yzhliu closed pull request #4381: Flaky floor div
URL: https://github.com/apache/incubator-tvm/pull/4381
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
yzhliu opened a new pull request #4381: Flaky floor div
URL: https://github.com/apache/incubator-tvm/pull/4381
#4378
@tqchen @kevinthesun am I understanding correctly?
This is an automated message from the Apache Git
yzhliu opened a new pull request #4380: [team] add Yizhi's pgp key
URL: https://github.com/apache/incubator-tvm/pull/4380
@tqchen
This is an automated message from the Apache Git Service.
To respond to the message,
JohnnyHopp commented on issue #4009: [RFC] 3D model support
URL: https://github.com/apache/incubator-tvm/issues/4009#issuecomment-555867510
Looking forward to seeing this new feature!!!
This is an automated message from the
tqchen commented on issue #4259: [DEV][DRAFT] TVM v0.6 Release candidate
URL: https://github.com/apache/incubator-tvm/issues/4259#issuecomment-555860717
That should be a quick fix(see the reference PR which fixed a related
problem. I would recommend doing so
This is an automated email from the ASF dual-hosted git repository.
liuyizhi pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from f8f4ceb [nvcc] enable multiple arch in one fatbin (#4377)
add d745d93 [CI] Avoid content-length
yzhliu merged pull request #4375: [CI] Avoid content-length request in test
data download
URL: https://github.com/apache/incubator-tvm/pull/4375
This is an automated message from the Apache Git Service.
To respond to the
yzhliu commented on issue #4259: [DEV][DRAFT] TVM v0.6 Release candidate
URL: https://github.com/apache/incubator-tvm/issues/4259#issuecomment-555853460
@tqchen should we wait till #4378 is fixed?
This is an automated message
apivovarov edited a comment on issue #4371: ImportError: cannot import name
'bilinear_sample_nchw'
URL: https://github.com/apache/incubator-tvm/issues/4371#issuecomment-555809592
@OriAlpha If I install `tvm`, `nnvm` and `topi` using `sudo python3
setup.py install` I do not have to set
FrozenGene commented on issue #4344: [ThreadPool] Solve thread transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#issuecomment-555816380
@vinx13 @tqchen @yidawang I wish I have understood your concerns / comments
completely.
@tqchen re Android, I don't provide
anijain2305 commented on issue #4351: [QNN] Lowering for Depthwise Convolution.
URL: https://github.com/apache/incubator-tvm/pull/4351#issuecomment-555814423
@zhiics Can you please review?
This is an automated message from
apivovarov commented on issue #4371: ImportError: cannot import name
'bilinear_sample_nchw'
URL: https://github.com/apache/incubator-tvm/issues/4371#issuecomment-555809592
@OriAlpha You do not understand. If I install `tvm`, `nnvm` and `topi` using
`sudo python3 setup.py install` I do not
hcho3 commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348241869
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def structural_hash(value):
tqchen commented on issue #4375: [CI] Avoid content-length request in test data
download
URL: https://github.com/apache/incubator-tvm/pull/4375#issuecomment-555782251
cc @yzhliu
This is an automated message from the Apache
yzhliu opened a new pull request #4379: [nvcc] enable multiple arch in one
fatbin
URL: https://github.com/apache/incubator-tvm/pull/4379
It enables compiling for multiple architecture. in mxnet we compile to one
binary which works for various types of gpus.
@Hzfengsy @reminisce
anijain2305 commented on a change in pull request #4351: [QNN] Lowering for
Depthwise Convolution.
URL: https://github.com/apache/incubator-tvm/pull/4351#discussion_r348231530
##
File path: src/relay/qnn/op/convolution.cc
##
@@ -417,23 +565,33 @@ Expr
merrymercy commented on issue #4151: [AutoTVM] Added an autotuning Config
Library to store autotune results
URL: https://github.com/apache/incubator-tvm/pull/4151#issuecomment-555778402
Sorry, I am busy this week. I will try to review later this week.
@mbarrett97 You can invite other
anijain2305 commented on a change in pull request #4351: [QNN] Lowering for
Depthwise Convolution.
URL: https://github.com/apache/incubator-tvm/pull/4351#discussion_r348233015
##
File path: src/relay/qnn/op/convolution.cc
##
@@ -166,6 +181,130 @@ Expr Conv2DPadInput(const
anijain2305 commented on a change in pull request #4351: [QNN] Lowering for
Depthwise Convolution.
URL: https://github.com/apache/incubator-tvm/pull/4351#discussion_r348231530
##
File path: src/relay/qnn/op/convolution.cc
##
@@ -417,23 +565,33 @@ Expr
anijain2305 commented on a change in pull request #4351: [QNN] Lowering for
Depthwise Convolution.
