inadob opened a new pull request #4805: [Frontend][TFlite] Add parser support
for relu6, leeaky_relu, relu_n1_to_1, log_softmax
URL: https://github.com/apache/incubator-tvm/pull/4805
**This patch is made on top of #4789.**
* Add helper functions in the parser to quantize and dequantize
wweic commented on a change in pull request #4628: [Object] Add String container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r373981359
##
File path: include/tvm/runtime/container.h
##
@@ -274,6 +276,137 @@ class ADT : public ObjectRef {
wweic commented on a change in pull request #4628: [Object] Add String container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r373976757
##
File path: include/tvm/runtime/container.h
##
@@ -274,6 +276,137 @@ class ADT : public ObjectRef {
wweic commented on a change in pull request #4628: [Object] Add String container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r373979652
##
File path: include/tvm/runtime/container.h
##
@@ -274,6 +276,137 @@ class ADT : public ObjectRef {
FrozenGene commented on a change in pull request #4747: [ThreadPool] Solve ARM
BIG.LITTLE heterogeneous multicores
URL: https://github.com/apache/incubator-tvm/pull/4747#discussion_r373962677
##
File path: src/runtime/threading_backend.cc
##
@@ -133,34 +133,42 @@ class
FrozenGene commented on a change in pull request #4747: [ThreadPool] Solve ARM
BIG.LITTLE heterogeneous multicores
URL: https://github.com/apache/incubator-tvm/pull/4747#discussion_r373962704
##
File path: src/runtime/threading_backend.cc
##
@@ -133,34 +133,42 @@ class
inadob opened a new pull request #4807: [Frontend][TFLite] Fix quantized pad
value for convolution
URL: https://github.com/apache/incubator-tvm/pull/4807
Zero-padding can't be done directly in case the model is quantized, so we
have to set
pad_value = zero_point
FrozenGene commented on issue #4803: [WIP][Frontend] Asymmetric padding of
convolution support
URL: https://github.com/apache/incubator-tvm/pull/4803#issuecomment-581527290
> I ran into this issue when I used a quantized model with convolution
#4807. Is this helpful?
Right! This is
FrozenGene 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_r374205954
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -384,7 +401,16 @@ def
anijain2305 commented on issue #4803: [WIP][Frontend] Asymmetric padding of
convolution support
URL: https://github.com/apache/incubator-tvm/pull/4803#issuecomment-581500134
@FrozenGene I will take a look in few hours (traveling right now). Happy to
help.
Coderx7 edited a comment on issue #4802: Building tvm using llvm on windows 10
fails (ninja: build stopped: subcommand failed.)
URL: https://github.com/apache/incubator-tvm/issues/4802#issuecomment-581227116
@tqchen Thanks a lot I removed the GCC from PATH and it detected clang/llvm.
comaniac commented on a change in pull request #4771: [Relay] Added Merge
Composite pass
URL: https://github.com/apache/incubator-tvm/pull/4771#discussion_r374225876
##
File path: python/tvm/relay/transform.py
##
@@ -513,6 +513,49 @@ def
u99127 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_r374233836
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -384,7 +401,16 @@ def
anijain2305 commented on a change in pull request #4797: [AutoTVM] Minor bug
fixes in AutoTVM for QNN graphs
URL: https://github.com/apache/incubator-tvm/pull/4797#discussion_r374311898
##
File path: python/tvm/autotvm/graph_tuner/utils/utils.py
##
@@ -73,7 +73,7 @@ def
jwfromm commented on issue #4756: [Docker] Update torch version to 1.4
URL: https://github.com/apache/incubator-tvm/pull/4756#issuecomment-581564923
I don't think I've tried that version combo. If it's working for you then
I'd say let's update!
tqchen merged pull request #4806: [TIR] Create a StringImm reference type
URL: https://github.com/apache/incubator-tvm/pull/4806
This is an automated message from the Apache Git Service.
