masahi commented on issue #4799: [QNN] Doc fix on convolution and dequantize
URL: https://github.com/apache/incubator-tvm/pull/4799#issuecomment-580980672
this is the error I get if I leave out kernel_size:
```
File "/home/masa/projects/dev/tvm/python/tvm/_ffi/_ctypes/function.py",
masahi opened a new pull request #4799: [QNN] Doc fix on convolution and
dequantize
URL: https://github.com/apache/incubator-tvm/pull/4799
cc @anijain2305 @vinx13 can you check if this change is correct? I'm not
completely sure if kernel_size is required, but I hit error if I leave it
icemelon9 merged pull request #4775: conv3d_ndhwc schedule
URL: https://github.com/apache/incubator-tvm/pull/4775
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
icemelon9 commented on issue #4775: conv3d_ndhwc schedule
URL: https://github.com/apache/incubator-tvm/pull/4775#issuecomment-580976944
Thanks @alexgl-github . This is now merged.
This is an automated message from the Apache
anijain2305 opened a new pull request #4798: [QNN] Optimize lowering for
requantize and FixedPointMultiply.
URL: https://github.com/apache/incubator-tvm/pull/4798
As Title.
Changes are verified through existing tests.
@jackwish @FrozenGene @yzhliu @vinx13
anijain2305 opened a new pull request #4797: [AutoTVM] Minor bug fixes in
AutoTVM for QNN graphs
URL: https://github.com/apache/incubator-tvm/pull/4797
I encountered many bugs during autotuning, both kernel and graph, for a QNN
graph. This PR adds fixes all the minor bugs.
There is
wweic commented on issue #4628: [Object] Add String container
URL: https://github.com/apache/incubator-tvm/pull/4628#issuecomment-580966470
@tqchen I'll send new revision soon.
This is an automated message from the Apache Git
tqchen commented on issue #4644: [WIP] Relay op strategy
URL: https://github.com/apache/incubator-tvm/pull/4644#issuecomment-580955527
cc @merrymercy @vinx13 @ZihengJiang @jwfromm please help review if you have
time
This is
icemelon9 commented on a change in pull request #4775: conv3d_ndhwc schedule
URL: https://github.com/apache/incubator-tvm/pull/4775#discussion_r373715263
##
File path: topi/python/topi/x86/conv3d.py
##
@@ -0,0 +1,85 @@
+# Licensed to the Apache Software Foundation (ASF)
This is an automated email from the ASF dual-hosted git repository.
liuyizhi pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git.
from 10f85d0 Dedup BindParamByName function in VM compiler (#4793)
add 90b2a1e [Relay][Topi] Use
yzhliu merged pull request #4795: [Relay][Topi] Use SimplifyInference for L2
Normalization.
URL: https://github.com/apache/incubator-tvm/pull/4795
This is an automated message from the Apache Git Service.
To respond to the
yzhliu commented on issue #4795: [Relay][Topi] Use SimplifyInference for L2
Normalization.
URL: https://github.com/apache/incubator-tvm/pull/4795#issuecomment-580939287
Thanks @anijain2305
This is an automated message from
tqchen commented on a change in pull request #4790: Fast exponent
URL: https://github.com/apache/incubator-tvm/pull/4790#discussion_r373702986
##
File path: topi/include/topi/elemwise.h
##
@@ -360,5 +359,66 @@ inline Tensor full_like(const Tensor& x,
}, name, tag);
}
tqchen commented on issue #4628: [Object] Add String container
URL: https://github.com/apache/incubator-tvm/pull/4628#issuecomment-580904651
gentle ping
This is an automated message from the Apache Git Service.
To respond to
anijain2305 opened a new pull request #4796: [QNN] Conv2D with dilation support.
URL: https://github.com/apache/incubator-tvm/pull/4796
Quantized SSD_VGG has a dilated conv. This PR supports better QNN lowering
for symmetric dilated conv.
Asymmetric dilated conv requires dilated
huajsj commented on a change in pull request #4791: [TOPI] upsample operator
'NCHWinic' format support.
