comaniac commented on a change in pull request #4899: [Relay][Frontend][Keras]
NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#discussion_r380301754
##
File path: python/tvm/relay/frontend/keras.py
##
@@ -751,6 +816,10 @@ def from_keras(model,
comaniac commented on a change in pull request #4899: [Relay][Frontend][Keras]
NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#discussion_r380300117
##
File path: python/tvm/relay/frontend/keras.py
##
@@ -274,8 +287,13 @@ def
comaniac commented on a change in pull request #4899: [Relay][Frontend][Keras]
NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#discussion_r380300684
##
File path: python/tvm/relay/frontend/keras.py
##
@@ -322,26 +353,39 @@ def
comaniac commented on a change in pull request #4899: [Relay][Frontend][Keras]
NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#discussion_r380299492
##
File path: python/tvm/relay/frontend/keras.py
##
@@ -234,18 +234,29 @@ def
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380296359
##
File path: topi/python/topi/arm_cpu/__init__.py
##
@@ -14,13 +14,14 @@
# KIND, either express or
tqchen merged pull request #4790: Fast exponent
URL: https://github.com/apache/incubator-tvm/pull/4790
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
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 a43e326 Update faq.md (#4893)
add 1314091 Fast exponent (#4790)
No new revisions were added by this
tqchen commented on issue #4790: Fast exponent
URL: https://github.com/apache/incubator-tvm/pull/4790#issuecomment-587092625
Thanks @alexgl-github @anijain2305 @masahi @FrozenGene !
This is an automated message from the
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380301022
##
File path: topi/python/topi/arm_cpu/conv2d_alter_op.py
##
@@ -0,0 +1,171 @@
+# Licensed to the Apache
yzhliu commented on a change in pull request #4644: [Relay][AutoTVM] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380008720
##
File path: topi/python/topi/arm_cpu/__init__.py
##
@@ -14,13 +14,14 @@
# KIND, either express or
yzhliu commented on a change in pull request #4644: [Relay][AutoTVM] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380275069
##
File path: topi/python/topi/cuda/conv3d.py
##
@@ -126,24 +78,55 @@ def schedule_conv3d_ncdhw_cuda(cfg,
yzhliu commented on a change in pull request #4644: [Relay][AutoTVM] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380255632
##
File path: topi/python/topi/arm_cpu/conv2d_alter_op.py
##
@@ -0,0 +1,171 @@
+# Licensed to the Apache
tqchen commented on a change in pull request #4891: [TEST][FLAKY]
topi/tests/python/test_topi_sort.py::test_argsort
URL: https://github.com/apache/incubator-tvm/pull/4891#discussion_r380298254
##
File path: topi/tests/python/test_topi_sort.py
##
@@ -27,6 +27,12 @@ def
cchung100m commented on issue #4891: [TEST][FLAKY]
topi/tests/python/test_topi_sort.py::test_argsort
URL: https://github.com/apache/incubator-tvm/pull/4891#issuecomment-587051868
Hi @tqchen
Thanks for the prompt reply and explanation.
I shuffle index and get data from the
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380302988
##
File path: topi/python/topi/cuda/conv3d.py
##
@@ -126,24 +78,55 @@ def schedule_conv3d_ncdhw_cuda(cfg,
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380295907
##
File path: src/relay/op/nn/convolution.h
##
@@ -153,6 +153,16 @@ bool Conv2DRel(const Array& types,
tqchen commented on issue #4546: [CODEGEN] Support cuda tensorcore subbyte int
data type in auto tensorcore
URL: https://github.com/apache/incubator-tvm/pull/4546#issuecomment-587091708
see if you can reprod the error locally. I wonder if it has things to do
with the join running of the
tqchen commented on issue #4884: [android_deploy] CRASH caused by `Module.load`
func while running App on Android Device with
URL: https://github.com/apache/incubator-tvm/issues/4884#issuecomment-587092114
Glad that the problem is resolved. For future trouble shooting questions,
you are
tmoreau89 merged pull request #4898: [DOCS] Introduce how to add hardware
backend to FAQ
URL: https://github.com/apache/incubator-tvm/pull/4898
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.
moreau pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 1314091 Fast exponent (#4790)
add 0b2d11a [DOCS] Introduce how to add hardware backend to FAQ (#4898)
jwfromm commented on issue #4899: [Relay][Frontend][Keras] NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#issuecomment-587108875
@masahi, Keras does use NHWC by default just like TensorFlow. However,
because TVM is better optimized for NCHW, the Keras frontend
zhiics commented on issue #4879: [Relay][Pass] Fix bug in re-processing call
node in MergeComposite pass
URL: https://github.com/apache/incubator-tvm/pull/4879#issuecomment-587125329
@mbaret PTAL. Let's land this if it looks good to you as well.
jwfromm commented on issue #4899: [Relay][Frontend][Keras] NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#issuecomment-587170389
@tqchen, you're right that would be a more direct way to apply
ConvertLayout. If everyone else would prefer it, I could change the
anijain2305 commented on issue #4899: [Relay][Frontend][Keras] NHWC import
support.
