Lunderberg commented on pull request #8532:
URL: https://github.com/apache/tvm/pull/8532#issuecomment-903756393
@areusch This PR is now rebased onto main to avoid conflicts with #8492, and
I've included the refactoring that you suggested in an earlier comment. All
integrations with
masahi opened a new pull request #8810:
URL: https://github.com/apache/tvm/pull/8810
Testing fp16 conversion on the EfficientDet D2 model from [TF2 detection
zoo](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md)
reveals that ops that
echuraev commented on a change in pull request #8636:
URL: https://github.com/apache/tvm/pull/8636#discussion_r693947325
##
File path: python/tvm/topi/gpu/conv2d_nhwc.py
##
@@ -85,15 +87,17 @@ def schedule_conv2d_nhwc_direct(cfg, s, Conv):
thread_yz = te.thread_axis((0,
mbaret commented on pull request #8795:
URL: https://github.com/apache/tvm/pull/8795#issuecomment-903770725
Also pinging some Arm people for review: @ekalda @NicolaLancellotti
@grant-arm @Leo-arm @tristan-arm @u99127
--
This is an automated message from the Apache Git Service.
To
Lunderberg commented on issue #8805:
URL: https://github.com/apache/tvm/issues/8805#issuecomment-903772650
Got it. In that case, can you try compiling with `USE_VULKAN=OFF`, and
`USE_LLVM=ON`?
--
This is an automated message from the Apache Git Service.
To respond to the message,
manupa-arm opened a new pull request #8811:
URL: https://github.com/apache/tvm/pull/8811
This commit introduces the TIR to Command Stream(CS) translation using Vela
API calls for conv2D and copy operations. It will create Vela npu_op objects
for each command.
This is blocked on
masahi commented on pull request #8285:
URL: https://github.com/apache/tvm/pull/8285#issuecomment-903641187
This change doesn't solve the issue in
https://github.com/apache/tvm/issues/8233, because `AllocDataSpace` can be
called from `NDArray::Empty`:
echuraev commented on a change in pull request #8636:
URL: https://github.com/apache/tvm/pull/8636#discussion_r693947325
##
File path: python/tvm/topi/gpu/conv2d_nhwc.py
##
@@ -85,15 +87,17 @@ def schedule_conv2d_nhwc_direct(cfg, s, Conv):
thread_yz = te.thread_axis((0,
Mousius opened a new pull request #8812:
URL: https://github.com/apache/tvm/pull/8812
I came across this file whilst looking at the C++ tests and realised it's a
duplicate of the PackedFunc tests which doesn't get invoked.
```
$ diff -u tests/cpp/contrib/bnns.cc
TaoCheng-CS commented on issue #7986:
URL: https://github.com/apache/tvm/issues/7986#issuecomment-903481534
> I have done the job, DLRM is already running on our NPU. If you have any
question, you can contact me.
您好, 我目前也遇到这个算子不支持的问题, 能给个联系方式吗?感谢!
--
This is an automated message from
comaniac edited a comment on pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#issuecomment-903496321
@tiandiao123 thanks for adding this support. Please add details to the PR
description.
@trevor-m it would be great if you could help sharpen this PR :)
--
This is
comaniac commented on pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#issuecomment-903496321
@tiandiao123 please add details to the PR description.
@trevor-m it would be great if you could help sharpen this PR :)
--
This is an automated message from the Apache Git
Mousius opened a new pull request #8809:
URL: https://github.com/apache/tvm/pull/8809
By using the `gtest_discover_tests` CMake macro the CPP and CRT tests can be
configured to build binaries with a single test runner each. Once CTest has
information about tests it can be used in IDE
masahi commented on a change in pull request #8781:
URL: https://github.com/apache/tvm/pull/8781#discussion_r693834511
##
File path: tests/python/frontend/pytorch/test_rnns.py
##
@@ -173,6 +253,115 @@ def compare(input, gold_data, rtol=1e-5, atol=1e-5):
leandron merged pull request #8762:
URL: https://github.com/apache/tvm/pull/8762
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
leeexyz commented on a change in pull request #8802:
URL: https://github.com/apache/tvm/pull/8802#discussion_r693698609
##
File path: src/relay/backend/te_compiler.h
##
@@ -166,30 +166,73 @@ void UpdateFunctionMetadata(Function relay_func,
/*!
