[tvm] branch main updated: [BUILD] Removed the duplicated MACROs in config.cmake (#14813)

This is an automated email from the ASF dual-hosted git repository. wuwei 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 42aaae143c [BUILD] Removed the duplicated MACROs in

[GitHub] [tvm] vinx13 merged pull request #14813: [BUILD] Removed the duplicated MACROs in config.cmake

vinx13 merged PR #14813: URL: https://github.com/apache/tvm/pull/14813 -- 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:

[GitHub] [tvm] jiangjiajun commented on a diff in pull request #14826: [Frontend][Paddle] [PaddlePaddle Hackathon 4] add convert support for affine_channel/p_norm/roi_align/softmax_with_cross_entropy

jiangjiajun commented on code in PR #14826: URL: https://github.com/apache/tvm/pull/14826#discussion_r1191923136 ## tests/python/frontend/paddlepaddle/test_forward.py: ## @@ -2302,5 +2302,112 @@ def forward(self, x, y): verify_model(Dist(), input_data=[y, v])

[GitHub] [tvm] f2013519 commented on pull request #14832: [relay][convert layout]: Infer correct layout for nn.batch_flatten

f2013519 commented on PR #14832: URL: https://github.com/apache/tvm/pull/14832#issuecomment-1545167028 Dropping as this can lead to issues with downstream ops such as dense layer whose weights assume original layout from the flatten layer. -- This is an automated message from the Apache

[GitHub] [tvm] f2013519 closed pull request #14832: [relay][convert layout]: Infer correct layout for nn.batch_flatten

f2013519 closed pull request #14832: [relay][convert layout]: Infer correct layout for nn.batch_flatten URL: https://github.com/apache/tvm/pull/14832 -- 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

[tvm] branch nightly updated (a1c1ccafa1 -> 2cafa87b10)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch nightly in repository https://gitbox.apache.org/repos/asf/tvm.git from a1c1ccafa1 [SUPPORT] Fix RingBuffer ReadWithCallback (#14743) add 48200fc3d7 [TOPI] Use f-strings for string

[GitHub] [tvm] MasterJH5574 opened a new issue, #14834: [Bug] Wrong lowered TIR with symbolic input buffer

MasterJH5574 opened a new issue, #14834: URL: https://github.com/apache/tvm/issues/14834 @wrongtest-intellif Hello, I noticed an abnormal issue after #14021 being merged. https://gist.github.com/MasterJH5574/64e9dca54cc08cd0131126efd55cfac7 In the gist above, the original TIR

[GitHub] [tvm] spectrometerHBH opened a new pull request, #14833: [Codegen][OpenCL] fix amibiguous selection operator call

spectrometerHBH opened a new pull request, #14833: URL: https://github.com/apache/tvm/pull/14833 Per OpenCL specification https://man.opencl.org/select.html -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the

[GitHub] [tvm] tvm-bot commented on pull request #14833: [Codegen][OpenCL] fix amibiguous selection operator call

tvm-bot commented on PR #14833: URL: https://github.com/apache/tvm/pull/14833#issuecomment-1544970331 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] kparzysz-quic commented on pull request #14830: [Frontend] Use f-strings for string formatting, NFC

kparzysz-quic commented on PR #14830: URL: https://github.com/apache/tvm/pull/14830#issuecomment-1544925398 > cutlass commit might be unintentional, other parts lgtm 臘 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub

[GitHub] [tvm] tqchen commented on pull request #14830: [Frontend] Use f-strings for string formatting, NFC

tqchen commented on PR #14830: URL: https://github.com/apache/tvm/pull/14830#issuecomment-1544828487 cutlass commit might be unintentional, other parts lgtm -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL

[GitHub] [tvm] tvm-bot commented on pull request #14832: [relay][convert layout]: Infer correct layout for nn.batch_flatten

tvm-bot commented on PR #14832: URL: https://github.com/apache/tvm/pull/14832#issuecomment-1544614779 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] f2013519 opened a new pull request, #14832: [relay][convert layout]: Infer correct layout for nn.batch_flatten

f2013519 opened a new pull request, #14832: URL: https://github.com/apache/tvm/pull/14832 Currently during the ConvertLayout pass, whenever there is a nn.batch_flatten op, additional layout transforms are inserted before the nn.batch_flatten op. This may not be efficient. In

