[GitHub] [incubator-tvm] kazum commented on issue #5283: [RUNTIME][CONTRIB] CoreML Runtime

kazum commented on issue #5283: [RUNTIME][CONTRIB] CoreML Runtime URL: https://github.com/apache/incubator-tvm/pull/5283#issuecomment-611642375 I've created RFC https://discuss.tvm.ai/t/rfc-coreml-runtime/6309. This is an

[GitHub] [incubator-tvm] mehrdadhe commented on a change in pull request #5293: [BUGFIX] Fix CRT static test bug

mehrdadhe commented on a change in pull request #5293: [BUGFIX] Fix CRT static test bug URL: https://github.com/apache/incubator-tvm/pull/5293#discussion_r406346840 ## File path: apps/bundle_deploy/demo_static.c ## @@ -103,6 +104,6 @@ int main(int argc, char **argv) {

[GitHub] [incubator-tvm] tqchen edited a comment on issue #5282: [LLVM] Fix generation of LLVM intrinsics

tqchen edited a comment on issue #5282: [LLVM] Fix generation of LLVM intrinsics URL: https://github.com/apache/incubator-tvm/pull/5282#issuecomment-611722366 The CI error was due to a flaky problem which is fixed now by https://github.com/apache/incubator-tvm/pull/5293

[GitHub] [incubator-tvm] tqchen commented on issue #5282: [LLVM] Fix generation of LLVM intrinsics

tqchen commented on issue #5282: [LLVM] Fix generation of LLVM intrinsics URL: https://github.com/apache/incubator-tvm/pull/5282#issuecomment-611722366 The CI error was due to a flaky problem which is fixed now This is an

[GitHub] [incubator-tvm] zhiics commented on a change in pull request #5277: [BYOC] Refine AnnotateTarget and MergeCompilerRegion Passes

zhiics commented on a change in pull request #5277: [BYOC] Refine AnnotateTarget and MergeCompilerRegion Passes URL: https://github.com/apache/incubator-tvm/pull/5277#discussion_r406409545 ## File path: src/relay/analysis/annotated_region_set.h ## @@ -184,11 +193,11 @@

[GitHub] [incubator-tvm] zhiics commented on a change in pull request #5277: [BYOC] Refine AnnotateTarget and MergeCompilerRegion Passes

zhiics commented on a change in pull request #5277: [BYOC] Refine AnnotateTarget and MergeCompilerRegion Passes URL: https://github.com/apache/incubator-tvm/pull/5277#discussion_r406408975 ## File path: src/relay/analysis/annotated_region_set.cc ## @@ -74,25 +75,21 @@

[GitHub] [incubator-tvm] alexbooth commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

alexbooth commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#discussion_r406382408 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -856,6

[GitHub] [incubator-tvm] siju-samuel opened a new pull request #5295: [PYTHON]Abs, Arange, Softplus ops

siju-samuel opened a new pull request #5295: [PYTHON]Abs, Arange, Softplus ops URL: https://github.com/apache/incubator-tvm/pull/5295 - Abs - Arange - Softplus @masahi please help to review and merget these pytorch ops. Thanks a lot.

[GitHub] [incubator-tvm] tqchen closed issue #5285: [TEST][FLAKY] Sphinx pre-check build error

tqchen closed issue #5285: [TEST][FLAKY] Sphinx pre-check build error URL: https://github.com/apache/incubator-tvm/issues/5285 This is an automated message from the Apache Git Service. To respond to the message, please log

[GitHub] [incubator-tvm] tqchen commented on issue #5285: [TEST][FLAKY] Sphinx pre-check build error

tqchen commented on issue #5285: [TEST][FLAKY] Sphinx pre-check build error URL: https://github.com/apache/incubator-tvm/issues/5285#issuecomment-611722116 closing for now. btw @zhiics perhaps we need to look into cuda driver of the rebooted machine, which reports cuda not found

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5277: [BYOC] Refine AnnotateTarget and MergeCompilerRegion Passes

comaniac commented on a change in pull request #5277: [BYOC] Refine AnnotateTarget and MergeCompilerRegion Passes URL: https://github.com/apache/incubator-tvm/pull/5277#discussion_r406463482 ## File path: src/relay/analysis/annotated_region_set.cc ## @@ -74,25 +75,21 @@

