icemelon9 commented on a change in pull request #5955:
URL: https://github.com/apache/incubator-tvm/pull/5955#discussion_r456753279
##
File path: python/tvm/relay/op/image/_image.py
##
@@ -64,6 +67,22 @@ def compute_crop_and_resize(attrs, inputs, out_type):
icemelon9 commented on pull request #6040:
URL: https://github.com/apache/incubator-tvm/pull/6040#issuecomment-660431091
Probably also worth checking with @kevinthesun. I remember last time he saw
some performance regression due to the injective schedule.
icemelon9 commented on pull request #6040:
URL: https://github.com/apache/incubator-tvm/pull/6040#issuecomment-660430944
@merrymercy Yes, I agree with @tqchen. If c++ version works well, we could
just remove the python implementation and use c++ one instead.
anijain2305 closed pull request #5354:
URL: https://github.com/apache/incubator-tvm/pull/5354
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
zhiics commented on pull request #6089:
URL: https://github.com/apache/incubator-tvm/pull/6089#issuecomment-660424453
@mbrookhart Yeah, I meant there is gpu context in the test. That's why the
ci fails. We can probably just disable check_result for this test ATM.
Heterogeneous execution
zhanghaohit commented on a change in pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#discussion_r456748692
##
File path: python/tvm/relay/op/strategy/generic.py
##
@@ -323,8 +331,10 @@ def compute_conv2d_transpose(attrs, inputs, out_dtype):
comaniac commented on issue #6090:
URL: https://github.com/apache/incubator-tvm/issues/6090#issuecomment-660422421
Could you provide more details for the crashing you encountered? DNNL
codegen is mainly for BYOC demonstration purpose and it's not in production for
now.
On the other
ZihengJiang commented on pull request #6079:
URL: https://github.com/apache/incubator-tvm/pull/6079#issuecomment-660420778
Hey Haibin , thanks for your help!
Could you try to enhance the test case with mixed `Tensor` and `Scalar`
inputs like:
```
zhanghaohit commented on a change in pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#discussion_r456745434
##
File path: python/tvm/autotvm/measure/measure_methods.py
##
@@ -186,6 +189,16 @@ def __init__(self,
timeout=10,
zhanghaohit edited a comment on pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#issuecomment-659790884
> Note that I've also uploaded the Pynq bitstream given the ISA changes
here: https://github.com/uwsampl/vta-distro/tree/master/bitstreams/pynq/0.0.2
>
>
yzhliu opened a new pull request #6091:
URL: https://github.com/apache/incubator-tvm/pull/6091
@tqchen
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
remotego commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456738642
##
File path: config/intelfocl_sample.json
##
@@ -0,0 +1,13 @@
+{
+ "TARGET" : "intelfocl",
Review comment:
I believe "Arria 10" is too
remotego edited a comment on pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660411735
> Thanks for the changes. Please apply the 0.0.2 and rename the vta target
to something more specific, e.g. "arria10". Also there are some CI errors
related to
remotego commented on pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660411735
> Thanks for the changes. Please apply the 0.0.2 and rename the vta target
to something more specific, e.g. "arria10". Also there are some CI errors
related to linting
mikeseven opened a new issue #6090:
URL: https://github.com/apache/incubator-tvm/issues/6090
Using the test_pass_partition_graph.py and whitelisting nn.conv2d to be
built with ccompiler crashes in codegen.cc because it seems to support only the
simple operations in this test!
Seems
remotego commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456738642
##
File path: config/intelfocl_sample.json
##
@@ -0,0 +1,13 @@
+{
+ "TARGET" : "intelfocl",
Review comment:
I believe "Arria 10" is too
fernchen commented on pull request #6023:
URL: https://github.com/apache/incubator-tvm/pull/6023#issuecomment-660408499
> > 1. We add caffe frontend based on SSD distribution, then this ci pr is
good enough to test;
> > 2. We add caffe frontend based on our maintaining tvm caffe, then I
mbrookhart commented on pull request #6089:
URL: https://github.com/apache/incubator-tvm/pull/6089#issuecomment-660406275
The GPU error seems to come from the check I added to fail early on GPU with
dynamic models. I believe dynamic shapes wont work on cuda until we get
heterogenous
zhiics edited a comment on pull request #6089:
URL: https://github.com/apache/incubator-tvm/pull/6089#issuecomment-660402821
Please check the error. We have gpu context in the check_result.
