[tvm] branch last-successful updated (c515241026 -> 3f304c8bc5)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from c515241026 fix type bug about topi test unitest (#12285) add 3f304c8bc5 [microTVM] Arduino: Fix

[GitHub] [tvm] cdljn2011 opened a new issue, #12302: [Bug] Can the virtual machine executor support inference on dynamic shape of OCR model?

cdljn2011 opened a new issue, #12302: URL: https://github.com/apache/tvm/issues/12302 Hi, I observe that tvm only support transform on Oneflow or Onnx dynamic shape model. I have tried to tranform so many OCR Onnx models to relay but all failed because the presence of this operator I

[GitHub] [tvm] mbrookhart commented on a diff in pull request #12287: [QNN] Add qnn op for abs to fix wrong scale on quantize

mbrookhart commented on code in PR #12287: URL: https://github.com/apache/tvm/pull/12287#discussion_r937315065 ## tests/python/relay/test_pass_fake_quantization_to_integer.py: ## @@ -27,6 +27,9 @@ def compare_fq_to_int(expr, args, allow_rounding_error=False): mod =

[tvm] branch main updated: [microTVM] Arduino: Fix f-strings on flash warning/error messages (#12175)

This is an automated email from the ASF dual-hosted git repository. junrushao 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 3f304c8bc5 [microTVM] Arduino: Fix f-strings on

[GitHub] [tvm] junrushao1994 merged pull request #12175: [microTVM] Arduino: Fix f-strings on flash warning/error messages

junrushao1994 merged PR #12175: URL: https://github.com/apache/tvm/pull/12175 -- 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:

[tvm] branch last-successful updated (46a8498ba9 -> c515241026)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from 46a8498ba9 [MetaSchedule] Enhance Conv2d NCHW Winograd Schedule Rules (#12127) add 2e02cf7cbe [LLVM]

[GitHub] [tvm] masahi commented on issue #12292: [Bug] Tensor advanced indexing with boolean mask not working

masahi commented on issue #12292: URL: https://github.com/apache/tvm/issues/12292#issuecomment-1204662978 There were related discussion on this topic. https://github.com/apache/tvm/issues/9237#issuecomment-992428813 https://github.com/apache/tvm/pull/7373 -- This is an automated

[GitHub] [tvm] wrongtest-intellif commented on issue #12298: [Bug][TIR] Split Error in Schedule Verification

wrongtest-intellif commented on issue #12298: URL: https://github.com/apache/tvm/issues/12298#issuecomment-1204648969 > use `canonical_simplify` here instead of `Simplify` Hi, the last modification to `canonical_simplify` is in https://github.com/apache/tvm/pull/9941. Where use

[GitHub] [tvm-rfcs] junrushao1994 commented on pull request #79: [RFC] TVMScript Metaprogramming

junrushao1994 commented on PR #79: URL: https://github.com/apache/tvm-rfcs/pull/79#issuecomment-1204646078 Thanks for following up! > the parser and IRBuilder are only called for statements and for function calls. Is this correct? In fact we allow registering visitors for any

[tvm] branch last-successful updated (df29e82629 -> 46a8498ba9)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from df29e82629 [TIR][CUDA] Fix sub-warp reduction using "max" (#12275) add 46a8498ba9 [MetaSchedule]

[GitHub] [tvm] AndrewZhaoLuo commented on pull request #12301: [QNN] Fix multi batch batch matmuls

AndrewZhaoLuo commented on PR #12301: URL: https://github.com/apache/tvm/pull/12301#issuecomment-1204614242 cc @mbrookhart -- 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] github-actions[bot] commented on pull request #12295: [ci][tvmbot] Enable re-run for GitHub Actions

github-actions[bot] commented on PR #12295: URL: https://github.com/apache/tvm/pull/12295#issuecomment-1204616079 Built docs for commit 0d35fbd227a14b2e9cbe7da58814e2a4857d310c can be found [here](https://pr-docs.tlcpack.ai/PR-12295/5/docs/index.html). -- This is an automated

[GitHub] [tvm] AndrewZhaoLuo opened a new pull request, #12301: [QNN] Fix multi batch batch matmuls