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5277: [BYOC] Refine AnnotateTarget and MergeCompilerRegion Passes

comaniac commented on a change in pull request #5277: [BYOC] Refine AnnotateTarget and MergeCompilerRegion Passes URL: https://github.com/apache/incubator-tvm/pull/5277#discussion_r406463527 ## File path: src/relay/analysis/annotated_region_set.h ## @@ -184,11 +193,11 @@

[GitHub] [incubator-tvm] wpan11nv commented on a change in pull request #5226: [CODEGEN][CUDA] Fix vector load

wpan11nv commented on a change in pull request #5226: [CODEGEN][CUDA] Fix vector load URL: https://github.com/apache/incubator-tvm/pull/5226#discussion_r406352686 ## File path: src/target/source/codegen_cuda.cc ## @@ -591,13 +591,17 @@ void CodeGenCUDA::VisitExpr_(const

[GitHub] [incubator-tvm] kparzysz-quic commented on issue #5282: [LLVM] Fix generation of LLVM intrinsics

kparzysz-quic commented on issue #5282: [LLVM] Fix generation of LLVM intrinsics URL: https://github.com/apache/incubator-tvm/pull/5282#issuecomment-611743041 Same failure again: https://ci.tvm.ai/blue/organizations/jenkins/tvm/detail/PR-5282/9/pipeline/152...

[GitHub] [incubator-tvm] mehrdadhe commented on a change in pull request #5293: [BUGFIX] Fix CRT static test bug

mehrdadhe commented on a change in pull request #5293: [BUGFIX] Fix CRT static test bug URL: https://github.com/apache/incubator-tvm/pull/5293#discussion_r406343537 ## File path: apps/bundle_deploy/demo_static.c ## @@ -103,6 +104,6 @@ int main(int argc, char **argv) {

[GitHub] [incubator-tvm] mehrdadhe commented on a change in pull request #5293: [BUGFIX] Fix CRT static test bug

mehrdadhe commented on a change in pull request #5293: [BUGFIX] Fix CRT static test bug URL: https://github.com/apache/incubator-tvm/pull/5293#discussion_r406343537 ## File path: apps/bundle_deploy/demo_static.c ## @@ -103,6 +104,6 @@ int main(int argc, char **argv) {

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

comaniac commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#discussion_r406401569 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -856,6

[GitHub] [incubator-tvm] shoubhik commented on issue #5230: Adding support for TFLite QnnSubtract operator.

shoubhik commented on issue #5230: Adding support for TFLite QnnSubtract operator. URL: https://github.com/apache/incubator-tvm/pull/5230#issuecomment-611724091 > > Correct. So here, TFLite converts Conv to depthwise conv, where depthwise conv has input_channels = groups = 1 and depth

[GitHub] [incubator-tvm] mehrdadhe commented on a change in pull request #5293: [BUGFIX] Fix CRT static test bug

mehrdadhe commented on a change in pull request #5293: [BUGFIX] Fix CRT static test bug URL: https://github.com/apache/incubator-tvm/pull/5293#discussion_r406347256 ## File path: apps/bundle_deploy/test_static.c ## @@ -118,6 +120,6 @@ int main(int argc, char **argv) {

[GitHub] [incubator-tvm] anijain2305 opened a new pull request #5296: Legalize - Use Non-recursive Rewriter.

anijain2305 opened a new pull request #5296: Legalize - Use Non-recursive Rewriter. URL: https://github.com/apache/incubator-tvm/pull/5296 One of my pre-quantized model parsing was seeing segfault because of stack overflow. Switching to non-recursive mutation resolves the issue, and I can

[GitHub] [incubator-tvm] alexwong commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

alexwong commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#discussion_r406382829 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -856,6

[GitHub] [incubator-tvm] alexbooth commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

alexbooth commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#discussion_r406382408 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -856,6

