Hanchen Ye
|
c73e43cf81
|
[Transforms] Add pytorch-pipeline-v2 with inter/intra-node memory optimizations; Remove axi-interf option from dse-pipeline as we haven't been able to scale it; [FuncDataflow] Support to specify target-func during initialization
|
2022-03-22 17:57:10 -05:00 |
Hanchen Ye
|
4a8b516e1d
|
[LoopPipelining] Stop to pipeline the loop if it is already dataflowed; [Samples] Add an new end-to-end pipeline for resnet18
|
2022-03-22 16:55:08 -05:00 |
Hanchen Ye
|
18cf114999
|
[Samples] Add the instructions to run the new torch flow
|
2022-03-20 17:02:11 -05:00 |
Hanchen Ye
|
564c413385
|
[Samples] Remove the separate readme under the rosetta directory
|
2022-03-08 01:21:53 -06:00 |
Hanchen Ye
|
81d48fe584
|
[Readme] Update c++ and pytorch test instructions; Fix some small errors
|
2022-02-23 18:03:46 -06:00 |
Hanchen Ye
|
a4b9e1fb7e
|
[Samples] Remove onnx-mlir folder completely
|
2022-02-23 17:38:50 -06:00 |
Hanchen Ye
|
4c1e8f177f
|
Remove onnx front-end option from scalehls pipeline; Deprecate onnx-mlir samples
|
2022-02-22 23:46:10 -06:00 |
Hanchen Ye
|
875f04a4c1
|
[LegalizeDataflow][SplitFunction] Support adopt TOSA IR from torch-mlir
|
2022-02-22 23:25:05 -06:00 |
Hanchen Ye
|
3bea951beb
|
[ArrayPartition][LegalizeHLSCpp] Handle the memref type of global op
|
2022-02-22 00:46:36 -06:00 |
Hanchen Ye
|
bdf72c8cc3
|
[LegalizeDataflow] Support linalg ops; Make this pass more robust; [SplitFunction] Support linalg ops
|
2022-02-21 19:41:03 -06:00 |
Hanchen Ye
|
c621ee0464
|
[Transforms] Add pipeline option to specify front-end: torch or onnx
|
2022-02-20 18:24:56 -06:00 |
Hanchen Ye
|
82eccaa915
|
[Samples] Update the structure of the samples folder
|
2022-02-20 17:15:16 -06:00 |
Hanchen Ye
|
0d51c997c5
|
[Readme] Separte instructions for onnx-mlir and torch-mlir to samples folder; Add bibtex for the scalehls paper; [DSE] Output the result after dse; [Samples] Add torchscript resnet18 example to torch-mlir folder
|
2022-02-20 14:56:05 -06:00 |
Hanchen Ye
|
5c8e1ebda2
|
[QoREstimator] Improve the accuracy when resource can be shared between multiple loop bands
|
2022-02-13 23:21:22 -06:00 |
Hanchen Ye
|
1be6af93d8
|
[QoREstimator] Support fexp op; Start to specify dsp usage in config.json
|
2022-02-11 05:56:38 -06:00 |
Hanchen Ye
|
baa212b473
|
[QoREstimation] Resolve an issue when estimating multi-loop functions with no_touch annotations
|
2022-02-11 04:51:42 -06:00 |
Hanchen Ye
|
61365028c1
|
[Samples] Add config file into rosetta benchmarks
|
2022-02-09 14:52:45 -06:00 |
Hanchen Ye
|
941d17bdf9
|
[Utils] Include reduce-initial-interval into the dse
|
2022-02-09 14:34:57 -06:00 |
Hanchen Ye
|
3c3c65a724
|
[Samples] Update array-partition and qor-estimation to have rosetta spam-filter work
|
2022-02-08 21:25:38 -06:00 |
Hanchen Ye
|
0ac214b1e7
|
[Samples] Add two rosetta benchmarks
|
2022-02-03 23:40:40 -06:00 |
Hanchen Ye
|
d356071a53
|
Improve error messeges of json parsing; Add comments to each item in config.json
|
2022-01-28 02:12:56 -06:00 |
Hanchen Ye
|
817620d5a0
|
[QoREstimation][MultipleLevelDSE] Use JSON file for configuration
|
2022-01-21 22:13:55 -06:00 |
Hanchen Ye
|
76e3980b9d
|
[Samples] Remove deprecated testing scripts
|
2021-11-12 02:35:53 -06:00 |
Hanchen Ye
|
36d5083315
|
[Samples] Update all onnx-mlir samples
|
2021-11-12 02:32:40 -06:00 |
Hanchen Ye
|
22e59877cb
|
[Samples] Update all polybench samples
|
2021-11-12 02:09:50 -06:00 |
Hanchen Ye
|
a033ec2640
|
update README; update polybench C test cases
|
2021-08-23 16:16:20 -05:00 |
Hanchen Ye
|
debc934b1b
|
[Samples] add baseline and opt c++ files for all computation kernels and dnn models shown in the paper
|
2021-04-08 17:59:36 -05:00 |
Hanchen Ye
|
