FrozenGene commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r563031514
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,431 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
This is an automated email from the ASF dual-hosted git repository.
vega pushed a change to branch ci-docker-staging
in repository https://gitbox.apache.org/repos/asf/tvm.git.
discard 198f5ed turn off test for now
discard dad91eb update images
discard 04947e1 update jenkins
discard
areusch opened a new pull request #7333:
URL: https://github.com/apache/tvm/pull/7333
Right now due to historical reasons we sometimes build the CRT from
build/standalone_crt and other times from src/runtime/crt. This PR consolidates
building to always be from `build/standalone_crt` and
tmoreau89 commented on a change in pull request #7331:
URL: https://github.com/apache/tvm/pull/7331#discussion_r563003136
##
File path: python/tvm/target/target.py
##
@@ -234,7 +234,10 @@ def micro(model="unknown", options=None):
trans_table = {
"host": [],
mbrookhart edited a comment on pull request #7332:
URL: https://github.com/apache/tvm/pull/7332#issuecomment-765811021
@masahi I agree the Expr will always be a var, but I wanted to let users
match on more complicated attributes like name/dtype/shape/etc, so I left it a
general in pattern
tmoreau89 commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562998116
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,431 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
mbrookhart commented on pull request #7332:
URL: https://github.com/apache/tvm/pull/7332#issuecomment-765811021
@masahi I agree the Expr will always be a var, but I wanted to let users
match on more complicated attributes like name/style/shape/etc, so I left it a
general in pattern
masahi edited a comment on pull request #7332:
URL: https://github.com/apache/tvm/pull/7332#issuecomment-765798515
I wonder, do we need to match against the let bound variable? Isn't it
always `Var`? Not sure if relay Let supports pattern destructing on let lhs,
like `let (x, y) = ... in
masahi edited a comment on pull request #7332:
URL: https://github.com/apache/tvm/pull/7332#issuecomment-765798515
I wonder, do we need to match against the let bound variable? Isn't it
always `Var`? Not sure if relay Let supports pattern destructing on let lhs,
like `let (x, y) = ... in
masahi edited a comment on pull request #7332:
URL: https://github.com/apache/tvm/pull/7332#issuecomment-765798515
I wonder, do we need to match against the let bound variable? Isn't it
always `Var`? Not sure if relay Let supports pattern destructing on let lhs,
like `let (x, y) = ... in
masahi commented on pull request #7332:
URL: https://github.com/apache/tvm/pull/7332#issuecomment-765798515
I wonder, do we need to match against the let bound variable? Isn't it
always `Var`?
This is an automated message
comaniac commented on a change in pull request #7332:
URL: https://github.com/apache/tvm/pull/7332#discussion_r562983099
##
File path: python/tvm/relay/dataflow_pattern/__init__.py
##
@@ -579,6 +602,27 @@ def __init__(self, cond: "DFPattern", true_branch:
"DFPattern",
comaniac commented on pull request #7317:
URL: https://github.com/apache/tvm/pull/7317#issuecomment-765793632
> Because this PR breaks compatibility, should we add some warning messages
and point users to this PR?
How about adding the message to the exsiting schedule not found
mdw-octoml commented on pull request #7331:
URL: https://github.com/apache/tvm/pull/7331#issuecomment-765793471
PTAL
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
mdw-octoml commented on a change in pull request #7331:
URL: https://github.com/apache/tvm/pull/7331#discussion_r562980165
##
File path: python/tvm/micro/contrib/zephyr.py
##
@@ -414,6 +417,16 @@ def flash(self, micro_binary):
build_dir = os.path.dirname(
mbrookhart opened a new pull request #7332:
URL: https://github.com/apache/tvm/pull/7332
I found myself needing a Let pattern for some object detection work. Thanks
for taking a look!
This is an automated message from
merrymercy commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562840117
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
mbrookhart closed pull request #7329:
URL: https://github.com/apache/tvm/pull/7329
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
mbrookhart commented on pull request #7329:
URL: https://github.com/apache/tvm/pull/7329#issuecomment-765792061
Hmm, probably, @masahi and i have been talking about cleaner ways to do this
for function/call/etc. Let me do some experiments and try to get back to you
next week. I'm going to
areusch commented on a change in pull request #7331:
URL: https://github.com/apache/tvm/pull/7331#discussion_r562976220
##
File path: python/tvm/micro/contrib/zephyr.py
##
@@ -414,6 +417,16 @@ def flash(self, micro_binary):
build_dir = os.path.dirname(
merrymercy commented on pull request #7317:
URL: https://github.com/apache/tvm/pull/7317#issuecomment-765780610
Because this PR breaks compatibility, should we add some warning messages
and point users to this PR?
