gigiblender opened a new pull request, #13776:
URL: https://github.com/apache/tvm/pull/13776
This PR is a prerequisite to #13714 and needs to be merged before. It
contains the changes to the ci_minimal docker image to support minimal
cross-compilation of TVM to aarch64.
--
This is an aut
gigiblender commented on code in PR #13714:
URL: https://github.com/apache/tvm/pull/13714#discussion_r1068650074
##
tests/scripts/task_config_build_minimal_cross_isa.sh:
##
@@ -0,0 +1,49 @@
+#!/usr/bin/env bash
+# Licensed to the Apache Software Foundation (ASF) under one
+# or
AndrewZhaoLuo merged PR #13758:
URL: https://github.com/apache/tvm/pull/13758
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.ap
slyubomirsky commented on PR #13769:
URL: https://github.com/apache/tvm/pull/13769#issuecomment-1380902667
The problem seems to be that this cast on mainline will not convince Python
that the result is an opaque Object rather than a PackedFunc. Not sure how to
change it:
```python
TV
NicolaLancellotti commented on code in PR #13762:
URL: https://github.com/apache/tvm/pull/13762#discussion_r1068501892
##
gallery/how_to/work_with_microtvm/install_cmsis.rst:
##
@@ -1,35 +0,0 @@
-.. Licensed to the Apache Software Foundation (ASF) under one
Review Comment:
driazati commented on code in PR #13714:
URL: https://github.com/apache/tvm/pull/13714#discussion_r1068498620
##
tests/scripts/task_config_build_minimal_cross_isa.sh:
##
@@ -0,0 +1,49 @@
+#!/usr/bin/env bash
+# Licensed to the Apache Software Foundation (ASF) under one
+# or mor
AndrewZhaoLuo commented on PR #13754:
URL: https://github.com/apache/tvm/pull/13754#issuecomment-1380867142
https://github.com/apache/tvm/pull/13775 <-- is probably the best solution
to my problem.
--
This is an automated message from the Apache Git Service.
To respond to the message, ple
slyubomirsky commented on PR #13769:
URL: https://github.com/apache/tvm/pull/13769#issuecomment-1380866581
Hm, this test worked in my fork. Wonder why it fails in mainline
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and us
tvm-bot commented on PR #13775:
URL: https://github.com/apache/tvm/pull/13775#issuecomment-1380865994
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
AndrewZhaoLuo opened a new pull request, #13775:
URL: https://github.com/apache/tvm/pull/13775
The `key` in a target is used in dispatching to different strategies.
In most of the code base `arm_cpu` --> `cpu`. E.g. in
`python/tvm/target/target.py`:
```
def arm_cpu(...):
driazati merged PR #13738:
URL: https://github.com/apache/tvm/pull/13738
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.
driazati merged PR #13689:
URL: https://github.com/apache/tvm/pull/13689
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.
vvchernov commented on code in PR #13621:
URL: https://github.com/apache/tvm/pull/13621#discussion_r1068445499
##
tests/python/integration/test_auto_tensorize.py:
##
@@ -278,6 +283,17 @@ def test_vnni_dense():
)
+# TODO(vvchernov): need schedule rules and postprocs for
mehrdadh commented on code in PR #13762:
URL: https://github.com/apache/tvm/pull/13762#discussion_r1068381448
##
gallery/how_to/work_with_microtvm/install_cmsis.rst:
##
@@ -1,35 +0,0 @@
-.. Licensed to the Apache Software Foundation (ASF) under one
Review Comment:
Please do
lhutton1 commented on code in PR #13653:
URL: https://github.com/apache/tvm/pull/13653#discussion_r1068313668
##
docker/install/ubuntu_install_ethosu_driver_stack.sh:
##
@@ -78,6 +78,17 @@ mkdir -p "${ethosu_dir}"
cd "${ethosu_dir}"
git clone --branch ${ethosu_driver_ver}
"ht
tvm-bot commented on PR #13774:
URL: https://github.com/apache/tvm/pull/13774#issuecomment-1380581889
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Lunderberg opened a new pull request, #13774:
URL: https://github.com/apache/tvm/pull/13774
Parametrization helped in the debugging of
https://github.com/apache/tvm/pull/13530, but is not otherwise related to that
PR.