[GitHub] [incubator-tvm] tqchen closed pull request #5292: [CI][DOCS] Make sure to refresh the cython part

tqchen closed pull request #5292: [CI][DOCS] Make sure to refresh the cython part URL: https://github.com/apache/incubator-tvm/pull/5292 This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [incubator-tvm] tqchen edited a comment on issue #5292: [CI][DOCS] Make sure to refresh the cython part

tqchen edited a comment on issue #5292: [CI][DOCS] Make sure to refresh the cython part URL: https://github.com/apache/incubator-tvm/pull/5292#issuecomment-611642942 I will fold this fix to #5293 given both might cause flaky errors (and we need to fix both to consistently pass ci)

[GitHub] [incubator-tvm] tqchen commented on issue #5292: [CI][DOCS] Make sure to refresh the cython part

tqchen commented on issue #5292: [CI][DOCS] Make sure to refresh the cython part URL: https://github.com/apache/incubator-tvm/pull/5292#issuecomment-611642942 I will fold this fix to #5293 given both might cause flaky errors

[GitHub] [incubator-tvm] zhiics commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

zhiics commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#discussion_r406404843 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -856,6

[GitHub] [incubator-tvm] tqchen merged pull request #5293: [BUGFIX] Fix CRT static test bug

tqchen merged pull request #5293: [BUGFIX] Fix CRT static test bug URL: https://github.com/apache/incubator-tvm/pull/5293 This is an automated message from the Apache Git Service. To respond to the message, please log on to

[incubator-tvm] branch master updated (ea06388 -> 4e05b47)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git. from ea06388 [BUGFIX][IR] Fix String SEqual (#5275) add 4e05b47 [BUGFIX] Fix CRT static test bug (#5293)

[GitHub] [incubator-tvm] tqchen commented on issue #5293: [BUGFIX] Fix CRT static test bug

tqchen commented on issue #5293: [BUGFIX] Fix CRT static test bug URL: https://github.com/apache/incubator-tvm/pull/5293#issuecomment-611720398 Thanks @masahi @zhiics @mehrdadhe @tmoreau89 this pr is now merged This is an

[GitHub] [incubator-tvm] yzhliu commented on a change in pull request #5171: [Arith] linear system and equation solver

yzhliu commented on a change in pull request #5171: [Arith] linear system and equation solver URL: https://github.com/apache/incubator-tvm/pull/5171#discussion_r406543797 ## File path: python/tvm/arith/int_solver.py ## @@ -0,0 +1,99 @@ +# Licensed to the Apache Software

[GitHub] [incubator-tvm] yzhliu commented on a change in pull request #5171: [Arith] linear system and equation solver

yzhliu commented on a change in pull request #5171: [Arith] linear system and equation solver URL: https://github.com/apache/incubator-tvm/pull/5171#discussion_r406543887 ## File path: tests/python/unittest/test_arith_solve_linear_system.py ## @@ -0,0 +1,224 @@ +#

[GitHub] [incubator-tvm] masahi edited a comment on issue #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

masahi edited a comment on issue #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#issuecomment-611758714 > Although I still feel this is a bit ad-hoc, it should be fine for now as we only use DNNL as an example to

[GitHub] [incubator-tvm] anijain2305 commented on a change in pull request #5296: Legalize - Use Non-recursive Rewriter.

anijain2305 commented on a change in pull request #5296: Legalize - Use Non-recursive Rewriter. URL: https://github.com/apache/incubator-tvm/pull/5296#discussion_r406507469 ## File path: src/relay/transforms/legalize.cc ## @@ -35,14 +35,14 @@ namespace legalize { //

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406523632 ## File path: src/relay/backend/vm/compiler.cc ## @@ -924,19 +924,20 @@ IRModule VMCompiler::OptimizeModule(const

[GitHub] [incubator-tvm] tqchen commented on issue #4312: [TOPI][Relay][OP] Dynamic NMS and strided_slice

tqchen commented on issue #4312: [TOPI][Relay][OP] Dynamic NMS and strided_slice URL: https://github.com/apache/incubator-tvm/pull/4312#issuecomment-611837123 any updated on this? This is an automated message from the Apache