This is an automated message from
zhiics commented on pull request #6089:
URL: https://github.com/apache/incubator-tvm/pull/6089#issuecomment-660402821
Please check the error. It cannot be executed on GPU.
This is an automated message from the Apache Git
MarisaKirisame commented on pull request #6080:
URL: https://github.com/apache/incubator-tvm/pull/6080#issuecomment-660367022
Thanks @mbrookhart for the work!
This is an automated message from the Apache Git Service.
To
MarisaKirisame merged pull request #6080:
URL: https://github.com/apache/incubator-tvm/pull/6080
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
This is an automated email from the ASF dual-hosted git repository.
marisa pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from ccacb1e Fixed point multiplication improvements for AArch64 (#5980)
add 3150db7 [Relay][Dyn] Add
tmoreau89 commented on pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#issuecomment-660364642
Thanks for the changes. Please apply the 0.0.2 and rename the vta target to
something more specific, e.g. "arria10". Also there are some CI errors related
to linting that
icemelon9 opened a new pull request #6089:
URL: https://github.com/apache/incubator-tvm/pull/6089
@jroesch @zhiics @mbrookhart Could you help review?
This is an automated message from the Apache Git Service.
To respond to
tmoreau89 commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456699804
##
File path: config/intelfocl_sample.json
##
@@ -0,0 +1,13 @@
+{
+ "TARGET" : "intelfocl",
+ "HW_VER" : "0.0.1",
Review comment:
Please
tmoreau89 commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456699307
##
File path: config/intelfocl_sample.json
##
@@ -0,0 +1,13 @@
+{
+ "TARGET" : "intelfocl",
Review comment:
SO I believe that the solution
junrushao1994 commented on a change in pull request #6077:
URL: https://github.com/apache/incubator-tvm/pull/6077#discussion_r456695376
##
File path: tests/python/unittest/test_auto_scheduler_measure.py
##
@@ -67,6 +67,26 @@ def test_measure_local_builder_runner():
assert
tmoreau89 commented on pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#issuecomment-660356504
I'm thinking that the `static auto-tune using profiling results` can also be
separated out in a standalone PR
tmoreau89 commented on pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#issuecomment-660355629
@zhanghaohit I took a second look at the changes in this PR, and I believe
it would be best to break the PR down into separate, smaller PRs:
(1) modifications to
tmoreau89 commented on a change in pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#discussion_r456673553
##
File path: python/tvm/relay/op/strategy/generic.py
##
@@ -323,8 +331,10 @@ def compute_conv2d_transpose(attrs, inputs, out_dtype):
tmoreau89 commented on a change in pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#discussion_r456673094
##
File path: python/tvm/autotvm/measure/measure_methods.py
##
@@ -186,6 +189,16 @@ def __init__(self,
timeout=10,
tmoreau89 commented on a change in pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#discussion_r456671336
##
File path: vta/python/vta/transform.py
##
@@ -381,9 +381,10 @@ def _fold_buffer_dim(buf, scope, elem_block):
def _get_2d_pattern(buf,
tmoreau89 commented on a change in pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#discussion_r456670755
##
File path: vta/python/vta/environment.py
##
@@ -241,7 +243,7 @@ def target_host(self):
return "llvm
jroesch commented on pull request #5892:
URL: https://github.com/apache/incubator-tvm/pull/5892#issuecomment-660317393
I have an open PR https://github.com/apache/incubator-tvm/pull/6011 that is
trying to move all Rust code to the updated bindings it might make sense to
land this against
anijain2305 commented on pull request #6039:
URL: https://github.com/apache/incubator-tvm/pull/6039#issuecomment-660275132
@icemelon9 Can you please manage this PR?
