srkreddy1238 commented on pull request #5617:
URL: https://github.com/apache/incubator-tvm/pull/5617#issuecomment-638617640
Thanks @deepakbabel23. PartionedCall support is very helpful with many
models from TF Hub.
This is
icemelon9 commented on a change in pull request #4312:
URL: https://github.com/apache/incubator-tvm/pull/4312#discussion_r43500
##
File path: src/relay/transforms/combine_parallel_conv2d.cc
##
@@ -164,19 +164,40 @@ class ParallelConv2DCombiner : public ParallelOpCombiner {
kevinthesun commented on a change in pull request #4312:
URL: https://github.com/apache/incubator-tvm/pull/4312#discussion_r435002980
##
File path: include/tvm/relay/attrs/transform.h
##
@@ -210,14 +210,22 @@ struct SplitAttrs : public tvm::AttrsNode {
/*! \brief Attributes
icemelon9 commented on a change in pull request #4312:
URL: https://github.com/apache/incubator-tvm/pull/4312#discussion_r434997152
##
File path: include/tvm/relay/attrs/transform.h
##
@@ -210,14 +210,22 @@ struct SplitAttrs : public tvm::AttrsNode {
/*! \brief Attributes
deepakbabel23 commented on pull request #5617:
URL: https://github.com/apache/incubator-tvm/pull/5617#issuecomment-638603018
Thank you Siva.
This is an automated message from the Apache Git Service.
To respond to the
zhiics commented on a change in pull request #4312:
URL: https://github.com/apache/incubator-tvm/pull/4312#discussion_r434987705
##
File path: src/relay/transforms/combine_parallel_conv2d.cc
##
@@ -164,19 +164,40 @@ class ParallelConv2DCombiner : public ParallelOpCombiner {
yongwww commented on a change in pull request #5699:
URL: https://github.com/apache/incubator-tvm/pull/5699#discussion_r434980636
##
File path: tests/python/frontend/tensorflow/test_forward.py
##
@@ -2238,6 +2258,40 @@ def test_forward_resnetv2():
srkreddy1238 merged pull request #5617:
URL: https://github.com/apache/incubator-tvm/pull/5617
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
This is an automated email from the ASF dual-hosted git repository.
srk pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 3d61dc8 [ONNX]ReduceL1, ReduceL2, ReduceSumSquare, ReduceLogSum ops
added (#5721)
add 43dcbc6
tqchen edited a comment on pull request #5601:
URL: https://github.com/apache/incubator-tvm/pull/5601#issuecomment-638583235
So to summarize, we will need to introduce two kinds of TypeCode here. See
the data type code defined here
tqchen commented on pull request #5601:
URL: https://github.com/apache/incubator-tvm/pull/5601#issuecomment-638583235
See the data type code defined here
https://github.com/apache/incubator-tvm/blob/master/include/tvm/runtime/data_type.h#L48
tqchen commented on pull request #5601:
URL: https://github.com/apache/incubator-tvm/pull/5601#issuecomment-638582535
I want to clarify the case of the TypeCode. Given that we are not trying to
making a consistent type code for arguments beyond kOpaqueHandle.
So in this case, we
tqchen commented on a change in pull request #5601:
URL: https://github.com/apache/incubator-tvm/pull/5601#discussion_r434972874
##
File path: include/tvm/runtime/data_type.h
##
@@ -372,7 +372,7 @@ inline DLDataType String2DLDataType(std::string s) {
t.lanes = 1;
yongwww commented on a change in pull request #4312:
URL: https://github.com/apache/incubator-tvm/pull/4312#discussion_r434968182
##
File path: include/tvm/relay/attrs/transform.h
##
@@ -210,14 +210,22 @@ struct SplitAttrs : public tvm::AttrsNode {
/*! \brief Attributes for
Menooker commented on a change in pull request #5601:
URL: https://github.com/apache/incubator-tvm/pull/5601#discussion_r434955796
##
File path: python/tvm/_ffi/runtime_ctypes.py
##
@@ -58,7 +58,8 @@ class DataType(ctypes.Structure):
0 : 'int',
1 : 'uint',
Menooker commented on a change in pull request #5601:
URL: https://github.com/apache/incubator-tvm/pull/5601#discussion_r434955653
##
File path: python/tvm/_ffi/runtime_ctypes.py
##
@@ -58,7 +58,8 @@ class DataType(ctypes.Structure):
0 : 'int',
1 : 'uint',
masahi commented on pull request #5721:
URL: https://github.com/apache/incubator-tvm/pull/5721#issuecomment-638557408
Thanks @siju-samuel
This is an automated message from the Apache Git Service.
