CoinCheung commented on issue #4361: I think the dependency of topi is broken
URL: https://github.com/apache/incubator-tvm/issues/4361#issuecomment-555379475
Thanks, the method of exporting `PYTHONPATH` is really good. I am closing
this.
CoinCheung closed issue #4361: I think the dependency of topi is broken
URL: https://github.com/apache/incubator-tvm/issues/4361
This is an automated message from the Apache Git Service.
To respond to the message, please log
jackwish commented on a change in pull request #4351: [QNN] Lowering for
Depthwise Convolution.
URL: https://github.com/apache/incubator-tvm/pull/4351#discussion_r347752773
##
File path: src/relay/qnn/op/convolution.cc
##
@@ -391,7 +526,20 @@ Expr Conv2DCombineTerms(const
jackwish commented on a change in pull request #4351: [QNN] Lowering for
Depthwise Convolution.
URL: https://github.com/apache/incubator-tvm/pull/4351#discussion_r347756284
##
File path: src/relay/qnn/op/convolution.cc
##
@@ -417,23 +565,33 @@ Expr
jackwish commented on a change in pull request #4351: [QNN] Lowering for
Depthwise Convolution.
URL: https://github.com/apache/incubator-tvm/pull/4351#discussion_r347745090
##
File path: src/relay/qnn/op/convolution.cc
##
@@ -84,26 +87,38 @@ WorkloadType GetWorkload(const
apivovarov commented on a change in pull request #4370: [WIP] Relay
visualization: exporter + visualizer
URL: https://github.com/apache/incubator-tvm/pull/4370#discussion_r347760145
##
File path: python/tvm/relay/analysis.py
##
@@ -22,11 +22,14 @@
"""
from . import
apivovarov opened a new pull request #4373: Compare all outputs in TFLite
test_forward_ssd_mobilenet_v1
URL: https://github.com/apache/incubator-tvm/pull/4373
`forward_ssd_mobilenet_v1` has two raw outputs. Currently we only test
output[0].
This PR makes the test to compare both
lsy643 opened a new pull request #4372: [Relay][Frontend][TF] Fix slice when
begin or size is not Const
URL: https://github.com/apache/incubator-tvm/pull/4372
For Relay Frontend of Tensorflow, def _slice can not handle begin or size is
not Const. _infer_value_simulated is used to handle
tmoreau89 commented on a change in pull request #4369: Add cusparse for sparse
dense
URL: https://github.com/apache/incubator-tvm/pull/4369#discussion_r347745819
##
File path: python/tvm/contrib/cusparse.py
##
@@ -0,0 +1,54 @@
+# Licensed to the Apache Software Foundation
tqchen commented on issue #4344: [ThreadPool] Solve thread transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#issuecomment-555332772
I think there seems to be two concerns:
- Whether we want to use cpu as worker0
- Whether we want bind the master thread to CPU0
minminsun 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_r347728939
##
File path: src/runtime/contrib/cublas/cublas.cc
##
@@ -124,35 +169,203 @@
FrozenGene commented on issue #4344: [ThreadPool] Solve thread transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#issuecomment-555329941
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
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 f1d6f33 add rule for clean (#4364)
add 26eb405 [Relay tests] AlterOpLayout - Temporary attr update
tqchen merged pull request #4357: [Relay tests] Temporary Attr Update for
Order-Independent Testing
URL: https://github.com/apache/incubator-tvm/pull/4357
This is an automated message from the Apache Git Service.
To respond
apivovarov commented on issue #4320: Fix TFLite RESHAPE assert
URL: https://github.com/apache/incubator-tvm/pull/4320#issuecomment-555319420
@FrozenGene All checks have passed - Can you approve?
This is an automated message
apivovarov opened a new issue #4371: ImportError: cannot import name
'bilinear_sample_nchw'
URL: https://github.com/apache/incubator-tvm/issues/4371
Same issue, tvm, nnvm and topi were installed after the build
```
~/tvm/python$ sudo python3 setup.py install
~/tvm/nnvm/python$
yzhliu commented on issue #4368: [tutorial][benchmark] nnvm -> relay
URL: https://github.com/apache/incubator-tvm/pull/4368#issuecomment-555317160
sorry I need to verify rpc once I get a device tomorrow.