* \brief Obtain the Target
tiandiao123 opened a new pull request #8808:
URL: https://github.com/apache/tvm/pull/8808
Thanks for contributing to TVM! Please refer to guideline
https://tvm.apache.org/docs/contribute/ for useful information and tips. After
the pull request is submitted, please request code reviews
This is an automated email from the ASF dual-hosted git repository.
leandron pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 0a1eae8 [TensorIR][M2a] Reorder (#8767)
add 10fca9c [microTVM] Fix platform name for qemu_x86 in Zephyr AOT
mehrdadh commented on pull request #8748:
URL: https://github.com/apache/tvm/pull/8748#issuecomment-903485972
@guberti could you please rebase this with main since the Arduino project
API landed now.
--
This is an automated message from the Apache Git Service.
To respond to the message,
mippos commented on issue #4102:
URL: https://github.com/apache/tvm/issues/4102#issuecomment-903548167
Hi! What is the status of Tensorflow2 support or schedule for it?
We use Tensorflow 2.x. Should TVM users still convert models back to TF1.12
or can we go with TF2 models?
(At
vvchernov commented on pull request #8781:
URL: https://github.com/apache/tvm/pull/8781#issuecomment-903498602
Seems that
https://github.com/apache/tvm/blob/main/tests/python/frontend/pytorch/test_forward.py#L4105
duplicates custom lstm test (pytest finds it by its-self). My lstm and gru
vvchernov commented on a change in pull request #8781:
URL: https://github.com/apache/tvm/pull/8781#discussion_r693711953
##
File path: python/tvm/relay/frontend/common.py
##
@@ -658,6 +658,92 @@ def unbind(data, axis=0):
return _expr.TupleWrapper(_expr.Tuple(ret),
wrongtest commented on pull request #8803:
URL: https://github.com/apache/tvm/pull/8803#issuecomment-903573759
@junrushao1994 Sorry for misleading information. What I tried is like
```python
func = param_in_arith_exprs.specialize({b: tir.decl_buffer([16])})
func =
masahi commented on pull request #8781:
URL: https://github.com/apache/tvm/pull/8781#issuecomment-903623984
Please try removing duplication between GRU and LSTM tests. For example, I
don't want to see
```
for batch_first in (True, False):
for use_bias in (True,
FrozenGene commented on a change in pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#discussion_r694443709
##
File path: src/runtime/contrib/tensorrt/tensorrt_runtime.cc
##
@@ -66,7 +78,16 @@ class TensorRTRuntime : public JSONRuntimeBase {
tiandiao123 commented on a change in pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#discussion_r694502009
##
File path: src/runtime/contrib/tensorrt/tensorrt_runtime.cc
##
@@ -66,7 +78,16 @@ class TensorRTRuntime : public JSONRuntimeBase {
junrushao1994 commented on a change in pull request #8817:
URL: https://github.com/apache/tvm/pull/8817#discussion_r694464837
##
File path: tests/python/unittest/test_tir_schedule_sampling.py
##
@@ -0,0 +1,95 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+#
This is an automated email from the ASF dual-hosted git repository.