[GitHub] [incubator-tvm] masahi commented on issue #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

masahi commented on issue #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#issuecomment-611758714 > Although I still feel this is a bit ad-hoc, it should be fine for now as we only use DNNL as an example to

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406501685 ## File path: src/relay/backend/vm/compiler.cc ## @@ -924,19 +924,20 @@ IRModule VMCompiler::OptimizeModule(const

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406501685 ## File path: src/relay/backend/vm/compiler.cc ## @@ -924,19 +924,20 @@ IRModule VMCompiler::OptimizeModule(const

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406501323 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -862,7 +863,7 @@ def expected():

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

comaniac commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406504773 ## File path: src/relay/backend/vm/compiler.cc ## @@ -924,19 +924,20 @@ IRModule VMCompiler::OptimizeModule(const

[GitHub] [incubator-tvm] apivovarov commented on issue #2858: [Relay/TOPI][OP] Add clip and wrap mode support in take

apivovarov commented on issue #2858: [Relay/TOPI][OP] Add clip and wrap mode support in take URL: https://github.com/apache/incubator-tvm/pull/2858#issuecomment-611849163 I prepared MXNet Hybrid block which uses `F.take(x, [1,2], axis=1)`. MXNet and TVM outputs are different. Script

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

masahi commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#discussion_r406487275 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -856,6

[GitHub] [incubator-tvm] FrozenGene commented on issue #5296: Legalize - Use Non-recursive Rewriter.

FrozenGene commented on issue #5296: Legalize - Use Non-recursive Rewriter. URL: https://github.com/apache/incubator-tvm/pull/5296#issuecomment-611850686 @LiangHao151941 I think you could benefit this in your pr https://github.com/apache/incubator-tvm/pull/4828

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5296: Legalize - Use Non-recursive Rewriter.

masahi commented on a change in pull request #5296: Legalize - Use Non-recursive Rewriter. URL: https://github.com/apache/incubator-tvm/pull/5296#discussion_r406489234 ## File path: src/relay/transforms/legalize.cc ## @@ -35,14 +35,14 @@ namespace legalize { // Call

[GitHub] [incubator-tvm] tqchen opened a new issue #5298: [CRT][FLAKY] Look into flaky failures on 32bit platforms

tqchen opened a new issue #5298: [CRT][FLAKY] Look into flaky failures on 32bit platforms URL: https://github.com/apache/incubator-tvm/issues/5298 We have some flaky tests for the CRT under the 32bit platforms(even after https://github.com/apache/incubator-tvm/pull/5293) , see example log

[GitHub] [incubator-tvm] alexbooth commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op

alexbooth commented on a change in pull request #5272: [BYOC] Add example of Composite + Annotate for DNNL fused op URL: https://github.com/apache/incubator-tvm/pull/5272#discussion_r406493347 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -856,6

[GitHub] [incubator-tvm] tqchen merged pull request #5190: Create loops according to storage scope and thread hierarchies

tqchen merged pull request #5190: Create loops according to storage scope and thread hierarchies URL: https://github.com/apache/incubator-tvm/pull/5190 This is an automated message from the Apache Git Service. To respond to

[incubator-tvm] branch master updated: Create loops according to storage scope and thread hierarchies (#5190)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git The following commit(s) were added to refs/heads/master by this push: new 3d09e64 Create loops according to

[GitHub] [incubator-tvm] redpanda3 commented on issue #4863: vta+nvdla integration

redpanda3 commented on issue #4863: vta+nvdla integration URL: https://github.com/apache/incubator-tvm/pull/4863#issuecomment-611842131 OK, recently a little bit busy with my cora project. This is an automated message from

[GitHub] [incubator-tvm] tqchen commented on issue #5282: [LLVM] Fix generation of LLVM intrinsics

tqchen commented on issue #5282: [LLVM] Fix generation of LLVM intrinsics URL: https://github.com/apache/incubator-tvm/pull/5282#issuecomment-611767350 OK https://github.com/apache/incubator-tvm/pull/5297 temporary disabled the crt test

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

comaniac commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406503687 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -862,7 +863,7 @@ def expected():