AndrewZhaoLuo opened a new pull request, #12301: URL: https://github.com/apache/tvm/pull/12301 The previous version incorrectly broadcasted across the batch dimensions which led to type errors. I have updated tests to handle multi batch examples and updated the documentation to note

[tvm] 01/01: initial commit

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a commit to branch aluo/fix-qnn-batch-matmul in repository https://gitbox.apache.org/repos/asf/tvm.git commit 24baa3e8e2d4fc89422866e3a5d9ebfcac777b47 Author: Andrew Zhao Luo AuthorDate: Wed Aug 3 17:12:12

[tvm] branch aluo/fix-qnn-batch-matmul created (now 24baa3e8e2)

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a change to branch aluo/fix-qnn-batch-matmul in repository https://gitbox.apache.org/repos/asf/tvm.git at 24baa3e8e2 initial commit This branch includes the following new commits: new

[tvm] branch nightly-docker-update updated (74f2736cd8 -> 3adfbf522c)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch nightly-docker-update in repository https://gitbox.apache.org/repos/asf/tvm.git discard 74f2736cd8 [ci][docker] Nightly Docker image update add 85624ff667 [ci][docker] Fix deploy to

[tvm] branch main updated (2e02cf7cbe -> c515241026)

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 2e02cf7cbe [LLVM] Create LLVM scope object for use with LLVM libraries (#12140) add c515241026 fix type bug about

[GitHub] [tvm] vinx13 merged pull request #12285: [Test]fix type bug about topi unitest

vinx13 merged PR #12285: URL: https://github.com/apache/tvm/pull/12285 -- 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-rfcs] tkonolige commented on pull request #79: [RFC] TVMScript Metaprogramming

tkonolige commented on PR #79: URL: https://github.com/apache/tvm-rfcs/pull/79#issuecomment-1204583687 Ok, I think I understand things a little bit better now. Thanks for the explanation! I can see how if the IRBuilder handles all state then there is not really much of a point to having

[GitHub] [tvm] kparzysz-quic commented on pull request #12140: [LLVM] Create LLVM scope object for use with LLVM libraries

kparzysz-quic commented on PR #12140: URL: https://github.com/apache/tvm/pull/12140#issuecomment-1204574709 I edited the commit message to use `LLVMInstance` instead of `LLVMScope` in the summary. -- This is an automated message from the Apache Git Service. To respond to the message,

[tvm] branch main updated (46a8498ba9 -> 2e02cf7cbe)

This is an automated email from the ASF dual-hosted git repository. kparzysz pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 46a8498ba9 [MetaSchedule] Enhance Conv2d NCHW Winograd Schedule Rules (#12127) add 2e02cf7cbe [LLVM] Create LLVM

[GitHub] [tvm] kparzysz-quic merged pull request #12140: [LLVM] Create LLVM scope object for use with LLVM libraries

kparzysz-quic merged PR #12140: URL: https://github.com/apache/tvm/pull/12140 -- 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] jwfromm commented on a diff in pull request #12284: [Relay][Op] Multinomial

jwfromm commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r937205819 ## python/tvm/topi/random/kernel.py: ## @@ -602,3 +602,57 @@ def normal(gen, mean, scale, out_shape, out_dtype): ) return new_gen, random_values + + +def

[GitHub] [tvm] junrushao1994 commented on a diff in pull request #12299: [TVMScript] TracedObject class that simplifies tracing ObjectPaths

junrushao1994 commented on code in PR #12299: URL: https://github.com/apache/tvm/pull/12299#discussion_r937164030 ## src/script/printer/traced_object.h: ## @@ -0,0 +1,410 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[tvm] branch main updated (df29e82629 -> 46a8498ba9)

This is an automated email from the ASF dual-hosted git repository. xiyou pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from df29e82629 [TIR][CUDA] Fix sub-warp reduction using "max" (#12275) add 46a8498ba9 [MetaSchedule] Enhance Conv2d NCHW