4a37ed13de
|
[QoREstimation] fixed a bug causing unaccurate estimation of syrk and syr2k; [Samples] update all testing scripts and samples
|
2021-04-07 22:44:38 -05:00 |
Hanchen Ye
|
d8420af618
|
[Samples] add mobilenet test case; split int and float ablation study
|
2021-04-07 18:17:13 -05:00 |
Hanchen Ye
|
d1786aeed7
|
[QuantizeOnnx] update the pass name
|
2021-04-06 20:35:32 -05:00 |
Hanchen Ye
|
a5cd433029
|
[EmitHLSCpp] fix an error of EmitSelectOp; update from val{n} to v{n}
|
2021-04-06 18:25:45 -05:00 |
Hanchen Ye
|
1c5e11940a
|
[Samples] update dse testing script
|
2021-04-02 23:54:28 -05:00 |
Hanchen Ye
|
d0b847c26a
|
[MultipleLevelDSE] support to specify the maximum exploration and loop parallelism; [Samples] add trmm test case, update the problem size of all benchmarks to 1024
|
2021-03-31 22:45:14 -05:00 |
Hanchen Ye
|
cd03a1def3
|
[MultipleLevelDSE] support to explore directive-only design space
|
2021-03-31 19:32:56 -05:00 |
Hanchen Ye
|
a9d93f0ed2
|
[MultipleLevelDSE] support to specify sample number; [EmitHLSCpp] stop to emit interface pragma if array is fully partitioned; [Samples] update polybench test script
|
2021-03-31 15:56:00 -05:00 |
Hanchen Ye
|
1d00ea9812
|
[ArrayPartition] fix a bug; add applyArrayPartition() to transform utils; [Samples] update dnn test target fpga
|
2021-03-30 15:54:53 -05:00 |
Hanchen Ye
|
6dc5b6785e
|
[Samples] update polybench dse and dnn ablation scripts
|
2021-03-29 23:56:35 -05:00 |
Hanchen Ye
|
d3c59a543e
|
[Samples] implement polybench dse test script
|
2021-03-29 21:50:53 -05:00 |
Hanchen Ye
|
6bf2845742
|
[MultipleLevelDSE] support to dump all design spaces (loops and funcs); begin of exporting pareto design points C++ files; [Samples] reconstruct the file structure
|
2021-03-29 18:53:55 -05:00 |
Hanchen Ye
|
58bd312356
|
[ArrayPartition] update to target FuncOp rather than ModuleOp; [MultipleLevelDSE] add resConstraint option; [Samples] update polybench benchmarks
|
2021-03-29 15:52:34 -05:00 |
Hanchen Ye
|
a040ce5009
|
[PartialAffineLoopTile] support loop order opt immediately after loop tiling; [AffineLoopOrderOpt] fix a small bug; [Samples] update the ablation study script
|
2021-03-26 12:23:28 -05:00 |
Hanchen Ye
|
75af36560c
|
[Samples] update onnx-mlir ablation study script; set bias to false for all conv layers
|
2021-03-24 11:25:05 -05:00 |
Hanchen Ye
|
2888f93d1d
|
[SimplifyAffineIf] fixed a major bug...; [Samples] replace pool layer with stride 2 conv layer in vgg16 and lenet; update ablation test script
|
2021-03-23 16:08:41 -05:00 |
Hanchen Ye
|
10fdc75234
|
[Samples] update onnx-mlir ablation study script; update lenet and vgg11 test case
|
2021-03-23 11:39:17 -05:00 |
Hanchen Ye
|
b372b731be
|
[Samples] add vgg11 benchmark; update ablation test scripts
|
2021-03-22 22:07:18 -05:00 |
Hanchen Ye
|
1fe0475195
|
update to LLVM f178c13fa899; [ArrayPartition] fixed a bug; [Samples] remove array initialization in atax, bicg, and gesummv
|
2021-03-22 17:29:43 -05:00 |
Hanchen Ye
|
dc3ef5ee67
|
[ArrayPartition] resolve an issue that non-pipelined loops are not considered; [Transforms] add an applyFullyUnrollAndPartition() until; [Samples] reverse gemm, syrk, syr2k back...
|
2021-03-16 23:57:18 -05:00 |
Hanchen Ye
|
6c9f415e92
|
[AffineStoreForward] fixed a small bug; [Samples] update gemm, syrk, syr2k to align with polybench
|
2021-03-16 15:10:24 -05:00 |
Hanchen Ye
|
1d39291a92
|
[MultipleLevelDSE] fixed a small issue; [Samples] update the problem sizes of polybench test cases
|
2021-03-04 19:58:46 -06:00 |
Hanchen Ye
|
17d34f0563
|
[MultipleLevelDSE] evaluate the tile config that fully unrolls all sequential loops in the design space initialization
|
2021-03-02 18:51:48 -06:00 |