This is an automated email from the ASF dual-hosted git repository.
zhaowu pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 9bbee96 [PYTORCH]Tensor creation ops support (#5347)
add 6dfcd37 [Hexagon] Add hexagon_posix.cc to TVM/
FrozenGene merged pull request #5346: [Hexagon] Add hexagon_posix.cc to TVM/RT
sources in the right place
URL: https://github.com/apache/incubator-tvm/pull/5346
This is an automated message from the Apache Git Service.
To re
FrozenGene commented on issue #5346: [Hexagon] Add hexagon_posix.cc to TVM/RT
sources in the right place
URL: https://github.com/apache/incubator-tvm/pull/5346#issuecomment-615066413
Thanks @kparzysz-quic It is merged now.
Th
MrCrazyCrab opened a new issue #5356: some bug about tvm
URL: https://github.com/apache/incubator-tvm/issues/5356
when i run my code, i met a problem about parsing my onnx model
`return self.input_dict[self.input_keys[item]]
IndexError: list index out of range`
i could't find th
zchuang11 commented on issue #5024: [OpenCL] Max Buildin Type Error
URL: https://github.com/apache/incubator-tvm/issues/5024#issuecomment-615033686
@tqchen I have pushed a unsupported op as [Lambda in
keras](https://discuss.tvm.ai/t/lambda-op-for-keras/6408) Thanks for your
replay!
-
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 0075d8c [CRT]Compilation warnings fixed for 32bit and 64bit
compilation (#5349)
add 9bbee96 [PYTORCH]T
masahi merged pull request #5347: [PYTORCH]Tensor creation ops support
URL: https://github.com/apache/incubator-tvm/pull/5347
This is an automated message from the Apache Git Service.
To respond to the message, please log on
kparzysz-quic commented on issue #5346: [Hexagon] Add hexagon_posix.cc to
TVM/RT sources in the right place
URL: https://github.com/apache/incubator-tvm/pull/5346#issuecomment-614982880
This was a mistake when copying the code changes over in the previous PR.
--
siju-samuel commented on issue #5355: Restructure imports in tflite frontend.
URL: https://github.com/apache/incubator-tvm/pull/5355#issuecomment-614980027
@u99127 I think this PR will create unnecessary dependancy with tflite
package and tvm. With this change you mandatorily need to instal
anijain2305 commented on a change in pull request #5354: [Tutorial] AutoTVM for
TFLite model on ARM CPUs.
URL: https://github.com/apache/incubator-tvm/pull/5354#discussion_r409919577
##
File path: tutorials/autotvm/tune_relay_tflite_arm.py
##
@@ -0,0 +1,427 @@
+# Licensed
anijain2305 commented on a change in pull request #5354: [Tutorial] AutoTVM for
TFLite model on ARM CPUs.
URL: https://github.com/apache/incubator-tvm/pull/5354#discussion_r409919577
##
File path: tutorials/autotvm/tune_relay_tflite_arm.py
##
@@ -0,0 +1,427 @@
+# Licensed
u99127 commented on issue #5355: Restructure imports in tflite frontend.
URL: https://github.com/apache/incubator-tvm/pull/5355#issuecomment-614961595
hmm, not sure how it passed my test run with linting before I submitted
this.
u99127 opened a new pull request #5355: Restructure imports in tflite frontend.
URL: https://github.com/apache/incubator-tvm/pull/5355
These python modules are needed for every tflite file parsed. Factorize out
imports of the common ones that I spotted. Now that the import of operator is
c
anijain2305 opened a new pull request #5354: [Tutorial] AutoTVM for TFLite
model on ARM CPUs.
URL: https://github.com/apache/incubator-tvm/pull/5354
Discuss.-
https://discuss.tvm.ai/t/topi-using-x86-schedules-for-arm-conv2d/6365
tqchen merged pull request #5349: [RUNTIME][CRT]Compilation warnings fixed for
32bit and 64bit compilation
URL: https://github.com/apache/incubator-tvm/pull/5349
This is an automated message from the Apache Git Service.
To r
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 baff99c enable tsim and fsim for GPU build (#5352)
add 0075d8c [CRT]Compilation warnings fixed for 32bi
liangfu commented on issue #5349: [RUNTIME][CRT]Compilation warnings fixed for
32bit and 64bit compilation
URL: https://github.com/apache/incubator-tvm/pull/5349#issuecomment-614952876
Thanks for your contribution. I think if this fixed the issue, we can then
re-enable the unit test in CI.