URL: https://github.com/apache/incubator-tvm/pull/4791#discussion_r373667310
##
File path: topi/python/topi/image/resize.py
##
@@ -18,8 +18,37 @@
"""TVM operator input
tmoreau89 commented on a change in pull request #4791: [TOPI] upsample operator
'NCHWinic' format support.
URL: https://github.com/apache/incubator-tvm/pull/4791#discussion_r373658550
##
File path: topi/python/topi/image/resize.py
##
@@ -18,8 +18,37 @@
"""TVM operator
tmoreau89 commented on issue #4791: [TOPI] upsample operator 'NCHWinic' format
support.
URL: https://github.com/apache/incubator-tvm/pull/4791#issuecomment-580886430
@srkreddy1238 could you help review?
This is an automated
huajsj commented on issue #4791: [TOPI] upsample operator 'NCHWinic' format
support.
URL: https://github.com/apache/incubator-tvm/pull/4791#issuecomment-580885374
Hi @jwfromm, if you have time could you help for a review? thanks.
anijain2305 opened a new pull request #4795: [Relay][Topi] Use
SimplifyInference for L2 Normazlization.
URL: https://github.com/apache/incubator-tvm/pull/4795
Reason - Observed that l2_normalize (which is seemingly a sequence of
element-wise op with one reduce op) was taking close to 15%
huajsj commented on issue #4791: [TOPI] upsample operator 'NCHWinic' format
support.
URL: https://github.com/apache/incubator-tvm/pull/4791#issuecomment-580881788
Hi @tmoreau89 , this patch related to let VTA to support Yolov3, if you have
time could you help for a review too?
icemelon9 commented on issue #4787: [Relay] Conv2D padding representation
URL: https://github.com/apache/incubator-tvm/pull/4787#issuecomment-580874675
You should also add `get_pad_tuple2d` to these contrib conv2d ops, e.g.,
contrib_conv2d_nchwc
icemelon9 commented on a change in pull request #4787: [Relay] Conv2D padding
representation
URL: https://github.com/apache/incubator-tvm/pull/4787#discussion_r373643774
##
File path: python/tvm/relay/op/nn/util.py
##
@@ -0,0 +1,56 @@
+# Licensed to the Apache Software
alexwong commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-580873337
> Reducing all the input sizes might help with memory issues, I don't think
theres any need to use big 224x224 test data.
I
icemelon9 commented on a change in pull request #4775: conv3d_ndhwc schedule
URL: https://github.com/apache/incubator-tvm/pull/4775#discussion_r373643228
##
File path: topi/python/topi/x86/conv3d.py
##
@@ -0,0 +1,92 @@
+# Licensed to the Apache Software Foundation (ASF)
jwfromm commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-580865226
Reducing all the input sizes might help with memory issues, I don't think
theres any need to use big 224x224 test data.
alexgl-github commented on a change in pull request #4790: Fast exponent
URL: https://github.com/apache/incubator-tvm/pull/4790#discussion_r373624561
##
File path: topi/include/topi/elemwise.h
##
@@ -360,5 +359,71 @@ inline Tensor full_like(const Tensor& x,
}, name,
comaniac commented on issue #4787: [Relay] Conv2D padding representation
URL: https://github.com/apache/incubator-tvm/pull/4787#issuecomment-580836237
LGTM.
@icemelon9 @yzhliu could you help review and merge? Thanks.
This
inadob closed pull request #4696: [Relay][Frontend][TFlite] Add support for
quantized LOGISTIC
URL: https://github.com/apache/incubator-tvm/pull/4696
This is an automated message from the Apache Git Service.
To respond to
tqchen commented on a change in pull request #4790: Fast exponent
URL: https://github.com/apache/incubator-tvm/pull/4790#discussion_r373583565
##
File path: topi/include/topi/elemwise.h
##
@@ -360,5 +359,71 @@ inline Tensor full_like(const Tensor& x,
}, name, tag);
}
vizero1 opened a new pull request #4794: Change color channel from BGR to RGB
for darknet preprocessing
URL: https://github.com/apache/incubator-tvm/pull/4794
Thanks for contributing to TVM! Please refer to guideline
https://docs.tvm.ai/contribute/ for useful information and tips. After
31 matches
Mail list logo
