[GitHub] [incubator-tvm] jcf94 edited a comment on pull request #6073: [Ansor][AutoTVM v2.0] Phase 1: Add annotation/compute_at/compute_root/compute_inline steps

jcf94 edited a comment on pull request #6073: URL: https://github.com/apache/incubator-tvm/pull/6073#issuecomment-660786809 Moved all the implementation to transform_step.h/cc, and updated the order of steps. In the current implementation, the next pr of steps adding will only need to c

[GitHub] [incubator-tvm] jcf94 edited a comment on pull request #6073: [Ansor][AutoTVM v2.0] Phase 1: Add annotation/compute_at/compute_root/compute_inline steps

jcf94 edited a comment on pull request #6073: URL: https://github.com/apache/incubator-tvm/pull/6073#issuecomment-660786809 Moved all the implementation to transform_step.h/cc, and updated the order of steps. In the current implementation, the next pr of steps adding will only need to c

[GitHub] [incubator-tvm] jcf94 commented on pull request #6073: [Ansor][AutoTVM v2.0] Phase 1: Add annotation/compute_at/compute_root/compute_inline steps

jcf94 commented on pull request #6073: URL: https://github.com/apache/incubator-tvm/pull/6073#issuecomment-660786809 Moved all the implementation to transform_step.h & transform_step.cc, and updated the order of steps. @merrymercy @junrushao1994 @comaniac @FrozenGene -

[GitHub] [incubator-tvm] FrozenGene commented on a change in pull request #5915: [BYOC][Contrib] Arm Compute Library integration

FrozenGene commented on a change in pull request #5915: URL: https://github.com/apache/incubator-tvm/pull/5915#discussion_r456996979 ## File path: docs/deploy/arm_compute_lib.rst ## @@ -0,0 +1,138 @@ +.. Licensed to the Apache Software Foundation (ASF) under one +or more c

[GitHub] [incubator-tvm] nhynes commented on pull request #5892: Add TVM application extension with WASM runtime

nhynes commented on pull request #5892: URL: https://github.com/apache/incubator-tvm/pull/5892#issuecomment-660776031 oh, also, it might be useful to have tensor contents be base64 serialized, if serde support is going to be upstreamed and json is the most common serializer -

[GitHub] [incubator-tvm] nhynes commented on a change in pull request #5892: Add TVM application extension with WASM runtime

nhynes commented on a change in pull request #5892: URL: https://github.com/apache/incubator-tvm/pull/5892#discussion_r457000729 ## File path: apps/wasm-standalone/wasm-graph/src/lib.rs ## @@ -0,0 +1,83 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * o

[GitHub] [incubator-tvm] lsy643 commented on a change in pull request #5955: Register Shape Func for Some Operators to Handle Dynamic Shapes

lsy643 commented on a change in pull request #5955: URL: https://github.com/apache/incubator-tvm/pull/5955#discussion_r456989467 ## File path: python/tvm/relay/op/image/_image.py ## @@ -64,6 +67,22 @@ def compute_crop_and_resize(attrs, inputs, out_type): reg.register_injecti

[GitHub] [incubator-tvm] ANSHUMAN87 commented on pull request #6066: [TIR][Transform] HoistIfThenElse added

ANSHUMAN87 commented on pull request #6066: URL: https://github.com/apache/incubator-tvm/pull/6066#issuecomment-660693140 > Since nearly all programs will be touched by this pass, potential bugs in this pass would be critical. Could you (maybe temporarily) add this pass into the default bu

[GitHub] [incubator-tvm-vta] remotego commented on a change in pull request #9: [Hardware][OpenCL] Intelfocl support

remotego commented on a change in pull request #9: URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456895729 ## File path: config/intelfocl_sample.json ## @@ -0,0 +1,13 @@ +{ + "TARGET" : "intelfocl", + "HW_VER" : "0.0.1", + "LOG_INP_WIDTH" : 3, Review c

[GitHub] [incubator-tvm-vta] remotego edited a comment on pull request #9: [Hardware][OpenCL] Intelfocl support

remotego edited a comment on pull request #9: URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660622675 > > The code should work for all devices supported by Intel OpenCL for FPGA, namely Intel Arria 10, Stratix V/10 and Cyclone V/10. > > Just my humble opinion,

[GitHub] [incubator-tvm-vta] remotego edited a comment on pull request #9: [Hardware][OpenCL] Intelfocl support

remotego edited a comment on pull request #9: URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660622675 > > The code should work for all devices supported by Intel OpenCL for FPGA, namely Intel Arria 10, Stratix V/10 and Cyclone V/10. > > Just my humble opinion,

[GitHub] [incubator-tvm-vta] remotego edited a comment on pull request #9: [Hardware][OpenCL] Intelfocl support

remotego edited a comment on pull request #9: URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660622675 > > The code should work for all devices supported by Intel OpenCL for FPGA, namely Intel Arria 10, Stratix V/10 and Cyclone V/10. > > Just my humble opinion,

[GitHub] [incubator-tvm-vta] remotego commented on pull request #9: [Hardware][OpenCL] Intelfocl support

remotego commented on pull request #9: URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660622675 > > The code should work for all devices supported by Intel OpenCL for FPGA, namely Intel Arria 10, Stratix V/10 and Cyclone V/10. > > Just my humble opinion, given t

[GitHub] [incubator-tvm-vta] liangfu commented on pull request #9: [Hardware][OpenCL] Intelfocl support

liangfu commented on pull request #9: URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660618430 > The code should work for all devices supported by Intel OpenCL for FPGA, namely Intel Arria 10, Stratix V/10 and Cyclone V/10. Just my humble opinion, given that both "I

[GitHub] [incubator-tvm-vta] liangfu edited a comment on pull request #9: [Hardware][OpenCL] Intelfocl support

liangfu edited a comment on pull request #9: URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660618430 > The code should work for all devices supported by Intel OpenCL for FPGA, namely Intel Arria 10, Stratix V/10 and Cyclone V/10. Just my humble opinion, given t

[GitHub] [incubator-tvm-vta] liangfu commented on a change in pull request #9: [Hardware][OpenCL] Intelfocl support

liangfu commented on a change in pull request #9: URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456884032 ## File path: config/intelfocl_sample.json ## @@ -0,0 +1,13 @@ +{ + "TARGET" : "intelfocl", + "HW_VER" : "0.0.1", + "LOG_INP_WIDTH" : 3, Review co

[GitHub] [incubator-tvm] MarisaKirisame commented on a change in pull request #6078: [Autodiff] Optimize and eliminate the Jacobian tensor for te.autodiff

MarisaKirisame commented on a change in pull request #6078: URL: https://github.com/apache/incubator-tvm/pull/6078#discussion_r456881913 ## File path: src/te/autodiff/ad_simplify.cc ## @@ -0,0 +1,1266 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or

[GitHub] [incubator-tvm] roastduck commented on pull request #6066: [TIR][Transform] HoistIfThenElse added

roastduck commented on pull request #6066: URL: https://github.com/apache/incubator-tvm/pull/6066#issuecomment-660601199 Since nearly all programs will be touched by this pass, potential bugs in this pass would be critical. Could you (maybe temporarily) add this pass into the default build