To respond to the message, please log
tqchen commented on issue #4806: [TIR] Create a StringImm reference type
URL: https://github.com/apache/incubator-tvm/pull/4806#issuecomment-581537448
Thanks @mbarrett97 This is Approved. However, we would recommend to switch
to #4628 once it is ready cc @wweic
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 bb1b7db [ThreadPool] Solve ARM BIG.LITTLE heterogeneous multicores
(#4747)
add c39ab93 [TIR] Create a
icemelon9 commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374229865
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
inadob commented on issue #4803: [WIP][Frontend] Asymmetric padding of
convolution support
URL: https://github.com/apache/incubator-tvm/pull/4803#issuecomment-581520870
I ran into this issue when I used a quantized model with convolution #4807.
Is this helpful?
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 c21d1ee Change color channel from BGR to RGB for darknet
preprocessing (#4794)
add bb1b7db
tqchen merged pull request #4747: [ThreadPool] Solve ARM BIG.LITTLE
heterogeneous multicores
URL: https://github.com/apache/incubator-tvm/pull/4747
This is an automated message from the Apache Git Service.
To respond to the
tmoreau89 merged pull request #4791: [TOPI] upsample operator 'NCHWinic' format
support.
URL: https://github.com/apache/incubator-tvm/pull/4791
This is an automated message from the Apache Git Service.
To respond to the
tmoreau89 commented on issue #4791: [TOPI] upsample operator 'NCHWinic' format
support.
URL: https://github.com/apache/incubator-tvm/pull/4791#issuecomment-581581745
Thanks @huajsj @Laurawly @jwfromm the PR has been merged.
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 c39ab93 [TIR] Create a StringImm reference type (#4806)
add 974195d [TOPI] upsample operator
tqchen commented on issue #4804: [LINT] Fix -Wextra
URL: https://github.com/apache/incubator-tvm/pull/4804#issuecomment-581581202
cc @zhiics @ZihengJiang
This is an automated message from the Apache Git Service.
To respond
mbarrett97 opened a new pull request #4806: [TIR] Create a StringImm reference
type
URL: https://github.com/apache/incubator-tvm/pull/4806
This is motivated by the want to send an array of strings across the
python/C++ boundary. Arrays only support ObjectRef types and so can't carry
FrozenGene commented on issue #4803: [WIP][Frontend] Asymmetric padding of
convolution support
URL: https://github.com/apache/incubator-tvm/pull/4803#issuecomment-581488095
Initial Investigation:
For QNN, before asymmetric padding support, we will handle it in tflite
frontend
mbarrett97 commented on issue #4771: [Relay] Added Merge Composite pass
URL: https://github.com/apache/incubator-tvm/pull/4771#issuecomment-581489473
I've updated the PR to account for the priority ordering of patterns. You no
longer specify a pattern map, but a list of tuples
inadob opened a new pull request #4696: [Relay][Frontend][TFlite] Add support
for quantized LOGISTIC
URL: https://github.com/apache/incubator-tvm/pull/4696
* add qnn implementation in the parser
* add test case for qnn logistic
FrozenGene 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_r374193768
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -384,7 +401,16 @@ def
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_r374201180
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -384,7 +401,16 @@ def
tqchen commented on issue #4764: [CI] ci-gpu update blockers
URL: https://github.com/apache/incubator-tvm/issues/4764#issuecomment-581628702
OK, seems the conclusion so far is not to use the MKL DNN version. If we
confirm that is the case, I we can go ahead and just update the pytorch and
tqchen closed pull request #4753: [Docker] Updating dependency on Mxnet-mkl
URL: https://github.com/apache/incubator-tvm/pull/4753
This is an automated message from the Apache Git Service.
To respond to the message, please
anijain2305 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_r374334979
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -384,7 +401,16 @@ def
masahi edited a comment on issue #4756: [Docker] Update torch version to 1.4
URL: https://github.com/apache/incubator-tvm/pull/4756#issuecomment-581640387
@jwfromm I posted what I think is the solution to "onnx segfault problem" at
soichih commented on issue #4061: nvcc fatal : Value 'sm_75' is not defined
for option 'gpu-architecture'
URL: https://github.com/apache/incubator-tvm/issues/4061#issuecomment-581649711
I tried tensorflow 1.13.1 and 1.14.0 but I am still seeing this error
message. Is this an issue
tom-gall opened a new issue #4808: relay.build fails with NHWC not supported
using OpenCL on intel
URL: https://github.com/apache/incubator-tvm/issues/4808
Hi!