URL: https://github.com/apache/incubator-tvm/pull/4899#issuecomment-587173870
I see. Yes, I meant that we should retain native framework format (NHWC in
this case). And use ConvertLayout to change the
Laurawly opened a new pull request #4901: [Fix] Fix get_valid_count flaky test
for cuda
URL: https://github.com/apache/incubator-tvm/pull/4901
Turned on get_valid_count test for cuda in topi. Used atomic operations in
this fix to replace previous block sync method.
@trevor-m
soiferj commented on issue #4879: [Relay][Pass] Fix bug in re-processing call
node in MergeComposite pass
URL: https://github.com/apache/incubator-tvm/pull/4879#issuecomment-587124344
@zhiics I merged your changes and updated the branch. Would you mind taking
another look?
masahi commented on a change in pull request #4644: [Relay][AutoTVM] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380376581
##
File path: src/relay/op/nn/convolution.h
##
@@ -153,6 +153,16 @@ bool Conv2DRel(const Array& types, int
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380386997
##
File path: python/tvm/autotvm/record.py
##
@@ -130,9 +135,17 @@ def decode(row, protocol='json'):
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380387035
##
File path: python/tvm/autotvm/task/relay_integration.py
##
@@ -67,27 +65,22 @@ def
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380386976
##
File path: python/tvm/autotvm/database.py
##
@@ -167,6 +167,7 @@ def filter(self, func):
alexwong commented on a change in pull request #4497: [Relay] Add a PyTorch to
Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#discussion_r380334499
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -0,0 +1,1026 @@
+# Licensed to the Apache
tmoreau89 commented on issue #4887: [VTA] YoloV3 Support
URL: https://github.com/apache/incubator-tvm/pull/4887#issuecomment-587124568
Hi @huajsj, thanks for adding YoloV3 support for VTA. Do you think that
along with this PR you can also construct a tutorial similar to this one:
tqchen edited a comment on issue #4899: [Relay][Frontend][Keras] NHWC import
support.
URL: https://github.com/apache/incubator-tvm/pull/4899#issuecomment-587154754
Given that the model is in native NHWC, perhaps it is better to do it in
another way around(note that the old approach
tqchen commented on issue #4899: [Relay][Frontend][Keras] NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#issuecomment-587154754
Given that the model is in native NHWC, perhaps it is better to do it in
another way around, ingest as NHWC as in this PR and then
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380383331
##
File path: python/tvm/autotvm/task/relay_integration.py
##
@@ -67,27 +65,22 @@ def
jwfromm commented on issue #4899: [Relay][Frontend][Keras] NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#issuecomment-587118058
@comaniac I think I've fixed all the style issues you pointed out.
zhiics merged pull request #4879: [Relay][Pass] Fix bug in re-processing call
node in MergeComposite pass
URL: https://github.com/apache/incubator-tvm/pull/4879
This is an automated message from the Apache Git Service.
To
This is an automated email from the ASF dual-hosted git repository.
zhic 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 27a0284 [Relay][Pass] Fix bug in
zhiics commented on issue #4879: [Relay][Pass] Fix bug in re-processing call
node in MergeComposite pass
URL: https://github.com/apache/incubator-tvm/pull/4879#issuecomment-587144682
Thanks @soiferj @mbaret @cbalint13
This
jwfromm commented on issue #4899: [Relay][Frontend][Keras] NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#issuecomment-587153715
@anijain2305, I think you're right that for most cases ConvertLayout will
achieve something similar. However, there are some
comaniac commented on a change in pull request #4644: [Relay][AutoTVM] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380382775
##
File path: python/tvm/autotvm/task/relay_integration.py
##
@@ -67,27 +65,22 @@ def
comaniac commented on a change in pull request #4644: [Relay][AutoTVM] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380340757
##
File path: python/tvm/autotvm/database.py
##
@@ -125,7 +125,7 @@ def load(self, inp, get_all=False):
comaniac commented on a change in pull request #4644: [Relay][AutoTVM] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380341368
##
File path: python/tvm/autotvm/database.py
##
@@ -167,6 +167,7 @@ def filter(self, func):
comaniac commented on a change in pull request #4644: [Relay][AutoTVM] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380342661
##
File path: python/tvm/autotvm/record.py
##
@@ -130,9 +135,17 @@ def decode(row, protocol='json'):
mbaret commented on issue #4879: [Relay][Pass] Fix bug in re-processing call
node in MergeComposite pass
URL: https://github.com/apache/incubator-tvm/pull/4879#issuecomment-587143463
Looks good.
