12-12 06:24:24 +0000
• 392427cca6 add transpsoe code and test wanghailu 2022-12-12 05:17:53 +0000
• 8cfe04e5b7 fix wanghailu 2022-12-08 07:41:15 +0000
• a8bd1a910c add convbpfilter wanghailu 2022-12-07 06:52:45 +0000
• b68dcf8b9a add test wanghailu 2022-12-06 07:09:51 +0000
• 111ff10df0 add test for activation_backward wanghailu 2022-12-06 04:38:26 +0000
• db9069f1b7 add activation backward operation wanghailu 2022-12-05 08:45:39 +0000
• 468ed541af commit for format wanghailu 2022-12-02 11:38:02 +0800
• 267bfa3a4b add activation operatiopn relu, tanh, sigmoid on mlu wanghailu 2022-12-02 03:24:09 +0000
• e991b3261b Add: clone for operators search_engine Liyan Zheng 2022-11-15 21:15:10 +0800
• f133f00478 [Intermediate state] Add: Graph ctor for OpVec Liyan Zheng 2022-11-15 21:09:03 +0800
• e549f21867 Add: tensor fuid Liyan Zheng 2022-11-15 15:53:16 +0800
• 89398a1c57 update fused kernels. power-fusion mazx 2022-11-15 15:13:50 +0800
• c5966f8d81
  Add: resize operator and cuda kernel,support nearest/linear coef. (#51) wendy12022 2022-11-14 09:30:22 +0800
• 6e3f0bbf9a update reduce kernel mazx 2022-11-02 16:08:49 +0800
• 2aadcb6e9c add: onnx ok. mazx 2022-10-30 23:08:16 +0800
• 3046dd5901 add: graph for bert. mazx 2022-10-29 00:18:00 +0800
• 5ed540be6e add: optimization pass for metaGraph. mazx 2022-10-28 20:42:10 +0800
• ec58c85505 add: add metaGaph for codegen. mazx 2022-10-27 00:42:23 +0800
• 18b79903ee add: codegen for all metaOps. mazx 2022-10-26 02:20:16 +0800
• 2c8bd3729b add: generate transpose, unary, and binary. mazx 2022-10-26 01:42:45 +0800
• 254d23b3c0 Add: transpose operator mazx 2022-10-21 16:51:12 +0800
• 05d39439db add: init power fusion. mazx 2022-10-19 14:13:52 +0800
• 48e986d377 add: fix bert to transpose. mazx 2022-10-30 21:46:46 +0800
• 09365c81f4 add: graph build for pf. mazx 2022-10-30 14:19:59 +0800
• 94faefb0ef Add: pytest for import_onnx Pairshoe 2022-10-26 23:57:29 +0800
• 970c77d0f4 Update: Rename GraphFactory -> GraphBuilder && Remove unnecessary outputs Pairshoe 2022-10-26 21:06:21 +0800
• 7cf2d8f78f Add: python interfaced for importing onnx Pairshoe 2022-10-15 11:15:24 +0800
• ff90c4c7d5 Add: test for class GraphFactoryObj Pairshoe 2022-10-13 23:14:28 +0800
• 9e45e51279 Add: class GraphFactory and pybind11 interfaces Pairshoe 2022-10-13 21:26:43 +0800
• 1b1fc2585b Add: save optime result op_timer Liyan Zheng 2022-11-02 17:38:08 +0800
• eb993f7829 Add: evaluate onnx script Liyan Zheng 2022-11-02 16:51:33 +0800
• 63e5df4227 Add: fused conv Liyan Zheng 2022-11-02 16:39:12 +0800
• a0e07199ff add: graph build for pf. graph-onnx mazx 2022-10-30 14:19:59 +0800
• f88aefb2ca Add: pytest for import_onnx graphFactory Pairshoe 2022-10-26 23:57:29 +0800
• 0ef74a2145 Update: Rename GraphFactory -> GraphBuilder && Remove unnecessary outputs Pairshoe 2022-10-26 21:06:21 +0800