[GitHub] [incubator-tvm] tqchen merged pull request #5297: [CI] Temporary disable CRT test

tqchen merged pull request #5297: [CI] Temporary disable CRT test URL: https://github.com/apache/incubator-tvm/pull/5297 This is an automated message from the Apache Git Service. To respond to the message, please log on to

[incubator-tvm] branch master updated (4e05b47 -> a4321e0)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git. from 4e05b47 [BUGFIX] Fix CRT static test bug (#5293) add a4321e0 [CI] Temporary disable CRT test (#5297)

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406502991 ## File path: src/relay/backend/contrib/dnnl/codegen.cc ## @@ -53,12 +53,19 @@ class CodegenDNNL : public

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

comaniac commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406505777 ## File path: src/relay/backend/contrib/dnnl/codegen.cc ## @@ -53,12 +53,19 @@ class CodegenDNNL : public

[incubator-tvm] branch master updated (4e05b47 -> a4321e0)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git. from 4e05b47 [BUGFIX] Fix CRT static test bug (#5293) add a4321e0 [CI] Temporary disable CRT test (#5297)

[GitHub] [incubator-tvm] tqchen commented on issue #5219: [WIP] Support int64 indices

tqchen commented on issue #5219: [WIP] Support int64 indices URL: https://github.com/apache/incubator-tvm/pull/5219#issuecomment-611836397 @kazum let us know if it is ready for review This is an automated message from the

[GitHub] [incubator-tvm] tqchen commented on issue #5190: Create loops according to storage scope and thread hierarchies

tqchen commented on issue #5190: Create loops according to storage scope and thread hierarchies URL: https://github.com/apache/incubator-tvm/pull/5190#issuecomment-611835958 Thanks @yongfeng-nv ! this PR is now merged This

[GitHub] [incubator-tvm] tqchen commented on issue #4857: Windows Support for cpp_rpc

tqchen commented on issue #4857: Windows Support for cpp_rpc URL: https://github.com/apache/incubator-tvm/pull/4857#issuecomment-611839443 @jmorrill @FrozenGene it would be great if we can followup :) This is an automated

[GitHub] [incubator-tvm] llehtahw commented on a change in pull request #5226: [CODEGEN][CUDA] Fix vector load

llehtahw commented on a change in pull request #5226: [CODEGEN][CUDA] Fix vector load URL: https://github.com/apache/incubator-tvm/pull/5226#discussion_r406589444 ## File path: src/target/source/codegen_cuda.cc ## @@ -591,13 +591,17 @@ void CodeGenCUDA::VisitExpr_(const

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5243: [Frontend][TensorFlow]Improve TensorFlow Static Shape Tensor Array

masahi commented on a change in pull request #5243: [Frontend][TensorFlow]Improve TensorFlow Static Shape Tensor Array URL: https://github.com/apache/incubator-tvm/pull/5243#discussion_r406498520 ## File path: python/tvm/relay/frontend/common.py ## @@ -483,6 +481,17 @@

[GitHub] [incubator-tvm] zhiics commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

zhiics commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406524552 ## File path: src/relay/backend/vm/compiler.cc ## @@ -924,19 +924,20 @@ IRModule VMCompiler::OptimizeModule(const

[GitHub] [incubator-tvm] yzhliu opened a new pull request #5300: [Node] Provide guide to user who has difficulty register SEqualReduce

yzhliu opened a new pull request #5300: [Node] Provide guide to user who has difficulty register SEqualReduce URL: https://github.com/apache/incubator-tvm/pull/5300 @tqchen This is an automated message from the Apache

[GitHub] [incubator-tvm] tqchen merged pull request #5180: [TENSORFLOW]reduce ops updated

tqchen merged pull request #5180: [TENSORFLOW]reduce ops updated URL: https://github.com/apache/incubator-tvm/pull/5180 This is an automated message from the Apache Git Service. To respond to the message, please log on to

[incubator-tvm] branch master updated: [TENSORFLOW]reduce ops updated (#5180)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git The following commit(s) were added to refs/heads/master by this push: new 99c4f9d [TENSORFLOW]reduce ops updated