--
This is an automated message from the Apache Git Service.
To respon
tvm-bot commented on PR #13773:
URL: https://github.com/apache/tvm/pull/13773#issuecomment-1380470400
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Lunderberg opened a new pull request, #13773:
URL: https://github.com/apache/tvm/pull/13773
The `crop_and_resize` operator uses floating-point arithmetic to determine
whether an index is within a view-box. This can cause the use of
`extrapolation_value` to depend on target-dependent roundi
tvm-bot commented on PR #13772:
URL: https://github.com/apache/tvm/pull/13772#issuecomment-1380438263
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
Alexey-Yazev opened a new pull request, #13772:
URL: https://github.com/apache/tvm/pull/13772
Does not fuse min and max operations with requantize if there are different
scales as it is not supported on NPU. Since there are hardware constraints, we
cannot perform min or max operation fused
fPecc opened a new issue, #13771:
URL: https://github.com/apache/tvm/issues/13771
I noticed that there is a problem with the docker image for Vitis AI, which
I also noticed was detected by @areusch in #13435, but I saw there is no
specific issue opened for this, so I wanted to open one with
NicolaLancellotti commented on PR #13762:
URL: https://github.com/apache/tvm/pull/13762#issuecomment-1380236687
> This looks good, but do these tutorials still need to include instructions
to install CMSIS? I'd love to remove this from `micro_aot.py`,
`micro_autotune.py`, and `micro_tflite.
fzi-peccia opened a new pull request, #13770:
URL: https://github.com/apache/tvm/pull/13770
Added integration to generate C code able to execute neural networks on the
Gemmini accelerator. Information about this can be found on [this
post](https://discuss.tvm.apache.org/t/presenting-the-gen
tvm-bot commented on PR #13770:
URL: https://github.com/apache/tvm/pull/13770#issuecomment-1380219284
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
ashutosh-arm commented on code in PR #13763:
URL: https://github.com/apache/tvm/pull/13763#discussion_r1068022670
##
tests/python/contrib/test_ethosn/test_conv2d.py:
##
@@ -228,10 +227,7 @@ def test_conv2d_depthwise(
)
),
}
-if ethosn_api_version()
petuca commented on issue #11691:
URL: https://github.com/apache/tvm/issues/11691#issuecomment-1380174305
Yes, @padreofthegame this issue has been fixed in PR #11760 .
Thank you for the notification.
--
This is an automated message from the Apache Git Service.
To respond to the message,
petuca closed issue #11691: [Bug][MXNet] MXNet dot not working with non 2D
tensors
URL: https://github.com/apache/tvm/issues/11691
--
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 specific comme
padreofthegame commented on issue #11691:
URL: https://github.com/apache/tvm/issues/11691#issuecomment-1380169837
Looks like this Issue may be closed?
@petuca @masahi
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
ekalda merged PR #13645:
URL: https://github.com/apache/tvm/pull/13645
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
ashutosh-arm merged PR #13764:
URL: https://github.com/apache/tvm/pull/13764
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apa
ashutosh-arm commented on PR #13764:
URL: https://github.com/apache/tvm/pull/13764#issuecomment-1380160011
Thanks @mehrdadh. LGTM!
--
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 specific co
lhutton1 commented on PR #13732:
URL: https://github.com/apache/tvm/pull/13732#issuecomment-1380066522
Thanks @arina-grovety for the explanation, just following up on some of the
questions...
