masahi commented on a change in pull request #7099:
URL: https://github.com/apache/tvm/pull/7099#discussion_r543096686
##
File path: python/tvm/driver/build_module.py
##
@@ -277,7 +277,7 @@ def _build_for_device(input_mod, target, target_host):
lambda f: "calli
masahi commented on a change in pull request #7099:
URL: https://github.com/apache/tvm/pull/7099#discussion_r543096686
##
File path: python/tvm/driver/build_module.py
##
@@ -277,7 +277,7 @@ def _build_for_device(input_mod, target, target_host):
lambda f: "calli
masahi commented on a change in pull request #7099:
URL: https://github.com/apache/tvm/pull/7099#discussion_r543096686
##
File path: python/tvm/driver/build_module.py
##
@@ -277,7 +277,7 @@ def _build_for_device(input_mod, target, target_host):
lambda f: "calli
mbrookhart commented on pull request #7099:
URL: https://github.com/apache/tvm/pull/7099#issuecomment-745088032
Many thanks to @masahi for helping me find an issue with heterogeneous
lowering and some overflow issues in how I was handling the threads. I think it
should be ready for review
tsupei commented on pull request #7111:
URL: https://github.com/apache/tvm/pull/7111#issuecomment-745079580
@insop thanks for your help! added!
This is an automated message from the Apache Git Service.
To respond to the messa
insop commented on pull request #7111:
URL: https://github.com/apache/tvm/pull/7111#issuecomment-745054063
1. @tsupei Good fix, you might want to `git am` [this patch
file](https://github.com/insop/incubator-tvm/commit/79e8cf5b24e5170fbfc51461249425ad6b868a7c.patch)
to fix other locations
roastduck commented on issue #848:
URL: https://github.com/apache/tvm/issues/848#issuecomment-745041181
Sorry for bringing up the issue again. I think the top-level-reduction-only
restriction still exists in TVM today.
I noticed that `relay.transform.transform.FuseOps` fuses a `conv2
ANSHUMAN87 commented on pull request #6889:
URL: https://github.com/apache/tvm/pull/6889#issuecomment-745028481
> @ANSHUMAN87 please go ahead
Thanks a lot @FrozenGene, @tkonolige!
This is an automated message from the
FrozenGene commented on pull request #6889:
URL: https://github.com/apache/tvm/pull/6889#issuecomment-745018756
@ANSHUMAN87 please go ahead
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 main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 054466b [ONNX] NMS in ONNX (#6839)
add 862655b [TOPI] sparse_dense Op sparse_data input added (#6889)
No new rev
FrozenGene merged pull request #6889:
URL: https://github.com/apache/tvm/pull/6889
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to t
tsupei commented on pull request #7111:
URL: https://github.com/apache/tvm/pull/7111#issuecomment-74501
cc @zhiics
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Git
tsupei opened a new pull request #7111:
URL: https://github.com/apache/tvm/pull/7111
This PR attempts to solve #7102
This is an automated message from the Apache Git Service.
To respond to the message, please log on to G
tsupei commented on issue #7102:
URL: https://github.com/apache/tvm/issues/7102#issuecomment-745008162
@zhiics Sure! I will send it later
This is an automated message from the Apache Git Service.
To respond to the message, pl
masahi commented on pull request #7109:
URL: https://github.com/apache/tvm/pull/7109#issuecomment-744987233
@mbrookhart please fix the lint issue
This is an automated message from the Apache Git Service.
To respond to the mes
codeislife99 opened a new pull request #7110:
URL: https://github.com/apache/tvm/pull/7110
Thanks for contributing to TVM! Please refer to guideline
https://tvm.apache.org/docs/contribute/ for useful information and tips. After
the pull request is submitted, please request code reviews f
lixiaoquan commented on pull request #6890:
URL: https://github.com/apache/tvm/pull/6890#issuecomment-744947668
@kevinthesun Sorry for late reply, I've enhanced get_name, please take a
look.
This is an automated message
tqchen edited a comment on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744864004
Thanks @tkonolige . First of all, I think it is very useful to be able to
support additional meta data like names when the user can provid eit.
On the other hand, attach
tqchen commented on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744864004
Thanks @tkonolige . First of all, I think it is very useful to be able to
support additional meta data like names when the user can provid eit.
On the other hand, attaching a S
zhiics commented on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744851482
BTW, can you add a unit test? I think you can add it to the existing gtest.
This is an automated message from the Apa
tkonolige commented on a change in pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#discussion_r542915733
##
File path: include/tvm/runtime/packed_func.h
##
@@ -293,6 +304,8 @@ class TypedPackedFunc {
friend class TVMRetValue;
/*! \brief The internal packe
d-smirnov commented on pull request #7106:
URL: https://github.com/apache/tvm/pull/7106#issuecomment-744758126
Yes, should be just OK. I believe.