[GitHub] [tvm] tvm-bot commented on pull request #14831: [Relay/Op] Use f-strings for string formatting, NFC

tvm-bot commented on PR #14831: URL: https://github.com/apache/tvm/pull/14831#issuecomment-1544610952 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] kparzysz-quic opened a new pull request, #14831: [Relay/Op] Use f-strings for string formatting, NFC

kparzysz-quic opened a new pull request, #14831: URL: https://github.com/apache/tvm/pull/14831 Replace uses of % and .format() with f-strings. -- 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

[GitHub] [tvm] tvm-bot commented on pull request #14830: [Frontend] Use f-strings for string formatting, NFC

tvm-bot commented on PR #14830: URL: https://github.com/apache/tvm/pull/14830#issuecomment-1544606170 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] kparzysz-quic opened a new pull request, #14830: [Frontend] Use f-strings for string formatting, NFC

kparzysz-quic opened a new pull request, #14830: URL: https://github.com/apache/tvm/pull/14830 Replace uses of % and .format() with f-strings. -- 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

[GitHub] [tvm] tqchen commented on issue #14829: [Bug][Unity] `LazyTransformParams` introduces impure actions inside `DataflowBlock`s

tqchen commented on issue #14829: URL: https://github.com/apache/tvm/issues/14829#issuecomment-1544576007 cc @jinhongyii I think ideally it should be done after ToNonDataflow -- This is an automated message from the Apache Git Service. To respond to the message, please log on to

[GitHub] [tvm] slyubomirsky commented on a diff in pull request #14394: [Unity][IR] Purity Tracking

slyubomirsky commented on code in PR #14394: URL: https://github.com/apache/tvm/pull/14394#discussion_r1191580313 ## python/tvm/relax/transform/lazy_transform_params.py: ## @@ -166,11 +165,15 @@ def visit_var_binding_(self, binding: relax.VarBinding) -> None:

[GitHub] [tvm] slyubomirsky opened a new issue, #14829: [Bug][Unity] `LazyTransformParams` introduces impure actions inside `DataflowBlock`s

slyubomirsky opened a new issue, #14829: URL: https://github.com/apache/tvm/issues/14829 The test case for the pass `LazyTransformParams` reveals a significant bug that should require rethinking how the pass works:

[GitHub] [tvm] jikechao commented on pull request #14821: [Bugfix][Relay] Fix softplus about the wrong calculation formula in Relay PyTorch frontend

jikechao commented on PR #14821: URL: https://github.com/apache/tvm/pull/14821#issuecomment-1544337258 The added test cases produce a wired flaky problem, I'll try to figure out it later. -- This is an automated message from the Apache Git Service. To respond to the message, please log

[GitHub] [tvm] tvm-bot commented on pull request #14828: [BugFix] Fix memory leak in Registry::Remove

tvm-bot commented on PR #14828: URL: https://github.com/apache/tvm/pull/14828#issuecomment-1544183784 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] lee-bin opened a new pull request, #14828: [BugFix] Fix memory leak in Registry::Remove

lee-bin opened a new pull request, #14828: URL: https://github.com/apache/tvm/pull/14828 The Registry should be deleted before being erased from the map. Otherwise, there will be a memory leak. -- This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] [tvm] aryamazaheri opened a new issue, #14827: [Bug] undeclared identifier 'RTLD_DI_LOAD_ADDR' while compiling for Hexagon

aryamazaheri opened a new issue, #14827: URL: https://github.com/apache/tvm/issues/14827 While compiling TVM for Hexagon backend, I get the following error: ``` /home/osboxes/tvm/src/runtime/dso_library.cc:129:32: error: use of undeclared identifier 'RTLD_DI_LOAD_ADDR' int rc =