I suspect this is a case of needing to expose this functionality from within
Vela, I'll see
lhutton1 commented on code in PR #13763:
URL: https://github.com/apache/tvm/pull/13763#discussion_r1067899742
##
tests/python/contrib/test_ethosn/test_conv2d.py:
##
@@ -228,10 +227,7 @@ def test_conv2d_depthwise(
)
),
}
-if ethosn_api_version() ==
Alexey-Yazev commented on code in PR #13645:
URL: https://github.com/apache/tvm/pull/13645#discussion_r1067796699
##
tests/python/contrib/test_ethosu/test_codegen.py:
##
@@ -1109,6 +1109,77 @@ def leaky_relu_func(x):
)
+# conv2d + relu_n1_to_1 is used because separate a
masahi merged PR #13733:
URL: https://github.com/apache/tvm/pull/13733
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
masahi commented on issue #13768:
URL: https://github.com/apache/tvm/issues/13768#issuecomment-1379810702
cc @mvermeulen , is rocm 5 known to work with TVM?
--
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
wangzy0327 commented on issue #13768:
URL: https://github.com/apache/tvm/issues/13768#issuecomment-1379809403
OK, Thank you. I try rocm 4.
--
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 spec
masahi commented on issue #13768:
URL: https://github.com/apache/tvm/issues/13768#issuecomment-1379808346
I don't remember, but we didn't touch any rocm stuff. So I think recent rocm
introduced some incompatibility with TVM. I'd try rocm 4 or even 3.
--
This is an automated message from t
wangzy0327 commented on issue #13768:
URL: https://github.com/apache/tvm/issues/13768#issuecomment-1379803549
> Don't create a new issue when you already have #13768
>
> We didn't change any rocm code recently and I'm sure we had a working rocm
backend before. So I have no idea what b
slyubomirsky commented on PR #13769:
URL: https://github.com/apache/tvm/pull/13769#issuecomment-1379802327
Please review @junrushao (or tag other reviewers)
--
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
slyubomirsky opened a new pull request, #13769:
URL: https://github.com/apache/tvm/pull/13769
This PR makes a few improvements to `py_converter` to make it more useful
for fuzz testing, especially when running larger modules.
In particular, these changes are to support returning the d
tvm-bot commented on PR #13769:
URL: https://github.com/apache/tvm/pull/13769#issuecomment-1379801686
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
masahi closed issue #13768: [Bug] how to fix rocm platform bugs?
URL: https://github.com/apache/tvm/issues/13768
--
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 specific comment.
To unsubscribe,
masahi commented on issue #13768:
URL: https://github.com/apache/tvm/issues/13768#issuecomment-1379801009
Don't create a new issue when you already have
https://github.com/apache/tvm/issues/13768
We didn't change any rocm code recently and I'm sure we had a working rocm
backend befor
masahi merged PR #13767:
URL: https://github.com/apache/tvm/pull/13767
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
wangzy0327 opened a new issue, #13768:
URL: https://github.com/apache/tvm/issues/13768
At present, my workmate and I use AMD GPU to support the SYCL platform on
tvm, and want to do related evaluations on opencl, rocm, and sycl on AMD GPU.
At present, I have made the sycl platform on tvm, bu
honghuichao commented on issue #13759:
URL: https://github.com/apache/tvm/issues/13759#issuecomment-1379733000
this is my demo extract from my network. @masahi
```
import torch
from pdb import set_trace as d
import onnx
from tvm import relay
class TestIndexPut(tor
honghuichao commented on issue #13759:
URL: https://github.com/apache/tvm/issues/13759#issuecomment-1379729882
@masahi
##this is test demo extract from my network. thanks.
import torch
from pdb import set_trace as d
import onnx
from tvm import relay
class TestIndexPut(
chunit-quic commented on PR #13767:
URL: https://github.com/apache/tvm/pull/13767#issuecomment-1379687912
Hi community,
This is a part of the RFC, TVM Explorer Infrastructure. After this PR, the
source node information of ONNX will be tagged to the converted relay.
Note that the n
tvm-bot commented on PR #13767:
URL: https://github.com/apache/tvm/pull/13767#issuecomment-1379687063
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
chunit-quic opened a new pull request, #13767:
URL: https://github.com/apache/tvm/pull/13767
- Set node name as the source name of span during the conversion of ONNX
model.
- Assign node name to a node based on op type when it is empty.
- To get the reference of renamed nodes. Add a fu
driazati opened a new issue, #13766:
URL: https://github.com/apache/tvm/issues/13766
Some tutorials that are defined in `gallery/` are not included in the final
built website since their enclosing folders are not listed in `docs/conf.py`,
for example the `work_with_pytorch` tutorials
https
AndrewZhaoLuo closed pull request #13754: [Generic] Forward MS and AS rewrites
for generic schedules
URL: https://github.com/apache/tvm/pull/13754
--
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
AndrewZhaoLuo commented on PR #13754:
URL: https://github.com/apache/tvm/pull/13754#issuecomment-1379581081
Ah so chatting with @tkonolige and others a bit more, this problem is a bit
more complicated and maybe exposes some past inconsistency with
tvm.target.Target. Anyway, here are my find
masahi merged PR #13688:
URL: https://github.com/apache/tvm/pull/13688
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
masahi commented on PR #13688:
URL: https://github.com/apache/tvm/pull/13688#issuecomment-1379531047
Hopefully we overhaul the current API to clean up mostly unused parameters.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub a
tkonolige commented on PR #13688:
URL: https://github.com/apache/tvm/pull/13688#issuecomment-1379524724
@masahi How about we merge this as #13470 is still in discussion?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
masahi commented on issue #13759:
URL: https://github.com/apache/tvm/issues/13759#issuecomment-1379482230
Please provide a reproducible script. Otherwise it is not clear what the
issue is, so I'll close it.
--
This is an automated message from the Apache Git Service.
To respond to the mes
masahi merged PR #13730:
URL: https://github.com/apache/tvm/pull/13730
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.or
cconvey opened a new issue, #13765:
URL: https://github.com/apache/tvm/issues/13765
This lies somewhere between feature-request and bug-report.
When writing TVMScript, it's sometimes necessary to specify certain PrimFunc
attribute values. E.g.:
``` python
@T.prim_func
driazati commented on PR #13689:
URL: https://github.com/apache/tvm/pull/13689#issuecomment-1379467892
I updated the PR to wrap the configs with ignores and made sure they are
inside a code block so they don't show up on their own, so now this PR should
make everything look fine from the fr
tkonolige merged PR #13755:
URL: https://github.com/apache/tvm/pull/13755
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache
cconvey commented on PR #13569:
URL: https://github.com/apache/tvm/pull/13569#issuecomment-1379456152
@Lunderberg @areusch : Mind giving this a look?
@JosephTheOctonaut : FYI
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Git
mehrdadh commented on PR #13760:
URL: https://github.com/apache/tvm/pull/13760#issuecomment-1379407755
https://github.com/apache/tvm/pull/13764 is updating the image.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
tvm-bot commented on PR #13764:
URL: https://github.com/apache/tvm/pull/13764#issuecomment-1379403908
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
mehrdadh opened a new pull request, #13764:
URL: https://github.com/apache/tvm/pull/13764
This PR updates the ci_cortexm image.
--
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 specific commen
tvm-bot commented on PR #13764:
URL: https://github.com/apache/tvm/pull/13764#issuecomment-1379403899
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
AndrewZhaoLuo commented on code in PR #13755:
URL: https://github.com/apache/tvm/pull/13755#discussion_r1067398808
##
python/tvm/topi/scatter.py:
##
@@ -268,63 +268,58 @@ def scatter_nd(data, indices, updates, mode):
_verify_scatter_nd_inputs(data, indices, updates)
mehrdadh commented on PR #13760:
URL: https://github.com/apache/tvm/pull/13760#issuecomment-137939
@ashutosh-arm I don't think we need https://github.com/apache/tvm/pull/13762
to merge this. Because Zephyr project API checks for dependency to CMSIS_NN and
if the project is not dependent
mehrdadh commented on PR #13760:
URL: https://github.com/apache/tvm/pull/13760#issuecomment-1379397695
Also we could separate updating the image from this PR, I think that should
be fine as well
--
This is an automated message from the Apache Git Service.
To respond to the message, please
mehrdadh commented on PR #13760:
URL: https://github.com/apache/tvm/pull/13760#issuecomment-1379391976
You need to update the image as well, otherwise it doesn't pass the tests.
You can use this image that I just pushed
https://hub.docker.com/layers/tlcpack/ci-cortexm/20230111-165944-a9c
tkonolige commented on code in PR #13755:
URL: https://github.com/apache/tvm/pull/13755#discussion_r1067385975
##
python/tvm/topi/scatter.py:
##
@@ -268,63 +268,58 @@ def scatter_nd(data, indices, updates, mode):
_verify_scatter_nd_inputs(data, indices, updates)
def
AndrewZhaoLuo commented on PR #13754:
URL: https://github.com/apache/tvm/pull/13754#issuecomment-1379375276
@tkonolige would you be fine if I made copies of the generic strategies for
my target, and properly forward rewrite info there?
We still have the problem with generic strategie
AndrewZhaoLuo commented on code in PR #13755:
URL: https://github.com/apache/tvm/pull/13755#discussion_r1067342210
##
python/tvm/topi/scatter.py:
##
@@ -268,63 +268,58 @@ def scatter_nd(data, indices, updates, mode):
_verify_scatter_nd_inputs(data, indices, updates)
jwfromm commented on PR #13761:
URL: https://github.com/apache/tvm/pull/13761#issuecomment-1379359559
Looks like this needs to be linted. I think its also worth throwing in a
comment explaining this explicit unpacking helps improve flexibility with other
incoming layouts.
--
This is an a
michalpiszczek commented on PR #13761:
URL: https://github.com/apache/tvm/pull/13761#issuecomment-1379357838
@tvm-bot rerun
--
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 specific comment.
ashutosh-arm opened a new pull request, #13763:
URL: https://github.com/apache/tvm/pull/13763
This commit removes support for NPU driver 22.05 now
that tlcpack contains the latest docker image for
ci_cpu that comprises support for NPU driver 22.11.
cc @lhutton1
--
This is
tvm-bot commented on PR #13763:
URL: https://github.com/apache/tvm/pull/13763#issuecomment-1379348748
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
junrushao commented on PR #13754:
URL: https://github.com/apache/tvm/pull/13754#issuecomment-1379342692
For context: tuning-based layout rewriting on Relay is quite hacky and buggy
because Relay is not designed to support this feature (even if it's necessary
to deliver better performance).
areusch commented on issue #12739:
URL: https://github.com/apache/tvm/issues/12739#issuecomment-1379302692
@mikepapadim has a few cycles to look at this
--
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
AndrewZhaoLuo commented on PR #13754:
URL: https://github.com/apache/tvm/pull/13754#issuecomment-1379293575
Hmm so my bug though has us hit a generic strategy, but metaschedule still
rewrites the layout during tuning.
During compilation step we get an error since the rewriting is not
tkonolige commented on PR #13754:
URL: https://github.com/apache/tvm/pull/13754#issuecomment-1379283603
Currently auto/metaschedule layout transform is enabled on a
target-by-target basis. You're changing the default to be enabled for all
targets that do not specify. In the past we've had s
AndrewZhaoLuo commented on PR #13754:
URL: https://github.com/apache/tvm/pull/13754#issuecomment-1379276998
> I don't think this is correct. Specifying `need_meta_schedule_layout`
should be set from the specific target strategy. Like here:
https://github.com/apache/tvm/blob/main/python/tvm/
ashutosh-arm commented on PR #13760:
URL: https://github.com/apache/tvm/pull/13760#issuecomment-1379183364
@mehrdadh we would need to merge https://github.com/apache/tvm/pull/13762
before building Cortex-M docker image. Without that change, some Zephyr tests
might fail.
--
This is an au
ekalda commented on PR #13761:
URL: https://github.com/apache/tvm/pull/13761#issuecomment-1379168078
I know it's a miniature PR, but
https://tvm.apache.org/docs/contribute/pull_request.html#commit-message-guideline
--
This is an automated message from the Apache Git Service.
To respond
mehrdadh commented on PR #13760:
URL: https://github.com/apache/tvm/pull/13760#issuecomment-1379152292
I'm gonna update the docker image to 20.04 once this is merged:
https://github.com/apache/tvm/pull/13736
then you can open this PR
--
This is an automated message from the Apache Gi
ekalda commented on code in PR #13645:
URL: https://github.com/apache/tvm/pull/13645#discussion_r1067227196
##
tests/python/contrib/test_ethosu/test_codegen.py:
##
@@ -1109,6 +1109,77 @@ def leaky_relu_func(x):
)
+# conv2d + relu_n1_to_1 is used because separate activat
Mousius merged PR #13736:
URL: https://github.com/apache/tvm/pull/13736
--
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 specific comment.
To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.o
mehrdadh commented on PR #13736:
URL: https://github.com/apache/tvm/pull/13736#issuecomment-1379137378
@Mousius I updated both PR description which will be used for commit message
and PR tilte
--
This is an automated message from the Apache Git Service.
To respond to the message, please l
areusch commented on issue #13586:
URL: https://github.com/apache/tvm/issues/13586#issuecomment-1379070667
we discussed the release in the community meeting this morning. here are
notes:
- release is running late. @leandron will update the schedule
- branch created, current status is p
gromero commented on code in PR #13751:
URL: https://github.com/apache/tvm/pull/13751#discussion_r1066285222
##
apps/microtvm/zephyr/template_project/microtvm_api_server.py:
##
@@ -657,7 +657,26 @@ def generate_project(self, model_library_format_path,
standalone_crt_dir, projec
NicolaLancellotti commented on PR #13762:
URL: https://github.com/apache/tvm/pull/13762#issuecomment-137905
@ashutosh-arm
--
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 specific comment
tvm-bot commented on PR #13762:
URL: https://github.com/apache/tvm/pull/13762#issuecomment-1379042534
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
NicolaLancellotti opened a new pull request, #13762:
URL: https://github.com/apache/tvm/pull/13762
Pr #13656 added support for CMSIS NN from a new GitHub location.
The project at the new location uses a new header `arm_acle.h` which is not
always present.
This pr removes the CMSIS depe
Alexey-Yazev commented on PR #13645:
URL: https://github.com/apache/tvm/pull/13645#issuecomment-1378820135
> so I suspect that if the graph consists only RELU and nothing else, it
doesn't get offloaded to NPU at all.
in this case requantize operation can be offloaded to NPU
>
michalpiszczek commented on PR #13761:
URL: https://github.com/apache/tvm/pull/13761#issuecomment-1378792355
@tvm-bot rerun
--
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 specific comment.
michalpiszczek commented on PR #13761:
URL: https://github.com/apache/tvm/pull/13761#issuecomment-1378791553
cc @jwfromm
--
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 specific comment.
michalpiszczek commented on PR #13761:
URL: https://github.com/apache/tvm/pull/13761#issuecomment-1378790845
cc @jfromm
--
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 specific comment.
To u
tvm-bot commented on PR #13761:
URL: https://github.com/apache/tvm/pull/13761#issuecomment-1378788292
Thanks for contributing to TVM! Please refer to the contributing guidelines
https://tvm.apache.org/docs/contribute/ for useful information and tips. Please
request code reviews from
301 - 400 of 19571 matches
Mail list logo