To respond to the message,
This is an automated email from the ASF dual-hosted git repository.
masahi pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 36aa895 Fix generating types like float44 and float88 (#5722)
add 3d61dc8 [ONNX]ReduceL1, ReduceL2,
masahi merged pull request #5721:
URL: https://github.com/apache/incubator-tvm/pull/5721
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
tqchen commented on pull request #5723:
URL: https://github.com/apache/incubator-tvm/pull/5723#issuecomment-638536740
@abergeron it would be great to understand where does the error comes from.
In this case, I guess that it comes from the near 1 rounding, we should do
an amendment
kevinthesun commented on a change in pull request #4312:
URL: https://github.com/apache/incubator-tvm/pull/4312#discussion_r434927495
##
File path: include/tvm/relay/attrs/transform.h
##
@@ -210,14 +210,22 @@ struct SplitAttrs : public tvm::AttrsNode {
/*! \brief Attributes
icemelon9 commented on a change in pull request #4312:
URL: https://github.com/apache/incubator-tvm/pull/4312#discussion_r434920515
##
File path: include/tvm/relay/attrs/transform.h
##
@@ -210,14 +210,22 @@ struct SplitAttrs : public tvm::AttrsNode {
/*! \brief Attributes
tqchen merged pull request #5722:
URL: https://github.com/apache/incubator-tvm/pull/5722
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
tqchen closed issue #5704:
URL: https://github.com/apache/incubator-tvm/issues/5704
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to
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 36aa895 Fix generating types like
tqchen commented on issue #5710:
URL: https://github.com/apache/incubator-tvm/issues/5710#issuecomment-638470243
fixed by #5722 thanks to @abergeron
This is an automated message from the Apache Git Service.
To respond to
tqchen closed issue #5710:
URL: https://github.com/apache/incubator-tvm/issues/5710
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to
tqchen commented on pull request #5716:
URL: https://github.com/apache/incubator-tvm/pull/5716#issuecomment-638447821
@mbrookhart @Huyuwei can you help to review the PR
This is an automated message from the Apache Git
tqchen commented on pull request #5722:
URL: https://github.com/apache/incubator-tvm/pull/5722#issuecomment-638448214
Thanks @abergeron !
This is an automated message from the Apache Git Service.
To respond to the message,
tqchen commented on pull request #5719:
URL: https://github.com/apache/incubator-tvm/pull/5719#issuecomment-638447450
Thanks @junrushao1994 !
This is an automated message from the Apache Git Service.
To respond to the
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 64ac555 [Object] Restore the StrMap
tqchen merged pull request #5719:
URL: https://github.com/apache/incubator-tvm/pull/5719
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
vinx13 opened a new pull request #5724:
URL: https://github.com/apache/incubator-tvm/pull/5724
We need to clear compiler engine cache after task extraction, otherwise
after tuning, we have the compilation result for task extraction in the cache
and we won't utilize the latest configs from
abergeron opened a new pull request #5723:
URL: https://github.com/apache/incubator-tvm/pull/5723
This fails more than 99% of the time on my machine with metal.
I've tested that seed on a couple of machines with OpenCL and CUDA to make
sure they still work and the seem to do.
junrushao1994 commented on a change in pull request #5719:
URL: https://github.com/apache/incubator-tvm/pull/5719#discussion_r434785865
##
File path: include/tvm/node/container.h
##
@@ -96,8 +98,7 @@ class MapNode : public Object {
* \tparam V The value NodeRef type.
*/
inadob opened a new pull request #4805:
URL: https://github.com/apache/incubator-tvm/pull/4805
**This patch is made on top of #4789.**
* Add helper functions in the parser to quantize and dequantize with Relay
* Add qnn tests for all 4 activation functions
kevinthesun commented on issue #4118:
URL: https://github.com/apache/incubator-tvm/issues/4118#issuecomment-638382843
@cloudhan Thanks for your info. @icemelon9 Do we have any work related to
dynamic axis range?
In terms of codegen, indeed efficiency(and also how to limit the
tqchen commented on a change in pull request #5719:
URL: https://github.com/apache/incubator-tvm/pull/5719#discussion_r434767483
##
File path: include/tvm/node/container.h
##
@@ -96,8 +98,7 @@ class MapNode : public Object {
* \tparam V The value NodeRef type.
