liangfu commented on a change in pull request #7059:
URL: https://github.com/apache/tvm/pull/7059#discussion_r539181597
##
File path: vta/tutorials/frontend/deploy_classification.py
##
@@ -56,9 +56,14 @@
from tvm.contrib.debugger import debug_runtime
from tvm.relay import tra
liangfu opened a new pull request #7069:
URL: https://github.com/apache/tvm/pull/7069
In XGBoost v1.3.0, the aggcv function has been moved from
[training.py](https://github.com/dmlc/xgboost/blob/release_1.2.0/python-package/xgboost/training.py#L337)
to
[callback.py](https://github.com/dml
yongwww commented on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741706021
@tqchen Thanks for the comment.
We chose to go with MLIR/HLO instead of XLA HLO, because HLO operations
works on tensors with static shape (dynamic ops like
[NMS](https://gi
yongwww edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741706021
@tqchen Thanks for the comment.
We chose to go with MLIR/HLO instead of XLA HLO, because HLO operations
works on tensors with static shape (dynamic ops like
[NMS](htt
masahi commented on pull request #6900:
URL: https://github.com/apache/tvm/pull/6900#issuecomment-741716260
@t-vi Please have a look at the CI issue. It is due to the recent change I
made.
This is an automated message from t
d-smirnov commented on pull request #6655:
URL: https://github.com/apache/tvm/pull/6655#issuecomment-741720201
@comaniac Ping. How can we make some progress here?
This is an automated message from the Apache Git Service.
To
d-smirnov commented on pull request #6970:
URL: https://github.com/apache/tvm/pull/6970#issuecomment-741720340
Ping. How can we make some progress here?
This is an automated message from the Apache Git Service.
To respond to
d-smirnov commented on pull request #6984:
URL: https://github.com/apache/tvm/pull/6984#issuecomment-741720499
Ping. How can we make some progress here?
This is an automated message from the Apache Git Service.
To respond to
d-smirnov commented on pull request #6998:
URL: https://github.com/apache/tvm/pull/6998#issuecomment-741720832
Ping. How can we make some progress here?
@comaniac Please advise should the commit be split? What to do with failed
unit tests?
-
t-vi commented on pull request #6900:
URL: https://github.com/apache/tvm/pull/6900#issuecomment-741734196
Oh, right, an undetected merge confict. Fixed. Thank you @masahi.
This is an automated message from the Apache Git Serv
yongwww edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741706021
@tqchen Thanks for the comment.
We chose to go with MLIR/HLO instead of XLA HLO, because HLO operations
works on tensors with static shape (dynamic ops like
[NMS](htt
fprotopapa commented on a change in pull request #7059:
URL: https://github.com/apache/tvm/pull/7059#discussion_r539306511
##
File path: vta/tutorials/frontend/deploy_classification.py
##
@@ -56,9 +56,14 @@
from tvm.contrib.debugger import debug_runtime
from tvm.relay import
FrozenGene commented on pull request #6889:
URL: https://github.com/apache/tvm/pull/6889#issuecomment-741807882
> @tqchen , @jroesch , @FrozenGene, @junrushao1994 : Would you please help
proceed with the PR. TIA!
Just see your mention and sorry for later reply. Will do one round of r
tqchen commented on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741822077
First of all, I do not disagree about the general goal of having a relay
dialect in MLIR. The main discussion pt is how can we get to that point.
Right now we have the need to
tqchen edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741822077
First of all, I do not disagree about the general goal of having a relay
dialect in MLIR. The main discussion pt is how can we get to that point.
Right now we have the n
tqchen edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741822077
First of all, I do not disagree about the general goal of having a relay
dialect in MLIR. The main discussion pt is how can we get to that point.
Right now we have the n
tqchen merged pull request #7064:
URL: https://github.com/apache/tvm/pull/7064
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
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 c31e338 fix nvcc compile option to be compatible with older cuda
(#7065)
add e848af1 [FIX] Improve error messages
tqchen edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741822077
Thanks @yongwww First of all, I do not disagree about the general goal of
having a relay dialect in MLIR. The main discussion pt is how can we get to
that point.