junrushao pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from e883dcb Run AOT tests against reference system (#8744)
add eabae30 [Rust] Fix memory leak #2 (#8725)
No new
junrushao1994 merged pull request #8725:
URL: https://github.com/apache/tvm/pull/8725
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
jroesch commented on pull request #8795:
URL: https://github.com/apache/tvm/pull/8795#issuecomment-904289451
cc @mbs-octoml @mikepapadim @electriclilies @csullivan
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
jroesch commented on a change in pull request #8795:
URL: https://github.com/apache/tvm/pull/8795#discussion_r694460440
##
File path: python/tvm/relay/backend/contrib/ethosu/__init__.py
##
@@ -0,0 +1,23 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or
FrozenGene commented on a change in pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#discussion_r694443709
##
File path: src/runtime/contrib/tensorrt/tensorrt_runtime.cc
##
@@ -66,7 +78,16 @@ class TensorRTRuntime : public JSONRuntimeBase {
jcf94 commented on a change in pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#discussion_r69768
##
File path: src/runtime/contrib/tensorrt/tensorrt_runtime.cc
##
@@ -66,7 +78,16 @@ class TensorRTRuntime : public JSONRuntimeBase {
kthready commented on issue #8805:
URL: https://github.com/apache/tvm/issues/8805#issuecomment-904294277
Verified, it works.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific
kthready closed issue #8805:
URL: https://github.com/apache/tvm/issues/8805
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
jroesch merged pull request #8744:
URL: https://github.com/apache/tvm/pull/8744
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
junrushao1994 commented on a change in pull request #8817:
URL: https://github.com/apache/tvm/pull/8817#discussion_r694394696
##
File path: src/tir/schedule/concrete_schedule.cc
##
@@ -208,6 +211,25 @@ Schedule ConcreteScheduleNode::Copy() const {
}
/ Schedule:
junrushao1994 commented on a change in pull request #8817:
URL: https://github.com/apache/tvm/pull/8817#discussion_r694394696
##
File path: src/tir/schedule/concrete_schedule.cc
##
@@ -208,6 +211,25 @@ Schedule ConcreteScheduleNode::Copy() const {
}
/ Schedule:
junrushao1994 commented on a change in pull request #8817:
URL: https://github.com/apache/tvm/pull/8817#discussion_r694394696
##
File path: src/tir/schedule/concrete_schedule.cc
##
@@ -208,6 +211,25 @@ Schedule ConcreteScheduleNode::Copy() const {
}
/ Schedule:
jroesch merged pull request #8796:
URL: https://github.com/apache/tvm/pull/8796
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from d1f19c4 Add LowerTEPass, and convert calls to LowerTE to application
of LowerTEPass (#8802)
add b1396a0 [FIX]
jroesch merged pull request #8819:
URL: https://github.com/apache/tvm/pull/8819
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
jroesch merged pull request #8818:
URL: https://github.com/apache/tvm/pull/8818
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from dfe21c5 Remove unused allocated memory in crt initialization (#8819)
add 4524567 Remove unnecessary memset in
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from b1396a0 [FIX] Remove leftover instances of USE_GRAPH_EXECUTOR_DEBUG
(#8796)
add dfe21c5 Remove unused allocated
tiandiao123 commented on a change in pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#discussion_r694501830
##
File path: src/runtime/contrib/tensorrt/tensorrt_runtime.cc
##
@@ -66,7 +78,16 @@ class TensorRTRuntime : public JSONRuntimeBase {
FrozenGene commented on pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#issuecomment-904273280
@jcf94 please help to review it too as you have implemented similar
functionality
--
This is an automated message from the Apache Git Service.
To respond to the message,
jcf94 commented on a change in pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#discussion_r69768
##
File path: src/runtime/contrib/tensorrt/tensorrt_runtime.cc
##
@@ -66,7 +78,16 @@ class TensorRTRuntime : public JSONRuntimeBase {
tiandiao123 commented on a change in pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#discussion_r694464676
##
File path: src/runtime/contrib/tensorrt/tensorrt_builder.cc
##
@@ -156,8 +161,18 @@ TensorRTEngineAndContext TensorRTBuilder::BuildEngine() {
tiandiao123 commented on a change in pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#discussion_r694464716
##
File path: src/runtime/contrib/tensorrt/tensorrt_runtime.cc
##
@@ -125,17 +147,26 @@ class TensorRTRuntime : public JSONRuntimeBase {
/*! \brief
jroesch commented on pull request #8823:
URL: https://github.com/apache/tvm/pull/8823#issuecomment-904290742
cc @zxybazh can you take a look at this one?
--
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
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 4524567 Remove unnecessary memset in TVMMutableFuncRegistry
initialization (#8818)
add e883dcb Run AOT tests
jroesch merged pull request #8802:
URL: https://github.com/apache/tvm/pull/8802
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
This is an automated email from the ASF dual-hosted git repository.
jroesch pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git
The following commit(s) were added to refs/heads/main by this push:
new d1f19c4 Add LowerTEPass, and convert calls to
jroesch edited a comment on pull request #8796:
URL: https://github.com/apache/tvm/pull/8796#issuecomment-904301596
Going to land this @comaniac to empty the PR queue feel free to request
follow up changes if I missed something.
