jackwish commented on a change in pull request #4798: [QNN] Optimize lowering
for requantize and FixedPointMultiply.
URL: https://github.com/apache/incubator-tvm/pull/4798#discussion_r376682469
##
File path: src/relay/qnn/op/requantize.cc
##
@@ -103,7 +103,11 @@ Expr
FrozenGene edited a comment on issue #4828: [QNN][TFLite] TFLite rounding mode
support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-583691425
> just modified add/mul/concat requantize rounding mode and tested, no luck.
will change the default rounding behavior for a
FrozenGene commented on issue #4828: [QNN][TFLite] TFLite rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-583691425
> just modified add/mul/concat requantize rounding mode and tested, no luck.
will change the default rounding behavior for a later
u99127 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_r376673796
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -244,6 +244,74 @@ def
u99127 commented on a change in pull request #4828: [QNN][TFLite] TFLite
rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#discussion_r376672825
##
File path: src/relay/qnn/util.cc
##
@@ -100,35 +134,54 @@ Expr FixedPointMultiply(Expr tensor,
u99127 commented on a change in pull request #4828: [QNN][TFLite] TFLite
rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#discussion_r376671194
##
File path: include/tvm/relay/qnn/attrs.h
##
@@ -44,14 +44,15 @@ struct RequantizeAttrs : public
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_r376657697
##
File path: tests/python/frontend/pytorch/test_forward.py
##
@@ -0,0 +1,767 @@
+# Licensed to the
tqchen commented on issue #4824: Tflite frontend needs to use zero point of
input tensor while lowering qnn.conv2d for padding
URL: https://github.com/apache/incubator-tvm/issues/4824#issuecomment-583661376
@u99127 perhaps we can open a discuss thread on the forum. I agree with most
of
u99127 commented on issue #4824: Tflite frontend needs to use zero point of
input tensor while lowering qnn.conv2d for padding
URL: https://github.com/apache/incubator-tvm/issues/4824#issuecomment-583660260
@tqchen , @FrozenGene
[NEED-BACKPORT-0.6] ? since there's no way of
icemelon9 commented on a change in pull request #4644: [WIP] Relay op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r376652263
##
File path: python/tvm/autotvm/task/space.py
##
@@ -944,11 +939,10 @@ def from_json_dict(json_dict):
ZihengJiang commented on a change in pull request #4644: [WIP] Relay op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r376651491
##
File path: python/tvm/autotvm/task/space.py
##
@@ -944,11 +939,10 @@ def from_json_dict(json_dict):
comaniac commented on issue #4187: [AutoTVM] Selective Tuning
URL: https://github.com/apache/incubator-tvm/pull/4187#issuecomment-583651669
Close for now as the feature is not in the development process anymore.
Will integrate this feature with a more comprehensive AutoTVM improvements
comaniac closed pull request #4187: [AutoTVM] Selective Tuning
URL: https://github.com/apache/incubator-tvm/pull/4187
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.
wuwei pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 0186ca4 [COMMUNITY] comaniac -> reviewer (#4841)
add e33de24 [Doc][AutoTVM] Fix bugs that override
icemelon9 commented on a change in pull request #4644: [WIP] Relay op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r376621668
##
File path: python/tvm/autotvm/task/space.py
##
@@ -944,11 +939,10 @@ def from_json_dict(json_dict):
ZihengJiang merged pull request #4841: [COMMUNITY] comaniac -> reviewer
URL: https://github.com/apache/incubator-tvm/pull/4841
This is an automated message from the Apache Git Service.
To respond to the message, please log
This is an automated email from the ASF dual-hosted git repository.
ziheng pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from e0122c0 [REFACTOR][PY][API-Change] Polish tvm.runtime,
tvm.runtime.module API update (#4837)
add
videetparekh opened a new issue #4843: Emscripten Runtime Error while linking
URL: https://github.com/apache/incubator-tvm/issues/4843
I'm trying to get the following tutorial
(https://github.com/apache/incubator-tvm/tree/master/web) up and running within
a Docker container. I'm using a
ZihengJiang commented on a change in pull request #4644: [WIP] Relay op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#discussion_r376570164
##
File path: python/tvm/autotvm/task/space.py
##
@@ -944,11 +939,10 @@ def from_json_dict(json_dict):
tqchen closed issue #4839: Nvcc fatal : Value ‘sm_75’ is not defined for option
‘gpu-architecture’
URL: https://github.com/apache/incubator-tvm/issues/4839
This is an automated message from the Apache Git Service.