Right
FrozenGene commented on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741841901
I agree TQ’s point. we introduce this pr’s main goal is to cover more models
(include dynamic). However, introducting tablegen will bring overhead to us. We
don’t have urgent req
FrozenGene edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741841901
I agree TQ @tqchen ’s point. we introduce this pr’s main goal is to cover
more models (include dynamic). However, introducting tablegen will bring
overhead to us. We don’t
masahi commented on pull request #6900:
URL: https://github.com/apache/tvm/pull/6900#issuecomment-741852553
Thanks @t-vi
This is an automated message from the Apache Git Service.
To respond to the message, please log on to G
masahi merged pull request #6900:
URL: https://github.com/apache/tvm/pull/6900
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
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 e848af1 [FIX] Improve error messages and docs (#7064)
add db0215e Incremental type inference (#6900)
No new revisi
ANSHUMAN87 commented on pull request #7048:
URL: https://github.com/apache/tvm/pull/7048#issuecomment-741855916
> Thanks @ANSHUMAN87. Overall looks pretty good, but I am a little confused
at exactly what is happening is this pass. Is it converting matrices from
sparse to dense at compile t
zhiics commented on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741866746
Thanks @yongwww for the great effort:) Both paths make sense to me. Adding
Relay as a dialect would benefit more other dialects including other projects
internally. But it does intro
zhiics edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741866746
Thanks @yongwww for the great effort:) Both paths make sense to me. Adding
Relay as a dialect would benefit more other dialects including other projects
internally. But it doe
zhiics edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741866746
Thanks @yongwww for the great effort:) Both paths make sense to me. Adding
Relay as a dialect would benefit more other dialects including other projects
internally. But it doe
zhiics edited a comment on pull request #7060:
URL: https://github.com/apache/tvm/pull/7060#issuecomment-741866746
Thanks @yongwww for the great effort:) Both paths make sense to me. Adding
Relay as a dialect would benefit more other dialects including other projects
internally. But it doe
giuseros opened a new pull request #7070:
URL: https://github.com/apache/tvm/pull/7070
- Add an autoschedule module to tvmc
- Extract common tuning option between autotuner and autoscheduler
- Add testing
This is an
t-vi commented on pull request #6900:
URL: https://github.com/apache/tvm/pull/6900#issuecomment-741894947
Thank you @masahi, for the guidance and discussion, and review!
This is an automated message from the Apache Git Servic
giuseros commented on pull request #7070:
URL: https://github.com/apache/tvm/pull/7070#issuecomment-741895859
cc @leandron @merrymercy
This is an automated message from the Apache Git Service.
To respond to the message, plea
tqchen merged pull request #6948:
URL: https://github.com/apache/tvm/pull/6948
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
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 db0215e Incremental type inference (#6900)
add f606637 [µTVM] Allow for platform-specific malloc in runtime (#6948)
merrymercy commented on pull request #7068:
URL: https://github.com/apache/tvm/pull/7068#issuecomment-741928470
Luckily, it does not affect the current correctness, because most conv layer
has the same H and W, right?
This i
merrymercy edited a comment on pull request #7068:
URL: https://github.com/apache/tvm/pull/7068#issuecomment-741928470
Luckily, this bug does not affect the correctness for current models,
because most conv layer has the same H and W, right?
---
merrymercy edited a comment on pull request #7068:
URL: https://github.com/apache/tvm/pull/7068#issuecomment-741928470
Luckily, this bug does not affect the correctness for current models,
because most conv layers have the same H and W, right?