URL: https://github.com/apache/incubator-tvm/pull/4351#discussion_r348231275
##
File path: src/relay/qnn/op/convolution.cc
##
@@ -417,23 +565,33 @@ Expr
anijain2305 commented on issue #4335: [Relay][WIP] ConvertLayout pass.
URL: https://github.com/apache/incubator-tvm/pull/4335#issuecomment-555768593
@tqchen Refactored to share code between AlterOpLayout and ConvertLayout
pass.
There is considerable refactoring. Following
tqchen opened a new issue #4378: [TESTS][FLAKY] test_floor_divide
URL: https://github.com/apache/incubator-tvm/issues/4378
http://ci.tvm.ai:8080/blue/rest/organizations/jenkins/pipelines/tvm/branches/PR-4375/runs/1/nodes/238/log/?start=0
This is an automated email from the ASF dual-hosted git repository.
wuwei pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from af52eba [PERF] Parallelize reduction for CPU (#4158)
add 500ff05 [Relay][Quantize] Integrate data-aware
vinx13 merged pull request #4295: [Relay][Quantize] Integrate data-aware
calibration into quantization
URL: https://github.com/apache/incubator-tvm/pull/4295
This is an automated message from the Apache Git Service.
To
cylinbao commented on a change in pull request #4369: [Runtime] Add cusparse
for sparse dense
URL: https://github.com/apache/incubator-tvm/pull/4369#discussion_r348194992
##
File path: docker/install/install_tvm_gpu.sh
##
@@ -29,6 +29,7 @@ git checkout
vinx13 commented on issue #4259: [DEV][DRAFT] TVM v0.6 Release candidate
URL: https://github.com/apache/incubator-tvm/issues/4259#issuecomment-555737813
@yzhliu let's get #4295 in, it's ready to merge once ci passed
This is
This is an automated email from the ASF dual-hosted git repository.
liuyizhi pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from fbeac5e [tutorial][benchmark] nnvm -> relay (#4368)
add af52eba [PERF] Parallelize reduction for CPU
yzhliu merged pull request #4158: [PERF] Parallelize reduction for CPU
URL: https://github.com/apache/incubator-tvm/pull/4158
This is an automated message from the Apache Git Service.
To respond to the message, please log on
yzhliu commented on issue #4158: [PERF] Parallelize reduction for CPU
URL: https://github.com/apache/incubator-tvm/pull/4158#issuecomment-555734227
Thanks @icemelon9 @anijain2305 @soiferj @kevinthesun
This is an automated
yzhliu opened a new pull request #4377: [nvcc] enable multiple arch in one
fatbin
URL: https://github.com/apache/incubator-tvm/pull/4377
Enable us to compile for multiple architectures. which in mxnet we use it to
build one binary for a series of nv gpus.
@Hzfengsy @reminisce
yzhliu commented on issue #4368: [tutorial][benchmark] nnvm -> relay
URL: https://github.com/apache/incubator-tvm/pull/4368#issuecomment-555729932
@masahi @petrex sorry typo - AMD gpu, as @petrex mentioned earlier. Thanks
for verifying:)
soiferj commented on a change in pull request #4372: [Relay][Frontend][TF] Fix
slice when begin or size is not Const
URL: https://github.com/apache/incubator-tvm/pull/4372#discussion_r348181332
##
File path: python/tvm/relay/frontend/tensorflow.py
##
@@ -626,8 +626,14 @@
masahi commented on issue #4368: [tutorial][benchmark] nnvm -> relay
URL: https://github.com/apache/incubator-tvm/pull/4368#issuecomment-555723116
I already tested it on AMD, @yzhliu maybe you mistook it for somebody else?
broune commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348175434
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def
yzhliu commented on issue #4368: [tutorial][benchmark] nnvm -> relay
URL: https://github.com/apache/incubator-tvm/pull/4368#issuecomment-555721377
@petrex could you also help to verify the arm results?
This is an automated
broune commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348170769
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def
broune commented on issue #4367: [Tests] Dump duration of Pytests.
URL: https://github.com/apache/incubator-tvm/pull/4367#issuecomment-555716642
Re @tqchen, which file does "the test script" refer to?