To
comaniac opened a new pull request #4842: [Doc][AutoTVM] Fix bugs that
overrides n_trials
URL: https://github.com/apache/incubator-tvm/pull/4842
In the loop of tuning a list of tasks, `n_trials` will be overwritten and
result in incorrect maximum trial number for rest tasks.
cc
tqchen commented on issue #4839: Nvcc fatal : Value ‘sm_75’ is not defined for
option ‘gpu-architecture’
URL: https://github.com/apache/incubator-tvm/issues/4839#issuecomment-583567644
Please open a new thread on https://discuss.tvm.ai/ for related questions.
In your case, it is due to
ZihengJiang opened a new pull request #4841: [COMMUNITY] comaniac -> reviewer
URL: https://github.com/apache/incubator-tvm/pull/4841
Please join me to welcome @comaniac as a new reviewer. He has been working
on improving operators in topi and quite active in code review for the
anijain2305 commented on a change in pull request #4816: [TFLite] Using real
image for QNN testing.
URL: https://github.com/apache/incubator-tvm/pull/4816#discussion_r376528698
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -1100,16 +1117,26 @@ def
anijain2305 commented on issue #4828: [QNN][TFLite] TFLite rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-583531400
> update: I forced FixedPointMultiply(PerChannel) rounding mode to be
TFLITE, but still unable to get bit-exact results.
:(
icemelon9 merged pull request #4837: [REFACTOR][PY][API-Change] Polish
tvm.runtime, tvm.runtime.module API update
URL: https://github.com/apache/incubator-tvm/pull/4837
This is an automated message from the Apache Git
This is an automated email from the ASF dual-hosted git repository.
haichen pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 30b7d83 [Frontend][TFLite] Add MIRROR_PAD operator (#4822)
add e0122c0 [REFACTOR][PY][API-Change]
jikechao removed a comment on issue #4764: [CI] ci-gpu update blockers
URL: https://github.com/apache/incubator-tvm/issues/4764#issuecomment-583411064
got it.
This is an automated message from the Apache Git Service.
To
jikechao commented on issue #4764: [CI] ci-gpu update blockers
URL: https://github.com/apache/incubator-tvm/issues/4764#issuecomment-583411064
got it.
This is an automated message from the Apache Git Service.
To respond to
uday60 commented on issue #4830: libnnvm_compiler.so file missing | Commands
inside
URL: https://github.com/apache/incubator-tvm/issues/4830#issuecomment-583401837
@tqchen
I have been using an earlier version of TVM. Now things have changed the
following script is giving me a
wyc-ruiker opened a new pull request #4840: [Frontend][TFlite] use qnn helper
function in softmax
URL: https://github.com/apache/incubator-tvm/pull/4840
Thanks for contributing to TVM! Please refer to guideline
https://docs.tvm.ai/contribute/ for useful information and tips. After the
inadob commented on a change in pull request #4816: [TFLite] Using real image
for QNN testing.
URL: https://github.com/apache/incubator-tvm/pull/4816#discussion_r376337649
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -1100,16 +1117,26 @@ def
inadob commented on a change in pull request #4816: [TFLite] Using real image
for QNN testing.
URL: https://github.com/apache/incubator-tvm/pull/4816#discussion_r375547063
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -1100,16 +1117,26 @@ def
LiangHao151941 edited a comment on issue #4828: [QNN][TFLite] TFLite rounding
mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-583328143
just modified add/mul/concat requantize rounding mode and tested, no luck.
will change the default rounding behavior for
jikechao opened a new issue #4839: Nvcc fatal : Value ‘sm_75’ is not defined
for option ‘gpu-architecture’
URL: https://github.com/apache/incubator-tvm/issues/4839
Thanks for participating in the TVM community! We use https://discuss.tvm.ai
for any general usage questions and discussions.
masahi opened a new pull request #4838: [Frontend, ONNX] Add Resize op converter
URL: https://github.com/apache/incubator-tvm/pull/4838
This is an automated message from the Apache Git Service.