[GitHub] [tvm] zxybazh merged pull request #12127: [MetaSchedule] Enhance Conv2d NCHW Winograd Schedule Rules

zxybazh merged PR #12127: URL: https://github.com/apache/tvm/pull/12127 -- 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] yelite commented on a diff in pull request #12299: [TVMScript] TracedObject class that simplifies tracing ObjectPaths

yelite commented on code in PR #12299: URL: https://github.com/apache/tvm/pull/12299#discussion_r937148401 ## src/script/printer/traced_object.h: ## @@ -0,0 +1,410 @@ +/* Review Comment: Can you move this file to the public headers (include/...)? Because it's used by the

[GitHub] [tvm] yelite commented on a diff in pull request #12299: [TVMScript] TracedObject class that simplifies tracing ObjectPaths

yelite commented on code in PR #12299: URL: https://github.com/apache/tvm/pull/12299#discussion_r937148401 ## src/script/printer/traced_object.h: ## @@ -0,0 +1,410 @@ +/* Review Comment: Can you move this to the public headers (include/...)? Because it's used by the

[GitHub] [tvm] yelite commented on a diff in pull request #12299: [TVMScript] TracedObject class that simplifies tracing ObjectPaths

yelite commented on code in PR #12299: URL: https://github.com/apache/tvm/pull/12299#discussion_r937148401 ## src/script/printer/traced_object.h: ## @@ -0,0 +1,410 @@ +/* Review Comment: Can you move this to the public headers (include/...)? Because it's used by the actual

[GitHub] [tvm] vinx13 opened a new pull request, #12300: [TIR] Add DeclBuffer IR node and functors

vinx13 opened a new pull request, #12300: URL: https://github.com/apache/tvm/pull/12300 This PR is part of https://github.com/apache/tvm/issues/11627. It added the IR node `DeclBufferNode`, IR functors and TVMScript printer / parser. cc @tqchen @Lunderberg @Hzfengsy @junrushao1994

[GitHub] [tvm] gbonik opened a new pull request, #12299: [TVMScript] TracedObject class that simplifies tracing ObjectPaths

gbonik opened a new pull request, #12299: URL: https://github.com/apache/tvm/pull/12299 Motivation: when printing a piece of TIR, we need to track an ObjectPath from the root TIR object to the currently printed object. This means that we need a convenient way to maintain an ObjectPath

[GitHub] [tvm] zxybazh opened a new issue, #12298: [Bug][TIR] TIR Split Error in Schedule Verification

zxybazh opened a new issue, #12298: URL: https://github.com/apache/tvm/issues/12298 Hi, I noticed a split error when schedule verification is enabled with `debug_mask="all"`, where I think the split is actually valid. Script to reproduce the issue is here:

[tvm] branch last-successful updated (6f831137c9 -> df29e82629)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from 6f831137c9 [microTVM] Refactor pytest fixtures (#12207) add df29e82629 [TIR][CUDA] Fix sub-warp

[tvm] branch last-successful updated (49587cfcea -> 6f831137c9)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from 49587cfcea [ci] Add retries to S3 uploads/downloads (#12221) add 6f831137c9 [microTVM] Refactor pytest

[GitHub] [tvm] joshherr-quic opened a new pull request, #12297: Quantized dense operator.

joshherr-quic opened a new pull request, #12297: URL: https://github.com/apache/tvm/pull/12297 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

[tvm] branch micro_rename_qemu updated (548f51ff97 -> 92430d708f)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch micro_rename_qemu in repository https://gitbox.apache.org/repos/asf/tvm.git from 548f51ff97 rebased with main add 92430d708f fix image name No new revisions were added by this

[GitHub] [tvm] tqchen commented on pull request #12140: [LLVM] Create LLVM scope object for use with LLVM libraries

tqchen commented on PR #12140: URL: https://github.com/apache/tvm/pull/12140#issuecomment-1204419044 LLVMInstance sounds good -- 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] lhutton1 commented on pull request #12296: [ETHOSN] Only use mock inference when whole graph is offloaded

lhutton1 commented on PR #12296: URL: https://github.com/apache/tvm/pull/12296#issuecomment-1204407562 This is expected to fix the docker image validation: https://ci.tlcpack.ai/job/docker-images-ci/job/daily-docker-image-validate/180/ -- This is an automated message from the Apache Git

