xqdan edited a comment on pull request #6443:
URL: https://github.com/apache/incubator-tvm/pull/6443#issuecomment-692519854
> @xqdan @Beya2019 I'm not quite sure about whether we really need a NHWC
layout topi implementation for roi_align on general purpose hardware, since IMO
this
xqdan commented on pull request #6443:
URL: https://github.com/apache/incubator-tvm/pull/6443#issuecomment-692519854
> @xqdan @Beya2019 I'm not quite sure about whether we really need a NHWC
layout topi implementation for roi_align on general purpose hardware, since IMO
this sh
Beya2019 commented on pull request #6443:
URL: https://github.com/apache/incubator-tvm/pull/6443#issuecomment-692521267
Hi, @kevinthesun
I have move the check for NCHW from relay type infer into op strategy for
corresponding targets. eg:
```
layout = attrs.layout
if
masahi opened a new issue #6474:
URL: https://github.com/apache/incubator-tvm/issues/6474
It seems the commonly used
[`torch.nn.LSTM`](https://pytorch.org/docs/stable/generated/torch.nn.LSTM.html)
module generates `aten::lstm` op, which we currently don't support. It is
reported in
https
yongwww commented on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692541301
> Would be interesting to run `torch._C._jit_pass_remove_inplace_ops(...)`
and see what happens
the graph is still the same as before
-
t-vi commented on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692546794
> Seems PyTorch already has a set of passes to remove mutation
Note fat warning at the top of the class, though.
A typical use of copy is something like this:
t-vi commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692548751
@zhiics @kevinthesun @masahi
Thank you, @kevinthesun for your summary and all the work in the
investigation and your PR.
I think using `if isinstance(..., _expr.Exp
masahi commented on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692548630
> the graph is still the same as before
Yeah, this pass only supports a very small subset of inplace ops, see
https://github.com/pytorch/pytorch/blob/61623430d32c0
masahi edited a comment on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692548630
> the graph is still the same as before
Yeah, this pass only supports a very small subset of inplace ops, see
https://github.com/pytorch/pytorch/blob/616234
masahi edited a comment on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692548630
> the graph is still the same as before
Yeah, this pass only supports a very small subset of inplace ops, see
https://github.com/pytorch/pytorch/blob/616234
masahi edited a comment on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692548630
> the graph is still the same as before
Yeah, this pass only supports a very small subset of inplace ops, see
https://github.com/pytorch/pytorch/blob/616234
Johnson9009 opened a new pull request #6475:
URL: https://github.com/apache/incubator-tvm/pull/6475
I found there are some broken web links from the "InferBound Pass" document,
and I have tried to find more broken web links in the repository, I only fix
the broken links that are easy to fi
yongwww commented on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692568164
> > Seems PyTorch already has a set of passes to remove mutation
>
> Note fat warning at the top of the class, though.
>
> A typical use of copy is something l
yongwww edited a comment on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692568164
> > Seems PyTorch already has a set of passes to remove mutation
>
> Note fat warning at the top of the class, though.
>
> A typical use of copy is some
leandron commented on pull request #6475:
URL: https://github.com/apache/incubator-tvm/pull/6475#issuecomment-692571731
Thanks for fixing this. LGTM
This is an automated message from the Apache Git Service.
To respond to the
t-vi commented on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692573786
@yongwww what does "the slicing assignment doesn't take effect in jit" mean?
The JITed function certainly does update `a` because `select` will return a
view that uses the sa
t-vi edited a comment on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692573786
@yongwww what does "the slicing assignment doesn't take effect in jit" mean?
The JITed function does update `a` because `select` will return a view that
uses the same
yongwww commented on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692589490
@t-vi I see, thanks for pointing the select out, yeah, the jit graph is
okay and works fine with PyTorch. But seems tvm doesn't support this case
without real in-place op
yongwww commented on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692594584
@masahi @t-vi @zhiics @kevinthesun
to sum up the discussion above. The mapping in this pr is able to support
some test cases (added in the test_forward) which are not
yongwww edited a comment on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692589490
@t-vi I see, thanks for pointing the select out, yeah, for your case, the
jit graph is okay and works fine with PyTorch. But seems tvm doesn't support
this case wi
wjliu1998 opened a new issue #6476:
URL: https://github.com/apache/incubator-tvm/issues/6476
When run the code
`input_tensor_placeholder = te.placeholder((224,224,3),
name="input_tensor_placeholder")
m_tensor_placeholder = te.placeholder((3,3), name="m_tensor_placeholder")
mid_res
BhushanIMG opened a new pull request #6477:
URL: https://github.com/apache/incubator-tvm/pull/6477
Hi,
I am Bhushan from Imagination Technologies.