*/
template
abergeron opened a new pull request #5722:
URL: https://github.com/apache/incubator-tvm/pull/5722
This fixes #5704 and #5710.
The existing tests already fail before this patch so I did not add any new
tests.
This is
yongwww commented on a change in pull request #4312:
URL: https://github.com/apache/incubator-tvm/pull/4312#discussion_r434764491
##
File path: src/relay/op/tensor/transform.cc
##
@@ -1772,73 +1788,161 @@ Array>
StridedSliceInferCorrectLayout(const Attrs& attrs,
}
junrushao1994 commented on a change in pull request #5719:
URL: https://github.com/apache/incubator-tvm/pull/5719#discussion_r434754470
##
File path: include/tvm/node/container.h
##
@@ -82,6 +82,21 @@ class MapNode : public Object {
/*! \brief the data content */
inadob closed pull request #4805:
URL: https://github.com/apache/incubator-tvm/pull/4805
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
abergeron commented on issue #5710:
URL: https://github.com/apache/incubator-tvm/issues/5710#issuecomment-638342817
This seems similar to what I've seen in #5704.
This is an automated message from the Apache Git Service.
To
tqchen commented on a change in pull request #5719:
URL: https://github.com/apache/incubator-tvm/pull/5719#discussion_r434726829
##
File path: include/tvm/node/container.h
##
@@ -82,6 +82,21 @@ class MapNode : public Object {
/*! \brief the data content */
ContainerType
wpan11nv commented on pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#issuecomment-638332458
> I'll just work on a fix.
Thanks! Will you fix it or just disable this schedule? Let me know if you
need any help to enable it.
t-vi commented on pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#issuecomment-638329923
I'll just work on a fix.
This is an automated message from the Apache Git Service.
To respond to the
siju-samuel opened a new pull request #5721:
URL: https://github.com/apache/incubator-tvm/pull/5721
- ReduceL1
- ReduceL2
- ReduceSumSquare
- ReduceLogSum
The above onnx frontend op support is added.
@masahi @FrozenGene Could you please help me to review and merge this
wpan11nv commented on a change in pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r434689864
##
File path: src/target/llvm/codegen_llvm.cc
##
@@ -736,7 +736,40 @@ llvm::Function*
CodeGenLLVM::GetIntrinsicDecl(llvm::Intrinsic::ID id,
wpan11nv commented on pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#issuecomment-638298924
> So ROCm uses the CUDA schedule, but warp reductions don't seem to
currently work (so arguably, ROCm would want to be improved). But so before
this PR, one could run
cloudhan edited a comment on issue #4118:
URL: https://github.com/apache/incubator-tvm/issues/4118#issuecomment-638269770
@kevinthesun Any timeframe?
Off topic, I want to mention TensorRT supports dynamic shape from 7.0. To
provide better performance, it supports multiple
cloudhan edited a comment on issue #4118:
URL: https://github.com/apache/incubator-tvm/issues/4118#issuecomment-638269770
@kevinthesun Any timeframe?
Off topic, I want to mention TensorRT supports dynamic shape from 7.0. To
provide better performance, it supports multiple
t-vi edited a comment on pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#issuecomment-638275567
So ROCm uses the CUDA schedule, but warp reductions don't seem to currently
work (so arguably, ROCm would want to be improved). But so before this PR, one
could run
t-vi commented on pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#issuecomment-638275567
So ROCm uses the CUDA schedule, but warp reductions don't seem to currently
work (so arguably, ROCm would want to be improved). But so before this PR, one
could run resnet18
ANSHUMAN87 commented on a change in pull request #5717:
URL: https://github.com/apache/incubator-tvm/pull/5717#discussion_r434657083
##
File path: src/relay/backend/compile_engine.cc
##
@@ -689,7 +689,7 @@ class CompileEngineImpl : public CompileEngineNode {
tqchen commented on a change in pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r434643666
##
File path: topi/python/topi/cuda/softmax.py
##
@@ -39,6 +39,7 @@ def schedule_softmax(outs):
outs = [outs] if isinstance(outs,
tqchen commented on a change in pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r434643666
##
File path: topi/python/topi/cuda/softmax.py
##
@@ -39,6 +39,7 @@ def schedule_softmax(outs):
outs = [outs] if isinstance(outs,
tqchen commented on pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#issuecomment-638261177
@t-vi can you elaborate?