--
Dmitriy Smirnov
On Mon, 14 Dec 2020 at 21:54, Zhi wrote:
> can we enable the runtime tests since I already added an AR
mbrookhart opened a new pull request #7109:
URL: https://github.com/apache/tvm/pull/7109
cc @jwfromm @masahi
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub an
zhiics commented on a change in pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#discussion_r542854713
##
File path: include/tvm/runtime/packed_func.h
##
@@ -293,6 +304,8 @@ class TypedPackedFunc {
friend class TVMRetValue;
/*! \brief The internal packed f
zhiics commented on a change in pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#discussion_r542854713
##
File path: include/tvm/runtime/packed_func.h
##
@@ -293,6 +304,8 @@ class TypedPackedFunc {
friend class TVMRetValue;
/*! \brief The internal packed f
junrushao1994 commented on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744734611
Oh nice! I didn't notice that :-)
This is an automated message from the Apache Git Service.
To respond to the
tkonolige commented on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744734355
That's what I do here:
https://github.com/apache/tvm/pull/7108/files#diff-0fe12f71cc9718f4bf446609180e9c13ca870f0a1ecf080f330b7af8743a0d43R61
zhiics commented on pull request #7106:
URL: https://github.com/apache/tvm/pull/7106#issuecomment-744734193
can we enable the runtime tests since I already added an ARM instance to the
CI?
This is an automated message from t
junrushao1994 edited a comment on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744728049
In file `runtime/registry.h`, Right now the macro is:
```C++
#define TVM_REGISTER_GLOBAL(OpName) \
TVM_STR_CONCAT(TVM_FUNC_REG_VAR_DEF, __COUNTER__) =
::t
junrushao1994 commented on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744728049
In file `runtime/registry.h`, Right now the macro is:
```C++
#define TVM_REGISTER_GLOBAL(OpName) \
TVM_STR_CONCAT(TVM_FUNC_REG_VAR_DEF, __COUNTER__) =
::tvm::run
junrushao1994 commented on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744726320
Shall we accomplish this more automatically by modifying the
`TVM_REGISTER_GLOBAL` macro itself?
This is an a
tkonolige commented on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744723606
This PR changes `TVM_REGISTER_GLOBAL` so that is sets the name of the packed
func. I can do the same on the python side.
-
junrushao1994 commented on pull request #7108:
URL: https://github.com/apache/tvm/pull/7108#issuecomment-744722859
Adding function name to packed functions definitely make sense to me. I
wonder if we can go a bit further:
* On C++ side, enhance `TVM_REGISTER_GLOBAL` to set the name of th
tkonolige opened a new pull request #7108:
URL: https://github.com/apache/tvm/pull/7108
This PR adds a `name` field to both `PackedFunc` and `TypedPackedFunc`. This
name defaults to ``, but I've also added code that sets the name for
`TVM_REGISTER_GLOBAL().set_body()` and
`TVM_REGISTER_GL
zhiics closed issue #7103:
URL: https://github.com/apache/tvm/issues/7103
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specif
zhiics commented on issue #7103:
URL: https://github.com/apache/tvm/issues/7103#issuecomment-744642013
Thanks for your interests in TVM. Please create a thread at the discuss
forum (https://discuss.tvm.apache.org/) and ask above questions there since
they are not issues. Thanks.
zhiics commented on issue #7104:
URL: https://github.com/apache/tvm/issues/7104#issuecomment-744640934
@mbaret @manupa-arm
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
zhiics commented on pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#issuecomment-744640011
Thanks everyone!
This is an automated message from the Apache Git Service.
To respond to the message, please log on t
This is an automated email from the ASF dual-hosted git repository.
zhic pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git
The following commit(s) were added to refs/heads/main by this push:
new 054466b [ONNX] NMS in ONNX (#6839)
054466b is described
zhiics merged pull request #6839:
URL: https://github.com/apache/tvm/pull/6839
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the s
zhiics commented on issue #7102:
URL: https://github.com/apache/tvm/issues/7102#issuecomment-744637958
Thanks for reporting the issue. I think it is `te.max`. Could you send a PR?
This is an automated message from the Apache
mbrookhart commented on pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#issuecomment-744636552
Awesome, thank you for the help @Laurawly. The idea for speeding up get
valid counts was especially helpful. I have an idea for parallelizing the last
serial function there, but
electriclilies commented on pull request #7086:
URL: https://github.com/apache/tvm/pull/7086#issuecomment-744635938
@u99127 I think this change won't remove any nodes from the relay IR, since
reverse_reshape just constructs a reshape node with the reverse attribute set
to true. There is no
This is an automated email from the ASF dual-hosted git repository.
moreau pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from c213ac2 [metal] support int64 (#7105)
add 6be4d0a Update tune_relay_vta.py to support single board (#7100)
No new
tmoreau89 commented on pull request #7100:
URL: https://github.com/apache/tvm/pull/7100#issuecomment-744615859
The PR has been merged.
This is an automated message from the Apache Git Service.