[GitHub] [tvm] lhutton1 opened a new pull request, #12296: [ETHOSN] Only use mock inference when whole graph is offloaded

lhutton1 opened a new pull request, #12296: URL: https://github.com/apache/tvm/pull/12296 The mock inference functionality is only supported when the whole graph is offloaded to the NPU, otherwise it can result in undefined behavior. This patch makes sure the mock inference functionality

[GitHub] [tvm] tqchen closed pull request #11320: [DRAFT][LLVM] Introduce RAII config option to config LLVM global config

tqchen closed pull request #11320: [DRAFT][LLVM] Introduce RAII config option to config LLVM global config URL: https://github.com/apache/tvm/pull/11320 -- 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

[GitHub] [tvm] zxybazh commented on a diff in pull request #12127: [MetaSchedule] Enhance Conv2d NCHW Winograd Schedule Rules

zxybazh commented on code in PR #12127: URL: https://github.com/apache/tvm/pull/12127#discussion_r937024824 ## tests/python/unittest/test_meta_schedule_space_cuda.py: ## @@ -1239,6 +1239,90 @@ def cbr_0(data: T.Buffer[(1, 224, 224, 3), "float32"], kernel: T.Buffer[(7, 7, 3

[GitHub] [tvm] vinx13 commented on a diff in pull request #12288: [TIR][Schedule] Support annotate dict typed value

vinx13 commented on code in PR #12288: URL: https://github.com/apache/tvm/pull/12288#discussion_r936993409 ## python/tvm/tir/schedule/_type_checker.py: ## @@ -177,6 +197,19 @@ def _type_check_list(v: List[Any], name: str, type_: Any) -> Optional[str]: return

[tvm] branch main updated: [TIR][CUDA] Fix sub-warp reduction using "max" (#12275)

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 df29e82629 [TIR][CUDA] Fix sub-warp reduction using

[GitHub] [tvm] junrushao1994 commented on pull request #12288: [TIR][Schedule] Support annotate dict typed value

junrushao1994 commented on PR #12288: URL: https://github.com/apache/tvm/pull/12288#issuecomment-1204306600 CC: @vinx13 -- 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] vinx13 closed issue #12274: [Bug] Incorrect result using sub-warp reduction

vinx13 closed issue #12274: [Bug] Incorrect result using sub-warp reduction URL: https://github.com/apache/tvm/issues/12274 -- 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] vinx13 merged pull request #12275: [TIR][CUDA] Fix sub-warp reduction using "max"

vinx13 merged PR #12275: URL: https://github.com/apache/tvm/pull/12275 -- 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:

[tvm] branch main updated: [microTVM] Refactor pytest fixtures (#12207)

This is an automated email from the ASF dual-hosted git repository. areusch 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 6f831137c9 [microTVM] Refactor pytest fixtures

[GitHub] [tvm] areusch merged pull request #12207: [microTVM] Refactor pytest fixtures

areusch merged PR #12207: URL: https://github.com/apache/tvm/pull/12207 -- 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] areusch commented on a diff in pull request #12207: [microTVM] Refactor pytest fixtures

areusch commented on code in PR #12207: URL: https://github.com/apache/tvm/pull/12207#discussion_r936982142 ## python/tvm/micro/testing/pytest_plugin.py: ## @@ -0,0 +1,111 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license

[GitHub] [tvm] elvin-n commented on a diff in pull request #12286: [Adreno][OpenCL] Get rid of extra memory copy

elvin-n commented on code in PR #12286: URL: https://github.com/apache/tvm/pull/12286#discussion_r936974898 ## src/relay/backend/build_module.cc: ## @@ -397,7 +397,18 @@ class RelayBuildModule : public runtime::ModuleNode { relay_module =

[GitHub] [tvm] mehrdadh commented on a diff in pull request #12207: [microTVM] Refactor pytest fixtures

mehrdadh commented on code in PR #12207: URL: https://github.com/apache/tvm/pull/12207#discussion_r936961362 ## python/tvm/micro/testing/pytest_plugin.py: ## @@ -0,0 +1,111 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license