This PR adds space_to_batch_nd and batch_to_space_nd operators to Relay and
TOPI.
I have added the tests for these operators an
roastduck opened a new issue #6478:
URL: https://github.com/apache/incubator-tvm/issues/6478
Please look at this line:
https://github.com/apache/incubator-tvm/blob/7b744b3f6b70e9323a6ea532c99e6aeb807950a2/python/tvm/topi/bifrost/conv2d.py#L321
It seems that `tile_k` has not be
lhutton1 commented on a change in pull request #6395:
URL: https://github.com/apache/incubator-tvm/pull/6395#discussion_r488545828
##
File path: cmake/modules/contrib/TensorRT.cmake
##
@@ -0,0 +1,55 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more con
mbaret commented on pull request #6355:
URL: https://github.com/apache/incubator-tvm/pull/6355#issuecomment-692747987
vta change reverted
This is an automated message from the Apache Git Service.
To respond to the message, pl
u99127 commented on issue #6421:
URL: https://github.com/apache/incubator-tvm/issues/6421#issuecomment-692754302
I'd like to see the tvmc work to be functional before we cut the release
please. I'd also like to see the Ethos-N PRs land as well.
---
tqchen merged pull request #6475:
URL: https://github.com/apache/incubator-tvm/pull/6475
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 g
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 7b744b3 [Relay]Some backend improvements for PT OD models (#6464)
add 72c48ff [DOC] Fix Some Broken Web
tqchen commented on issue #6473:
URL: https://github.com/apache/incubator-tvm/issues/6473#issuecomment-692759775
good catch, please send a PR to fix the problem. It is due to how pytest
running everything together and file being imported elsewere
--
roastduck opened a new pull request #6479:
URL: https://github.com/apache/incubator-tvm/pull/6479
#6473
Added the missing import.
I did not any new tests since I cannot fix pytest.
@tqchen please have a look.
---
zhiics commented on pull request #6351:
URL: https://github.com/apache/incubator-tvm/pull/6351#issuecomment-692790307
@jwfromm @electriclilies please take another look
This is an automated message from the Apache Git Service.
roastduck commented on pull request #6479:
URL: https://github.com/apache/incubator-tvm/pull/6479#issuecomment-692792601
Why does the CI complaint about something I did not modify?
This is an automated message from the Apache
rkimball opened a new pull request #6480:
URL: https://github.com/apache/incubator-tvm/pull/6480
Add a target making it easier to apply style, just type `make style` and you
code is styled
This is an automated message fr
rkimball commented on pull request #6480:
URL: https://github.com/apache/incubator-tvm/pull/6480#issuecomment-692806664
@tqchen
This is an automated message from the Apache Git Service.
To respond to the message, please
rkimball closed pull request #6480:
URL: https://github.com/apache/incubator-tvm/pull/6480
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
rkimball opened a new pull request #6481:
URL: https://github.com/apache/incubator-tvm/pull/6481
Add a new style target so style can be applied using the command `make style`
This is an automated message from the Apac
tqchen merged pull request #6334:
URL: https://github.com/apache/incubator-tvm/pull/6334
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 g
rkimball commented on pull request #6481:
URL: https://github.com/apache/incubator-tvm/pull/6481#issuecomment-692818356
@tqchen
This is an automated message from the Apache Git Service.
To respond to the message, please log
tqchen commented on pull request #6334:
URL: https://github.com/apache/incubator-tvm/pull/6334#issuecomment-692818512
Thansk @liangfu @areusch !
This is an automated message from the Apache Git Service.
To respond to the mess
tqchen commented on pull request #6481:
URL: https://github.com/apache/incubator-tvm/pull/6481#issuecomment-692834196
Thanks @rkimball Given that we already have a script for the style
application, I feel we don't need to add the duplicated command here. Notably
the check can differ depend
rkimball closed pull request #6481:
URL: https://github.com/apache/incubator-tvm/pull/6481
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
tqchen opened a new pull request #6482:
URL: https://github.com/apache/incubator-tvm/pull/6482
Also added document about the encouraged usage.