--
This is an automated message from the Apache Git
jroesch commented on pull request #8796:
URL: https://github.com/apache/tvm/pull/8796#issuecomment-904301596
Going to land this @comaniac feel free to request follow up changes if I
missed something.
--
This is an automated message from the Apache Git Service.
To respond to the
zxybazh commented on a change in pull request #8823:
URL: https://github.com/apache/tvm/pull/8823#discussion_r694504182
##
File path: python/tvm/target/target.py
##
@@ -510,23 +548,27 @@ def validate_hvx_length(codegen_hvx, sim_args):
+ cpu_attr["post"]
Lunderberg opened a new pull request #8813:
URL: https://github.com/apache/tvm/pull/8813
Was added in #8127, should have been removed in #8171 along with the rest of
the references outside of libtvm_runtime.so. This didn't impact the Vulkan+g++
builds, because no symbols were accessed
manupa-arm commented on a change in pull request #11:
URL: https://github.com/apache/tvm-rfcs/pull/11#discussion_r694117407
##
File path: rfcs/0011_Arm_Ethos-U_Integration.md
##
@@ -192,42 +191,42 @@ TVM_DLL int32_t ethosu_0(TVMValue* args, int* type_code,
int num_args,
kthready commented on issue #8805:
URL: https://github.com/apache/tvm/issues/8805#issuecomment-903886458
OK, thx, I will verify it tmr.
--
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
jtuyls opened a new pull request #8814:
URL: https://github.com/apache/tvm/pull/8814
This PR updates the CI Vitis AI PyXIR version to v0.3.1. This is needed for
testing a subsequent PR that updates the Vitis AI docker and documentation to
Vitis AI 1.4. The ReLU test is removed as it fails
junrushao1994 commented on pull request #8809:
URL: https://github.com/apache/tvm/pull/8809#issuecomment-903950679
Hey I sort of understand ctest is the standard cmake-way to run tests, but
would you mind giving some instructions on this? Thanks a lot!
--
This is an automated message
huanmei9 commented on pull request #8770:
URL: https://github.com/apache/tvm/pull/8770#issuecomment-903801393
Hi, @Mousius .
The situation you describe is indeed possible, Thanks for your suggestion.
(●'◡'●)
--
This is an automated message from the Apache Git Service.
To
huanmei9 closed pull request #8770:
URL: https://github.com/apache/tvm/pull/8770
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
jtuyls opened a new pull request #8815:
URL: https://github.com/apache/tvm/pull/8815
This PR updates the Vitis AI docker and documentation to the Vitis AI 1.4
release. The following PR should be merged first for tests to pass:
https://github.com/apache/tvm/pull/8814.
@comaniac
Lunderberg commented on issue #8805:
URL: https://github.com/apache/tvm/issues/8805#issuecomment-903847237
Found the issue. I had a `#include ` that should have
been removed. #8813 removes the include, and should resolve the issue. Can
you test and verify?
--
This is an automated
vinx13 commented on pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#issuecomment-903853319
cc @FrozenGene
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
ekiwi commented on pull request #34:
URL: https://github.com/apache/tvm-vta/pull/34#issuecomment-903853739
Thanks @vegaluisjose ! Unfortunately the tests that I knew about did not
trigger these bugs.
--
This is an automated message from the Apache Git Service.
To respond to the message,
manupa-arm commented on pull request #11:
URL: https://github.com/apache/tvm-rfcs/pull/11#issuecomment-903921134
Hi all,
I think all the comments are addressed now.
Waiting for approval or comments.
--
This is an automated message from the Apache Git Service.
To respond to the
Laurawly commented on pull request #8808:
URL: https://github.com/apache/tvm/pull/8808#issuecomment-903949096
> Hi, this commit is used to add calibration and tensorrt int8 mode in the
original src/runtime/contrib/tensorrt. Currently, I tested several models
including mobilenet_v2 and
Lunderberg edited a comment on issue #8805:
URL: https://github.com/apache/tvm/issues/8805#issuecomment-903772650
~~Got it. In that case, can you try compiling with `USE_VULKAN=OFF`, and
`USE_LLVM=ON`?~~ Never mind, see below.