[GitHub] [tvm] vinx13 commented on a diff in pull request #12127: [MetaSchedule] Enhance Conv2d NCHW Winograd Schedule Rules

vinx13 commented on code in PR #12127: URL: https://github.com/apache/tvm/pull/12127#discussion_r936956096 ## tests/python/unittest/test_meta_schedule_space_cuda.py: ## @@ -1239,6 +1239,90 @@ def cbr_0(data: T.Buffer[(1, 224, 224, 3), "float32"], kernel: T.Buffer[(7, 7, 3

[tvm] branch last-successful updated (39ffe0a5ce -> 49587cfcea)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from 39ffe0a5ce [TVMScript] Add object path tracing to StructuralEqual (#12101) add 49587cfcea [ci] Add

[GitHub] [tvm] kparzysz-quic commented on pull request #12140: [LLVM] Create LLVM scope object for use with LLVM libraries

kparzysz-quic commented on PR #12140: URL: https://github.com/apache/tvm/pull/12140#issuecomment-1204259018 What do you think about `LLVMInstance`? -- 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

[GitHub] [tvm] nverke opened a new pull request, #12294: [Hexagon] Add tests for 2 operators that utilize Qualcomm HVX intrinsics.

nverke opened a new pull request, #12294: URL: https://github.com/apache/tvm/pull/12294 **Background** This PR adds 2 tests for unused HVX implementations of mmul and conv2d. These were created to experiment with the performance of HVX intrinsics in tvm and each test outputs the runtime

[GitHub] [tvm] nverke closed pull request #12293: [Hexagon] Add tests for HVX implementations of conv2d and mmul.

nverke closed pull request #12293: [Hexagon] Add tests for HVX implementations of conv2d and mmul. URL: https://github.com/apache/tvm/pull/12293 -- 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

[tvm] branch micro_rename_qemu updated (f2fb18b5a2 -> 548f51ff97)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch micro_rename_qemu in repository https://gitbox.apache.org/repos/asf/tvm.git omit f2fb18b5a2 fake name for ci_cortexm omit e3e7aa77f3 Jenkins omit 2eeb16a428 more ci omit

[GitHub] [tvm] kparzysz-quic commented on pull request #12140: [LLVM] Create LLVM scope object for use with LLVM libraries

kparzysz-quic commented on PR #12140: URL: https://github.com/apache/tvm/pull/12140#issuecomment-1204233776 > [...] perhaps we can change name of LLVMScope to LLVMContext I'm a bit concerned that it could create confusion given that LLVMContext is name of LLVM type. Let me think a

[GitHub] [tvm] nverke opened a new pull request, #12293: [Hexagon] Add tests for HVX implementations of conv2d and mmul.

nverke opened a new pull request, #12293: URL: https://github.com/apache/tvm/pull/12293 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

[GitHub] [tvm] octoJon commented on a diff in pull request #12284: [Relay][Op] Multinomial

octoJon commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r936925295 ## python/tvm/topi/random/kernel.py: ## @@ -602,3 +602,57 @@ def normal(gen, mean, scale, out_shape, out_dtype): ) return new_gen, random_values + + +def

[GitHub] [tvm] octoJon commented on a diff in pull request #12284: [Relay][Op] Multinomial

octoJon commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r936918746 ## tests/python/topi/python/test_topi_prng.py: ## @@ -157,8 +185,27 @@ def test_uniform(target, dev): assert np.max(rands) <= 10.0

[GitHub] [tvm] octoJon commented on a diff in pull request #12284: [Relay][Op] Multinomial

octoJon commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r936918746 ## tests/python/topi/python/test_topi_prng.py: ## @@ -157,8 +185,27 @@ def test_uniform(target, dev): assert np.max(rands) <= 10.0

[GitHub] [tvm] octoJon commented on a diff in pull request #12284: [Relay][Op] Multinomial

octoJon commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r936912646 ## python/tvm/topi/random/kernel.py: ## @@ -602,3 +602,57 @@ def normal(gen, mean, scale, out_shape, out_dtype): ) return new_gen, random_values + + +def