In particular, we encourage the following usage.
lib = relay.build(...)
gmod = graph_runtime.GraphModule(lib["default"](ctx))
I h
trevor-m commented on pull request #6395:
URL: https://github.com/apache/incubator-tvm/pull/6395#issuecomment-692843402
> Just a couple of minor suggestions and comments - looks good overall,
without regard to comments/concerns above.
>
> One overall suggestion from a users perspecti
tqchen commented on pull request #6479:
URL: https://github.com/apache/incubator-tvm/pull/6479#issuecomment-692844205
@roastduck you can try to apply `tests/script/git-black.sh -i
upstream/master`
This is an automated messag
tqchen commented on issue #6483:
URL: https://github.com/apache/incubator-tvm/issues/6483#issuecomment-692844819
cc @comaniac @jroesch @areusch
This is an automated message from the Apache Git Service.
To respond to the mess
tqchen opened a new issue #6483:
URL: https://github.com/apache/incubator-tvm/issues/6483
see http://ci.tvm.ai:8080/job/tvm/job/master/1588/execution/node/22/log/
This is an automated message from the Apache Git Service.
To r
tqchen commented on a change in pull request #6456:
URL: https://github.com/apache/incubator-tvm/pull/6456#discussion_r488820408
##
File path: CMakeLists.txt
##
@@ -144,7 +144,7 @@ else(MSVC)
CMAKE_CXX_COMPILER_VERSION VERSION_GREATER 7.0)
set(CMAKE_CXX_FLAGS "-fali
comaniac commented on issue #6483:
URL: https://github.com/apache/incubator-tvm/issues/6483#issuecomment-692847204
I can take a look.
This is an automated message from the Apache Git Service.
To respond to the message, please
tqchen commented on pull request #6482:
URL: https://github.com/apache/incubator-tvm/pull/6482#issuecomment-692849038
cc @anijain2305 @ZihengJiang @jroesch @FrozenGene @jroesch @zhiics @yzhliu
This is an automated message fr
lhutton1 commented on pull request #6395:
URL: https://github.com/apache/incubator-tvm/pull/6395#issuecomment-692850735
Np @trevor-m, it seems that way, at least that's where I added the Arm
Compute Library doc
This is an au
trevor-m commented on a change in pull request #6395:
URL: https://github.com/apache/incubator-tvm/pull/6395#discussion_r488828728
##
File path: tests/python/contrib/test_tensorrt.py
##
@@ -0,0 +1,573 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more c
comaniac opened a new pull request #6484:
URL: https://github.com/apache/incubator-tvm/pull/6484
Ignore the deleted files in Python formatter. Per requested by #6483
Suggestions to the filter combination are welcome.
Reference
(https://git-scm.com/docs/git-diff#Documentation/git-
areusch opened a new pull request #6485:
URL: https://github.com/apache/incubator-tvm/pull/6485
Adding a ci_qemu docker image. This image contains the external dependencies
needed to run µTVM end-to-end tests with simulated hardware.
---
areusch commented on pull request #6485:
URL: https://github.com/apache/incubator-tvm/pull/6485#issuecomment-692893056
cc @tqchen how should we proceed here? I didn't modify Jenkinsfile yet in
this PR
This is an automated me
tqchen commented on pull request #6485:
URL: https://github.com/apache/incubator-tvm/pull/6485#issuecomment-692896086
I think we can go ahead and merge it in, it would be great @areusch if you
can also verify locally it works
---
areusch commented on pull request #6485:
URL: https://github.com/apache/incubator-tvm/pull/6485#issuecomment-692906435
i'll verify it again and see if others are interested in trying it as well.
we don't currently have an integration test that uses Zephyr + qemu, so this
may be a little pr
kevinthesun commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692917238
> @zhiics @kevinthesun @masahi
> Thank you, @kevinthesun for your summary and all the work in the
investigation and your PR.
>
> I think using `if isinstance(
kevinthesun edited a comment on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692917238
> @zhiics @kevinthesun @masahi
> Thank you, @kevinthesun for your summary and all the work in the
investigation and your PR.