This is an automated message
tqchen opened a new pull request #4900: [REFACTOR][PY] Establish tvm.te and
tvm.driver
URL: https://github.com/apache/incubator-tvm/pull/4900
- Move the related files to tvm.te
- Move build_module.py to tvm.driver
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380386912
##
File path: include/tvm/relay/op_attr_types.h
##
@@ -207,13 +216,137 @@ enum AnyCodegenStrategy {
icemelon9 commented on a change in pull request #4644: [Relay][AutoTVM] Relay
op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r380386961
##
File path: python/tvm/autotvm/database.py
##
@@ -125,7 +125,7 @@ def load(self, inp, get_all=False):
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 27a0284 [Relay][Pass] Fix bug in re-processing call node in
MergeComposite pass (#4879)
add 08338dd
tqchen merged pull request #4892: [Bugfix] Fixed: Bitwise ops on floats causing
wrong code generation and crashes.
URL: https://github.com/apache/incubator-tvm/pull/4892
This is an automated message from the Apache Git
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 08338dd [REFACTOR][PY] Establish tvm.te and tvm.driver (#4900)
add 976c08a Fixed bugs that occured
tqchen commented on issue #4892: [Bugfix] Fixed: Bitwise ops on floats causing
wrong code generation and crashes.
URL: https://github.com/apache/incubator-tvm/pull/4892#issuecomment-587248067
Thanks @dpankratz !
This is an
tqchen opened a new pull request #4903: [CI] Update ci docker to add autodocsumm
URL: https://github.com/apache/incubator-tvm/pull/4903
source https://github.com/apache/incubator-tvm/pull/4902
This is an automated message
tqchen opened a new pull request #4902: [CI] Add autodocsum as dep
URL: https://github.com/apache/incubator-tvm/pull/4902
autodocsumm can generate a summary table for functions in a module
automatically.
This is an automated
tqchen merged pull request #4900: [REFACTOR][PY] Establish tvm.te and tvm.driver
URL: https://github.com/apache/incubator-tvm/pull/4900
This is an automated message from the Apache Git Service.
To respond to the message,
FrozenGene commented on a change in pull request #4847: Return empty
CSourceModule when no lowered_funcs exists in Relay mod
URL: https://github.com/apache/incubator-tvm/pull/4847#discussion_r380467063
##
File path: src/relay/backend/build_module.cc
##
@@ -437,28 +441,51
tqchen opened a new pull request #4904: [REFACTOR][PY] Establish tvm.arith
URL: https://github.com/apache/incubator-tvm/pull/4904
This is an automated message from the Apache Git Service.
To respond to the message, please
huajsj commented on issue #4887: [VTA] YoloV3 Support
URL: https://github.com/apache/incubator-tvm/pull/4887#issuecomment-587283268
Hi @tmoreau89 , Thanks for the follow up , sure, would work for a tutorial
and upload soon.
Regards
Hua
> Hi @huajsj, thanks for adding
soiferj commented on a change in pull request #4857: Windows Support for cpp_rpc
URL: https://github.com/apache/incubator-tvm/pull/4857#discussion_r380402254
##
File path: apps/cpp_rpc/rpc_env.cc
##
@@ -20,141 +20,139 @@
* \file rpc_env.cc
* \brief Server environment
jmorrill commented on a change in pull request #4857: Windows Support for
cpp_rpc
URL: https://github.com/apache/incubator-tvm/pull/4857#discussion_r380410740
##
File path: apps/cpp_rpc/rpc_env.cc
##
@@ -20,141 +20,139 @@
* \file rpc_env.cc
* \brief Server environment
tqchen commented on issue #4899: [Relay][Frontend][Keras] NHWC import support.
URL: https://github.com/apache/incubator-tvm/pull/4899#issuecomment-587224358
That sounds good.
This is an automated message from the Apache Git
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 8310b25 [CI] Update ci docker to add autodocsumm (#4903)
add 38d1dd2 [CI] Add autodocsum as dep
tqchen merged pull request #4903: [CI] Update ci docker to add autodocsumm
URL: https://github.com/apache/incubator-tvm/pull/4903
This is an automated message from the Apache Git Service.