--
This is an automated message from the Apache Git
mdw-octoml commented on a change in pull request #8807:
URL: https://github.com/apache/tvm/pull/8807#discussion_r694123503
##
File path: python/tvm/runtime/module.py
##
@@ -30,8 +31,65 @@
from . import _ffi_api
-# profile result of time evaluator
-ProfileResult =
tkonolige commented on a change in pull request #8807:
URL: https://github.com/apache/tvm/pull/8807#discussion_r694170135
##
File path: python/tvm/runtime/module.py
##
@@ -230,8 +288,7 @@ def evaluator(*args):
blob = feval(*args)
fmt = "@" +
mbs-octoml commented on a change in pull request #8788:
URL: https://github.com/apache/tvm/pull/8788#discussion_r694195744
##
File path: src/relay/backend/aot_executor_codegen.cc
##
@@ -562,32 +558,35 @@ class AOTExecutorCodegen : public ExprVisitor {
zxybazh opened a new pull request #8817:
URL: https://github.com/apache/tvm/pull/8817
This PR upstreams the first sampling primitive for meta schedule,
`SampleCategorical`, which samples an `Integer` with given probability
distribution. The implementation utilizes the linear congruential
csullivan commented on a change in pull request #23:
URL: https://github.com/apache/tvm-rfcs/pull/23#discussion_r694239596
##
File path: rfcs/0023-associating-allocate-nodes-with-pinned-memories.md
##
@@ -0,0 +1,158 @@
+
+- Feature Name: TIR Pinned Memory Representation
+-
electriclilies commented on pull request #8802:
URL: https://github.com/apache/tvm/pull/8802#issuecomment-904068910
@Mousius Thanks so much for figuring that out, I really appreciate it! I
will put these changes in a follow-up PR-- switching between my two branches is
getting a bit
This is an automated email from the ASF dual-hosted git repository.
wuwei pushed a change to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git.
from 10fca9c [microTVM] Fix platform name for qemu_x86 in Zephyr AOT tests
(#8762)
add aafc2d5 Remove duplicated
tkonolige commented on a change in pull request #8807:
URL: https://github.com/apache/tvm/pull/8807#discussion_r694264620
##
File path: python/tvm/runtime/module.py
##
@@ -230,8 +288,7 @@ def evaluator(*args):
blob = feval(*args)
fmt = "@" +
vinx13 merged pull request #8812:
URL: https://github.com/apache/tvm/pull/8812
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail:
masahi commented on a change in pull request #8810:
URL: https://github.com/apache/tvm/pull/8810#discussion_r694304633
##
File path: tests/python/relay/test_to_mixed_precision.py
##
@@ -241,6 +241,28 @@ def test_do_not_convert_arange():
assert tvm.ir.structural_equal(mod,
comaniac commented on a change in pull request #8810:
URL: https://github.com/apache/tvm/pull/8810#discussion_r694305322
##
File path: tests/python/relay/test_to_mixed_precision.py
##
@@ -241,6 +241,28 @@ def test_do_not_convert_arange():
assert
vegaluisjose commented on a change in pull request #8797:
URL: https://github.com/apache/tvm/pull/8797#discussion_r694316957
##
File path: cmake/modules/VTA.cmake
##
@@ -73,15 +73,30 @@ elseif(PYTHON)
# Cycle accurate simulator driver build
if(USE_VTA_TSIM)
+if(NOT
vegaluisjose commented on a change in pull request #8797:
URL: https://github.com/apache/tvm/pull/8797#discussion_r694316957
##
File path: cmake/modules/VTA.cmake
##
@@ -73,15 +73,30 @@ elseif(PYTHON)
# Cycle accurate simulator driver build
if(USE_VTA_TSIM)
+if(NOT
junrushao1994 commented on pull request #8809:
URL: https://github.com/apache/tvm/pull/8809#issuecomment-904145694
CC: @areusch @jroesch @Lunderberg for a second look
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
vinx13 commented on a change in pull request #8820:
URL: https://github.com/apache/tvm/pull/8820#discussion_r694322843
##
File path: python/tvm/autotvm/tuner/xgboost_cost_model.py
##
@@ -327,14 +329,9 @@ def _get_feature(self, indexes):
if need_extract:
mbs-octoml commented on pull request #8382:
URL: https://github.com/apache/tvm/pull/8382#issuecomment-904099250
Hi Michalis, is this one ready to go in? I think if you push to get past the
Jenkins hickup it should proceed through ci as usual.
