zhen-jia opened a new pull request #5000: And opt out operator for
has_multiple_inputs for graph tuner
URL: https://github.com/apache/incubator-tvm/pull/5000
If layout_transform is in the graph, graph tuner’s has_multiple_inputs will
give wrong answer. This PR is to solve this problem.
FrozenGene commented on a change in pull request #4990: [TF][Relay] BatchNorm
support with run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#discussion_r388734661
##
File path: tests/python/frontend/tensorflow/test_bn_trainingmod.py
FrozenGene commented on a change in pull request #4990: [TF][Relay] BatchNorm
support with run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#discussion_r388734493
##
File path: tests/python/frontend/tensorflow/test_bn_trainingmod.py
FrozenGene commented on a change in pull request #4990: [TF][Relay] BatchNorm
support with run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#discussion_r388735248
##
File path: python/tvm/relay/frontend/tensorflow.py
##
@@
zhen-jia closed pull request #5000: And opt out operator for
has_multiple_inputs for graph tuner
URL: https://github.com/apache/incubator-tvm/pull/5000
This is an automated message from the Apache Git Service.
To respond to
FrozenGene commented on a change in pull request #4990: [TF][Relay] BatchNorm
support with run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#discussion_r388734798
##
File path: tests/python/frontend/tensorflow/test_bn_trainingmod.py
FrozenGene commented on a change in pull request #4990: [TF][Relay] BatchNorm
support with run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#discussion_r388734523
##
File path: tests/python/frontend/tensorflow/test_bn_trainingmod.py
FrozenGene commented on a change in pull request #4990: [TF][Relay] BatchNorm
support with run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#discussion_r388735082
##
File path: tests/python/frontend/tensorflow/test_bn_trainingmod.py
FrozenGene commented on a change in pull request #4990: [TF][Relay] BatchNorm
support with run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#discussion_r388734566
##
File path: tests/python/frontend/tensorflow/test_bn_trainingmod.py
zhen-jia opened a new pull request #5000: And opt out operator for
has_multiple_inputs for graph tuner
URL: https://github.com/apache/incubator-tvm/pull/5000
If layout_transform is in the graph, graph tuner’s has_multiple_inputs will
give wrong answer. This PR is to solve this problem.
liangfu edited a comment on issue #4998: [VTA][Chisel] Change Scala Linter
scalafmt => scalastyle
URL: https://github.com/apache/incubator-tvm/pull/4998#issuecomment-595612183
In addition, based on the suggestions in the linter, I made some final
changes in fixing indentation errors, here
liangfu commented on issue #4998: [VTA][Chisel] Change Scala Linter scalafmt =>
scalastyle
URL: https://github.com/apache/incubator-tvm/pull/4998#issuecomment-595612183
In addition, I made some final changes in fixing indentation errors, here is
a typical example:
```diff
diff --git
lfengad edited a comment on issue #4990: [TF][Relay] BatchNorm support with
run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595611551
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty
lfengad commented on issue #4990: [TF][Relay] BatchNorm support with run-time
mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595611551
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty `VarNode`
lfengad edited a comment on issue #4990: [TF][Relay] BatchNorm support with
run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595562012
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty
lfengad edited a comment on issue #4990: [TF][Relay] BatchNorm support with
run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595562012
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty
lfengad edited a comment on issue #4990: [TF][Relay] BatchNorm support with
run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595562012
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty
lfengad edited a comment on issue #4990: [TF][Relay] BatchNorm support with
run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595562012
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty
lfengad edited a comment on issue #4990: [TF][Relay] BatchNorm support with
run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595562012
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty
MarisaKirisame commented on issue #4979: [Relay/TOPI/TF][Op] Add support for
randomuniform Op in TVM
URL: https://github.com/apache/incubator-tvm/pull/4979#issuecomment-595598371
Because if there is only one seed, every random uniform call will return the
same result for the same tensor
liangfu commented on issue #4998: [VTA][Chisel] Change Scala Linter scalafmt =>
scalastyle
URL: https://github.com/apache/incubator-tvm/pull/4998#issuecomment-595597598
@tmoreau89 @vegaluisjose Thanks for the review. I think this is now ready
for a merge, and let's wait for the CI testing
tmoreau89 commented on issue #4998: [VTA][Chisel] Change Scala Linter scalafmt
=> scalastyle
URL: https://github.com/apache/incubator-tvm/pull/4998#issuecomment-595596630
Thanks @liangfu ; let us know when it's ready for a merge.
liangfu removed a comment on issue #4998: [VTA][Chisel] Change Scala Linter
scalafmt => scalastyle
URL: https://github.com/apache/incubator-tvm/pull/4998#issuecomment-595594406
vta/apps still need to be updated.