masahi commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409902806
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,424 @@
+/*
+ * Licensed to the Apa
masahi commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409902404
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,424 @@
+/*
+ * Licensed to the Apa
masahi commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409902168
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,424 @@
+/*
+ * Licensed to the Apa
masahi commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409900411
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,424 @@
+/*
+ * Licensed to the Apa
masahi commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409900186
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,424 @@
+/*
+ * Licensed to the Apa
masahi commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409900186
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,424 @@
+/*
+ * Licensed to the Apa
masahi commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409899250
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,434 @@
+/*
+ * Licensed to the Apa
kparzysz-quic commented on issue #5353: [RUNTIME] FastRPC interface for Hexagon
runtime
URL: https://github.com/apache/incubator-tvm/pull/5353#issuecomment-614931719
@FrozenGene
This is an automated message from the Apache G
kparzysz-quic opened a new pull request #5353: [RUNTIME] FastRPC interface for
Hexagon runtime
URL: https://github.com/apache/incubator-tvm/pull/5353
Co-authored-by: Ravishankar Kolachana
Co-authored-by: Krzysztof Parzyszek
-
zhiics closed issue #5351: [Jenkinsfile] Should we include libvta_tsim.so at
the top level
URL: https://github.com/apache/incubator-tvm/issues/5351
This is an automated message from the Apache Git Service.
To respond to the
zhiics commented on issue #5351: [Jenkinsfile] Should we include libvta_tsim.so
at the top level
URL: https://github.com/apache/incubator-tvm/issues/5351#issuecomment-614929802
close by #5352
This is an automated message fro
mbrookhart commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409879725
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,440 @@
+/*
+ * Licensed to the
tqchen merged pull request #5352: [CI] Enable tsim and fsim for GPU build to
avoid pack_lib error
URL: https://github.com/apache/incubator-tvm/pull/5352
This is an automated message from the Apache Git Service.
To respond 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 c9c [BYOC][FIX] Fix typo in "default" (#5348)
add baff99c enable tsim and fsim for GPU build (#5352
masahi commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409842978
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,440 @@
+/*
+ * Licensed to the Apa
tqchen merged pull request #5348: [BYOC][FIX] Fix typo in "default"
URL: https://github.com/apache/incubator-tvm/pull/5348
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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 37bd812 [RUNTIME][CRT] support DLTensor whose ndim == 0 (#5344)
add c9c [BYOC][FIX] Fix typo in "de
zhiics opened a new pull request #5352: [CI] Enable tsim and fsim for GPU build
to avoid pack_lib error
URL: https://github.com/apache/incubator-tvm/pull/5352
#5351
@tqchen
This is an automated message from the A
tqchen merged pull request #5344: [RUNTIME][CRT] scalar's ndim is 0
URL: https://github.com/apache/incubator-tvm/pull/5344
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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 84d1eec [RELAY][BYOC] Register pattern tables from external codegens
(#5262)
add 37bd812 [RUNTIME][CRT
tqchen commented on issue #5344: [RUNTIME][CRT] scalar's ndim is 0
URL: https://github.com/apache/incubator-tvm/pull/5344#issuecomment-614866193
Thanks @windclarion
This is an automated message from the Apache Git Service.
To
tqchen commented on issue #5351: [Jenkinsfile] Should we include libvta_tsim.so
at the top level
URL: https://github.com/apache/incubator-tvm/issues/5351#issuecomment-614846263
I agree, please send a patch :)
This is an autom
icemelon9 commented on issue #5319: [cuDNN] Add cuDNN grouped convolution
support
URL: https://github.com/apache/incubator-tvm/pull/5319#issuecomment-614845264
@wpan11nv The test cases for strategy mostly reside in the relay op tests.
When an op is compiled for cuda target, it will use th
icemelon9 commented on a change in pull request #5319: [cuDNN] Add cuDNN
grouped convolution support
URL: https://github.com/apache/incubator-tvm/pull/5319#discussion_r409790621
##
File path: python/tvm/relay/op/strategy/cuda.py
##
@@ -91,6 +91,9 @@ def schedule_lrn_cuda(a
zhiics edited a comment on issue #5351: [Jenkinsfile] Should we include
libvta_tsim.so at the top level
URL: https://github.com/apache/incubator-tvm/issues/5351#issuecomment-614841035
I see. Thanks. Then should we add `set(USE_VTA_TSIM/FSIM, ON)` in the
build_gpu.sh? We see errors when pac
zhiics commented on issue #5351: [Jenkinsfile] Should we include libvta_tsim.so
at the top level
URL: https://github.com/apache/incubator-tvm/issues/5351#issuecomment-614841035
I see. Thanks. Then should we add `set(USE_VTA_TSIM/FSIM, ON)` in the
build_gpu.sh? We see errors when packing th
tqchen commented on issue #5351: [Jenkinsfile] Should we include libvta_tsim.so
at the top level
URL: https://github.com/apache/incubator-tvm/issues/5351#issuecomment-614839000
we still need them for building vta tutorials, which is in the last docs
stage
-
zhiics opened a new issue #5351: [Jenkinsfile] Should we include libvta_tsim.so
at the top level
URL: https://github.com/apache/incubator-tvm/issues/5351
@tqchen @tmoreau89
Should we include libvta_tsim.so and libvta_fsim.so here?