[GitHub] [incubator-tvm] hlu1 opened a new pull request #5301: [NDArray] Set NDArray::Container.shape_ in NDArray::FromDLPack

hlu1 opened a new pull request #5301: [NDArray] Set NDArray::Container.shape_ in NDArray::FromDLPack URL: https://github.com/apache/incubator-tvm/pull/5301 In some cases, the shape info in DLTensor.shape might not be dynamically allocated and may cease to exist after passing the DLTensor

[GitHub] [incubator-tvm] tqchen commented on issue #5296: Legalize - Use Non-recursive Rewriter.

tqchen commented on issue #5296: Legalize - Use Non-recursive Rewriter. URL: https://github.com/apache/incubator-tvm/pull/5296#issuecomment-611760427 @mbrookhart can you help to review the PR This is an automated message from

[GitHub] [incubator-tvm] tqchen opened a new pull request #5297: [CI] Temporary disable CRT test

tqchen opened a new pull request #5297: [CI] Temporary disable CRT test URL: https://github.com/apache/incubator-tvm/pull/5297 Due to problems under the 32bit platform, need more investigation cc @liangfu @mehrdadhe

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

masahi commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406501323 ## File path: tests/python/relay/test_pass_partition_graph.py ## @@ -862,7 +863,7 @@ def expected():

[GitHub] [incubator-tvm] zhiics commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen

zhiics commented on a change in pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288#discussion_r406505427 ## File path: src/relay/backend/vm/compiler.cc ## @@ -924,19 +924,20 @@ IRModule VMCompiler::OptimizeModule(const

[GitHub] [incubator-tvm] tqchen opened a new pull request #5299: [NODE] General serialzation of leaf objects into bytes.

tqchen opened a new pull request #5299: [NODE] General serialzation of leaf objects into bytes. URL: https://github.com/apache/incubator-tvm/pull/5299 This PR refactors the serialization mechanism to support general serialization of leaf objects into bytes. The new feature

[GitHub] [incubator-tvm] tqchen commented on issue #5299: [NODE] General serialzation of leaf objects into bytes.

tqchen commented on issue #5299: [NODE] General serialzation of leaf objects into bytes. URL: https://github.com/apache/incubator-tvm/pull/5299#issuecomment-611788162 cc @zhiics @yzhliu @ZihengJiang @wweic This is an

[GitHub] [incubator-tvm] tqchen commented on issue #4863: vta+nvdla integration

tqchen commented on issue #4863: vta+nvdla integration URL: https://github.com/apache/incubator-tvm/pull/4863#issuecomment-611839115 closing for now as the hardware related parts are moved to https://github.com/apache/incubator-tvm-vta

[GitHub] [incubator-tvm] tqchen closed pull request #4863: vta+nvdla integration

tqchen closed pull request #4863: vta+nvdla integration URL: https://github.com/apache/incubator-tvm/pull/4863 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and

[GitHub] [incubator-tvm] tqchen commented on issue #5289: add tensorflow cumsum

tqchen commented on issue #5289: add tensorflow cumsum URL: https://github.com/apache/incubator-tvm/pull/5289#issuecomment-611838733 THanks @adobay please fix the CI error This is an automated message from the Apache Git

[GitHub] [incubator-tvm] zhiics merged pull request #5299: [NODE] General serialzation of leaf objects into bytes.

zhiics merged pull request #5299: [NODE] General serialzation of leaf objects into bytes. URL: https://github.com/apache/incubator-tvm/pull/5299 This is an automated message from the Apache Git Service. To respond to the

[GitHub] [incubator-tvm] tqchen commented on issue #5296: Legalize - Use Non-recursive Rewriter.

tqchen commented on issue #5296: Legalize - Use Non-recursive Rewriter. URL: https://github.com/apache/incubator-tvm/pull/5296#issuecomment-611867101 Thanks @FrozenGene @anijain2305 @mbrookhart @tmoreau89 This is an

[GitHub] [incubator-tvm] anijain2305 commented on issue #5230: Adding support for TFLite QnnSubtract operator.