--
This is an automated message from the
aasorokiin commented on a change in pull request #8797:
URL: https://github.com/apache/tvm/pull/8797#discussion_r694297979
##
File path: cmake/modules/VTA.cmake
##
@@ -73,15 +73,30 @@ elseif(PYTHON)
# Cycle accurate simulator driver build
if(USE_VTA_TSIM)
+if(NOT
shingjan opened a new pull request #8820:
URL: https://github.com/apache/tvm/pull/8820
Thanks for contributing to TVM! Please refer to guideline
https://tvm.apache.org/docs/contribute/ for useful information and tips. After
the pull request is submitted, please request code reviews from
junrushao1994 commented on pull request #8809:
URL: https://github.com/apache/tvm/pull/8809#issuecomment-904145240
@Mousius Thanks for the detailed explanation! That's exactly I am interested
in.
--
This is an automated message from the Apache Git Service.
To respond to the message,
vinx13 commented on a change in pull request #8820:
URL: https://github.com/apache/tvm/pull/8820#discussion_r694323839
##
File path: python/tvm/autotvm/tuner/xgboost_cost_model.py
##
@@ -144,6 +146,12 @@ def __init__(
self._sample_size = 0
shingjan commented on a change in pull request #8820:
URL: https://github.com/apache/tvm/pull/8820#discussion_r694326004
##
File path: python/tvm/autotvm/tuner/xgboost_cost_model.py
##
@@ -144,6 +146,12 @@ def __init__(
self._sample_size = 0
junrushao1994 edited a comment on pull request #8645:
URL: https://github.com/apache/tvm/pull/8645#issuecomment-904074969
Thanks @leandron @areusch! It looks like ci-cpu:v0.76 does include the
paddlepaddle dependency, according to the `ci-docker-staging` branch, but I am
not 100% sure.
guberti commented on pull request #8748:
URL: https://github.com/apache/tvm/pull/8748#issuecomment-904150262
@mehrdadh PR has been rebased, and is now ready for review.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
vinx13 commented on pull request #8799:
URL: https://github.com/apache/tvm/pull/8799#issuecomment-904153346
Merging these functions sounds good to me. Note that we have unified
operators in Relay level, it might be fine to lower to different topi operators
if this is simpler
--
This is
kparzysz-quic opened a new pull request #8821:
URL: https://github.com/apache/tvm/pull/8821
The TVM runtime is not linked with LLVM libraries, so using LLVM in it
carries a risk of referencing undefined symbols. This may
work for objects defined in header files, but it then relies on
AndrewZhaoLuo commented on pull request #8810:
URL: https://github.com/apache/tvm/pull/8810#issuecomment-904159480
Hmm, did you get a chance to try FP32 accumulators for these operations
@masahi ? If not, no worries but I'm curious if turning that on will make the
data loss less.
--
kparzysz-quic opened a new pull request #8822:
URL: https://github.com/apache/tvm/pull/8822
Different versions of the Hexagon SDK may have different directory
structures. Extract the directory identification code into a separate cmake
module. Use that module in `Hexagon.cmake` and in the
mdw-octoml commented on a change in pull request #8807:
URL: https://github.com/apache/tvm/pull/8807#discussion_r694180020
##
File path: python/tvm/runtime/module.py
##
@@ -230,8 +288,7 @@ def evaluator(*args):
blob = feval(*args)
fmt = "@" +
mbs-octoml commented on a change in pull request #8788:
URL: https://github.com/apache/tvm/pull/8788#discussion_r694194725
##
File path: include/tvm/runtime/logging.h
##
@@ -409,6 +411,57 @@ inline bool DebugLoggingEnabled() {
return state == 1;
}
+/*!
+ * \brief Returns
1 - 100 of 165 matches
Mail list logo