This is an automated
anijain2305 commented on issue #4357: [Relay tests] Temporary Attr Update for
Order-Independent Testing
URL: https://github.com/apache/incubator-tvm/pull/4357#issuecomment-555316537
@tqchen comments addressed
This is an
FrozenGene commented on issue #4344: [ThreadPool] Solve thread transitions issue
URL: https://github.com/apache/incubator-tvm/pull/4344#issuecomment-555313133
ping @vinx13 @yidawang @tqchen have any comments?
This is an
comaniac edited a comment on issue #4280: [TVM][RUNTIME] A minimum example to
generate external library wrappers for DSOModule
URL: https://github.com/apache/incubator-tvm/pull/4280#issuecomment-555303900
@tqchen we have moved the example JSON runtime as well as the other two
examplese to
cylinbao opened a new pull request #4369: Add cusparse for sparse dense
URL: https://github.com/apache/incubator-tvm/pull/4369
Add a simple gpu scheduling and cusparse support for topi.nn.sparse_dense()
operator.
Please review. Thanks!
@ZihengJiang @tmoreau89 @yuluny2
vinx13 commented on issue #4295: [Relay][Quantize] Integrate data-aware
calibration into quantization
URL: https://github.com/apache/incubator-tvm/pull/4295#issuecomment-555281603
@yzhliu comments addressed
This is an
masahi commented on issue #4368: [tutorial][benchmark] nnvm -> relay
URL: https://github.com/apache/incubator-tvm/pull/4368#issuecomment-555262197
Shouldn't we update get_network() too?
This is an automated message from the
yzhliu opened a new pull request #4368: [tutorial][benchmark] nnvm -> relay
URL: https://github.com/apache/incubator-tvm/pull/4368
@tqchen @masahi
This is an automated message from the Apache Git Service.
To respond to the
yzhliu commented on issue #4365: reminding message for TVM_REGISTER_NODE_TYPE
URL: https://github.com/apache/incubator-tvm/pull/4365#issuecomment-555253086
Thanks guys @hcho3 @anijain2305 @tqchen @junrushao1994
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 3914be5 fix Android and OpenCL docker install (#4363)
add 7efb72e reminding message for
yzhliu merged pull request #4365: reminding message for TVM_REGISTER_NODE_TYPE
URL: https://github.com/apache/incubator-tvm/pull/4365
This is an automated message from the Apache Git Service.
To respond to the message,
anijain2305 commented on a change in pull request #4357: [Relay tests]
Temporary Attr Update for Order-Independent Testing
URL: https://github.com/apache/incubator-tvm/pull/4357#discussion_r347653451
##
File path: python/tvm/relay/testing/temp_op_attr.py
##
@@ -0,0 +1,37
tqchen commented on a change in pull request #4357: [Relay tests] Temporary
Attr Update for Order-Independent Testing
URL: https://github.com/apache/incubator-tvm/pull/4357#discussion_r347649506
##
File path: python/tvm/relay/testing/temp_op_attr.py
##
@@ -0,0 +1,37 @@
+#
tqchen commented on a change in pull request #4357: [Relay tests] Temporary
Attr Update for Order-Independent Testing
URL: https://github.com/apache/incubator-tvm/pull/4357#discussion_r347649668
##
File path: src/relay/ir/op.cc
##
@@ -152,6 +169,15 @@
tqchen commented on a change in pull request #4357: [Relay tests] Temporary
Attr Update for Order-Independent Testing
URL: https://github.com/apache/incubator-tvm/pull/4357#discussion_r347649770
##
File path: tests/python/relay/test_ir_op.py
##
@@ -27,6 +27,32 @@ def
anijain2305 opened a new pull request #4367: [Tests] Dump duration of Pytests.
URL: https://github.com/apache/incubator-tvm/pull/4367
Dumping test durations can help identify any misbehaving tests.