[GitHub] [tvm] echuraev commented on pull request #14751: [CUDA] Fixed the call of the min function in the schedule for cuda

echuraev commented on PR #14751: URL: https://github.com/apache/tvm/pull/14751#issuecomment-1544120446 @tvm-bot rerun -- 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

[GitHub] [tvm] tvm-bot commented on pull request #14826: [Frontend][Paddle] [PaddlePaddle Hackathon 4] add convert support for affine_channel/p_norm/roi_align/softmax_with_cross_entropy

tvm-bot commented on PR #14826: URL: https://github.com/apache/tvm/pull/14826#issuecomment-1544119639 Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews

[GitHub] [tvm] AndPuQing opened a new pull request, #14826: [Frontend][Paddle] [PaddlePaddle Hackathon 4] add convert support for affine_channel/p_norm/roi_align/softmax_with_cross_entropy

AndPuQing opened a new pull request, #14826: URL: https://github.com/apache/tvm/pull/14826 Add support for converting the following Paddle OPs. - [affine_channel](https://www.paddlepaddle.org.cn/documentation/docs/en/1.8/api/layers/affine_channel.html) -

[GitHub] [tvm] jikechao closed issue #14805: [Bug][relay][pytorch] Wrong implementation logic in Threshold operator

jikechao closed issue #14805: [Bug][relay][pytorch] Wrong implementation logic in Threshold operator URL: https://github.com/apache/tvm/issues/14805 -- 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

[GitHub] [tvm] jikechao commented on issue #14805: [Bug][relay][pytorch] Wrong implementation logic in Threshold operator

jikechao commented on issue #14805: URL: https://github.com/apache/tvm/issues/14805#issuecomment-1543859346 This issue has been fixed by the [PR-14820](https://github.com/apache/tvm/pull/14820) -- This is an automated message from the Apache Git Service. To respond to the message, please

[tvm] branch main updated: [Bugfix][Relay] Fix threshold calculation logic in PyTorch frontend (#14820)

This is an automated email from the ASF dual-hosted git repository. masahi 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 2cafa87b10 [Bugfix][Relay] Fix threshold calculation

[GitHub] [tvm] masahi merged pull request #14820: [Bugfix][Relay] Fix threshold calculation logic in PyTorch frontend

masahi merged PR #14820: URL: https://github.com/apache/tvm/pull/14820 -- 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:

[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #14819: [TFLite][Frontend] Generate name when tensor name is missing

ashutosh-arm commented on code in PR #14819: URL: https://github.com/apache/tvm/pull/14819#discussion_r1190947596 ## python/tvm/relay/frontend/tflite.py: ## @@ -4116,7 +4116,12 @@ def get_tensor_name(subgraph, tensor_idx): --- tensor name in UTF-8 encoding

[GitHub] [tvm] ashutosh-arm commented on a diff in pull request #14819: [TFLite][Frontend] Generate name when tensor name is missing

ashutosh-arm commented on code in PR #14819: URL: https://github.com/apache/tvm/pull/14819#discussion_r1190920015 ## python/tvm/relay/frontend/tflite.py: ## @@ -4150,7 +4155,8 @@ def _input_type(model): tensor = subgraph.Tensors(input_) input_shape =

[GitHub] [tvm] lhutton1 commented on a diff in pull request #14819: [TFLite][Frontend] Generate name when tensor name is missing

lhutton1 commented on code in PR #14819: URL: https://github.com/apache/tvm/pull/14819#discussion_r1190843066 ## python/tvm/relay/frontend/tflite.py: ## @@ -4116,7 +4116,12 @@ def get_tensor_name(subgraph, tensor_idx): --- tensor name in UTF-8 encoding

[tvm] branch main updated: [Relay][BugFix] fix a bug about ReLu in the threshold attribute which causes a different results with keras (#14824)

This is an automated email from the ASF dual-hosted git repository. echuraev 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 fd2a510ccf [Relay][BugFix] fix a bug about ReLu in

[GitHub] [tvm] echuraev merged pull request #14824: [Relay][BugFix] fix a bug about ReLu in the threshold attribute which causes a different results with keras

echuraev merged PR #14824: URL: https://github.com/apache/tvm/pull/14824 -- 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:

