FrozenGene commented on issue #4803: [WIP][Frontend] Asymmetric padding of
convolution support
URL: https://github.com/apache/incubator-tvm/pull/4803#issuecomment-581793013
Thanks for the quick response @anijain2305 !
> I can handle them in 2 separate PRs. For Conv2D attrs, I will ha
FrozenGene edited a comment on issue #4803: [WIP][Frontend] Asymmetric padding
of convolution support
URL: https://github.com/apache/incubator-tvm/pull/4803#issuecomment-581793013
Thanks for the quick response @anijain2305 !
> I can handle them in 2 separate PRs. For Conv2D attrs, I
FrozenGene commented on issue #4714: Mxnet parser for Qnn dialect
URL: https://github.com/apache/incubator-tvm/pull/4714#issuecomment-581800500
> @tmoreau89 @yzhliu @tqchen @jackwish @FrozenGene @liangfu @vinx13
@anijain2305 I want to discuss this PR from testability POV. Mxnet suggests to
ghostplant opened a new issue #4810: `tvm.tensor.HybridOp` doesn't have
reduce_axis
URL: https://github.com/apache/incubator-tvm/issues/4810
Is hyrbid script supporting scheduling its reduce_axis?
This is an automated message
FrozenGene commented on a change in pull request #4749: [AutoTVM] Support to
pass additional compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#discussion_r374597354
##
File path: python/tvm/module.py
##
@@ -162,13 +162,20 @@ def export_library(self,
inadob commented on a change in pull request #4704: [Relay][Frontend][TFLite]
Add parser support for arg_min_max
URL: https://github.com/apache/incubator-tvm/pull/4704#discussion_r374611987
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -826,6 +828,50 @@ def _conv
inadob commented on a change in pull request #4704: [Relay][Frontend][TFLite]
Add parser support for arg_min_max
URL: https://github.com/apache/incubator-tvm/pull/4704#discussion_r374611987
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -826,6 +828,50 @@ def _conv
wyc-ruiker commented on a change in pull request #4789: [Frontend][TFLite]
Dynamically calculate input_stats of any fake_quant range
URL: https://github.com/apache/incubator-tvm/pull/4789#discussion_r374618096
##
File path: tests/python/frontend/tflite/test_forward.py
##
@
inadob commented on a change in pull request #4789: [Frontend][TFLite]
Dynamically calculate input_stats of any fake_quant range
URL: https://github.com/apache/incubator-tvm/pull/4789#discussion_r374622386
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -7
MarisaKirisame commented on a change in pull request #4644: [WIP] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r374619441
##
File path: python/tvm/relay/backend/compile_engine.py
##
@@ -63,6 +83,316 @@ def _get_cache_key(source_func,
inadob commented on a change in pull request #4789: [Frontend][TFLite]
Dynamically calculate input_stats of any fake_quant range
URL: https://github.com/apache/incubator-tvm/pull/4789#discussion_r374629200
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -1
inadob commented on a change in pull request #4789: [Frontend][TFLite]
Dynamically calculate input_stats of any fake_quant range
URL: https://github.com/apache/incubator-tvm/pull/4789#discussion_r374629200
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -1
mbarrett97 commented on issue #4771: [Relay] Added Merge Composite pass
URL: https://github.com/apache/incubator-tvm/pull/4771#issuecomment-581887507
I've done a big clean up + simplified the code a bit and added some comments.
---
inadob commented on a change in pull request #4696: [Relay][Frontend][TFlite]
Add support for quantized LOGISTIC
URL: https://github.com/apache/incubator-tvm/pull/4696#discussion_r374712987
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -143,11 +144,13 @@
wyc-ruiker commented on a change in pull request #4805: [Frontend][TFlite] Add
parser support for relu6, leaky_relu, relu_n1_to_1, log_softmax
URL: https://github.com/apache/incubator-tvm/pull/4805#discussion_r374723352
##
File path: python/tvm/relay/frontend/tflite.py
##
wyc-ruiker commented on a change in pull request #4805: [Frontend][TFlite] Add
parser support for relu6, leaky_relu, relu_n1_to_1, log_softmax
URL: https://github.com/apache/incubator-tvm/pull/4805#discussion_r374724376
##
File path: python/tvm/relay/frontend/tflite.py
##
vinx13 commented on issue #4798: [QNN] Optimize lowering for requantize and
FixedPointMultiply.