This is an automated
hcho3 commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348167717
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def structural_hash(value):
hcho3 commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348166313
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def structural_hash(value):
hcho3 commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348166313
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def structural_hash(value):
hcho3 commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348164395
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def structural_hash(value):
hcho3 commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348164809
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def structural_hash(value):
hcho3 commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348164395
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def structural_hash(value):
tmoreau89 commented on a change in pull request #4295: [Relay][Quantize]
Integrate data-aware calibration into quantization
URL: https://github.com/apache/incubator-tvm/pull/4295#discussion_r348163524
##
File path: python/tvm/relay/quantize/quantize.py
##
@@ -143,9
broune commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348141345
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def
broune commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348141743
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def
broune commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348141393
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def
broune commented on a change in pull request #4370: [WIP] Relay visualization:
exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r348139803
##
File path: python/tvm/relay/analysis.py
##
@@ -408,3 +411,116 @@ def
tqchen opened a new pull request #4375: [CI] Avoid content-length request in
test data download
URL: https://github.com/apache/incubator-tvm/pull/4375
To avoid CI out-rage due to download limitations which happened today cc
@icemelon9
Perhaps we need a smarter way to verify the
Laurawly commented on a change in pull request #4353: [Perf] Enhance cudnn and
cublas backend and enable TensorCore
URL: https://github.com/apache/incubator-tvm/pull/4353#discussion_r348137728
##
File path: src/runtime/contrib/cublas/cublas.cc
##
@@ -124,35 +169,203 @@
tmoreau89 commented on a change in pull request #4369: [Runtime] Add cusparse
for sparse dense
URL: https://github.com/apache/incubator-tvm/pull/4369#discussion_r348135715
##
File path: docker/install/install_tvm_gpu.sh
##
@@ -29,6 +29,7 @@ git checkout
yidawang commented on a change in pull request #4344: [ThreadPool] Solve thread
transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#discussion_r348132452
##
File path: src/runtime/threading_backend.cc
##
@@ -134,21 +134,28 @@ class ThreadGroup::Impl {
cylinbao commented on a change in pull request #4369: [Runtime] Add cusparse
for sparse dense
URL: https://github.com/apache/incubator-tvm/pull/4369#discussion_r348130505
##
File path: docker/install/install_tvm_gpu.sh
##
@@ -29,6 +29,7 @@ git checkout
comaniac commented on a change in pull request #4369: [Runtime] Add cusparse
for sparse dense
URL: https://github.com/apache/incubator-tvm/pull/4369#discussion_r348129443
##
File path: docker/install/install_tvm_gpu.sh
##
@@ -29,6 +29,7 @@ git checkout
yidawang commented on a change in pull request #4344: [ThreadPool] Solve thread
transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#discussion_r348118387
##
File path: src/runtime/threading_backend.cc
##
@@ -134,21 +134,28 @@ class ThreadGroup::Impl {
broune commented on issue #4367: [Tests] Dump duration of Pytests.
URL: https://github.com/apache/incubator-tvm/pull/4367#issuecomment-555662599
tldr: how about having a run_pytets.sh file that takes care of how to run
pytest?
1) I agree with Ramana that it is unfortunate to
tqchen commented on a change in pull request #4344: [ThreadPool] Solve thread
transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#discussion_r348097761
##
File path: src/runtime/threading_backend.cc
##
@@ -134,21 +134,28 @@ class ThreadGroup::Impl {
tqchen commented on a change in pull request #4344: [ThreadPool] Solve thread
transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#discussion_r348105649
##
File path: src/runtime/threading_backend.cc
##
@@ -134,21 +134,28 @@ class ThreadGroup::Impl {
yidawang commented on a change in pull request #4344: [ThreadPool] Solve thread
transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#discussion_r348078074
##
File path: src/runtime/threading_backend.cc
##
@@ -134,21 +134,28 @@ class ThreadGroup::Impl {
yongwww commented on a change in pull request #4372: [Relay][Frontend][TF] Fix
slice when begin or size is not Const
URL: https://github.com/apache/incubator-tvm/pull/4372#discussion_r348083252
##
File path: python/tvm/relay/frontend/tensorflow.py
##
@@ -626,8 +626,14 @@
zhiics commented on issue #4339: Added tflite frontend support for quantized
mean
URL: https://github.com/apache/incubator-tvm/pull/4339#issuecomment-555612886
@tristan-arm CI is consistently failing. Shouldn't relate to your change.
yzhliu commented on issue #4372: [Relay][Frontend][TF] Fix slice when begin or
size is not Const
URL: https://github.com/apache/incubator-tvm/pull/4372#issuecomment-555612171
@yongwww could you take a look?
This is an
This is an automated email from the ASF dual-hosted git repository.
liuyizhi pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 26eb405 [Relay tests] AlterOpLayout - Temporary attr update (#4357)
add 331f6fd Fix TFLite RESHAPE
yzhliu merged pull request #4320: Fix TFLite RESHAPE assert
URL: https://github.com/apache/incubator-tvm/pull/4320
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
yzhliu commented on issue #4320: Fix TFLite RESHAPE assert
URL: https://github.com/apache/incubator-tvm/pull/4320#issuecomment-555611755
Thanks @apivovarov @FrozenGene @u99127
This is an automated message from the Apache Git
anijain2305 commented on issue #4367: [Tests] Dump duration of Pytests.