This is
This is an automated email from the ASF dual-hosted git repository.
lmzheng 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 3ec67f0 [AutoScheduler] Fix conv3d's op strategy for
merrymercy merged pull request #7328:
URL: https://github.com/apache/tvm/pull/7328
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
mdw-octoml commented on pull request #7331:
URL: https://github.com/apache/tvm/pull/7331#issuecomment-765759262
@areusch
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
mdw-octoml opened a new pull request #7331:
URL: https://github.com/apache/tvm/pull/7331
This PR makes a few small changes to uTVM and its tests to support the
nRF5340 dev board out of the box.
This is an automated
comaniac commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562972265
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,431 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
tmoreau89 commented on pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#issuecomment-765742685
Marking the PR as ready to review, thank you for the input so far @leandron
@comaniac @merrymercy
This is an
slyubomirsky edited a comment on pull request #7329:
URL: https://github.com/apache/tvm/pull/7329#issuecomment-765737215
Q: Can you also have this for function calls? (Matching a call with any
number of arguments.) It seems like pretty much the same principle. @jroesch
can affirm that I
slyubomirsky commented on pull request #7329:
URL: https://github.com/apache/tvm/pull/7329#issuecomment-765737215
Q: Can you also have this for function calls? (Matching a call with any
number of arguments.)
This is an
tkonolige closed pull request #7309:
URL: https://github.com/apache/tvm/pull/7309
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
tkonolige opened a new pull request #7330:
URL: https://github.com/apache/tvm/pull/7330
This PR fixes error messages that arise when a runtime array does not have
the correct element types for a function. For example:
```
# before
Check failed: ObjectTypeChecker::Check(ptr) ==
tmoreau89 commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562946754
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
tmoreau89 merged pull request #7286:
URL: https://github.com/apache/tvm/pull/7286
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
This is an automated email from the ASF dual-hosted git repository.
moreau pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 6787d74 get_top_results works on a copy of output (#7327)
add af9d1d2 [BYOC][Verilator] add support to dynamically
masahi commented on a change in pull request #7329:
URL: https://github.com/apache/tvm/pull/7329#discussion_r562913049
##
File path: src/relay/ir/dataflow_matcher.cc
##
@@ -400,12 +400,16 @@ bool DFPatternMatcher::VisitDFPattern_(const
TupleGetItemPatternNode* op, const
bool
mbrookhart opened a new pull request #7329:
URL: https://github.com/apache/tvm/pull/7329
Fixes issue raised in
https://discuss.tvm.apache.org/t/tuplepattern-for-any-number-of-inputs-to-the-tuple/8926
I'm not sure I love the fix, the specialization of paritioning to support
ops that
tkonolige commented on pull request #7308:
URL: https://github.com/apache/tvm/pull/7308#issuecomment-765643886
I've added a todo (and put your name on it @icemelon9, hope that's ok).
This is an automated message from the
tkonolige commented on pull request #7153:
URL: https://github.com/apache/tvm/pull/7153#issuecomment-765624123
@tqchen @junrushao1994 @antinucleon @areusch dmlc-core has merged the
changes needed for the PR. I've update this PR to inject logging macros into
dmlc-core, so we now use our
merrymercy commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562840117
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
merrymercy commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562840117
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
merrymercy opened a new pull request #7328:
URL: https://github.com/apache/tvm/pull/7328
We do not implement kernel layout rewrite for `topi.nn.conv3d_ncdhw`, so we
should not pass `auto_scheduler_layout` to it.