[GitHub] [tvm] octoJon commented on a diff in pull request #12284: [Relay][Op] Multinomial

octoJon commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r936912331 ## python/tvm/topi/random/kernel.py: ## @@ -602,3 +602,57 @@ def normal(gen, mean, scale, out_shape, out_dtype): ) return new_gen, random_values + + +def

[GitHub] [tvm] jwfromm commented on a diff in pull request #12284: [Relay][Op] Multinomial

jwfromm commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r936902713 ## tests/python/topi/python/test_topi_prng.py: ## @@ -157,8 +185,27 @@ def test_uniform(target, dev): assert np.max(rands) <= 10.0

[GitHub] [tvm] jwfromm commented on a diff in pull request #12284: [Relay][Op] Multinomial

jwfromm commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r936900244 ## python/tvm/relay/frontend/pytorch.py: ## @@ -3386,6 +3386,19 @@ def trilu(self, inputs, input_types, mode): upper = True if mode == "triu" else False

[GitHub] [tvm] tkonolige commented on a diff in pull request #12284: [Relay][Op] Multinomial

tkonolige commented on code in PR #12284: URL: https://github.com/apache/tvm/pull/12284#discussion_r936845448 ## tests/python/topi/python/test_topi_prng.py: ## @@ -157,8 +185,27 @@ def test_uniform(target, dev): assert np.max(rands) <= 10.0

[GitHub] [tvm] tkonolige commented on issue #10842: [make error] 3rdparty/libbacktrace/configure Syntax error

tkonolige commented on issue #10842: URL: https://github.com/apache/tvm/issues/10842#issuecomment-1204140164 @wolfworld6 Can you check that all submodules are up to date by running `git submodule init; git submodule update`. If that does not work, can you post the full configure + build

[GitHub] [tvm] tqchen commented on pull request #12140: [LLVM] Create LLVM scope object for use with LLVM libraries

tqchen commented on PR #12140: URL: https://github.com/apache/tvm/pull/12140#issuecomment-1204140133 Thanks @kparzysz-quic , my guess is that we just need naming and documentation clarifications. e.g. the perhaps we can change name of LLVMScope to LLVMContext to capture the fact that it is

[GitHub] [tvm] gromero commented on pull request #11406: Fix command line input process issue.

gromero commented on PR #11406: URL: https://github.com/apache/tvm/pull/11406#issuecomment-1204128171 @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] kparzysz-quic commented on pull request #12140: [LLVM] Create LLVM scope object for use with LLVM libraries

kparzysz-quic commented on PR #12140: URL: https://github.com/apache/tvm/pull/12140#issuecomment-1204077887 Originally the scope took effect immediately, but there was an ordering problem with deserializing LLVM IR: LLVM flags would only be known once we've made LLVM calls, and once we've

[GitHub] [tvm] tqchen commented on pull request #12140: [LLVM] Create LLVM scope object for use with LLVM libraries

tqchen commented on PR #12140: URL: https://github.com/apache/tvm/pull/12140#issuecomment-1204060680 Sorry for delayed reply. Approving as it is a good chunk of improvement. It would be great to discuss naming and documentation a bit. When I see LLVMScope, I was expecting LLVM to

[GitHub] [tvm] blackkker commented on pull request #12028: [WIP][Pylint] Making frontend tests pylint compliant

blackkker commented on PR #12028: URL: https://github.com/apache/tvm/pull/12028#issuecomment-1204050200 > > > > > > > > > > > > Not work casue it will result in `RuntimeError: Boolean value of Tensor with more than one value is ambiguous` as

[GitHub] [tvm] mbrookhart commented on pull request #12290: [FQ2I] Add attrs to adaptive_avg_pool1d

mbrookhart commented on PR #12290: URL: https://github.com/apache/tvm/pull/12290#issuecomment-1204050108 Can you add a test? -- 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.