URL: https://github.com/apache/incubator-tvm/pull/4798#issuecomment-581966405
LGTM, possible to have a test?
This is an automated
wyc-ruiker commented on a change in pull request #4788:
[FRONTEND][TFLITE]Gather, StridedSlice op support added
URL: https://github.com/apache/incubator-tvm/pull/4788#discussion_r374733608
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -792,6 +794,147 @@ def conve
wyc-ruiker commented on a change in pull request #4788:
[FRONTEND][TFLITE]Gather, StridedSlice op support added
URL: https://github.com/apache/incubator-tvm/pull/4788#discussion_r374734446
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -792,6 +794,147 @@ def conve
wyc-ruiker commented on a change in pull request #4788:
[FRONTEND][TFLITE]Gather, StridedSlice op support added
URL: https://github.com/apache/incubator-tvm/pull/4788#discussion_r374744847
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -244,6 +244,74 @@ d
masahi commented on issue #4809: [DOCS] Fix vta tutorial
URL: https://github.com/apache/incubator-tvm/pull/4809#issuecomment-581995303
fyi I have the same fix in #4756 as part of pillow update. Without it
sphinx complained.
anijain2305 commented on issue #4807: [Frontend][TFLite] Fix quantized pad
value for convolution
URL: https://github.com/apache/incubator-tvm/pull/4807#issuecomment-582013859
Let me take a look at this. You might have to wait for sometime as I am
little busy, but I will get to it within 2
tmoreau89 merged pull request #4809: [DOCS] Fix vta tutorial
URL: https://github.com/apache/incubator-tvm/pull/4809
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
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 6f7d6fa [LINT] Fix -Wextra (#4804)
add 3fb937f [DOCS] Fix vta tutorial (#4809)
No new revisions were a
anijain2305 commented on issue #4798: [QNN] Optimize lowering for requantize
and FixedPointMultiply.
URL: https://github.com/apache/incubator-tvm/pull/4798#issuecomment-582020388
Done. Thanks!
This is an automated message fro
tqchen closed issue #4810: `tvm.tensor.HybridOp` doesn't have reduce_axis
URL: https://github.com/apache/incubator-tvm/issues/4810
This is an automated message from the Apache Git Service.
To respond to the message, please lo
tqchen commented on issue #4810: `tvm.tensor.HybridOp` doesn't have reduce_axis
URL: https://github.com/apache/incubator-tvm/issues/4810#issuecomment-582022090
Please open a new thread on the discuss forum https://discuss.tvm.ai/ I
guess the current answer to the question is no. However, we
kevinthesun commented on issue #4797: [AutoTVM] Minor bug fixes in AutoTVM for
QNN graphs
URL: https://github.com/apache/incubator-tvm/pull/4797#issuecomment-582025432
Thanks @anijain2305
This is an automated message from th
This is an automated email from the ASF dual-hosted git repository.
kevinthesun pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 3fb937f [DOCS] Fix vta tutorial (#4809)
add d279991 [AutoTVM] Minor bug fixes in AutoTVM for QNN g
kevinthesun merged pull request #4797: [AutoTVM] Minor bug fixes in AutoTVM for
QNN graphs
URL: https://github.com/apache/incubator-tvm/pull/4797
This is an automated message from the Apache Git Service.