@tqchen @icemelon9 @wweic @broune
tqchen commented on issue #4346: [Runtime] Make ADTObject POD container type
URL: https://github.com/apache/incubator-tvm/pull/4346#issuecomment-555223817
Interesting, sorry I didn't realize that due to injection the child class
was not complete. It is fine to turn into runtime constant
anijain2305 commented on issue #4357: [Relay tests] Temporary Attr Update for
Order-Independent Testing
URL: https://github.com/apache/incubator-tvm/pull/4357#issuecomment-555221188
@tqchen Can you take a look now? Thanks for your patience.
tqchen commented on a change in pull request #4346: [Runtime] Make ADTObject
POD container type
URL: https://github.com/apache/incubator-tvm/pull/4346#discussion_r347622247
##
File path: include/tvm/runtime/container.h
##
@@ -0,0 +1,220 @@
+/*
+ * Licensed to the Apache
anwang2009 opened a new pull request #4366: add benchmark log format doc
URL: https://github.com/apache/incubator-tvm/pull/4366
Adds benchmark log format as discussed in
https://discuss.tvm.ai/t/rfc-benchmark-performance-log-format/4610
wweic commented on issue #4346: [Runtime] Make ADTObject POD container type
URL: https://github.com/apache/incubator-tvm/pull/4346#issuecomment-555198920
@tqchen The c++ compiler in CI does not support defining static constexpr
from sizeof of ADTObj, since its definition is not complete. I
wweic commented on a change in pull request #4346: [Runtime] Make ADTObject POD
container type
URL: https://github.com/apache/incubator-tvm/pull/4346#discussion_r347595585
##
File path: include/tvm/runtime/container.h
##
@@ -0,0 +1,220 @@
+/*
+ * Licensed to the Apache
yzhliu opened a new pull request #4365: reminding message for
TVM_REGISTER_NODE_TYPE
URL: https://github.com/apache/incubator-tvm/pull/4365
per https://discuss.tvm.ai/t/i-spent-5hr-today-add-a-new-node/4754/9
@tqchen @hcho3
vmiheer commented on issue #4364: add rule for clean
URL: https://github.com/apache/incubator-tvm/pull/4364#issuecomment-555179355
cc: @tqchen please review.
This is an automated message from the Apache Git Service.
To
yzhliu merged pull request #4363: [Docker] Fix TVM folder name for installing
on Android and OpenCL
URL: https://github.com/apache/incubator-tvm/pull/4363
This is an automated message from the Apache Git Service.
To respond
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 00521fa [SOURCE] Add ASF header to __init__.py files (#4359)
add 3914be5 fix Android and OpenCL
yzhliu commented on issue #4363: [Docker] Fix TVM folder name for installing on
Android and OpenCL
URL: https://github.com/apache/incubator-tvm/pull/4363#issuecomment-555169679
Thanks @comaniac @tqchen
This is an automated
tqchen commented on issue #4362: About model conversion to TVM
URL: https://github.com/apache/incubator-tvm/issues/4362#issuecomment-555154249
Thanks for bringing this up. Please open threads on https://discuss.tvm.ai/
:)
tqchen closed issue #4362: About model conversion to TVM
URL: https://github.com/apache/incubator-tvm/issues/4362
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
Hzfengsy 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_r347546810
##
File path: src/runtime/contrib/cublas/cublas.cc
##
@@ -124,35 +169,203 @@
tqchen commented on issue #4304: [RFC] Benchmark Performance Log Format WIP
URL: https://github.com/apache/incubator-tvm/issues/4304#issuecomment-555150731
Awesome, let us bring it to the official document by sending a pr to
docs/dev/benchmark.rst
tqchen commented on a change in pull request #4346: [Runtime] Make ADTObject
POD container type
URL: https://github.com/apache/incubator-tvm/pull/4346#discussion_r347543087
##
File path: include/tvm/runtime/memory.h
##
@@ -73,6 +74,26 @@ class ObjAllocatorBase {
tqchen commented on a change in pull request #4346: [Runtime] Make ADTObject
POD container type
URL: https://github.com/apache/incubator-tvm/pull/4346#discussion_r347542358
##
File path: include/tvm/runtime/container.h
##
@@ -0,0 +1,220 @@
+/*
+ * Licensed to the Apache