This is an
liangfu commented on issue #4998: [VTA][Chisel] Change Scala Linter scalafmt =>
scalastyle
URL: https://github.com/apache/incubator-tvm/pull/4998#issuecomment-595594406
vta/apps still need to be updated.
This is an automated
tmoreau89 commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595593538
Agree with @vegaluisjose on the separate PR; if you submit it, we'll work to
merge it quickly so you can rebase
vegaluisjose opened a new issue #4999: [VTA][Chisel] build in Catalina (OSX)
URL: https://github.com/apache/incubator-tvm/issues/4999
Hey @liangfu and @tmoreau89 ,
I was trying to figure out today why building the Chisel backend in VTA is
failing in Catalina:
*
liangfu opened a new pull request #4998: [VTA][Chisel] Change Scala Linter
scalafmt => scalastyle
URL: https://github.com/apache/incubator-tvm/pull/4998
As scalafmt changes the code even with `--test` argument in the latest
version, this PR switches scala linter for Chisel VTA from
vegaluisjose commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595590569
Hey @pasqoc ,
I just ran/build everything in both Linux/Mac but it seems to be failing
consistently in
vegaluisjose commented on a change in pull request #4986:
[VTA][Chisel,de10nano] Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388703764
##
File path: vta/hardware/chisel/Makefile
##
@@ -109,7 +133,7 @@ else
liangfu commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388688246
##
File path: vta/hardware/chisel/Makefile
##
@@ -109,7 +133,7 @@ else
lib_path =
kevinthesun commented on issue #4938: [topi][relay] add operation tan to TVM
URL: https://github.com/apache/incubator-tvm/pull/4938#issuecomment-595568771
@tqchen Do we need to do something for this PR?
This is an automated
masahi commented on issue #4995: [AutoTVM] Avoid using RPC for LocalRunner
URL: https://github.com/apache/incubator-tvm/pull/4995#issuecomment-595563667
AMDGPU also needs RPC runner for autotvm, if I remember correctly
This
lfengad edited a comment on issue #4990: [TF][Relay] BatchNorm support with
run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595562012
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty
lfengad edited a comment on issue #4990: [TF][Relay] BatchNorm support with
run-time mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595562012
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty
lfengad commented on issue #4990: [TF][Relay] BatchNorm support with run-time
mean and variance calculation
URL: https://github.com/apache/incubator-tvm/pull/4990#issuecomment-595562012
> > Yeah, our current implementation is just to check whether `mean` /
`variance` is empty `VarNode`
tqchen commented on issue #4938: [topi][relay] add operation tan to TVM
URL: https://github.com/apache/incubator-tvm/pull/4938#issuecomment-595541216
NOTE: the git commit message messed up today due to github issue
kevinthesun merged pull request #4938: [topi][relay] add operation tan to TVM
URL: https://github.com/apache/incubator-tvm/pull/4938
This is an automated message from the Apache Git Service.
To respond to the message, please
This is an automated email from the ASF dual-hosted git repository.
kevinthesun pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git
The following commit(s) were added to refs/heads/master by this push:
new d992468 [topi][relay] add
kevinthesun commented on issue #4938: [topi][relay] add operation tan to TVM
URL: https://github.com/apache/incubator-tvm/pull/4938#issuecomment-595537526
@notoraptor Thanks!