To respond to the message,
FrozenGene commented on issue #4822: [Frontend][TFLite] Add MIRROR_PAD operator
URL: https://github.com/apache/incubator-tvm/pull/4822#issuecomment-583360475
Thanks @wyc-ruiker @inadob It is merged now.
This is an automated
FrozenGene merged pull request #4822: [Frontend][TFLite] Add MIRROR_PAD operator
URL: https://github.com/apache/incubator-tvm/pull/4822
This is an automated message from the Apache Git Service.
To respond to the message,
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 5a13575 [Relay][Frontend][TFlite] Add support for quantized LOGISTIC
(#4696)
add 30b7d83
FrozenGene commented on issue #4822: [Frontend][TFLite] Add MIRROR_PAD operator
URL: https://github.com/apache/incubator-tvm/pull/4822#issuecomment-583360008
> LGTM
@inadob Could you do the approve operation as this guide?
inadob commented on issue #4822: [Frontend][TFLite] Add MIRROR_PAD operator
URL: https://github.com/apache/incubator-tvm/pull/4822#issuecomment-583359091
LGTM
This is an automated message from the Apache Git Service.
To
LiangHao151941 commented on issue #4828: [QNN][TFLite] TFLite rounding mode
support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-583328143
> cc @LiangHao151941 Could you try this way to do bit-extract testing? For
conveniently, you could simply change the C++
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 4c961dd [Doc] ConvertLayout - Call RemoveUnunsedFunctions.
add 5a13575 [Relay][Frontend][TFlite] Add
FrozenGene merged pull request #4696: [Relay][Frontend][TFlite] Add support for
quantized LOGISTIC
URL: https://github.com/apache/incubator-tvm/pull/4696
This is an automated message from the Apache Git Service.
To respond
FrozenGene commented on issue #4696: [Relay][Frontend][TFlite] Add support for
quantized LOGISTIC
URL: https://github.com/apache/incubator-tvm/pull/4696#issuecomment-583327668
Thanks @inadob @anijain2305 @u99127 This is merged.
FrozenGene commented on issue #4828: [QNN][TFLite] TFLite rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-583326430
> I like this way
>
> > Another way is we provide rounding args for qnn.add / qnn.mul /
qnn.conconcate, because they use
LiangHao151941 commented on a change in pull request #4828: [QNN][TFLite]
TFLite rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#discussion_r376313583
##
File path: include/tvm/relay/qnn/attrs.h
##
@@ -44,14 +44,16 @@ struct RequantizeAttrs :
LiangHao151941 commented on a change in pull request #4828: [QNN][TFLite]
TFLite rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#discussion_r376311994
##
File path: src/relay/qnn/util.cc
##
@@ -22,13 +22,49 @@
* \brief Utility functions for
FrozenGene commented on a change in pull request #4828: [QNN][TFLite] TFLite
rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#discussion_r376286808
##
File path: include/tvm/relay/qnn/attrs.h
##
@@ -44,14 +44,16 @@ struct RequantizeAttrs :
anijain2305 commented on issue #4828: [QNN][TFLite] TFLite rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-583292788
I like this way
> Another way is we provide rounding args for qnn.add / qnn.mul /
qnn.conconcate, because they use requantize in
wweic merged pull request #4834: [Doc] ConvertLayout - Call
RemoveUnunsedFunctions.
URL: https://github.com/apache/incubator-tvm/pull/4834
This is an automated message from the Apache Git Service.
To respond to the message,
This is an automated email from the ASF dual-hosted git repository.
wweic pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from e578777 Improve tol to resolve flaky case (#4836)
add 4c961dd [Doc] ConvertLayout - Call
FrozenGene commented on issue #4828: [QNN][TFLite] TFLite rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-583285949
> I thought little more about the bit exact problem. One source of
discrepancy for certain is QNN add, and QNN concatenate ops.
kazum 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_r376260979
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -0,0 +1,1036 @@
+# Licensed to the Apache
FrozenGene commented on a change in pull request #4828: [QNN][TFLite] TFLite
rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#discussion_r376258392
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -1812,7 +1816,7 @@ def
FrozenGene commented on a change in pull request #4828: [QNN][TFLite] TFLite
rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#discussion_r376257846
##
File path: src/relay/qnn/util.cc
##
@@ -22,13 +22,49 @@
* \brief Utility functions for QNN.
56 matches
Mail list logo