To respond to the me
icemelon9 commented on a change in pull request #4644: [WIP] Relay op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r374821520
##
File path: python/tvm/relay/backend/compile_engine.py
##
@@ -63,6 +83,316 @@ def _get_cache_key(source_func, target
merrymercy edited a comment on issue #4749: [AutoTVM] Support to pass
additional compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582055130
1. We should not put `options` in `MeasureResult`. It is not the result
(output). It is input.
2. If you change
merrymercy commented on issue #4749: [AutoTVM] Support to pass additional
compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582055130
1. We should not put `options` in `MeasureResult`. It is not result
(output). It is input.
2. If you change MeasureInp
icemelon9 opened a new pull request #4811: [RUNTIME] Fix memory leak when using
openMP
URL: https://github.com/apache/incubator-tvm/pull/4811
https://discuss.tvm.ai/t/memory-leak-running-in-cpu-mode-on-windows-tvm-0-6-0/5462/3
--
icemelon9 commented on issue #4787: [Relay] Conv2D padding representation
URL: https://github.com/apache/incubator-tvm/pull/4787#issuecomment-582063672
@zxy844288792 I see that you didn't modify the
`contrib_depthwise_conv2d_nchwc`. Should we also use `get_pad_tuple2d` for it?
merrymercy edited a comment on issue #4749: [AutoTVM] Support to pass
additional compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582055130
1. We should not put `options` in `MeasureResult`. It is not the result
(output). It is input.
2. If you change
tqchen opened a new issue #4812: [RFC] TVM Python Project Organization after
Unified IR
URL: https://github.com/apache/incubator-tvm/issues/4812
This RFC summarizes discussion in
https://discuss.tvm.ai/t/tvm-python-project-organization-after-unified-ir/5383
We will perform the follo
tqchen opened a new pull request #4813: [REFACTOR][PY] tvm._ffi
URL: https://github.com/apache/incubator-tvm/pull/4813
- Remove reverse dependency of _ctypes _cython to object_generic.
- function.py -> packed_func.py
- Function -> PackedFunc
- all registry related logics goes to tvm
tqchen commented on issue #4813: [REFACTOR][PY] tvm._ffi
URL: https://github.com/apache/incubator-tvm/pull/4813#issuecomment-582075338
cc @ZihengJiang @yzhliu @zhiics @icemelon9
This is an automated message from the Apache Gi
anijain2305 commented on issue #4786: [WIP] [TOPI][x86] Injective schedule
improvement
URL: https://github.com/apache/incubator-tvm/pull/4786#issuecomment-582095984
@yzhliu @tqchen
One test fails with this
~~~
E tvm._ffi.base.TVMError: Traceback (most recent call
MarisaKirisame commented on a change in pull request #4644: [WIP] Relay op
strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r374903457
##
File path: python/tvm/relay/backend/compile_engine.py
##
@@ -63,6 +83,316 @@ def _get_cache_key(source_func,
anijain2305 commented on issue #4786: [WIP] [TOPI][x86] Injective schedule
improvement
URL: https://github.com/apache/incubator-tvm/pull/4786#issuecomment-582111043
Yizhi helped. Vectorize works only with const extents. Added a split to make
it work.
--
tqchen commented on issue #4811: [RUNTIME] Fix memory leak when using openMP
URL: https://github.com/apache/incubator-tvm/pull/4811#issuecomment-582118143
Thanks @icemelon9
This is an automated message from the Apache Git Ser
tqchen merged pull request #4811: [RUNTIME] Fix memory leak when using openMP
URL: https://github.com/apache/incubator-tvm/pull/4811
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 change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from d279991 [AutoTVM] Minor bug fixes in AutoTVM for QNN graphs (#4797)
add 60c52e1 fix memory leak (#4811)
icemelon9 opened a new pull request #4814: [Contrib] Allow customize MKLDNN
library location
URL: https://github.com/apache/incubator-tvm/pull/4814
Thanks for contributing to TVM! Please refer to guideline
https://docs.tvm.ai/contribute/ for useful information and tips. After the pull
r
shoubhik commented on issue #4714: Mxnet parser for Qnn dialect
URL: https://github.com/apache/incubator-tvm/pull/4714#issuecomment-582138102
@tqchen @yzhliu I think the code can be merged now.