anijain2305 commented on issue #5230: Adding support for TFLite QnnSubtract operator. URL: https://github.com/apache/incubator-tvm/pull/5230#issuecomment-611885516 Thanks @shoubhik @FrozenGene This is merged! This is an

[GitHub] [incubator-tvm] anijain2305 merged pull request #5230: Adding support for TFLite QnnSubtract operator.

anijain2305 merged pull request #5230: Adding support for TFLite QnnSubtract operator. URL: https://github.com/apache/incubator-tvm/pull/5230 This is an automated message from the Apache Git Service. To respond to the

[incubator-tvm] branch master updated: Adding support for TFLite QnnSub operator. (#5230)

This is an automated email from the ASF dual-hosted git repository. anijain2305 pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git The following commit(s) were added to refs/heads/master by this push: new 6ecfaaf Adding support for TFLite

[incubator-tvm] branch master updated: Legalize - Use Non-recursive Rewriter. (#5296)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git The following commit(s) were added to refs/heads/master by this push: new 7d670b0 Legalize - Use Non-recursive

[GitHub] [incubator-tvm] tqchen commented on issue #5301: [NDArray] Set NDArray::Container.shape_ in NDArray::FromDLPack

tqchen commented on issue #5301: [NDArray] Set NDArray::Container.shape_ in NDArray::FromDLPack URL: https://github.com/apache/incubator-tvm/pull/5301#issuecomment-611866774 Good catch, thanks @hlu1 ! This is an automated

[GitHub] [incubator-tvm] FrozenGene commented on issue #5230: Adding support for TFLite QnnSubtract operator.

FrozenGene commented on issue #5230: Adding support for TFLite QnnSubtract operator. URL: https://github.com/apache/incubator-tvm/pull/5230#issuecomment-611884419 > > > Correct. So here, TFLite converts Conv to depthwise conv, where depthwise conv has input_channels = groups = 1 and depth

[GitHub] [incubator-tvm] jwfromm commented on a change in pull request #5284: [Topi] Tensorcore support for Conv3D

jwfromm commented on a change in pull request #5284: [Topi] Tensorcore support for Conv3D URL: https://github.com/apache/incubator-tvm/pull/5284#discussion_r406608704 ## File path: python/tvm/relay/op/strategy/cuda.py ## @@ -268,11 +268,24 @@ def

[GitHub] [incubator-tvm] tqchen merged pull request #5296: Legalize - Use Non-recursive Rewriter.

tqchen merged pull request #5296: Legalize - Use Non-recursive Rewriter. URL: https://github.com/apache/incubator-tvm/pull/5296 This is an automated message from the Apache Git Service. To respond to the message, please log

[GitHub] [incubator-tvm] tqchen edited a comment on issue #5301: [NDArray] Set NDArray::Container.shape_ in NDArray::FromDLPack

tqchen edited a comment on issue #5301: [NDArray] Set NDArray::Container.shape_ in NDArray::FromDLPack URL: https://github.com/apache/incubator-tvm/pull/5301#issuecomment-611866774 THanks @hlu1 however, according to the DLPack conventionm, the DLManagedTensor won't be deleted until the

[incubator-tvm] branch master updated (99c4f9d -> 2b96820)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git. from 99c4f9d [TENSORFLOW]reduce ops updated (#5180) add 2b96820 [Node] Provide guide to user who has

[GitHub] [incubator-tvm] tqchen merged pull request #5300: [Node] Provide guide to user who has difficulty register SEqualReduce

tqchen merged pull request #5300: [Node] Provide guide to user who has difficulty register SEqualReduce URL: https://github.com/apache/incubator-tvm/pull/5300 This is an automated message from the Apache Git Service. To

[GitHub] [incubator-tvm] adobay opened a new pull request #5287: add tensorflow cumsum

adobay opened a new pull request #5287: add tensorflow cumsum URL: https://github.com/apache/incubator-tvm/pull/5287 Thanks for contributing to TVM! Please refer to guideline https://tvm.apache.org/docs/contribute/ for useful information and tips. After the pull request is submitted,