[GitHub] [tvm] Mousius closed issue #12136: [ci] Add retries to upload to S3

Mousius closed issue #12136: [ci] Add retries to upload to S3 URL: https://github.com/apache/tvm/issues/12136 -- 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,

[GitHub] [tvm] Mousius merged pull request #12221: [ci] Add retries to S3 uploads/downloads

Mousius merged PR #12221: URL: https://github.com/apache/tvm/pull/12221 -- 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:

[tvm] branch main updated: [ci] Add retries to S3 uploads/downloads (#12221)

This is an automated email from the ASF dual-hosted git repository. mousius 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 49587cfcea [ci] Add retries to S3 uploads/downloads

[GitHub] [tvm] SebastianBoblest commented on pull request #12028: [WIP][Pylint] Making frontend tests pylint compliant

SebastianBoblest commented on PR #12028: URL: https://github.com/apache/tvm/pull/12028#issuecomment-1203989596 > > > > > Not work casue it will result in `RuntimeError: Boolean value of Tensor with more than one value is ambiguous` as

[GitHub] [tvm] blackkker commented on pull request #12028: [WIP][Pylint] Making frontend tests pylint compliant

blackkker commented on PR #12028: URL: https://github.com/apache/tvm/pull/12028#issuecomment-1203968943 > Not work casue it will result in `RuntimeError: Boolean value of Tensor with more than one value is ambiguous` as

[GitHub] [tvm] SebastianBoblest commented on pull request #12028: [WIP][Pylint] Making frontend tests pylint compliant

SebastianBoblest commented on PR #12028: URL: https://github.com/apache/tvm/pull/12028#issuecomment-1203964598 > Got some problem when fixing `dangerous-default-value` in `pytorch/test_forward.py`. > >

[GitHub] [tvm] NVukobrat opened a new issue, #12292: [Bug] Tensor advanced indexing with boolean mask not working

NVukobrat opened a new issue, #12292: URL: https://github.com/apache/tvm/issues/12292 Hi, I’m trying to index the tensor in the following fashion: ```py x[0, mask] ``` where x is the input tensor (e.g. (2, 3) shape), while the mask is a boolean array (e.g. (3,)

[GitHub] [tvm] gigiblender closed pull request #12291: [CI] Add retries to upload to S3

gigiblender closed pull request #12291: [CI] Add retries to upload to S3 URL: https://github.com/apache/tvm/pull/12291 -- 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] gigiblender commented on pull request #12291: [CI] Add retries to upload to S3

gigiblender commented on PR #12291: URL: https://github.com/apache/tvm/pull/12291#issuecomment-1203946269 > hi @gigiblender, what is the relation between this PR and #12221 ? Oh, my bad. Seems like there's been a misunderstanding, and I wasn't aware of the existence of #12221. Thanks

[GitHub] [tvm] Mousius commented on pull request #12291: [CI] Add retries to upload to S3

Mousius commented on PR #12291: URL: https://github.com/apache/tvm/pull/12291#issuecomment-1203927149 hi @gigiblender, what is the relation between this PR and https://github.com/apache/tvm/pull/12221 ? -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] gigiblender opened a new pull request, #12291: [CI] Add retries to upload to S3

gigiblender opened a new pull request, #12291: URL: https://github.com/apache/tvm/pull/12291 This PR adds a wrapper function to retry up to 3 times the `aws s3 cp` call in case it fails. @driazati @areusch -- This is an automated message from the Apache Git Service. To respond

[GitHub] [tvm] blackkker commented on pull request #12028: [WIP][Pylint] Making frontend tests pylint compliant

blackkker commented on PR #12028: URL: https://github.com/apache/tvm/pull/12028#issuecomment-1203913515 Got some problem when fixing `dangerous-default-value` in `pytorch/test_forward.py`.

[tvm] branch last-successful updated (85624ff667 -> 39ffe0a5ce)

This is an automated email from the ASF dual-hosted git repository. github-bot pushed a change to branch last-successful in repository https://gitbox.apache.org/repos/asf/tvm.git from 85624ff667 [ci][docker] Fix deploy to tlcpackstaging on Docker Hub (#12282) add 39ffe0a5ce

[GitHub] [tvm] wrongtest-intellif commented on a diff in pull request #12266: [TIR] Add tir::builtin::undef

wrongtest-intellif commented on code in PR #12266: URL: https://github.com/apache/tvm/pull/12266#discussion_r936538582 ## src/tir/transforms/remove_store_undef.cc: ## @@ -0,0 +1,173 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [tvm] nrdmtt opened a new issue, #12289: Can't compile from source with TFLite support

nrdmtt opened a new issue, #12289: URL: https://github.com/apache/tvm/issues/12289 I am trying to complite TVM from source in order to have microTVM support and GPU acceleration using Cuda and Tensorflow. ### Expected behavior TVM compiled from source, microTVM enabled, with