This is an automated message fr
masahi commented on a change in pull request #4714: Mxnet parser for Qnn dialect
URL: https://github.com/apache/incubator-tvm/pull/4714#discussion_r374951699
##
File path: tests/python/frontend/mxnet/test_forward.py
##
@@ -28,68 +28,173 @@
import model_zoo
+def get_glu
masahi commented on issue #4714: Mxnet parser for Qnn dialect
URL: https://github.com/apache/incubator-tvm/pull/4714#issuecomment-582140804
@shoubhik I'll have a look and merge. I'm interested in this PR.
This is an automated
abergeron opened a new pull request #4815: [TOPI][Relay] Add bitwise ops
URL: https://github.com/apache/incubator-tvm/pull/4815
We found we have a need for bitwise ops in certain cases (mostly for
implementing random number generators on the GPU), so I've added them.
There are no tes
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-582157850
CI is green now, great! Is it ready for final review? @alexwong
This i
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 60c52e1 fix memory leak (#4811)
add 4a39e52 [TOPI][x86] Injective schedule improvement (#4786)
No ne
yzhliu commented on issue #4786: [TOPI][x86] Injective schedule improvement
URL: https://github.com/apache/incubator-tvm/pull/4786#issuecomment-582165310
looks good to me. Thanks @anijain2305
This is an automated message from
yzhliu merged pull request #4786: [TOPI][x86] Injective schedule improvement
URL: https://github.com/apache/incubator-tvm/pull/4786
This is an automated message from the Apache Git Service.
To respond to the message, please l
tqchen closed issue #4764: [CI] ci-gpu update blockers
URL: https://github.com/apache/incubator-tvm/issues/4764
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
tqchen commented on issue #4813: [REFACTOR][PY] tvm._ffi
URL: https://github.com/apache/incubator-tvm/pull/4813#issuecomment-582171301
@yzhliu can you help with the name change in other related frontends?
This is an automated
tqchen edited a comment on issue #4813: [REFACTOR][PY] tvm._ffi
URL: https://github.com/apache/incubator-tvm/pull/4813#issuecomment-582171301
@yzhliu can you help with the name change in other related frontends? NOTE
we can do it per frontend
--
anijain2305 opened a new pull request #4816: [TFLite] Using real image for QNN
testing.
URL: https://github.com/apache/incubator-tvm/pull/4816
This PR does 2 things
* Test QNN on a real image. QNN and TFLite outputs differ due to rounding
differences, making random input not a good cand
yzhliu commented on issue #4813: [REFACTOR][PY] tvm._ffi
URL: https://github.com/apache/incubator-tvm/pull/4813#issuecomment-582173274
@tqchen sure
This is an automated message from the Apache Git Service.