To respond to the message, please
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 8310b25 [CI] Update ci docker to add
FrozenGene commented on a change in pull request #4847: Return empty
CSourceModule when no lowered_funcs exists in Relay mod
URL: https://github.com/apache/incubator-tvm/pull/4847#discussion_r380467063
##
File path: src/relay/backend/build_module.cc
##
@@ -437,28 +441,51
jmorrill commented on a change in pull request #4857: Windows Support for
cpp_rpc
URL: https://github.com/apache/incubator-tvm/pull/4857#discussion_r380492019
##
File path: apps/cpp_rpc/rpc_env.cc
##
@@ -20,141 +20,139 @@
* \file rpc_env.cc
* \brief Server environment
jmorrill commented on a change in pull request #4857: Windows Support for
cpp_rpc
URL: https://github.com/apache/incubator-tvm/pull/4857#discussion_r380492019
##
File path: apps/cpp_rpc/rpc_env.cc
##
@@ -20,141 +20,139 @@
* \file rpc_env.cc
* \brief Server environment
cchung100m commented on a change in pull request #4891: [TEST][FLAKY]
topi/tests/python/test_topi_sort.py::test_argsort
URL: https://github.com/apache/incubator-tvm/pull/4891#discussion_r380464642
##
File path: topi/tests/python/test_topi_sort.py
##
@@ -27,6 +27,12 @@ def
KindleHe commented on issue #4884: [android_deploy] CRASH caused by
`Module.load` func while running App on Android Device with
URL: https://github.com/apache/incubator-tvm/issues/4884#issuecomment-586919994
Thanks for your quick reply!
As you said in [[REFACTOR][PY][API-Change]
KindleHe edited a comment on issue #4884: [android_deploy] CRASH caused by
`Module.load` func while running App on Android Device with
URL: https://github.com/apache/incubator-tvm/issues/4884#issuecomment-586919994
Thanks for your quick reply!
As you said in
FrozenGene commented on issue #4857: Windows Support for cpp_rpc
URL: https://github.com/apache/incubator-tvm/pull/4857#issuecomment-586882243
> @FrozenGene Thanks for taking a peek. I totally understand limited time.
>
> You are right, I do not include a case for Android to skip
masahi commented on issue #4497: [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-586890441
ping @zhiics @kazum @FrozenGene please review the change you requested.
Your approvals are required to move forward.
Orion34C commented on a change in pull request #4546: [CODEGEN] Support cuda
tensorcore subbyte int data type in auto tensorcore
URL: https://github.com/apache/incubator-tvm/pull/4546#discussion_r380072928
##
File path: tutorials/autotvm/tensor_core_matmul_subbyte_int.py
masahi commented on issue #4790: Fast exponent
URL: https://github.com/apache/incubator-tvm/pull/4790#issuecomment-586890882
ping @tqchen
This is an automated message from the Apache Git Service.
To respond to the message,
masahi commented on issue #4821: [WINDOWS][AutoTVM] OSError: [WinError 10048]
Only one usage of each socket address (protocol/network address/port) is
normally permitted and OSError: [WinError 10049] The requested address is not
valid in its context
URL:
masahi commented on a change in pull request #4497: [Relay] Add a PyTorch to
Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#discussion_r380058573
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -0,0 +1,1026 @@
+# Licensed to the Apache
masahi commented on a change in pull request #4497: [Relay] Add a PyTorch to
Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#discussion_r380058573
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -0,0 +1,1026 @@
+# Licensed to the Apache
KindleHe closed issue #4884: [android_deploy] CRASH caused by `Module.load`
func while running App on Android Device with
URL: https://github.com/apache/incubator-tvm/issues/4884
This is an automated message from the
Orion34C commented on issue #4546: [CODEGEN] Support cuda tensorcore subbyte
int data type in auto tensorcore
URL: https://github.com/apache/incubator-tvm/pull/4546#issuecomment-586994714
@vinx13 Hi, I re-run the tests several times, the same error happened in the
test_workspace_add with
KindleHe commented on issue #4884: [android_deploy] CRASH caused by
`Module.load` func while running App on Android Device with
URL: https://github.com/apache/incubator-tvm/issues/4884#issuecomment-586994657
@tqchen Yes! I find the answer finally! Your answer is right!
TVM4J is
mbaret commented on issue #4847: Return empty CSourceModule when no
lowered_funcs exists in Relay mod
URL: https://github.com/apache/incubator-tvm/pull/4847#issuecomment-587005445
I've tested the most recent commit with my use-case (aarch64 with offloading
to external codegen) and it
81 matches
Mail list logo