[GitHub] [tvm] jikechao commented on a diff in pull request #14824: [Relay][BugFix] fix a bug about ReLu in the threshold attribute which causes a different results with keras

jikechao commented on code in PR #14824: URL: https://github.com/apache/tvm/pull/14824#discussion_r1190833344 ## tests/python/frontend/keras/test_forward.py: ## @@ -227,6 +227,7 @@ def test_forward_activations_except(self, keras_mod): act_funcs = [

[GitHub] [tvm] vinx13 commented on a diff in pull request #14766: [TIR][Schedule] Improve blockize to support blockizing multiple blocks

vinx13 commented on code in PR #14766: URL: https://github.com/apache/tvm/pull/14766#discussion_r1190791916 ## python/tvm/tir/schedule/schedule.py: ## @@ -2691,13 +2691,15 @@ def after_set_dtype( ## Schedule: Blockize & Tensorize ## @type_checked -

[GitHub] [tvm] echuraev commented on pull request #14751: [CUDA] Fixed the call of the min function in the schedule for cuda

echuraev commented on PR #14751: URL: https://github.com/apache/tvm/pull/14751#issuecomment-1543545223 cc: @junrushao, @masahi -- 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

[GitHub] [tvm] echuraev commented on a diff in pull request #14751: [CUDA] Fixed the call of the min function in the schedule for cuda

echuraev commented on code in PR #14751: URL: https://github.com/apache/tvm/pull/14751#discussion_r1190795868 ## tests/python/relay/test_any.py: ## @@ -59,6 +59,7 @@ def check_result( continue if kind == "debug" and (only_vm or dev.device_type !=

[tvm] branch unity updated (3b33caf757 -> 28567bdc24)

This is an automated email from the ASF dual-hosted git repository. wuwei pushed a change to branch unity in repository https://gitbox.apache.org/repos/asf/tvm.git from 3b33caf757 [Unity] Cover all Relax functions in implicit attention rewrite (#14818) add 28567bdc24 [Unity] Improve

[GitHub] [tvm] vinx13 merged pull request #14823: [Unity] Improve Error Message loading NDArray Shards

vinx13 merged PR #14823: URL: https://github.com/apache/tvm/pull/14823 -- 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:

[GitHub] [tvm] echuraev commented on pull request #14757: [Unity][Training] Categorical cross entropy loss

echuraev commented on PR #14757: URL: https://github.com/apache/tvm/pull/14757#issuecomment-1543500765 @tvm-bot rerun -- 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

[GitHub] [tvm] echuraev commented on a diff in pull request #14824: [Relay][BugFix] fix a bug about ReLu in the threshold attribute which causes a different results with keras

echuraev commented on code in PR #14824: URL: https://github.com/apache/tvm/pull/14824#discussion_r1190755313 ## tests/python/frontend/keras/test_forward.py: ## @@ -227,6 +227,7 @@ def test_forward_activations_except(self, keras_mod): act_funcs = [

[GitHub] [tvm] echuraev commented on pull request #14821: [Bugfix][Relay] Fix softplus about the wrong calculation formula in Relay PyTorch frontend

echuraev commented on PR #14821: URL: https://github.com/apache/tvm/pull/14821#issuecomment-1543485078 @tvm-bot rerun -- 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

[GitHub] [tvm] echuraev commented on pull request #14757: [Unity][Training] Categorical cross entropy loss

echuraev commented on PR #14757: URL: https://github.com/apache/tvm/pull/14757#issuecomment-1543458535 @tvm-bot rerun -- 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

[tvm] branch main updated: [TOPI] Use f-strings for string formatting, NFC (#14822)

This is an automated email from the ASF dual-hosted git repository. masahi 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 48200fc3d7 [TOPI] Use f-strings for string formatting,

[GitHub] [tvm] masahi merged pull request #14822: [TOPI] Use f-strings for string formatting, NFC

masahi merged PR #14822: URL: https://github.com/apache/tvm/pull/14822 -- 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: