zhaiyi000 commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943021024
I still get an error when testing on the iOS sumulator, but it is not the
same as the above error, prompting `unsupported case in deploy.dylib`
```
(py38)
PhilippvK commented on pull request #9256:
URL: https://github.com/apache/tvm/pull/9256#issuecomment-943046720
@masahi I first tried to follow the same approach as in the PyTorch
frontend, just passing a dict to overwrite the convert map. However in the
TFLite frontend, the `convert_*`
masahi commented on a change in pull request #9261:
URL: https://github.com/apache/tvm/pull/9261#discussion_r728768117
##
File path: .gitmodules
##
@@ -13,3 +13,6 @@
[submodule "3rdparty/libbacktrace"]
path = 3rdparty/libbacktrace
url =
mikepapadim commented on pull request #9284:
URL: https://github.com/apache/tvm/pull/9284#issuecomment-943307587
LGMT, thanks @Mousius
--
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
mikepapadim commented on a change in pull request #9284:
URL: https://github.com/apache/tvm/pull/9284#discussion_r728926349
##
File path: python/tvm/meta_schedule/utils.py
##
@@ -20,7 +20,7 @@
import shutil
from typing import Any, Callable, List, Optional, Union
-import
lucagrementieri edited a comment on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943113815
The `TracerWarning` simply says that the output does not match.
> TracerWarning: Output nr 1. of the traced function does not match the
corresponding output of the
lucagrementieri commented on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943113815
The `TracerWarning` simply says that the output does not match.
> TracerWarning: Output nr 1. of the traced function does not match the
corresponding output of the Python
jiangjiajun closed pull request #9267:
URL: https://github.com/apache/tvm/pull/9267
--
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:
Mousius opened a new pull request #9285:
URL: https://github.com/apache/tvm/pull/9285
Spun out from https://github.com/apache/tvm/pull/9284 as I think we need
to update the image to get this in?
--
This is an automated message from the Apache Git Service.
To respond to the
lucagrementieri edited a comment on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943090057
--
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
lucagrementieri edited a comment on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943090057
I am facing a similar problem running the script `from_torch.py` published
with the tutorial
https://tvm.apache.org/docs/how_to/compile_models/from_pytorch.html.
masahi edited a comment on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943122022
Interesting. The warning says the output is corrupted already at the traced
model, so if we convert from the traced model TVM result would also be
corrupted. It is likely that
masahi commented on issue #8412:
URL: https://github.com/apache/tvm/issues/8412#issuecomment-943168765
See
https://discuss.tvm.apache.org/t/relay-nn-does-relay-nn-dense-supports-multi-dimensional-input/10343/7.
I realized that supporting more than 2D input for `dense` is extremely
Mousius commented on pull request #9285:
URL: https://github.com/apache/tvm/pull/9285#issuecomment-943295282
Closing as we can do this at the same time as #9284 and I'll remove the `#
type : ignore` later!
--
This is an automated message from the Apache Git Service.
To respond to the
Mousius closed pull request #9285:
URL: https://github.com/apache/tvm/pull/9285
--
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:
lucagrementieri commented on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943090057
I am facing a similar problem running ,the script `from_torch.py` published
with the tutorial
https://tvm.apache.org/docs/how_to/compile_models/from_pytorch.html.
When I
lucagrementieri edited a comment on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943090057
I am facing a similar problem running the script `from_torch.py` published
with the tutorial
https://tvm.apache.org/docs/how_to/compile_models/from_pytorch.html.
This is an automated email from the ASF dual-hosted git repository.
jroesch 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 d153676 Update TVM_LOG_DEBUG for IR tracing. (#9278)
jroesch merged pull request #9278:
URL: https://github.com/apache/tvm/pull/9278
--
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:
masahi edited a comment on issue #8412:
URL: https://github.com/apache/tvm/issues/8412#issuecomment-943168765
See
https://discuss.tvm.apache.org/t/relay-nn-does-relay-nn-dense-supports-multi-dimensional-input/10343/7.
I realized that supporting more than 2D input for `dense` is extremely
PhilippvK commented on pull request #9243:
URL: https://github.com/apache/tvm/pull/9243#issuecomment-943236871
To fix the CI run i will very likely need to rebase on top of `main`. This
will eventually screw up open discussions in this PR. Please let me know if
should just merge `main`
masahi commented on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943106737
@lucagrementieri That's very odd and not reproducible on my end (linux, PT
1.7 or 1.9). Note that the tutorial on the doc and many PyTorch tests run on
every CI job, so our PyTorch
masahi commented on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943122022
Interesting. The warning says the output is corrupted already at the traced
model, so if we convert from the traced model TVM result would also be
corrupted. It is likely that some of
jroesch commented on pull request #9263:
URL: https://github.com/apache/tvm/pull/9263#issuecomment-943112722
cc @electriclilies @mikepapadim
--
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
This is an automated email from the ASF dual-hosted git repository.
jroesch 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 594f23d [Core][Build] Move build module
mkroening opened a new pull request #9281:
URL: https://github.com/apache/tvm/pull/9281
This fixes a few issues (formatting, dead links, grammar), I encountered
when reading the tutorial.
@hogepodge, could you review this?
--
This is an automated message from the Apache Git
apeskov commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943299463
Hi, @zhaiyi000 thank you
This is a known issue related with latest update to Xcode 13. We are working
on fix for main. I hope it will be available soon.
Meanwhile I can suggest
Mousius commented on pull request #9284:
URL: https://github.com/apache/tvm/pull/9284#issuecomment-943299116
cc @mikepapadim
--
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
mikepapadim commented on pull request #9284:
URL: https://github.com/apache/tvm/pull/9284#issuecomment-943308662
cc @areusch
--
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
lucagrementieri edited a comment on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943090057
I am facing a similar problem running the script `from_torch.py` published
with the tutorial
https://tvm.apache.org/docs/how_to/compile_models/from_pytorch.html.
masahi edited a comment on issue #9237:
URL: https://github.com/apache/tvm/issues/9237#issuecomment-943106737
@lucagrementieri That's very odd and not reproducible on my end (linux, PT
1.7 or 1.9). Note that the tutorial on the doc and many PyTorch tests run on
every CI job, so our
jroesch commented on a change in pull request #9103:
URL: https://github.com/apache/tvm/pull/9103#discussion_r728738291
##
File path: src/driver/driver_api.cc
##
@@ -373,97 +394,96 @@ TVM_REGISTER_GLOBAL("driver.lower_schedule")
return LowerSchedule(std::move(sch),
jroesch merged pull request #9103:
URL: https://github.com/apache/tvm/pull/9103
--
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:
Mousius opened a new pull request #9280:
URL: https://github.com/apache/tvm/pull/9280
Seeing:
```
ERROR: Could not find a version that satisfies the requirement
sphinx-gallery==0.4.1 (from versions: 0.0.4, 0.0.5, 0.0.6, 0.0.7, 0.0.8,
0.0.10, 0.0.11.post1, 0.1.0, 0.1.1, 0.1.2, 0.1.3,
masahi commented on issue #8412:
URL: https://github.com/apache/tvm/issues/8412#issuecomment-943180691
I think a good way to add N-D input support to `topi.dense` is to do
`topi.reshape` on the input before doing dense compute, and inside dense
schedules we do `compute_inline(...)` on the
PhilippvK commented on pull request #9243:
URL: https://github.com/apache/tvm/pull/9243#issuecomment-943234803
Thank you for your review @huajsj. I will try to apply those patches soon.
Regarding unit tests, to did not want to introduce a new but otherwise
useless BYOC codegen just
mikepapadim opened a new pull request #9282:
URL: https://github.com/apache/tvm/pull/9282
This is a follow-up work for #8775 to address the pending comments to remove
the `compile_engine`.
As this is an intermediate step towards completely moving away from the
engine, we exposed
jiangjiajun opened a new pull request #9283:
URL: https://github.com/apache/tvm/pull/9283
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
Mousius opened a new pull request #9284:
URL: https://github.com/apache/tvm/pull/9284
There's a few errors that come up when type checking that aren't triggering
any failures:
```
Checking MyPy Type defs in the meta schedule package.
python/tvm/meta_schedule/utils.py:23:1: error:
Lunderberg commented on pull request #9263:
URL: https://github.com/apache/tvm/pull/9263#issuecomment-943340978
Rebased to resolve conflicts with main.
--
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
mikepapadim commented on a change in pull request #9103:
URL: https://github.com/apache/tvm/pull/9103#discussion_r728774603
##
File path: src/driver/driver_api.cc
##
@@ -373,97 +394,96 @@ TVM_REGISTER_GLOBAL("driver.lower_schedule")
return LowerSchedule(std::move(sch),
Lunderberg commented on pull request #9194:
URL: https://github.com/apache/tvm/pull/9194#issuecomment-943406846
Updated to resolve breaks this caused in ethos unit tests.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
echuraev edited a comment on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943452534
@zhaiyi000, I tried to reproduce your issue connected to the Metal locally
and I wasn't able to do that. Also, I have some doubts about this warning
message in your log:
echuraev commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943452534
@zhaiyi000, I tried to reproduce your issue connected to the Metal locally
and I wasn't able to do that. Also, I have some doubts about this warning
message in your log: `Warning:
zhaiyi000 commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943346320
@apeskov Thank you very much for your reply. I followed your suggestion and
modified the following code, but I got a new error.
```
clang = xcrun(["-sdk", sdk, "-find",
zhaiyi000 commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943420373
@apeskov Encountered another error when executing the following code.
```
run(annotate(model, "coremlcompiler"), target_host)
```
```
(py38) zy@diyideMacBook-Pro
PhilippvK commented on pull request #9255:
URL: https://github.com/apache/tvm/pull/9255#issuecomment-943468171
@areusch CI should be fixed now
--
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
ccjoechou edited a comment on pull request #9235:
URL: https://github.com/apache/tvm/pull/9235#issuecomment-943494906
Hi Wuwei, (or any other who can help?)
I have spent lots time on this pooling issue, which I believe it is a
fundamental-concept-related issue.
Similar to what we
apeskov commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943416403
@echuraev could you please take a look on issue with shader language
mismatch?
Looks like `ios_rpc_test.py` script from main branch doesn't work with
pointed device configuration.
This is an automated email from the ASF dual-hosted git repository.
moreau 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 2177632 [Tutorial] Fix vta vision detection tutorial
tmoreau89 merged pull request #9279:
URL: https://github.com/apache/tvm/pull/9279
--
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:
apeskov commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943393716
Looks like some mismatch between requested metal shader language(2.3) and
generated kernel code. We will also take a look on it.
If you are not interested in iOS GPU/Metal code
Hzfengsy opened a new pull request #9286:
URL: https://github.com/apache/tvm/pull/9286
This PR is for updating block syntax. Please see RFC:
[here](https://github.com/apache/tvm-rfcs/blob/main/rfcs/0041-script-block-syntax.md)
cc @junrushao1994 @tqchen @tkonolige
--
This is an
echuraev edited a comment on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943423144
Hi @zhaiyi000!
> @apeskov Encountered another error when executing the following code.
It looks like you should compile your runtime with CoreML support. Please,
echuraev commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943423144
Hi @zhaiyi000!
> @apeskov Encountered another error when executing the following code.
It looks like you should compile your runtime with CoreML support. Please,
use this
areusch commented on pull request #9040:
URL: https://github.com/apache/tvm/pull/9040#issuecomment-943492288
@Raghav-Chakravarthy please fix the lint error when you get a minute!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
areusch commented on pull request #7919:
URL: https://github.com/apache/tvm/pull/7919#issuecomment-943495292
ok, if you want to do this i'd strongly encourage you to consider whether
it's worth the complexity of multi-threading in c++ (it certainly could be, but
the GIL is a pretty easy
areusch edited a comment on pull request #7919:
URL: https://github.com/apache/tvm/pull/7919#issuecomment-943495292
@junrushao1994 ok, if you want to do this i'd strongly encourage you to
consider whether it's worth the complexity of multi-threading in c++ (it
certainly could be, but the
zhaiyi000 commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943507414
@echuraev, I recompiled libtvm.dylib and libtvm_runtime.dylib, specifying
`-DUSE_METAL=ON -DUSE_COREML=ON`. But still have the same error.
My xcode version is `Version 13.0
zhaiyi000 commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943351479
I tried the following code but it didn’t work
```
cmd += ["-target", "arm64-apple-ios14.7.1"] # any iOS version <15
```
```
cmake ..\
-DCMAKE_BUILD_TYPE=Debug\
zhaiyi000 commented on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943397933
Thank you for your reply. I need both CPU and GPU. I will write the code for
the CPU part first.
Thank you once again.
--
This is an automated message from the Apache Git
Lunderberg commented on pull request #9275:
URL: https://github.com/apache/tvm/pull/9275#issuecomment-943374796
Thank you for catching that, and I've updated to expose the
`ReportNode::AsTable`, explicitly disable the column summations, and use it
instead of `.csv()` to have a consistent
echuraev edited a comment on issue #9268:
URL: https://github.com/apache/tvm/issues/9268#issuecomment-943452534
@zhaiyi000, I tried to reproduce your issue connected to the Metal locally
and I wasn't able to do that.
Steps how I tried to reproduce your problem:
1. `cmake ..
ccjoechou commented on pull request #9235:
URL: https://github.com/apache/tvm/pull/9235#issuecomment-943494906
Hi Wuwei, (or any other who can help?)
I have spent lots time on this pooling issue, which I believe it is a
fundamental-concept-related issue.
Similar to what we have
areusch commented on pull request #34:
URL: https://github.com/apache/tvm-rfcs/pull/34#issuecomment-943491380
@tqchen in this case we prefer not to pay the cost of retrieving such a
pointer as tracking global state on an µC can be tricky. When the pointer is on
the stack, it is easier to
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 59b3cf7 Reset sphinx-gallery version to 0.4.0 (#9280)
add 95a2031 [Tests] Ensure MyPy type checks pass (#9284)
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 2177632 [Tutorial] Fix vta vision detection tutorial 'sphinx' style
error. (#9279)
add 59b3cf7 Reset
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a commit to branch fix-rust-doc
in repository https://gitbox.apache.org/repos/asf/tvm.git
commit 52c716b1b5e84a87b66a50bd32c4bc937db6741c
Author: Jared Roesch
AuthorDate: Thu Oct 14 10:36:03 2021 -0700
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a change to branch fix-rust-doc
in repository https://gitbox.apache.org/repos/asf/tvm.git.
at 52c716b Re-enable Rust documentation build
This branch includes the following new commits:
new 52c716b
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a change to branch fix-rust-doc
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 52c716b Re-enable Rust documentation build
add 45486eb Address Chris's comment
No new revisions were
This is an automated email from the ASF dual-hosted git repository.
wuwei pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 95a2031 [Tests] Ensure MyPy type checks pass (#9284)
add 08018ea [TIR] Add support for 0-dim buffer (#9224)
No new
This is an automated email from the ASF dual-hosted git repository.
masahi pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from acff61c [Frontend][PaddlePaddle] Remove unused parameters and fix doc
string (#9283)
add c00ce37 [Profiler] Do
This is an automated email from the ASF dual-hosted git repository.
masahi pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 08018ea [TIR] Add support for 0-dim buffer (#9224)
add f4db899 [TFLite] Add option to overwrite OperatorConverter
This is an automated email from the ASF dual-hosted git repository.
masahi pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from f4db899 [TFLite] Add option to overwrite OperatorConverter class in
relay.frontend.from_tflite (#9256)
add acff61c
74 matches
Mail list logo