This is an automated message from the Apache Git Service.
csullivan commented on a change in pull request #6069:
URL: https://github.com/apache/incubator-tvm/pull/6069#discussion_r456595165
##
File path: src/tir/ir/data_layout.cc
##
@@ -323,7 +323,12 @@ inline Array TransformShape(const
Array& src_shape,
if
merrymercy commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456549272
##
File path: src/auto_scheduler/compute_dag.cc
##
@@ -276,10 +276,18 @@ std::pair>
ComputeDAG::ApplySteps(
// return value, so the
anijain2305 commented on pull request #5980:
URL: https://github.com/apache/incubator-tvm/pull/5980#issuecomment-660197332
Thanks @giuseros @tmoreau89 @kparzysz-quic This is merged!
This is an automated message from the
This is an automated email from the ASF dual-hosted git repository.
anijain2305 pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 3c12a5e [Test] Add missing test for fast erf (#6058)
add ccacb1e Fixed point multiplication
icemelon9 merged pull request #6058:
URL: https://github.com/apache/incubator-tvm/pull/6058
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
tmoreau89 commented on pull request #5842:
URL: https://github.com/apache/incubator-tvm/pull/5842#issuecomment-660185671
@zhanghaohit thanks for the follow up. I recommend to echo @liangfu to
rebase so the PR can be in a mergeable state. In addition, if the target is a
custom FPGA board,
tmoreau89 commented on pull request #5980:
URL: https://github.com/apache/incubator-tvm/pull/5980#issuecomment-660183790
@anijain2305 please let us know if your requested changes have been
addressed! thanks!
This is an
merrymercy edited a comment on pull request #6040:
URL: https://github.com/apache/incubator-tvm/pull/6040#issuecomment-660172683
Currently, both c++ and python version work well. This PR is ready to be
merged from my side. cc @icemelon9
I have no idea how to solve the mess of
merrymercy commented on pull request #6040:
URL: https://github.com/apache/incubator-tvm/pull/6040#issuecomment-660172510
Currently, both c++ and python versions work well. This PR is ready to be
merged from my side.
I have no idea how to solve the mess of duplicated python and c++ topi
merrymercy commented on pull request #6040:
URL: https://github.com/apache/incubator-tvm/pull/6040#issuecomment-660172683
Currently, both c++ and python versions work well. This PR is ready to be
merged from my side. cc @icemelon9
I have no idea how to solve the mess of duplicated
merrymercy removed a comment on pull request #6040:
URL: https://github.com/apache/incubator-tvm/pull/6040#issuecomment-660172510
Currently, both c++ and python versions work well. This PR is ready to be
merged from my side.
I have no idea how to solve the mess of duplicated python and
tqchen commented on pull request #6040:
URL: https://github.com/apache/incubator-tvm/pull/6040#issuecomment-660150798
If c++ version is good we can call into the c++ version from the pyone
This is an automated message from
tqchen commented on pull request #6083:
URL: https://github.com/apache/incubator-tvm/pull/6083#issuecomment-660150229
Thanks @tkonolige !
This is an automated message from the Apache Git Service.
To respond to the message,
tqchen merged pull request #6083:
URL: https://github.com/apache/incubator-tvm/pull/6083
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
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 3ddce13 [Fix] Add missing expr visitor for any (#6082)
add 378a85a Fix LocalBuilder on macos with
tqchen commented on a change in pull request #6069:
URL: https://github.com/apache/incubator-tvm/pull/6069#discussion_r456491809
##
File path: src/tir/ir/data_layout.cc
##
@@ -323,7 +323,12 @@ inline Array TransformShape(const
Array& src_shape,
if
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 6e1b09e [TOPI] Fix the filter width parameter in depthwise_conv2d
(#6081)
add 3ddce13 [Fix] Add
tqchen merged pull request #6082:
URL: https://github.com/apache/incubator-tvm/pull/6082
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
tqchen commented on pull request #6082:
URL: https://github.com/apache/incubator-tvm/pull/6082#issuecomment-660148843
Thanks @icemelon9 @junrushao1994
This is an automated message from the Apache Git Service.