tqchen commented on a change in pull request #4346: [Runtime] Make ADTObject
POD container type
URL: https://github.com/apache/incubator-tvm/pull/4346#discussion_r347541893
##
File path: include/tvm/runtime/container.h
##
@@ -0,0 +1,220 @@
+/*
+ * Licensed to the Apache
tqchen commented on a change in pull request #4346: [Runtime] Make ADTObject
POD container type
URL: https://github.com/apache/incubator-tvm/pull/4346#discussion_r347540754
##
File path: include/tvm/runtime/container.h
##
@@ -0,0 +1,220 @@
+/*
+ * Licensed to the Apache
tqchen commented on a change in pull request #4346: [Runtime] Make ADTObject
POD container type
URL: https://github.com/apache/incubator-tvm/pull/4346#discussion_r347541168
##
File path: include/tvm/runtime/container.h
##
@@ -0,0 +1,220 @@
+/*
+ * Licensed to the Apache
tqchen edited a comment on issue #4280: [TVM][RUNTIME] A minimum example to
generate external library wrappers for DSOModule
URL: https://github.com/apache/incubator-tvm/pull/4280#issuecomment-555146621
Comments about json runtime. I would consider move the json runtime to
tqchen commented on a change in pull request #4280: [TVM][RUNTIME] A minimum
example to generate external library wrappers for DSOModule
URL: https://github.com/apache/incubator-tvm/pull/4280#discussion_r347537299
##
File path: tests/cpp/external_runtime_test.cc
##
@@
tqchen commented on a change in pull request #4280: [TVM][RUNTIME] A minimum
example to generate external library wrappers for DSOModule
URL: https://github.com/apache/incubator-tvm/pull/4280#discussion_r347536983
##
File path: include/tvm/runtime/module.h
##
@@ -111,7
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 a226973 [Frontend]Add TensorFlow FloorMod (#4308)
add 00521fa [SOURCE] Add ASF header to __init__.py
tqchen merged pull request #4359: [SOURCE] Add ASF header to __init__.py files
URL: https://github.com/apache/incubator-tvm/pull/4359
This is an automated message from the Apache Git Service.
To respond to the message,
comaniac commented on issue #4360: Build docker container failed
URL: https://github.com/apache/incubator-tvm/issues/4360#issuecomment-555135914
PR filed #4363
This is an automated message from the Apache Git Service.
To
comaniac opened a new pull request #4363: [Docker] Fix TVM folder name for
installing on Android and OpenCL
URL: https://github.com/apache/incubator-tvm/pull/4363
After cloning `incubator-tvm`, we need to rename the folder back to `tvm` to
make the script working.
junrushao1994 commented on issue #4361: I think the dependency of topi is broken
URL: https://github.com/apache/incubator-tvm/issues/4361#issuecomment-555135543
Hmmm I am not sure, but I think probably the best way to use TVM is just
setup PYTHONPATH, but not to pip install it, as it is
OriAlpha commented on issue #4361: I think the dependency of topi is broken
URL: https://github.com/apache/incubator-tvm/issues/4361#issuecomment-555010443
ya it's true, but there is a temporary fix for that
export
OriAlpha opened a new issue #4362: About model conversion to TVM
URL: https://github.com/apache/incubator-tvm/issues/4362
Hallo,
I have tried to compile model to TVM, and the model is consists of a
1SimpleRNN layer and dense layers. but when I try to compile it gives out me
CoinCheung opened a new issue #4361: I think the dependency of topi is break
URL: https://github.com/apache/incubator-tvm/issues/4361
After building following the
(documentation)[https://docs.tvm.ai/install/from_source.html]. I tried to
import topi, and I got the error:
```python
cszhz opened a new issue #4360: Build docker container failed
URL: https://github.com/apache/incubator-tvm/issues/4360
Build docker container failed with the file "docker\Dockerfile.demo_android"
please add below in line 60:
mv incubator-tvm tvm && \
minminsun 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_r347291047
##
File path: src/runtime/contrib/cublas/cublas.cc
##
@@ -124,35 +169,203 @@
65 matches
Mail list logo