[GitHub] [incubator-tvm] comaniac opened a new pull request #5288: [BYOC] Refine DNNL Codegen

comaniac opened a new pull request #5288: [BYOC] Refine DNNL Codegen URL: https://github.com/apache/incubator-tvm/pull/5288 With multiple output and merge compiler region support, we can now refine the DNNL codegen to support ops with multiple outputs such as `batch_norm`. This PR:

[GitHub] [incubator-tvm] adobay opened a new pull request #5289: add tensorflow cumsum

adobay opened a new pull request #5289: add tensorflow cumsum URL: https://github.com/apache/incubator-tvm/pull/5289 Thanks for contributing to TVM! Please refer to guideline https://tvm.apache.org/docs/contribute/ for useful information and tips. After the pull request is submitted,

[GitHub] [incubator-tvm] adobay closed pull request #5287: add tensorflow cumsum

adobay closed pull request #5287: add tensorflow cumsum URL: https://github.com/apache/incubator-tvm/pull/5287 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and

[GitHub] [incubator-tvm] FrozenGene commented on issue #5284: [Topi] Tensorcore support for Conv3D

FrozenGene commented on issue #5284: [Topi] Tensorcore support for Conv3D URL: https://github.com/apache/incubator-tvm/pull/5284#issuecomment-611427467 @yangjunpro @minminsun Please also have a look, because you have done tensorcore support ever.

[GitHub] [incubator-tvm] Shawn-Inspur commented on a change in pull request #5284: [Topi] Tensorcore support for Conv3D

Shawn-Inspur commented on a change in pull request #5284: [Topi] Tensorcore support for Conv3D URL: https://github.com/apache/incubator-tvm/pull/5284#discussion_r406157545 ## File path: python/tvm/relay/op/strategy/cuda.py ## @@ -268,11 +268,24 @@ def

[GitHub] [incubator-tvm] tqchen opened a new pull request #5292: [CI][DOCS] Make sure to refresh the cython part

tqchen opened a new pull request #5292: [CI][DOCS] Make sure to refresh the cython part URL: https://github.com/apache/incubator-tvm/pull/5292 https://github.com/apache/incubator-tvm/issues/5285 This is an automated message

[GitHub] [incubator-tvm] tqchen edited a comment on issue #5292: [CI][DOCS] Make sure to refresh the cython part

tqchen edited a comment on issue #5292: [CI][DOCS] Make sure to refresh the cython part URL: https://github.com/apache/incubator-tvm/pull/5292#issuecomment-611616205 cc @zhiics @yzhliu @ziheng @masahi This is an automated

[GitHub] [incubator-tvm] tqchen commented on issue #5285: [TEST][FLAKY] Sphinx pre-check build error

tqchen commented on issue #5285: [TEST][FLAKY] Sphinx pre-check build error URL: https://github.com/apache/incubator-tvm/issues/5285#issuecomment-611616607 Should be fixed by https://github.com/apache/incubator-tvm/pull/5292 Thanks @zhiics for looking into it

[GitHub] [incubator-tvm] tqchen opened a new pull request #5293: [BUGFIX] Fix CRT static test bug

tqchen opened a new pull request #5293: [BUGFIX] Fix CRT static test bug URL: https://github.com/apache/incubator-tvm/pull/5293 Fixed a segfault caused by wrongly assigning the address of the array pointer to the shape instead of the array. This problem was revealed by a compiler warning

[GitHub] [incubator-tvm] tqchen commented on issue #5293: [BUGFIX] Fix CRT static test bug

tqchen commented on issue #5293: [BUGFIX] Fix CRT static test bug URL: https://github.com/apache/incubator-tvm/pull/5293#issuecomment-611626829 cc @tmoreau89 @mehrdadhe @masahi @zhiics This is an automated message from the

[GitHub] [incubator-tvm] tqchen opened a new issue #5294: [CRT] Clear Compiler Warnings

tqchen opened a new issue #5294: [CRT] Clear Compiler Warnings URL: https://github.com/apache/incubator-tvm/issues/5294 During the investigation of https://github.com/apache/incubator-tvm/pull/5293, we find that the CRT source currently generate multiple warnings in both 64 bit and 32 bit

  1   2   >