>
> I think using `if isin
kevinthesun commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692921717
@masahi It looks like certain recent change causes this error. I'm
investigating.
This is an automate
kevinthesun commented on pull request #6443:
URL: https://github.com/apache/incubator-tvm/pull/6443#issuecomment-692922971
@xqdan That's what I meant.
This is an automated message from the Apache Git Service.
To respond to th
kevinthesun commented on pull request #6443:
URL: https://github.com/apache/incubator-tvm/pull/6443#issuecomment-692926700
@Beya2019 Thanks for adding nhwc implementation. However, this also adds
maintenance burden if there is no obvious performance benefit. Recently we are
adding dynamic
masahi commented on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692943139
@yongwww to be clear, this translation works for box_coder from
maskrcnn-benchmark you mentioned above?
https://github.com/facebookresearch/maskrcnn-benchmark/blob/4ac0a3b4
masahi edited a comment on pull request #6472:
URL: https://github.com/apache/incubator-tvm/pull/6472#issuecomment-692943139
@yongwww to be clear, this translation works for box_coder from
maskrcnn-benchmark you mentioned above?
https://github.com/facebookresearch/maskrcnn-benchmark/blob/4
t-vi commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692944886
I don't actually get why we need try: ... except: ... what case does it not
handle that is handled properly by the except part?
-
tqchen merged pull request #6484:
URL: https://github.com/apache/incubator-tvm/pull/6484
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 g
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 745e542 µTVM RPC server and Part 1 of AutoTVM compilation
infrastructure (#6334)
add 69e4f44 add git d
ZihengJiang opened a new issue #6486:
URL: https://github.com/apache/incubator-tvm/issues/6486
# New Features
## Accelerator
## Ansor
## Frontend and User Interface
## Relay
## Runtime
## TIR
## Quantization
# Feature Improvement
ZihengJiang commented on issue #6486:
URL: https://github.com/apache/incubator-tvm/issues/6486#issuecomment-692966769
I have opened a gist for monthly reports from last v0.6 release:
https://gist.github.com/ZihengJiang/6d3440ec22852dc9baae2e3f278ad8b4
We can start organizing the mont
kevinthesun commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692975419
Original pt frontend just handles limited cases, mostly static
shape/attributes. It is fine we just keep input as it is. For more dynamic
models, we need to do some ex
kevinthesun edited a comment on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692975419
Original pt frontend just handles limited cases, mostly static
shape/attributes. It is fine we just keep input as it is. For more dynamic
models, we need to do
kevinthesun edited a comment on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692975419
Original pt frontend just handles limited cases, mostly static
shape/attributes. It is fine we just keep input as it is for static models. For
more dynamic mode
t-vi commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692981543
I think I'm slowly starting to understand. But couldn't one have something
like `e_new = _try_to_make_static(e)` that takes an expression, makes it as
static as possible and
kevinthesun commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692985102
As I have discussed with @masahi, the problem of having a try interface is
that there is no common logic between different dynamic ops when dealing with
dynamic attrib
masahi commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692989903
@kevinthesun this mighe be related to the error
https://github.com/apache/incubator-tvm/pull/6316
This is
t-vi commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692994588
But which part raises the exception? infer_value_ or the casting that
follows?
This is an automated message
zhiics opened a new pull request #6487:
URL: https://github.com/apache/incubator-tvm/pull/6487
This PR makes saving and loading tensor size consistent for DLTensor
serialization. This somehow was triggered by recent dynamic operator changes
but only caught in some large models like TF SSD
kevinthesun commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692998861
@masahi I rolled back topk to still _infer_value. Directly use dyn topk will
cause a certain shape dim to be dynamic and cause get_valid_count issue.
Compiling pt mask
kevinthesun commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-692999233
> But which part raises the exception? infer_value_ or the casting that
follows?
It's for ```_infer_value```.
-
t-vi commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693002214
but wouldn't then a pattern of
```
inferred_val = _try_infer_value(inp)
if inferred_val is not None:
res = foo(inferred_val)
else:
res = bar(inp)
```
mbaret commented on pull request #6355:
URL: https://github.com/apache/incubator-tvm/pull/6355#issuecomment-693003547
@zhiics @masahi is there anything left to do here?
This is an automated message from the Apache Git Service
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 69e4f44 add git diff filter (#6484)
add 4c5f0c0 [BYOC][ETHOSN] Introduce further operator support (#6355)
zhiics commented on pull request #6355:
URL: https://github.com/apache/incubator-tvm/pull/6355#issuecomment-693004646
Thanks everyone!
This is an automated message from the Apache Git Service.
To respond to the message, pleas
zhiics merged pull request #6355:
URL: https://github.com/apache/incubator-tvm/pull/6355
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 g
mbaret commented on pull request #6355:
URL: https://github.com/apache/incubator-tvm/pull/6355#issuecomment-693005891
Thanks for all the reviews everyone :)
This is an automated message from the Apache Git Service.
To respond
kevinthesun edited a comment on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693007551
@t-vi For now IMO it doesn't bring obvious benefit of adding an extra common
api to just wrap ```try ... except``` for ```_infer_value```, considering the
lack
kevinthesun commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693007551
@t-vi For now IMO it doesn't bring obvious benefit of adding an extra common
api to just wrap ```try ... except``` for ```_infer_value```, considering the
lack of gene
kevinthesun edited a comment on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693007551
@t-vi For now IMO it doesn't bring obvious benefit of adding an extra common
api to just wrap ```try ... except``` for ```_infer_value```, considering the
lack
tqchen closed issue #6483:
URL: https://github.com/apache/incubator-tvm/issues/6483
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
kevinthesun edited a comment on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693007551
@t-vi For now IMO it doesn't bring obvious benefit of adding an extra common
api to just wrap ```try ... except``` for ```_infer_value```, considering the
lack
masahi commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693013272
@kevinthesun Although having a wrapper api doesn't make the code shorter, I
do believe there is a value in isolating such "toxic" programming idiom into
one place (e.g. it
masahi edited a comment on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693013272
@kevinthesun Although having a wrapper api doesn't make the code shorter, I
do believe there is a value in isolating such "toxic" programming idiom into
one place (e
tqchen commented on pull request #6482:
URL: https://github.com/apache/incubator-tvm/pull/6482#issuecomment-693016488
would be great @zhiics @anijain2305 if you can do a followup pass to pick up
the places in the tests. VSCode search for `graph_runtime.create` can usually
reveal the code.
kevinthesun commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693016775
@masahi Yeah. It probably worth a forum discussion thread before we proceed
with any solution. In tf frontend, there are even more ```try .. except```
blocks which doe
tqchen commented on pull request #6349:
URL: https://github.com/apache/incubator-tvm/pull/6349#issuecomment-693017408
@kongroo @vinx13 please followup
This is an automated message from the Apache Git Service.
To respond to t
zhiics commented on a change in pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#discussion_r489041322
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -2961,6 +3293,7 @@ def from_pytorch(script_module, input_shapes,
custom_convert_map=None, d
zhiics commented on a change in pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#discussion_r489041666
##
File path: tests/python/frontend/pytorch/test_object_detection.py
##
@@ -0,0 +1,137 @@
+# Licensed to the Apache Software Foundation (ASF) under o
trevor-m commented on pull request #6395:
URL: https://github.com/apache/incubator-tvm/pull/6395#issuecomment-693018923
There appears to be an inconsistency in the CI between
`tests/lint/cpplint.sh` and `tests/lint/clang_format.sh` for this macro
definition:
```
#define TRT_VERSION_G
masahi commented on a change in pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#discussion_r489046236
##
File path: tests/python/frontend/pytorch/test_object_detection.py
##
@@ -0,0 +1,136 @@
+# Licensed to the Apache Software Foundation (ASF) under o
masahi commented on pull request #6449:
URL: https://github.com/apache/incubator-tvm/pull/6449#issuecomment-693023530
@kevinthesun The maskrcnn test worked for me, great!
But unfortunately under torch 1.6, conversion fails with the following
error. Most likely they come from pytorch
comaniac commented on pull request #6395:
URL: https://github.com/apache/incubator-tvm/pull/6395#issuecomment-693024970
Looks like a conflict between cpplint and clang-format-10. clang-format-10
result seems more reasonable, so we may need to fix the cpplint.
cc @areusch
-
1 - 100 of 152 matches
Mail list logo