[tvm] 01/02: Update TensorFlow to release 2.9

This is an automated email from the ASF dual-hosted git repository. leandron pushed a commit to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git commit ff30307397e83d1d4afc965439f1269538d0cefe Author: Leandro Nunes AuthorDate: Mon May 23 11:55:08 2022 +0100

[tvm] 02/02: Add #12130 PR for testing.

This is an automated email from the ASF dual-hosted git repository. leandron pushed a commit to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git commit 1beb0c19c5e128eaf2e2d0eaaffcd9cf6ba1abb2 Author: Leandro Nunes AuthorDate: Wed Aug 3 11:43:10 2022 +0100

[tvm] branch ci-docker-staging updated (112c37eda3 -> 1beb0c19c5)

This is an automated email from the ASF dual-hosted git repository. leandron pushed a change to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git discard 112c37eda3 Update Jenkins ci_arm, ci_cpu, ci_gpu, ci_qemu to use 20220725-085822-8ae520b1b discard

[GitHub] [tvm] wrongtest-intellif opened a new pull request, #12288: [TIR][Schedule] Support annotate dict typed value

wrongtest-intellif opened a new pull request, #12288: URL: https://github.com/apache/tvm/pull/12288 Allow `s.annotate(loop_rv, pragma_key, {"cfg0": 1, "cfg1": 2, ...})` . -- This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] [tvm] mikepapadim opened a new pull request, #12287: [QNN] Add qnn op for abs to solve wrong scale on quantize

mikepapadim opened a new pull request, #12287: URL: https://github.com/apache/tvm/pull/12287 Prior implementation caused the FQI pass to produce the following graph when `abs` was used: ``` %14 = qnn.dequantize(%13, 6.4712f /* ty=float32 */, 119 /* ty=int32 */, axis=1) /*

[GitHub] [tvm] wolfworld6 commented on issue #7857: pynq z1 fails to make the VTA, caused by the libbacktrace

wolfworld6 commented on issue #7857: URL: https://github.com/apache/tvm/issues/7857#issuecomment-1203711120 > Have you solved this problem？ Have you solved this problem？thx -- This is an automated message from the Apache Git Service. To respond to the message, please log on to

[GitHub] [tvm] billishyahao commented on pull request #11966: [DNNL][BYOC] Enable Altering Dense Weight Layout

billishyahao commented on PR #11966: URL: https://github.com/apache/tvm/pull/11966#issuecomment-1203700770 @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.

[GitHub] [tvm] wolfworld6 commented on issue #10842: [make error] 3rdparty/libbacktrace/configure Syntax error

wolfworld6 commented on issue #10842: URL: https://github.com/apache/tvm/issues/10842#issuecomment-1203697890 > @jackyflame Are you still having this issue? Happy to help if it is still an issue. > > If so, could you provide some more information to help us debug? For example: what

[GitHub] [tvm] sfvaroglu commented on pull request #12284: [Relay][Op] Multinomial

sfvaroglu commented on PR #12284: URL: https://github.com/apache/tvm/pull/12284#issuecomment-1203674033 LGTM, thanks @jwfromm! Would be nice to have this in the onnx importer, too :) -- This is an automated message from the Apache Git Service. To respond to the message, please log on

[GitHub] [tvm] echuraev opened a new pull request, #12286: [Adreno][OpenCL] Get rid of extra memory copy

echuraev opened a new pull request, #12286: URL: https://github.com/apache/tvm/pull/12286 - Added device_copy pass - Support textures for all injective ops - Avoid extra memory copy cc: @csullivan, @elvin-n -- This is an automated message from the Apache Git Service. To

[GitHub] [tvm] junrushao1994 merged pull request #12101: [TVMScript] Add object path tracing to StructuralEqual

junrushao1994 merged PR #12101: URL: https://github.com/apache/tvm/pull/12101 -- 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:

  1   2   >