• bef4c422a0 Add: improve conv2dreduce kernel case-fsrcnn Liyan Zheng 2022-10-22 13:26:41 +0800
• 67c06733e6 Chore: format Liyan Zheng 2022-10-22 13:24:42 +0800
• beba9c16c4 add code for test resnet testAccuracy wanghailu 2022-10-21 14:46:42 +0800
• 53594b2ebc test resnet wanghailu 2022-10-20 16:36:55 +0800
• aa552b5bd2 Add: batch size in test Liyan Zheng 2022-10-19 15:37:09 +0800
• 36755c3160 convNHWC+pReLU+biasPReLU huangshuhong 2022-10-18 14:01:45 +0800
• 74e998e262 Add: use Runtime stream in Copy Liyan Zheng 2022-10-13 22:22:09 +0800
• 7abe7da0e4 Add: CUDA graph for fsrcnn Liyan Zheng 2022-10-13 21:41:17 +0800
• 133513be34 fsrcnn test huangshuhong 2022-10-12 15:29:26 +0800
• 78425c3209 Add: fsrcnn Liyan Zheng 2022-10-07 21:50:25 +0800
• 63d8aff985
  Fix: cuCtxCreate before other initialization (#49) zhengly123 2022-10-19 15:41:48 +0800
• 00b2f18c17
  Fix: unsigned compare in test (#50) Zixuan Ma 2022-10-19 15:03:03 +0800
• 4e0040c8a0
  Add: connection among tensors and operators (#45) zhengly123 2022-10-18 22:02:51 +0800
• d50fecc132 Add: python interfaced for importing onnx Pairshoe 2022-10-15 11:15:24 +0800
• 4836decc69 Add: test for class GraphFactoryObj Pairshoe 2022-10-13 23:14:28 +0800
• fe535e72a7 Add: class GraphFactory and pybind11 interfaces Pairshoe 2022-10-13 21:26:43 +0800
• d1c913010f
  ADD:reduce_mean operator and cuda kernel. (#47) wendy12022 2022-10-15 16:53:58 +0800
• a4d6426589
  ADD: batch norm operator and cuda kernel. (#44) wendy12022 2022-10-15 16:29:28 +0800
• 7382a94243 add code for training solution train_wanghailu_1010 wanghailu 2022-10-10 17:11:41 +0800
• 1152adc94a
  Add: python API for timing ConvTranspose (#46) zhengly123 2022-10-07 16:03:11 +0800
• b0c2a08252
  Support bang c kernel wanghailu 0927 (#43) Hardy 2022-09-30 11:01:52 +0800
• 26cee55e81
  ADD:extend operator and cuda kernel. (#40) wendy12022 2022-09-29 14:52:50 +0800
• fe14c91f54
  ADD: Gather operator and cuda kernel. (#41) wendy12022 2022-09-29 14:44:20 +0800
• 3c6e208f42
  ADD:concat/split operator and cuda kernels (#29) wendy12022 2022-09-29 11:01:30 +0800
• 5560d0f2fb
  ADD:pad/slice operator and cuda kernel. (#39) wendy12022 2022-09-29 10:29:24 +0800
• 1aefc1b27e
  Add python interface for CUDA operator evaluation (#42) zhengly123 2022-09-27 10:41:12 +0800
• 11d5aa1ccc
  Add TVM codegen for MemboundOp (#35) deathwings602 2022-09-22 18:06:45 +0800
• ba0b11a499
  Update README.md zhengly123 2022-09-22 17:38:15 +0800
• c7c974f07a
  Add bangc runtime and element-wise kernels Hardy 2022-09-22 16:57:39 +0800
• 90eb9d05a8
  Json perfrecord (#32) Anmuliar 2022-09-22 15:34:34 +0800
• 9032cbb973
  Add: reshape/flatten/identity OP and cuda kernel (#34) wendy12022 2022-09-21 14:04:30 +0800
• 2f8f706f1c
  Fix CMake USE_CUDA (#36) zhengly123 2022-09-21 12:28:00 +0800
• 8f67a5cc76
  Add: ConvTransposed (#33) zhengly123 2022-09-19 15:05:39 +0800