New to TVM yet, so pointers to the fine manual are certainly much
appreciated.
I'm trying to use OpenCL on intel.
anijain2305 commented on issue #4642: [Relay][Frontend][TFLite] Add parser
support for logical operators
URL: https://github.com/apache/incubator-tvm/pull/4642#issuecomment-581611590
@inadob Can you please rebase?
@yzhliu Is conditional check for TFLite version good for now?
anijain2305 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_r374358802
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -826,6 +828,50 @@ def
anijain2305 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_r374357079
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -826,6 +828,50 @@ def
masahi commented on issue #4756: [Docker] Update torch version to 1.4
URL: https://github.com/apache/incubator-tvm/pull/4756#issuecomment-581640387
@jwfromm I posted what I think is the solution to "onnx segfault problem" at
https://github.com/onnx/onnx/issues/2394. I encourage you to try
shoubhik commented on issue #4714: Mxnet parser for Qnn dialect
URL: https://github.com/apache/incubator-tvm/pull/4714#issuecomment-581672885
@tmoreau89 @yzhliu @tqchen @jackwish @FrozenGene @liangfu @vinx13
@anijain2305 I want to discuss this PR from testability POV. Mxnet suggests to
anijain2305 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_r374335873
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -143,11
anijain2305 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_r374350076
##
File path: tests/python/frontend/tflite/test_forward.py
##
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374448684
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374447787
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374450303
##
File path: tests/cpp/container_test.cc
##
@@ -226,6 +226,63 @@ TEST(Map, Iterator) {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374450428
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
zhiics merged pull request #4804: [LINT] Fix -Wextra
URL: https://github.com/apache/incubator-tvm/pull/4804
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
This is an automated email from the ASF dual-hosted git repository.
zhic pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 974195d [TOPI] upsample operator 'NCHWinic' format support. (#4791)
add 6f7d6fa [LINT] Fix -Wextra
wweic commented on a change in pull request #4628: [Object] Add String container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374420184
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
anijain2305 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_r374425432
##
File path: src/relay/qnn/op/requantize.cc
##
@@ -103,7 +103,11 @@ Expr
anijain2305 commented on a change in pull request #4797: [AutoTVM] Minor bug
fixes in AutoTVM for QNN graphs
URL: https://github.com/apache/incubator-tvm/pull/4797#discussion_r374441091
##
File path: python/tvm/autotvm/graph_tuner/utils/utils.py
##
@@ -73,7 +73,7 @@ def
jwfromm commented on issue #4756: [Docker] update onnx to 1.6 and torch to 1.4
URL: https://github.com/apache/incubator-tvm/pull/4756#issuecomment-581691232
You're right that importing onnx first resolves the issue. I tried it out
with a few different pytorch and onnx version combinations
tqchen commented on issue #4756: [Docker] update onnx to 1.6 and torch to 1.4
URL: https://github.com/apache/incubator-tvm/pull/4756#issuecomment-581712105
OK, will report back in one to two days
This is an automated message
tqchen closed issue #4808: relay.build fails with NHWC not supported using
OpenCL on intel
URL: https://github.com/apache/incubator-tvm/issues/4808
This is an automated message from the Apache Git Service.
To respond to the
tqchen commented on issue #4808: relay.build fails with NHWC not supported
using OpenCL on intel
URL: https://github.com/apache/incubator-tvm/issues/4808#issuecomment-581718440
Thannks for reporting the problem, please open a new troubleshooting thread
on https://discuss.tvm.ai/
Coderx7 edited a comment on issue #4802: Building tvm using llvm on windows 10
fails (ninja: build stopped: subcommand failed.)
URL: https://github.com/apache/incubator-tvm/issues/4802#issuecomment-581227116
@tqchen Thanks a lot I removed the GCC from PATH and it detected clang/llvm.