https://github.com/apache/incubator-tvm/blob
icemelon9 commented on a change in pull request #5319: [cuDNN] Add cuDNN
grouped convolution support
URL: https://github.com/apache/incubator-tvm/pull/5319#discussion_r409777883
##
File path: topi/python/topi/cuda/conv2d.py
##
@@ -67,7 +67,7 @@ def _callback(op):
@autot
anijain2305 opened a new pull request #5350: [TOPI-ARM] Do not alter layout if
layout is NHWC
URL: https://github.com/apache/incubator-tvm/pull/5350
@icemelon9 I think we missed this piece while working on op strategy.
This i
siju-samuel opened a new pull request #5349: [RUNTIME][CRT]Compilation warnings
fixed for 32bit and 64bit compilation
URL: https://github.com/apache/incubator-tvm/pull/5349
@liangfu Please help to review this PR. TIA
Thanks for contributing to TVM! Please refer to guideline
https:
comaniac commented on issue #5348: [BYOC][FIX] Fix typo in "default"
URL: https://github.com/apache/incubator-tvm/pull/5348#issuecomment-614780555
Thanks for the fix. Apparently our unit tests didn't cover this case. Would
you be able to improve the unit test a bit to cover it?
--
mbrookhart commented on a change in pull request #5231: [POC] Pattern Language,
Matcher, and Rewriter V0
URL: https://github.com/apache/incubator-tvm/pull/5231#discussion_r409681253
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -0,0 +1,440 @@
+/*
+ * Licensed to the
mbaret commented on issue #5348: [BYOC][FIX] Fix typo in "default"
URL: https://github.com/apache/incubator-tvm/pull/5348#issuecomment-614723215
cc @comaniac @zhiics @manupa-arm
This is an automated message from the Apache Gi
mbaret opened a new pull request #5348: [BYOC][FIX] Fix typo in "default"
URL: https://github.com/apache/incubator-tvm/pull/5348
Default annotations were incorrectly being named 'defualt' which results in
them not being removed in PartitionGraph.
---
mbaret commented on issue #5345: [RELAY] Move frontend utils
URL: https://github.com/apache/incubator-tvm/pull/5345#issuecomment-614706054
cc @anijain2305 @zhiics
This is an automated message from the Apache Git Service.
To r
siju-samuel opened a new pull request #5347: [PYTORCH]Tensor creation ops
support
URL: https://github.com/apache/incubator-tvm/pull/5347
- ones
- ones_like
- zeros
- zeros_like
- full
- full_like
- linspace
@masahi Please help me to review and merge this PR. Thanks.
kparzysz-quic opened a new pull request #5346: [Hexagon] Add hexagon_posix.cc
to TVM/RT sources in the right place
URL: https://github.com/apache/incubator-tvm/pull/5346
This file was added before the variable with TVM/RT was initialized. The
initialization overwrote the addition.
---
mbaret opened a new pull request #5345: [RELAY] Move frontend utils
URL: https://github.com/apache/incubator-tvm/pull/5345
The util file currently under frontend is used from outside of frontend (in
qnn/op/legalizations). This suggests that the file should be pushed up to a
higher level.
masahi commented on issue #5262: [RELAY][BYOC] Register pattern tables from
external codegens
URL: https://github.com/apache/incubator-tvm/pull/5262#issuecomment-614501472
Thanks @mbaret
This is an automated message from the
masahi merged pull request #5262: [RELAY][BYOC] Register pattern tables from
external codegens
URL: https://github.com/apache/incubator-tvm/pull/5262
This is an automated message from the Apache Git Service.
To respond to th
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 6e36da3 [TOPI][PYTORCH]Logical & Bitwise operator support (#5341)
add 84d1eec [RELAY][BYOC] Register pa
masahi commented on issue #5341: [TOPI][PYTORCH]Logical & Bitwise operator
support
URL: https://github.com/apache/incubator-tvm/pull/5341#issuecomment-614500334
Thanks @siju-samuel
This is an automated message from the Apach
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 cc8cacb [DOCS] Bring relay docs to the top-level flat view (#5343)
add 6e36da3 [TOPI][PYTORCH]Logical &
masahi merged pull request #5341: [TOPI][PYTORCH]Logical & Bitwise operator
support
URL: https://github.com/apache/incubator-tvm/pull/5341
This is an automated message from the Apache Git Service.
To respond to the message,
62 matches
Mail list logo