To respond to the me
tqchen commented on issue #4749: [AutoTVM] Support to pass additional compiler
options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582188780
Likely you can create a build function(via functor that captures these
additional options) instead of directly passing ndk
-
tqchen edited a comment on issue #4749: [AutoTVM] Support to pass additional
compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582188780
Likely you can create a build function(via functor that captures these
additional options) instead of directly passing
tqchen commented on issue #4749: [AutoTVM] Support to pass additional compiler
options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582190081
Try this:
```python
from tvm.contrib import cc, ndk
build_func = cc.cross_compiler(
ndk.create_shared,
tqchen edited a comment on issue #4749: [AutoTVM] Support to pass additional
compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582190081
Try something along the following line:
```python
from tvm.contrib import cc, ndk
build_func = cc.cros
tqchen merged pull request #4813: [REFACTOR][PY] tvm._ffi
URL: https://github.com/apache/incubator-tvm/pull/4813
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
tqchen commented on issue #4813: [REFACTOR][PY] tvm._ffi
URL: https://github.com/apache/incubator-tvm/pull/4813#issuecomment-582191479
Thanks @ZihengJiang @yzhliu
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 4a39e52 [TOPI][x86] Injective schedule improvement (#4786)
add f9b46c4 [REFACTOR][PY] tvm._ffi (#4813)
tqchen edited a comment on issue #4749: [AutoTVM] Support to pass additional
compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582190081
Try something along the following line:
```python
from tvm.contrib import cc, ndk
build_func = cc.cros
FrozenGene commented on issue #4749: [AutoTVM] Support to pass additional
compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582194982
> Try something along the following line:
>
> ```python
> from tvm.contrib import cc, ndk
>
> build_func
FrozenGene commented on issue #4749: [AutoTVM] Support to pass additional
compiler options
URL: https://github.com/apache/incubator-tvm/pull/4749#issuecomment-582195632
> 1. We should not put `options` in `MeasureResult`. It is not the result
(output). It is input.
> 2. If you change Me
tqchen opened a new pull request #4817: [CONTRIB][CC] Enhance cc.cross_compiler
URL: https://github.com/apache/incubator-tvm/pull/4817
- Enhance cc.cross_compiler to take str argument.
- Remove cc.build_create_shared_func as it is dupilicated with cross_compiler
- Add examples to cc.cr
tqchen opened a new pull request #4818: [REFACTOR][PY] Establish tvm.runtime
URL: https://github.com/apache/incubator-tvm/pull/4818
This PR establishes the tvm.runtime namespace that contains the core runtime
data structures.
The top-level API are kept inact for now via re-exporting.
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from f9b46c4 [REFACTOR][PY] tvm._ffi (#4813)
add 3e7bd70 allow customize mkldnn library location (#4814)
N
jroesch merged pull request #4814: [Contrib] Allow customize MKLDNN library
location
URL: https://github.com/apache/incubator-tvm/pull/4814
This is an automated message from the Apache Git Service.
To respond to the message,
Coderx7 opened a new issue #4819: AttributeError: module 'os' has no attribute
'setsid'
URL: https://github.com/apache/incubator-tvm/issues/4819
Under windows,trying to run the example : [Auto-tuning a convolutional
network for x86
CPU](https://docs.tvm.ai/tutorials/autotvm/tune_relay_x86
Coderx7 opened a new pull request #4820: Fixed subprocess creation under windows
URL: https://github.com/apache/incubator-tvm/pull/4820
This addresses #4819
This is an automated message from the Apache Git Service.
To
Coderx7 commented on issue #4820: Fixed subprocess creation under windows
URL: https://github.com/apache/incubator-tvm/pull/4820#issuecomment-582260857
@tqchen Please review.
Thanks
This is an automated message from the Apa
kevinthesun commented on issue #4241: [Dynamic Shape] Add Graph Dispatcher
URL: https://github.com/apache/incubator-tvm/pull/4241#issuecomment-582261468
@zhiics Thank you for your review. I'll update when I get time.
This is a
alexwong commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-582263501
> CI is green now, great! Is it ready for final review? @alexwong
Yes I believe so. I think I resolved all of the smaller change
Coderx7 edited a comment on issue #4820: Fixed subprocess creation under windows
URL: https://github.com/apache/incubator-tvm/pull/4820#issuecomment-582260857
@eqy Please review.
Thanks
This is an automated message from the
FrozenGene commented on a change in pull request #4817: [CONTRIB][CC] Enhance
cc.cross_compiler
URL: https://github.com/apache/incubator-tvm/pull/4817#discussion_r375099901
##
File path: python/tvm/contrib/cc.py
##
@@ -131,14 +116,43 @@ def cross_compiler(compile_func, bas
Coderx7 opened a new issue #4821: 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: https://github.com/apache/incubator-tvm/issues/4821
T
81 matches
Mail list logo