URL: https://github.com/apache/incubator-tvm/pull/4367#issuecomment-555610907
@u99127 I can close this PR then. I think your PR will supersede this.
This
tqchen commented on issue #4371: ImportError: cannot import name
'bilinear_sample_nchw'
URL: https://github.com/apache/incubator-tvm/issues/4371#issuecomment-555601224
Thanks @OriAlpha @apivovarov please use the discuss forum for future
trouble shooting
tqchen closed issue #4360: Build docker container failed
URL: https://github.com/apache/incubator-tvm/issues/4360
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
tqchen closed issue #4371: ImportError: cannot import name
'bilinear_sample_nchw'
URL: https://github.com/apache/incubator-tvm/issues/4371
This is an automated message from the Apache Git Service.
To respond to the message,
tristan-arm commented on issue #4339: Added tflite frontend support for
quantized mean
URL: https://github.com/apache/incubator-tvm/pull/4339#issuecomment-82843
CI failure is unrelated to this patch - I'll kick it off again once the
issue is resolved
FrozenGene commented on issue #4344: [ThreadPool] Solve thread transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#issuecomment-66529
@vinx13 @yidawang @tqchen I use `pthread_atfork` function to avoid master
thread affinity derived by child threads. I find this
u99127 commented on issue #4367: [Tests] Dump duration of Pytests.
URL: https://github.com/apache/incubator-tvm/pull/4367#issuecomment-08562
We need to measure this I agree.
There's a simpler way of achieving this this - I would suggest using
PYTEST_ADDOPTS instead in
snowolfhawk commented on issue #3859: [WIP] [AutoTVM][TOPI] NHWC conv2d
templates for ARM
URL: https://github.com/apache/incubator-tvm/pull/3859#issuecomment-555484476
@jackwish
Do you still use winograd for NHWC?
This is
tristan-arm commented on a change in pull request #4339: Added tflite frontend
support for quantized mean
URL: https://github.com/apache/incubator-tvm/pull/4339#discussion_r347878614
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -659,7 +659,23 @@ def
tristan-arm commented on a change in pull request #4339: Added tflite frontend
support for quantized mean
URL: https://github.com/apache/incubator-tvm/pull/4339#discussion_r347878384
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -710,16 +710,23 @@ def
masahi commented on a change in pull request #4368: [tutorial][benchmark] nnvm
-> relay
URL: https://github.com/apache/incubator-tvm/pull/4368#discussion_r347848575
##
File path: apps/benchmark/util.py
##
@@ -46,38 +47,32 @@ def get_network(name, batch_size,
masahi commented on a change in pull request #4368: [tutorial][benchmark] nnvm
-> relay
URL: https://github.com/apache/incubator-tvm/pull/4368#discussion_r347846634
##
File path: apps/benchmark/util.py
##
@@ -46,38 +47,32 @@ def get_network(name, batch_size,
hgt312 opened a new pull request #4374: [Typo][Doc]Fix a typo in dev doc
URL: https://github.com/apache/incubator-tvm/pull/4374
Thanks for contributing to TVM! Please refer to guideline
https://docs.tvm.ai/contribute/ for useful information and tips. After the pull
request is submitted,
OriAlpha commented on issue #4371: ImportError: cannot import name
'bilinear_sample_nchw'
URL: https://github.com/apache/incubator-tvm/issues/4371#issuecomment-555418529
It's just a broken path error, you can do
export
comaniac commented on a change in pull request #4369: [Runtime] Add cusparse
for sparse dense
URL: https://github.com/apache/incubator-tvm/pull/4369#discussion_r347800352
##
File path: cmake/config.cmake
##
@@ -151,6 +151,9 @@ set(USE_CUDNN OFF)
# Whether use cuBLAS
comaniac commented on a change in pull request #4369: [Runtime] Add cusparse
for sparse dense
URL: https://github.com/apache/incubator-tvm/pull/4369#discussion_r347797396
##
File path: cmake/config.cmake
##
@@ -151,6 +151,9 @@ set(USE_CUDNN OFF)
# Whether use cuBLAS
cylinbao commented on a change in pull request #4369: [Runtime] Add cusparse
for sparse dense
URL: https://github.com/apache/incubator-tvm/pull/4369#discussion_r347792620
##
File path: topi/tests/python/test_topi_sparse.py
##
@@ -218,22 +220,42 @@ def test_dense():
yidawang commented on issue #4344: [ThreadPool] Solve thread transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#issuecomment-555386180
> Hmm...I think Auto Tuning and OpenCV + TVM is very common and many users
will forget this things, so I prefer it be `false` by
83 matches
Mail list logo