This is an automated message from the Apache Git
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 a2c7f52 hotfix gcn tutorial fail (#4994)
add a198c9f Adding Hua Jiang as reviewer. (#4993)
No new
tqchen merged pull request #4993: [COMMUNITY] @huajsj -> Reviewer
URL: https://github.com/apache/incubator-tvm/pull/4993
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
vegaluisjose commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595522233
I think it would be nice to do that in a separate PR.
pasqoc commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595517463
Do you want me to remove python from
find_program(PYTHON NAMES python python3 python3.6)
in
vegaluisjose commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595516892
This is an error on that cmake file actually because everything should be
python3 by now. In fact, f-strings are
pasqoc commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595514049
Just removed them, but f strings are so much better .
Any particular reason why python 2 is still used?
vegaluisjose commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595512678
I was running into the same issue with the `f-strings` in your code. I think
the problem is that we are running
comaniac commented on issue #4995: [AutoTVM] Avoid using RPC for LocalRunner
URL: https://github.com/apache/incubator-tvm/pull/4995#issuecomment-595512579
> Did you test GPU tuning? There are runtime issues like we cannot do fork
after initializing CUDA runtime
That's a good point.
merrymercy commented on issue #4995: [AutoTVM] Avoid using RPC for LocalRunner
URL: https://github.com/apache/incubator-tvm/pull/4995#issuecomment-595508666
Did you test GPU tuning? There are runtime issues like we cannot do fork
after initializing CUDA runtime
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 f63b249 refactor build module to take IRModule (#4988)
add a2c7f52 hotfix gcn tutorial fail (#4994)
No
zhiics merged pull request #4994: [Hotfix] Fix gcn tutorial failure
URL: https://github.com/apache/incubator-tvm/pull/4994
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
pasqoc commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595504375
Any chance the docker image maybe using python2 instead of python3.
Not running docker so not able to reproduce, but it
comaniac commented on issue #4995: [AutoTVM] Avoid using RPC for LocalRunner
URL: https://github.com/apache/incubator-tvm/pull/4995#issuecomment-595503657
> The main reason to adopt the RPC runner is for the process isolation --
because we cannot guarantee that the local runner won't crash
pasqoc commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595502268
Right now the build is failing in runtime.cc, not sure why.
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388629955
##
File path: vta/tutorials/matrix_multiply.py
##
@@ -52,7 +52,7 @@
# We configure
abergeron opened a new issue #4997: [relay][VM] Miscompile with double-if
URL: https://github.com/apache/incubator-tvm/issues/4997
This relay code:
```
def @main(%b: bool) -> ((int32,), (int32,)) {
let %v0: (int32,) = (0 /* ty=int32 */,);
let %v1: (int32,) = (1 /*
zhiics commented on issue #4996: [relay][external codegen] outline and inline
lifted functions for external codegen
URL: https://github.com/apache/incubator-tvm/pull/4996#issuecomment-595489617
cc @comaniac @mbaret @soiferj @tqchen @icemelon9 @masahi
zhiics opened a new pull request #4996: [relay][external codegen] outline and
inline lifted functions for external codegen
URL: https://github.com/apache/incubator-tvm/pull/4996
This PR outlines the functions that will be handled by external codegen and
then inline them back to avoid the
alexwong commented on a change in pull request #4964: [Torch] Add initial
control flow support
URL: https://github.com/apache/incubator-tvm/pull/4964#discussion_r388613725
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -614,6 +615,55 @@ def _impl(inputs,
tqchen commented on issue #4995: [AutoTVM] Avoid using RPC for LocalRunner
URL: https://github.com/apache/incubator-tvm/pull/4995#issuecomment-595486181
The main reason to adopt the RPC runner is for the process isolation --
because we cannot guarantee that the local runner won't crash and
masahi commented on a change in pull request #4964: [Torch] Add initial control
flow support
URL: https://github.com/apache/incubator-tvm/pull/4964#discussion_r388606332
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -842,7 +898,13 @@ def
tmoreau89 commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595474164
It looks like at the moment the cpp lint test is failing:
tmoreau89 commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388597868
##
File path: vta/tutorials/matrix_multiply.py
##
@@ -52,7 +52,7 @@
# We
masahi commented on a change in pull request #4964: [Torch] Add initial control
flow support
URL: https://github.com/apache/incubator-tvm/pull/4964#discussion_r388596458
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -614,6 +615,55 @@ def _impl(inputs,
comaniac opened a new pull request #4995: [AutoTVM] Avoid using RPC for
LocalRunner
URL: https://github.com/apache/incubator-tvm/pull/4995
### Motivation and Summary
`LocalRunner`, which measures the runtime of an op with a certain schedule
config on the host machine directly, is
pasqoc commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595457508
General question.