To respond to
cam145 opened a new issue #6088:
URL: https://github.com/apache/incubator-tvm/issues/6088
Thanks for participating in the TVM community! We use https://discuss.tvm.ai
for any general usage questions and discussions. The issue tracker is used for
actionable items such as feature proposals
siju-samuel commented on pull request #6023:
URL: https://github.com/apache/incubator-tvm/pull/6023#issuecomment-660085200
> 1. We add caffe frontend based on SSD distribution, then this ci pr is
good enough to test;
> 2. We add caffe frontend based on our maintaining tvm caffe, then I
giuseros commented on pull request #5980:
URL: https://github.com/apache/incubator-tvm/pull/5980#issuecomment-660067570
Hi @anijain2305 , @tmoreau89 , @kparzysz-quic
I addressed the name change, reformatted the help text in `math.py` and
addressed an integration change in the arm
giuseros commented on a change in pull request #5980:
URL: https://github.com/apache/incubator-tvm/pull/5980#discussion_r456394839
##
File path: topi/python/topi/math.py
##
@@ -612,6 +612,31 @@ def _compute(*indices):
return tvm.te.max(tvm.te.min(value, const_max),
HakanKumdakci opened a new issue #6087:
URL: https://github.com/apache/incubator-tvm/issues/6087
I used this commmand in new cloned repository of incubator-tvm.
`docker build -t tvm.demo_android -f docker/Dockerfile.demo_android ./docker`
And got this error
FrozenGene edited a comment on pull request #5913:
URL: https://github.com/apache/incubator-tvm/pull/5913#issuecomment-660027353
> see comments, we don't need to expose the(non_empty) function as a
primitive function, given that it is not a numpy function, and is only used in
AutoTVM,
FrozenGene commented on pull request #5913:
URL: https://github.com/apache/incubator-tvm/pull/5913#issuecomment-660027353
> see comments, we don't need to expose the(non_empty) function as a
primitive function, given that it is not a numpy function, and is only used in
AutoTVM, instead we
lhutton1 commented on a change in pull request #5915:
URL: https://github.com/apache/incubator-tvm/pull/5915#discussion_r456353670
##
File path: src/relay/backend/contrib/arm_compute_lib/codegen.cc
##
@@ -0,0 +1,186 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF)
manupa-arm commented on a change in pull request #5915:
URL: https://github.com/apache/incubator-tvm/pull/5915#discussion_r456314032
##
File path: src/relay/backend/contrib/arm_compute_lib/codegen.cc
##
@@ -0,0 +1,186 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF)
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456321361
##
File path: src/auto_scheduler/utils.h
##
@@ -89,6 +89,15 @@ inline int GetIndex(const Array& array, const T&
to_locate) {
return -1;
}
+/*!
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456321060
##
File path: tests/python/unittest/test_auto_scheduler_loop_state.py
##
@@ -61,5 +61,79 @@ def test_split_fuse_reorder():
assert
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456320683
##
File path: tests/python/unittest/test_auto_scheduler_loop_state.py
##
@@ -61,5 +61,79 @@ def test_split_fuse_reorder():
assert
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456319743
##
File path: src/auto_scheduler/loop_state.cc
##
@@ -143,12 +265,67 @@ void State::DoReorderStep(const ReorderStep& step) {
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456319452
##
File path: src/auto_scheduler/loop_state.cc
##
@@ -90,12 +90,69 @@ Stage::Stage(te::Operation op, StageKind op_type, const
Array& iters,
data_
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456319021
##
File path: src/auto_scheduler/loop_state.h
##
@@ -217,6 +196,68 @@ class Stage : public ObjectRef {
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456318718
##
File path: python/tvm/auto_scheduler/loop_state.py
##
@@ -161,16 +202,116 @@ def fuse(self, stage, iters):
The Stage to be fused, can
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456318375
##
File path: python/tvm/auto_scheduler/loop_state.py