This is an
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 676026a [AutoScheduler] Fix conv3d's op strategy for auto-scheduler
This branch includes the following
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 676026a959427d1f05ef7ab44886d5cb98ad2d10
Author: Lianmin Zheng
AuthorDate: Fri Jan 22 10:38:13 2021 -0800
tkonolige commented on a change in pull request #7267:
URL: https://github.com/apache/tvm/pull/7267#discussion_r562824062
##
File path: python/tvm/relay/frontend/tensorflow.py
##
@@ -941,9 +934,48 @@ def _impl(inputs, attr, params, mod):
(values_tensor, (rows,
tmoreau89 commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562820343
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
tmoreau89 commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562816620
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
tmoreau89 commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562816483
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
tmoreau89 commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562815910
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
tmoreau89 commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562815223
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
tmoreau89 commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562814613
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
comaniac commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562801935
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
ANSHUMAN87 commented on a change in pull request #7267:
URL: https://github.com/apache/tvm/pull/7267#discussion_r562791091
##
File path: python/tvm/relay/frontend/tensorflow.py
##
@@ -942,7 +942,10 @@ def _impl(inputs, attr, params, mod):
)
if sparse_lhs:
-
tkonolige commented on a change in pull request #7152:
URL: https://github.com/apache/tvm/pull/7152#discussion_r562750692
##
File path: include/tvm/runtime/packed_func.h
##
@@ -229,13 +229,13 @@ class TypedPackedFunc {
* \endcode
*
* \param typed_lambda typed
tkonolige commented on a change in pull request #7152:
URL: https://github.com/apache/tvm/pull/7152#discussion_r562750127
##
File path: include/tvm/runtime/packed_func.h
##
@@ -1240,16 +1293,19 @@ struct unpack_call_dispatcher {
template
struct unpack_call_dispatcher {
tkonolige commented on a change in pull request #7152:
URL: https://github.com/apache/tvm/pull/7152#discussion_r562750127
##
File path: include/tvm/runtime/packed_func.h
##
@@ -1240,16 +1293,19 @@ struct unpack_call_dispatcher {
template
struct unpack_call_dispatcher {
ANSHUMAN87 commented on a change in pull request #7308:
URL: https://github.com/apache/tvm/pull/7308#discussion_r562748580
##
File path: python/tvm/topi/cuda/sparse.py
##
@@ -295,7 +295,14 @@ def is_valid_for_sparse_dense_padded(data, weight_data):
"""
#
ANSHUMAN87 commented on a change in pull request #7267:
URL: https://github.com/apache/tvm/pull/7267#discussion_r562744658
##
File path: python/tvm/relay/frontend/tensorflow.py
##
@@ -942,7 +942,10 @@ def _impl(inputs, attr, params, mod):
)
if sparse_lhs:
-
tkonolige commented on a change in pull request #7152:
URL: https://github.com/apache/tvm/pull/7152#discussion_r562744494
##
File path: include/tvm/runtime/packed_func.h
##
@@ -1377,7 +1449,7 @@ inline TObjectRef TVMPODValue_::AsObjectRef() const {
using ContainerType =
ANSHUMAN87 commented on a change in pull request #7308:
URL: https://github.com/apache/tvm/pull/7308#discussion_r562730799
##
File path: python/tvm/topi/cuda/sparse.py
##
@@ -295,7 +295,14 @@ def is_valid_for_sparse_dense_padded(data, weight_data):
"""
#
tkonolige commented on a change in pull request #7308:
URL: https://github.com/apache/tvm/pull/7308#discussion_r562740167
##
File path: python/tvm/topi/cuda/sparse.py
##
@@ -295,7 +295,14 @@ def is_valid_for_sparse_dense_padded(data, weight_data):
"""
#
ANSHUMAN87 commented on a change in pull request #7308:
URL: https://github.com/apache/tvm/pull/7308#discussion_r562730799
##
File path: python/tvm/topi/cuda/sparse.py
##
@@ -295,7 +295,14 @@ def is_valid_for_sparse_dense_padded(data, weight_data):
"""
#
tqchen commented on pull request #7327:
URL: https://github.com/apache/tvm/pull/7327#issuecomment-765414996
THanks @SebastianBoblestETAS @leandron !
This is an automated message from the Apache Git Service.
To respond to the
This is an automated email from the ASF dual-hosted git repository.
tqchen 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 6787d74 get_top_results works on a copy of output
tqchen merged pull request #7327:
URL: https://github.com/apache/tvm/pull/7327
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
tqchen closed issue #7322:
URL: https://github.com/apache/tvm/issues/7322
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
leandron commented on a change in pull request #7326:
URL: https://github.com/apache/tvm/pull/7326#discussion_r562527228
##
File path: tutorials/auto_scheduler/tune_network_arm.py
##
@@ -0,0 +1,420 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more
SebastianBoblestETAS opened a new pull request #7327:
URL: https://github.com/apache/tvm/pull/7327
Closes #7322
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
67 matches
Mail list logo