I see the CI jenkins merge-hook build fails for reasons other than the PR
itself (not right now, I
zhiics commented on issue #4988: [Refactor][Relay Build] refactor build module
to take IRModule
URL: https://github.com/apache/incubator-tvm/pull/4988#issuecomment-595457431
hmm, this actually revealed that the gcn tutorial didn't pass mod. We need
to bind params when adding the function
zhiics opened a new pull request #4994: [Hotfix] Fix gcn tutorial failure
URL: https://github.com/apache/incubator-tvm/pull/4994
Fix gcn tutorial fail in the upstream CI.
cc @tqchen
This is an automated message
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388575975
##
File path: vta/hardware/chisel/Makefile
##
@@ -109,7 +133,7 @@ else
lib_path =
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388572722
##
File path: vta/tutorials/matrix_multiply.py
##
@@ -52,7 +52,7 @@
# We configure
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388572033
##
File path: vta/tests/python/de10nano/test_program_rpc.py
##
@@ -0,0 +1,45 @@
+#
alexwong commented on a change in pull request #4964: [Torch] Add initial
control flow support
URL: https://github.com/apache/incubator-tvm/pull/4964#discussion_r388557544
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -842,7 +898,13 @@ def
zhiics commented on issue #4988: [Refactor][Relay Build] refactor build module
to take IRModule
URL: https://github.com/apache/incubator-tvm/pull/4988#issuecomment-595436882
Okay. I will. Thanks for reminding.
This is an
masahi commented on issue #4992: [Frontend][Torch] Check graph inputs match
expected
URL: https://github.com/apache/incubator-tvm/pull/4992#issuecomment-595431009
Let's wait until the github bug is fixed
This is an automated
tqchen commented on issue #4988: [Refactor][Relay Build] refactor build module
to take IRModule
URL: https://github.com/apache/incubator-tvm/pull/4988#issuecomment-595428966
@zhiics seems this change breaks the master ci, please look into it
tmoreau89 commented on issue #3934: [Runtime] MISRA-C compliant TVM runtime
URL: https://github.com/apache/incubator-tvm/pull/3934#issuecomment-595428050
Also @u99127 and @weberlo may be interested in this PR
This is an
tmoreau89 commented on issue #3934: [Runtime] MISRA-C compliant TVM runtime
URL: https://github.com/apache/incubator-tvm/pull/3934#issuecomment-595427659
Excellent, thanks for adding this to CI so quickly! I was able to reproduce
the demo by typing in `make demo`; it ran for the most part
tmoreau89 commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388535249
##
File path: vta/tutorials/matrix_multiply.py
##
@@ -52,7 +52,7 @@
# We
tmoreau89 commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388533974
##
File path: vta/hardware/chisel/Makefile
##
@@ -109,7 +133,7 @@ else
lib_path
tmoreau89 commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388532585
##
File path: vta/tests/python/de10nano/test_program_rpc.py
##
@@ -0,0 +1,45 @@
+#
maheshambule commented on issue #4979: [Relay/TOPI/TF][Op] Add support for
randomuniform Op in TVM
URL: https://github.com/apache/incubator-tvm/pull/4979#issuecomment-595393925
Thanks @MarisaKirisame. I understand that We want to keep TVM operator pure
and Relay can have side effects. But
pasqoc edited a comment on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595390387
Last commit addresses previous comments.
sbt scalafmt --test is still open since it changes code.
Please let
pasqoc commented on issue #4986: [VTA][Chisel,de10nano] Chisel fixes and
de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#issuecomment-595390387
Last commit address previous comments.
sbt scalafmt --test is still open since it changes code.
Please let me know
This is an automated email from the ASF dual-hosted git repository.
moreau pushed a change to branch tmoreau89-reviewer-huajsj
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
at 976fccd Adding Hua Jiang as reviewer.
This branch includes the following new commits:
This is an automated email from the ASF dual-hosted git repository.
moreau pushed a commit to branch tmoreau89-reviewer-huajsj
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git
commit 976fccd1a7cb9be17c815a9025cd7cf943fec74b
Author: Thierry Moreau
AuthorDate: Thu Mar 5
tmoreau89 opened a new pull request #4993: [COMMUNITY] @huajsj -> Reviewer
URL: https://github.com/apache/incubator-tvm/pull/4993
Congratulations to @huajsj for joining TVM's list of reviewers.
This is an automated
anijain2305 commented on issue #4828: [QNN][TFLite] TFLite rounding mode support
URL: https://github.com/apache/incubator-tvm/pull/4828#issuecomment-595347242
Good finding @LiangHao151941
I don't think we should add tflite_mode attr to Pool2D op. It seems adhoc to
me.
I
inadob commented on issue #4805: [Frontend][TFlite] Add parser support for
relu6, leaky_relu, relu_n1_to_1, log_softmax
URL: https://github.com/apache/incubator-tvm/pull/4805#issuecomment-595344454
> Can you show me code example here? I have typically used float numbers
like 1.0 or 6.0 to
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388434267
##
File path: vta/hardware/chisel/Makefile
##
@@ -109,7 +133,7 @@ else
lib_path =
MarisaKirisame commented on issue #4979: [Relay/TOPI/TF][Op] Add support for
randomuniform Op in TVM
URL: https://github.com/apache/incubator-tvm/pull/4979#issuecomment-595330595
I am proposing the following design:
the tvm operator randomuniform take a seed, and return a random uniform
maheshambule commented on issue #4979: [Relay/TOPI/TF][Op] Add support for
randomuniform Op in TVM
URL: https://github.com/apache/incubator-tvm/pull/4979#issuecomment-595320437
@MarisaKirisame, could you please elaborate more on outputting the new seed?
Does it mean that in case if seed
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388403925
##
File path: vta/tests/python/de10nano/test_program_rpc.py
##
@@ -0,0 +1,45 @@
+#
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388402428
##
File path: vta/src/de10nano/de10nano_mgr.h
##
@@ -0,0 +1,554 @@
+/*
+ * Licensed to
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388402066
##
File path: vta/src/de10nano/de10nano_mgr.h
##
@@ -0,0 +1,554 @@
+/*
+ * Licensed to
zhiics commented on issue #4988: [Refactor][Relay Build] refactor build module
to take IRModule
URL: https://github.com/apache/incubator-tvm/pull/4988#issuecomment-595313695
Thanks @comaniac @tqchen @wweic
This is an
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388400268
##
File path: vta/hardware/chisel/Makefile
##
@@ -32,16 +32,19 @@ ifeq (,
pasqoc commented on a change in pull request #4986: [VTA][Chisel,de10nano]
Chisel fixes and de10nano support
URL: https://github.com/apache/incubator-tvm/pull/4986#discussion_r388398027
##
File path: vta/hardware/chisel/Makefile
##
@@ -109,7 +133,7 @@ else
lib_path =
greysteil commented on issue #4977: [Torch, QNN] Add support for quantized
models via QNN
URL: https://github.com/apache/incubator-tvm/pull/4977#issuecomment-595305194
FYI, this issue (the change in commit author) got escalated to me at GitHub.
We have a bug in our squash and merge logic
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 fe74b37 Conditions updated to cover better user scenarios (#4951)
add f63b249 refactor build module to
zhiics merged pull request #4988: [Refactor][Relay Build] refactor build module
to take IRModule
URL: https://github.com/apache/incubator-tvm/pull/4988
This is an automated message from the Apache Git Service.
To respond to
jjohnson-arm commented on a change in pull request #4992: [Frontend][Torch]
Check graph inputs match expected
URL: https://github.com/apache/incubator-tvm/pull/4992#discussion_r388323372
##
File path: python/tvm/relay/frontend/pytorch.py
##
@@ -902,6 +902,21 @@ def
1 - 100 of 114 matches
Mail list logo