To respond to the message, pleas
tmoreau89 merged pull request #7100:
URL: https://github.com/apache/tvm/pull/7100
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to th
csullivan commented on a change in pull request #7099:
URL: https://github.com/apache/tvm/pull/7099#discussion_r54203
##
File path: python/tvm/topi/cuda/sort.py
##
@@ -94,64 +96,155 @@ def sort_ir(data, values_out, axis, is_ascend,
indices_out=None):
axis_mul_
ANSHUMAN87 commented on pull request #7107:
URL: https://github.com/apache/tvm/pull/7107#issuecomment-744606857
> I'm a little confused as to what is going on here. It seems like you've
moved the input data generation from one function to another? Could you provide
a little description of
This is an automated email from the ASF dual-hosted git repository.
junrushao pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 28647f2 clean standalone CRT files in microTVM VM rebuild script
(#7095)
add c213ac2 [metal] support int64 (#71
ANSHUMAN87 commented on a change in pull request #7107:
URL: https://github.com/apache/tvm/pull/7107#discussion_r542585227
##
File path: tutorials/frontend/deploy_sparse.py
##
@@ -314,13 +310,20 @@ def run_sparse(mod, params, shape_dict, target, ctx,
bs_r, sparsity, gen_weight
tkonolige commented on a change in pull request #7107:
URL: https://github.com/apache/tvm/pull/7107#discussion_r542582066
##
File path: tutorials/frontend/deploy_sparse.py
##
@@ -314,13 +310,20 @@ def run_sparse(mod, params, shape_dict, target, ctx,
bs_r, sparsity, gen_weights
junrushao1994 merged pull request #7105:
URL: https://github.com/apache/tvm/pull/7105
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go t
ANSHUMAN87 opened a new pull request #7107:
URL: https://github.com/apache/tvm/pull/7107
cc @tkonolige , @jwfromm !
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Git
tkonolige commented on a change in pull request #7099:
URL: https://github.com/apache/tvm/pull/7099#discussion_r542534776
##
File path: python/tvm/topi/cuda/sort.py
##
@@ -62,7 +62,9 @@ def traverse(op):
return s
-def sort_ir(data, values_out, axis, is_ascend, indices_
ANSHUMAN87 commented on pull request #6889:
URL: https://github.com/apache/tvm/pull/6889#issuecomment-744495232
Gentle ping @antinucleon , @FrozenGene , @tqchen , @jroesch !
Unless if any further comments, I think we can proceed with the merge, as we
have greens from @tkonolige , @Frozen
euntaik commented on a change in pull request #7093:
URL: https://github.com/apache/tvm/pull/7093#discussion_r542391750
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -1991,20 +1994,29 @@ def convert_conv(self, op, conv_type):
weight_tensor_type_str = self.get
tqchen edited a comment on pull request #7098:
URL: https://github.com/apache/tvm/pull/7098#issuecomment-744129790
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub an
giuseros commented on a change in pull request #7093:
URL: https://github.com/apache/tvm/pull/7093#discussion_r542390418
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -1991,20 +1994,29 @@ def convert_conv(self, op, conv_type):
weight_tensor_type_str = self.ge
euntaik commented on a change in pull request #7093:
URL: https://github.com/apache/tvm/pull/7093#discussion_r542268005
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -1991,20 +1994,29 @@ def convert_conv(self, op, conv_type):
weight_tensor_type_str = self.get
leandron commented on pull request #7106:
URL: https://github.com/apache/tvm/pull/7106#issuecomment-744343419
cc @tqchen, as we need to re-create the Docker images.
This is an automated message from the Apache Git Service.
To
d-smirnov opened a new pull request #7106:
URL: https://github.com/apache/tvm/pull/7106
This PR switches ACL version from 20.05 to 20.11
This is an automated message from the Apache Git Service.
To respond to the message, ple
antinucleon opened a new pull request #7105:
URL: https://github.com/apache/tvm/pull/7105
Thanks for contributing to TVM! Please refer to guideline
https://tvm.apache.org/docs/contribute/ for useful information and tips. After
the pull request is submitted, please request code reviews fr
giuseros commented on a change in pull request #7093:
URL: https://github.com/apache/tvm/pull/7093#discussion_r542233725
##
File path: python/tvm/relay/frontend/tflite.py
##
@@ -1991,20 +1994,29 @@ def convert_conv(self, op, conv_type):
weight_tensor_type_str = self.ge
dlexplorer opened a new issue #7104:
URL: https://github.com/apache/tvm/issues/7104
The [ONNX inception v1
](https://github.com/onnx/models/tree/master/vision/classification/inception_and_googlenet/inception_v1)
been compiled with ACL integration crashes on the Android
The same network
jay-karan opened a new issue #7103:
URL: https://github.com/apache/tvm/issues/7103
Hi, I am new to this space. I am looking forward to contribute after a deep
view.
1. How do I analyze the operators of a particular framework (say onnx,
tensorflow) are being supported in Relay? Is th
65 matches
Mail list logo