##
@@ -161,16 +202,116 @@ def fuse(self, stage, iters):
The Stage to be fused, can
lhutton1 commented on a change in pull request #5915:
URL: https://github.com/apache/incubator-tvm/pull/5915#discussion_r456313568
##
File path: src/runtime/contrib/arm_compute_lib/acl_runtime.cc
##
@@ -0,0 +1,310 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF)
lhutton1 commented on a change in pull request #5915:
URL: https://github.com/apache/incubator-tvm/pull/5915#discussion_r456312016
##
File path: src/runtime/contrib/arm_compute_lib/acl_runtime.cc
##
@@ -0,0 +1,310 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF)
lhutton1 commented on a change in pull request #5915:
URL: https://github.com/apache/incubator-tvm/pull/5915#discussion_r456310867
##
File path: src/relay/backend/contrib/arm_compute_lib/codegen.cc
##
@@ -0,0 +1,186 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF)
mbaret commented on a change in pull request #5915:
URL: https://github.com/apache/incubator-tvm/pull/5915#discussion_r456115777
##
File path: python/tvm/relay/op/contrib/arm_compute_lib.py
##
@@ -0,0 +1,122 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+#
remotego commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456267226
##
File path: src/intelfocl/intelfocl_device.cc
##
@@ -0,0 +1,185 @@
+#include
+#include
+#include
+#include "intelfocl_device.h"
+#include
remotego commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456264532
##
File path: config/intelfocl_sample.json
##
@@ -0,0 +1,13 @@
+{
+ "TARGET" : "intelfocl",
+ "HW_VER" : "0.0.1",
Review comment:
Agreed.
remotego commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456264177
##
File path: src/intelfocl/intelfocl_device.cc
##
@@ -0,0 +1,185 @@
+#include
+#include
+#include
+#include "intelfocl_device.h"
+#include
hzfan commented on a change in pull request #5235:
URL: https://github.com/apache/incubator-tvm/pull/5235#discussion_r456261216
##
File path: tests/python/relay/test_pass_fuse_ops.py
##
@@ -621,6 +621,81 @@ def expected():
after = run_opt_pass(expected(),
remotego commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456259647
##
File path: config/intelfocl_sample.json
##
@@ -0,0 +1,13 @@
+{
+ "TARGET" : "intelfocl",
+ "HW_VER" : "0.0.1",
+ "LOG_INP_WIDTH" : 3,
Review
remotego commented on a change in pull request #9:
URL: https://github.com/apache/incubator-tvm-vta/pull/9#discussion_r456257771
##
File path: hardware/intelfocl/src/vta.cl
##
@@ -0,0 +1,341 @@
+#pragma OPENCL EXTENSION cl_intel_channels: enable
Review comment:
The
junrushao1994 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456257744
##
File path: src/auto_scheduler/loop_state.h
##
@@ -217,6 +196,68 @@ class Stage : public ObjectRef {
hzfan commented on a change in pull request #5235:
URL: https://github.com/apache/incubator-tvm/pull/5235#discussion_r456257305
##
File path: tests/python/relay/test_pass_fuse_ops.py
##
@@ -621,6 +621,81 @@ def expected():
after = run_opt_pass(expected(),
comaniac commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456255907
##
File path: src/auto_scheduler/loop_state.cc
##
@@ -90,12 +90,69 @@ Stage::Stage(te::Operation op, StageKind op_type, const
Array& iters,
hzfan commented on a change in pull request #5235:
URL: https://github.com/apache/incubator-tvm/pull/5235#discussion_r456252011
##
File path: src/te/schedule/operation_inline.cc
##
@@ -63,7 +63,9 @@ class OperationInliner final : public StmtExprMutator {
} else {
jcf94 commented on a change in pull request #6073:
URL: https://github.com/apache/incubator-tvm/pull/6073#discussion_r456249617
##
File path: src/auto_scheduler/loop_state.cc
##
@@ -90,12 +90,69 @@ Stage::Stage(te::Operation op, StageKind op_type, const
Array& iters,
data_
lixiaoquan opened a new pull request #6086:
URL: https://github.com/apache/incubator-tvm/pull/6086
@icemelon9 Could you please take a look?
This is an automated message from the Apache Git Service.
To respond to the
88 matches
Mail list logo