-
jwfromm commented on a change in pull request #7063:
URL: https://github.com/apache/tvm/pull/7063#discussion_r539509778
##
File path: python/tvm/contrib/nvcc.py
##
@@ -269,15 +270,24 @@ def have_int8(compute_version):
return False
-def have_tensorcore(compute_version):
This is an automated email from the ASF dual-hosted git repository.
lmzheng pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from f606637 [µTVM] Allow for platform-specific malloc in runtime (#6948)
add a72bdd3 [Auto Scheduler][Auto TVM] Fix in
merrymercy merged pull request #7068:
URL: https://github.com/apache/tvm/pull/7068
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
tqchen merged pull request #7069:
URL: https://github.com/apache/tvm/pull/7069
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
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 a72bdd3 [Auto Scheduler][Auto TVM] Fix infer tile size for NHWC
winograd (#7068)
add 94b2e44 [AutoSchedule] Compat
merrymercy commented on pull request #7069:
URL: https://github.com/apache/tvm/pull/7069#issuecomment-741935197
Could you also update this place
https://github.com/apache/tvm/blob/a72bdd3fe74017ff7284691d96fc043ef67cb511/python/tvm/autotvm/tuner/xgboost_cost_model.py#L472
?
-
merrymercy edited a comment on pull request #7069:
URL: https://github.com/apache/tvm/pull/7069#issuecomment-741935197
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHu
This is an automated email from the ASF dual-hosted git repository.
lmzheng pushed a change to branch merrymercy-patch-1
in repository https://gitbox.apache.org/repos/asf/tvm.git.
at 85e71db [AutoScheduler] Delete deprecated file auto_schedule.py
This branch includes the following new co
This is an automated email from the ASF dual-hosted git repository.
lmzheng pushed a commit to branch merrymercy-patch-1
in repository https://gitbox.apache.org/repos/asf/tvm.git
commit 85e71db9fb5533ea5e04f029ba180bfc8387f80a
Author: Lianmin Zheng
AuthorDate: Wed Dec 9 09:44:14 2020 -0800
merrymercy opened a new pull request #7071:
URL: https://github.com/apache/tvm/pull/7071
`auto_schedule.py` is deprecated in #7028.
But I forget to delete this file.
This is an automated message from the Apache Git Service
comaniac commented on a change in pull request #7063:
URL: https://github.com/apache/tvm/pull/7063#discussion_r539516443
##
File path: python/tvm/contrib/nvcc.py
##
@@ -269,15 +270,24 @@ def have_int8(compute_version):
return False
-def have_tensorcore(compute_version)
merrymercy commented on a change in pull request #7053:
URL: https://github.com/apache/tvm/pull/7053#discussion_r539517512
##
File path: python/tvm/auto_scheduler/auto_schedule.py
##
@@ -117,14 +132,21 @@ def __init__(
num_measure_trials=0,
early_stopping=None
comaniac commented on a change in pull request #7070:
URL: https://github.com/apache/tvm/pull/7070#discussion_r539522022
##
File path: python/tvm/driver/tvmc/autoscheduler.py
##
@@ -0,0 +1,212 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contribut
comaniac commented on pull request #6655:
URL: https://github.com/apache/tvm/pull/6655#issuecomment-741988621
> @comaniac Ping. How can we make some progress here?
I still don't think we should make AnnotateTarget pass even more complicate,
so I proposed to have a separate ACL specif
comaniac commented on pull request #6998:
URL: https://github.com/apache/tvm/pull/6998#issuecomment-741989718
> Ping. How can we make some progress here?
> @comaniac Please advise should the commit be split? What to do with failed
unit tests?
I'm not familiar with TFLite frontend.
tqchen commented on pull request #6984:
URL: https://github.com/apache/tvm/pull/6984#issuecomment-742000393
cc @mbaret @FrozenGene
This is an automated message from the Apache Git Service.
To respond to the message, please l
tqchen edited a comment on pull request #6984:
URL: https://github.com/apache/tvm/pull/6984#issuecomment-742000393
cc @mbaret @FrozenGene @siju-samuel
This is an automated message from the Apache Git Service.
To respond to t
tkonolige opened a new pull request #7072:
URL: https://github.com/apache/tvm/pull/7072
Somehow a unnecessary print statement was left in the codebase.
@tqchen
This is an automated message from the Apache Git Service.
tqchen merged pull request #6970:
URL: https://github.com/apache/tvm/pull/6970
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
tqchen commented on pull request #6970:
URL: https://github.com/apache/tvm/pull/6970#issuecomment-742001092
Thanks @d-smirnov this is merged. Please tag more related reviewers(by AT
them) next time when you raised the pr
Thi
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 94b2e44 [AutoSchedule] Compatibility improvement with XGBoost v1.3.0
(#7069)
add db8edc1 [TFLite] added scalar axi
mbrookhart commented on pull request #6978:
URL: https://github.com/apache/tvm/pull/6978#issuecomment-742003328
@kevinthesun Can you re-review without the topk legalization?
This is an automated message from the Apache Git Se
tqchen commented on pull request #6777:
URL: https://github.com/apache/tvm/pull/6777#issuecomment-742003795
@comaniac consider close this PR as it seems to be stale?
This is an automated message from the Apache Git Service.
T
tqchen closed issue #6332:
URL: https://github.com/apache/tvm/issues/6332
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
tqchen opened a new issue #6332:
URL: https://github.com/apache/tvm/issues/6332
Dear Community:
Thanks to everyone who participated in the discussion about graduation[1].
This is a formal voting thread for Apache TVM’s graduation.
If this vote passes, the next step would be t
tqchen closed issue #6299:
URL: https://github.com/apache/tvm/issues/6299
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
tqchen closed issue #6350:
URL: https://github.com/apache/tvm/issues/6350
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
comaniac commented on pull request #6777:
URL: https://github.com/apache/tvm/pull/6777#issuecomment-742006769
> @comaniac consider close this PR as it seems to be stale?
Ah sorry we don't have bandwidth to work on this recently. I'll close the PR
for now.
--
comaniac closed pull request #6777:
URL: https://github.com/apache/tvm/pull/6777
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
merrymercy closed issue #6133:
URL: https://github.com/apache/tvm/issues/6133
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 sp
merrymercy commented on issue #6133:
URL: https://github.com/apache/tvm/issues/6133#issuecomment-742009953
The functionality of Ansor is complete. Tutorials on tuning
operators/networks on CPU/GPU are available on
https://tvm.apache.org/docs/tutorials/#autoscheduler-template-free-auto-sch
merrymercy edited a comment on issue #6133:
URL: https://github.com/apache/tvm/issues/6133#issuecomment-742009953
The functionality of Ansor is complete. Tutorials on tuning
operators/networks on CPU/GPU are available on
https://tvm.apache.org/docs/tutorials/#autoscheduler-template-free-a
csullivan opened a new pull request #7073:
URL: https://github.com/apache/tvm/pull/7073
Rollback changes to SSA begin/end scope for Store in C codegen. Instead,
scope binary operator codegen in CUDA to fix the issue originally addressed by
5f4b9a9.
Thanks for contributing to TVM!
mbrookhart commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539624790
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffe
jwfromm commented on a change in pull request #7063:
URL: https://github.com/apache/tvm/pull/7063#discussion_r539661492
##
File path: python/tvm/contrib/nvcc.py
##
@@ -269,15 +270,24 @@ def have_int8(compute_version):
return False
-def have_tensorcore(compute_version):
comaniac commented on a change in pull request #7063:
URL: https://github.com/apache/tvm/pull/7063#discussion_r539685119
##
File path: python/tvm/contrib/nvcc.py
##
@@ -269,15 +269,34 @@ def have_int8(compute_version):
return False
-def have_tensorcore(compute_version)
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539693783
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539693783
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539693783
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539699599
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539699599
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
mbrookhart commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539700585
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffe
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539699599
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
mbrookhart commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539700714
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffe
mbrookhart commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539701202
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffe
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539702878
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539704122
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
Laurawly commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r539704122
##
File path: python/tvm/topi/cuda/nms.py
##
@@ -97,47 +97,44 @@ def get_valid_counts_ir(
valid_count = ib.buffer_ptr(valid_count)
out = ib.buffer_
antinucleon commented on pull request #7053:
URL: https://github.com/apache/tvm/pull/7053#issuecomment-742134344
I find out this solution only work for a single task, but doesn't work with
TaskScheduler. Will update it later.
---
mbrookhart opened a new pull request #7074:
URL: https://github.com/apache/tvm/pull/7074
@masahi @anijain2305
Fix for #7067
This is an automated message from the Apache Git Service.
To respond to the message, ple
masahi commented on pull request #7074:
URL: https://github.com/apache/tvm/pull/7074#issuecomment-742146711
@anijain2305 So after https://github.com/apache/tvm/pull/6704, it seems type
inferencer can pass `IncompleteType` to QNN type rel functions, which by itself
is not wrong. @mbrookhart
masahi edited a comment on pull request #7074:
URL: https://github.com/apache/tvm/pull/7074#issuecomment-742146711
@anijain2305 So after https://github.com/apache/tvm/pull/6704, it seems type
inferencer can pass `IncompleteType` to QNN type rel functions, which by itself
is not wrong. Prev
jwfromm commented on a change in pull request #7074:
URL: https://github.com/apache/tvm/pull/7074#discussion_r539753995
##
File path: src/relay/qnn/op/op_common.h
##
@@ -171,6 +171,11 @@ static inline bool QnnBroadcastRel(const Array&
types, int num_inputs, con
ICHECK_EQ(ty
jwfromm commented on pull request #7074:
URL: https://github.com/apache/tvm/pull/7074#issuecomment-742155528
I'm a little confused what exactly is being type checked. The comments say
its scale and zero points but the number of checked types in the loop don't
match up. A little better docu
corehalt commented on issue #7067:
URL: https://github.com/apache/tvm/issues/7067#issuecomment-742158829
@mbrookhart thank you for fixing the issue so fast, appreciate it.
This is an automated message from the Apache Git Serv
comaniac opened a new pull request #7075:
URL: https://github.com/apache/tvm/pull/7075
PR #6999 added a TOPI compute for deformable Conv2D NHWC. This PR adds the
required supports to lower NHWC deformable Conv2D from Relay so that it can be
tuned by auto_scheduler.
Notes:
- I did
vinx13 commented on a change in pull request #7075:
URL: https://github.com/apache/tvm/pull/7075#discussion_r539762513
##
File path: src/relay/op/nn/convolution.h
##
@@ -1128,7 +1169,8 @@ bool DeformableConv2DRel(const Array& types, int
num_inputs, const Attrs&
} else {
dsteger opened a new pull request #21:
URL: https://github.com/apache/tvm-vta/pull/21
An update is needed in the pynq driver to control
VTA. Once VTA is started it is necessary to wait
for fetch ap_done signal prior to reading compute
done flag. Also, the autorestart IPs need to be
masahi commented on a change in pull request #7074:
URL: https://github.com/apache/tvm/pull/7074#discussion_r539764667
##
File path: src/relay/qnn/op/op_common.h
##
@@ -171,6 +171,11 @@ static inline bool QnnBroadcastRel(const Array&
types, int num_inputs, con
ICHECK_EQ(typ
masahi commented on a change in pull request #7074:
URL: https://github.com/apache/tvm/pull/7074#discussion_r539764667
##
File path: src/relay/qnn/op/op_common.h
##
@@ -171,6 +171,11 @@ static inline bool QnnBroadcastRel(const Array&
types, int num_inputs, con
ICHECK_EQ(typ
tqchen commented on pull request #6890:
URL: https://github.com/apache/tvm/pull/6890#issuecomment-742164379
ping @kevinthesun please followup
This is an automated message from the Apache Git Service.
To respond to the message
tqchen edited a comment on pull request #6890:
URL: https://github.com/apache/tvm/pull/6890#issuecomment-742164379
ping @lixiaoquan @kevinthesun please followup
This is an automated message from the Apache Git Service.
To res
masahi commented on a change in pull request #7074:
URL: https://github.com/apache/tvm/pull/7074#discussion_r539766653
##
File path: tests/python/frontend/pytorch/qnn_test.py
##
@@ -32,17 +32,58 @@
from tvm.relay.frontend.pytorch_utils import is_version_greater_than
from tvm.
1 - 100 of 135 matches
Mail list logo