Coderx7 edited a comment on issue #4802: Building tvm using llvm on windows 10
fails (ninja: build stopped: subcommand failed.)
URL: https://github.com/apache/incubator-tvm/issues/4802#issuecomment-581227116
@tqchen Thanks a lot I removed the GCC from PATH and it detected clang/llvm.
Coderx7 edited a comment on issue #4802: Building tvm using llvm on windows 10
fails (ninja: build stopped: subcommand failed.)
URL: https://github.com/apache/incubator-tvm/issues/4802#issuecomment-581227116
@tqchen Thanks a lot I removed the GCC from PATH and it detected clang/llvm.
siju-samuel commented on issue #4788: [FRONTEND][TFLITE]Gather, StridedSlice op
support added
URL: https://github.com/apache/incubator-tvm/pull/4788#issuecomment-581737945
@FrozenGene @u99127 Thanks for the review and could you please review
again.
I have updated as per your review
siju-samuel 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_r374469548
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -244,6 +244,57 @@
Coderx7 removed a comment on issue #754: How to set the llvm-config & cuda
path for windows
URL: https://github.com/apache/incubator-tvm/issues/754#issuecomment-581104886
@prob1995 may I ask, how you compiled LLVM? after downloading the source,
what sets of commands did you use?
anijain2305 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_r374425432
##
File path: src/relay/qnn/op/requantize.cc
##
@@ -103,7 +103,11 @@ Expr
masahi commented on a change in pull request #4790: Fast exponent
URL: https://github.com/apache/incubator-tvm/pull/4790#discussion_r374431619
##
File path: topi/include/topi/elemwise.h
##
@@ -360,5 +359,66 @@ inline Tensor full_like(const Tensor& x,
}, name, tag);
}
masahi commented on issue #4756: [Docker] update onnx to 1.6 and torch to 1.4
URL: https://github.com/apache/incubator-tvm/pull/4756#issuecomment-581697770
@tqchen we are ready to upgrade!
This is an automated message from
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374449514
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374448205
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374448894
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374448310
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374448029
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374447406
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374450097
##
File path: tests/cpp/container_test.cc
##
@@ -226,6 +226,63 @@ TEST(Map, Iterator) {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374448735
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374449717
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374449910
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374447314
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen commented on a change in pull request #4628: [Object] Add String
container
URL: https://github.com/apache/incubator-tvm/pull/4628#discussion_r374448792
##
File path: include/tvm/runtime/container.h
##
@@ -274,7 +276,181 @@ class ADT : public ObjectRef {
tqchen opened a new pull request #4809: [DOCS] Fix vta tutorial
URL: https://github.com/apache/incubator-tvm/pull/4809
cc @tmoreau89
This is an automated message from the Apache Git Service.
To respond to the message, please
anijain2305 commented on issue #4803: [WIP][Frontend] Asymmetric padding of
convolution support
URL: https://github.com/apache/incubator-tvm/pull/4803#issuecomment-581764770
So, if I understand correctly there are 2 issues here
* If you try to pass the asymmetric padding via QNN
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_r374497747
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -244,6 +244,74 @@ def
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_r374498740
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -244,6 +244,74 @@ def
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_r374499522
##
File path: tests/python/frontend/tflite/test_forward.py
##
@@ -244,6 +244,74 @@ def
anijain2305 commented on issue #4714: Mxnet parser for Qnn dialect
URL: https://github.com/apache/incubator-tvm/pull/4714#issuecomment-581769804
Given the constraints, I am ok with MxNet-parser code w/o any MxNet-MKLDNN
CI testing, as long as we have a test script linked in the PR, that
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_r374500086
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -792,6 +794,147 @@ def
anijain2305 commented on issue #4807: [Frontend][TFLite] Fix quantized pad
value for convolution
URL: https://github.com/apache/incubator-tvm/pull/4807#issuecomment-581766259
Excellent find! I missed this somehow.
I think the current test case failure might be because we give random
87 matches
Mail list logo