@wpan11nv please also followup. One thing that get missed during review is
that we should move the nvptx related intrinsics into
tqchen commented on a change in pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r434642922
##
File path: src/target/llvm/codegen_llvm.cc
##
@@ -736,7 +736,40 @@ llvm::Function*
CodeGenLLVM::GetIntrinsicDecl(llvm::Intrinsic::ID id,
tqchen commented on a change in pull request #5601:
URL: https://github.com/apache/incubator-tvm/pull/5601#discussion_r434640906
##
File path: python/tvm/_ffi/runtime_ctypes.py
##
@@ -58,7 +58,8 @@ class DataType(ctypes.Structure):
0 : 'int',
1 : 'uint',
tqchen commented on a change in pull request #5717:
URL: https://github.com/apache/incubator-tvm/pull/5717#discussion_r434639667
##
File path: src/relay/backend/compile_engine.cc
##
@@ -689,7 +689,7 @@ class CompileEngineImpl : public CompileEngineNode {
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 927510a Avoid downloading when TOPHUB_LOCATION is NONE (#5720)
add 9151a51 [Object][FFI] Introduce
tqchen merged pull request #5718:
URL: https://github.com/apache/incubator-tvm/pull/5718
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
tqchen merged pull request #5720:
URL: https://github.com/apache/incubator-tvm/pull/5720
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
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 c1f3b2f [MXNET]Softmin, trunc op support added (#5715)
add 927510a Avoid downloading when
zxy844288792 commented on pull request #5720:
URL: https://github.com/apache/incubator-tvm/pull/5720#issuecomment-638080331
> This PR won't affect what #4671 want to achieve, if TOPHUB_LOCATION is not
set(which is default behavior), config will be downloaded if not exsit.
I agree
t-vi commented on pull request #5600:
URL: https://github.com/apache/incubator-tvm/pull/5600#issuecomment-638068589
This broke the ROCm backend.
This is an automated message from the Apache Git Service.
To respond to the
vinx13 commented on issue #5710:
URL: https://github.com/apache/incubator-tvm/issues/5710#issuecomment-638058179
I think printing `t.lanes()` here is unnecessary, a patch is welcomed
This is an automated message from the
This is an automated email from the ASF dual-hosted git repository.
masahi pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 4347b41 [Object] Unify StrMapNode and MapNode (#5687)
add c1f3b2f [MXNET]Softmin, trunc op support
masahi merged pull request #5715:
URL: https://github.com/apache/incubator-tvm/pull/5715
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
lixiaoquan commented on pull request #5720:
URL: https://github.com/apache/incubator-tvm/pull/5720#issuecomment-638034528
This PR won't affect what #4671 want to achieve, if TOPHUB_LOCATION is not
set(which is default behavior), config will be downloaded if not exsit.
comaniac commented on pull request #5720:
URL: https://github.com/apache/incubator-tvm/pull/5720#issuecomment-638020499
I see. The behavior was removed by #4671. Will take a closer look tomorrow
morning.
Also cc @zxy844288792
junrushao1994 commented on a change in pull request #5718:
URL: https://github.com/apache/incubator-tvm/pull/5718#discussion_r434366562
##
File path: src/ir/attrs.cc
##
@@ -37,7 +37,7 @@ void DictAttrsNode::InitByPackedArgs(const runtime::TVMArgs&
args, bool allow_un
lixiaoquan edited a comment on pull request #5720:
URL: https://github.com/apache/incubator-tvm/pull/5720#issuecomment-638014252
> Could you provide the background of making this change?
TOPHUB_LOCATION is introduced by #4015 to disable downloading from tophub,
but behavior of this
lixiaoquan commented on pull request #5720:
URL: https://github.com/apache/incubator-tvm/pull/5720#issuecomment-638014252
> Could you provide the background of making this change?
TOPHUB_LOCATION is introduced by #4015 to disable downloading from tophub,
but behavior of this
comaniac commented on pull request #5720:
URL: https://github.com/apache/incubator-tvm/pull/5720#issuecomment-637998049
Could you provide the background of making this change?
This is an automated message from the Apache Git
lixiaoquan opened a new pull request #5720:
URL: https://github.com/apache/incubator-tvm/pull/5720
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
zhiics commented on a change in pull request #5718:
URL: https://github.com/apache/incubator-tvm/pull/5718#discussion_r434329930
##
File path: src/ir/attrs.cc
##
@@ -37,7 +37,7 @@ void DictAttrsNode::InitByPackedArgs(const runtime::TVMArgs&
args, bool allow_un
77 matches
Mail list logo