[GitHub] [tvm] leandron commented on pull request #12541: [ACL] Adjust mobilenet test for Keras 2.9

leandron commented on PR #12541: URL: https://github.com/apache/tvm/pull/12541#issuecomment-1223576754 Thanks @ashutosh-arm @lhutton1 ! -- 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 main updated (e252d7f3ab -> d26bf809e4)

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 e252d7f3ab [docs] Add CI contribution instructions (#12551) add d26bf809e4 [ACL] Adjust mobilenet test for Keras

[GitHub] [tvm] leandron merged pull request #12541: [ACL] Adjust mobilenet test for Keras 2.9

leandron merged PR #12541: URL: https://github.com/apache/tvm/pull/12541 -- 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] masahi commented on issue #12524: [Bug] Lower Accuracy with a model compared to pytorch

masahi commented on issue #12524: URL: https://github.com/apache/tvm/issues/12524#issuecomment-1223566223 Yeah, that PR is very old, but their current implementation in

[GitHub] [tvm] Lucien0 opened a new issue, #12553: [Bug][MicroTVM][Model Library Format] Cannot export when whole graph is partitioned to another compiler

Lucien0 opened a new issue, #12553: URL: https://github.com/apache/tvm/issues/12553 Hi, I constructed a simple example like that def test_export_mlf(): x = relay.var("x", shape=(10, 10)) w0 = relay.var("w0", shape=(10, 10)) w1 = relay.var("w1",

[GitHub] [tvm] crawlingcub commented on issue #12524: [Bug] Lower Accuracy with a model compared to pytorch

crawlingcub commented on issue #12524: URL: https://github.com/apache/tvm/issues/12524#issuecomment-1223556813 There was this fix in pytorch for numerically stable logsigmoid: https://github.com/pytorch/pytorch/pull/2211 Maybe something similar needs to be done here? -- This is an

[GitHub] [tvm] masahi commented on issue #12524: [Bug] Lower Accuracy with a model compared to pytorch

masahi commented on issue #12524: URL: https://github.com/apache/tvm/issues/12524#issuecomment-1223554340 Interesting. Is there a known numerical issue if we translate PT `log_sigmoid` simply by `log(sigmoid(x))`? Otherwise I have no idea where the difference is coming from. -- This is

[GitHub] [tvm] crawlingcub commented on issue #12524: [Bug] Lower Accuracy with a model compared to pytorch

crawlingcub commented on issue #12524: URL: https://github.com/apache/tvm/issues/12524#issuecomment-1223549827 Yes, the outputs changed only when I replaced relu with logsigmoid. They were exactly same before that. -- This is an automated message from the Apache Git Service. To respond

[tvm] branch last-successful updated (8e95bba697 -> 58f2139ffd)

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 8e95bba697 Remove mutable defaults in mlp_model (#12546) add 3bd168194f check for CMSIS_PATH in

[GitHub] [tvm] yelite opened a new pull request, #12552: [TVMScript] TIR Printer: Expr

yelite opened a new pull request, #12552: URL: https://github.com/apache/tvm/pull/12552 This PR: - Adds printing function for all expression types in TIR This PR is in draft state because it's branched off from a pending PR #12506 Tracking issue:

[GitHub] [tvm] masahi commented on issue #12524: [Bug] Lower Accuracy with a model compared to pytorch

masahi commented on issue #12524: URL: https://github.com/apache/tvm/issues/12524#issuecomment-1223543413 You said you replaced relu with logsigmoid. I have a feeling that huge values in the PT output come from logarithm. If you keep relu, do the two outputs agree? -- This is an

[GitHub] [tvm] Lucien0 commented on pull request #12519: [MicroTVM] fix compile error when the compiler implements char as unsigned

Lucien0 commented on PR #12519: URL: https://github.com/apache/tvm/pull/12519#issuecomment-1223530278 done -- 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] crawlingcub commented on issue #12524: [Bug] Lower Accuracy with a model compared to pytorch

crawlingcub commented on issue #12524: URL: https://github.com/apache/tvm/issues/12524#issuecomment-1223524294 Yes, thats what I am seeing as well. With argmax, torch's output label is 39, while for tvm it is 33. Seems like some numerical error is happening here? -- This is an automated

[tvm] branch last-successful updated (902343a817 -> 8e95bba697)

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 902343a817 [BugFix][UMA] Fix order issue in uma_lower (#12447) add 1e399fa451 [TIR] Add pass to check

[GitHub] [tvm] HaoKang-Timmy commented on issue #12536: [Bug][Parser] Uable to parse 'broadcast_to' operator

HaoKang-Timmy commented on issue #12536: URL: https://github.com/apache/tvm/issues/12536#issuecomment-1223484129 Could you please look at this problem, please? @jroesch -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and

[GitHub] [tvm] alter-xp commented on pull request #12534: [Docker][CI][RISC-V] Build riscv-isa-sim (spike) in ci_riscv Docker image to enable RISC-V unit testing

alter-xp commented on PR #12534: URL: https://github.com/apache/tvm/pull/12534#issuecomment-1223472292 @PhilippvK @areusch > cc @alter-xp can you take a look at the compiler used here and see if it would also work for you (or whether there's one in Zephyr we should consider)? for

[tvm] branch main updated (58f2139ffd -> e252d7f3ab)

This is an automated email from the ASF dual-hosted git repository. driazati pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 58f2139ffd [microTVM] Replace static fixtures with parameterization (#12530) add e252d7f3ab [docs] Add CI

[GitHub] [tvm] driazati merged pull request #12551: Add instructions on how to add a new CI docker image

driazati merged PR #12551: URL: https://github.com/apache/tvm/pull/12551 -- 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] Sunny-Island commented on pull request #12144: [Auto Scheduler] Upgrade autoscheduler xgboost callback

Sunny-Island commented on PR #12144: URL: https://github.com/apache/tvm/pull/12144#issuecomment-1223467861 > I don't think a unit test is viable specifically for xgboost version compatibility because (IIUC) our CI instance could only install one version of xgboost... Hi, sorry I am

[GitHub] [tvm] mkatanbaf commented on pull request #12390: [MicroTVM] add heap-size to project options

mkatanbaf commented on PR #12390: URL: https://github.com/apache/tvm/pull/12390#issuecomment-1223453759 @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] PhilippvK commented on pull request #12534: [Docker][CI][RISC-V] Build riscv-isa-sim (spike) in ci_riscv Docker image to enable RISC-V unit testing

PhilippvK commented on PR #12534: URL: https://github.com/apache/tvm/pull/12534#issuecomment-1223448196 @areusch > cc @alter-xp can you take a look at the compiler used here and see if it would also work for you (or whether there's one in Zephyr we should consider)? I got in

[GitHub] [tvm] PhilippvK commented on pull request #12534: [Docker][CI][RISC-V] Build riscv-isa-sim (spike) in ci_riscv Docker image to enable RISC-V unit testing

PhilippvK commented on PR #12534: URL: https://github.com/apache/tvm/pull/12534#issuecomment-1223439906 > turns out the network issue was actually a forgotten step in adding the image, which i've now hopefully rectified. @areusch it is probably a fault on my side. Seems like the

[GitHub] [tvm] PhilippvK commented on a diff in pull request #12534: [Docker][CI][RISC-V] Build riscv-isa-sim (spike) in ci_riscv Docker image to enable RISC-V unit testing

PhilippvK commented on code in PR #12534: URL: https://github.com/apache/tvm/pull/12534#discussion_r952081941 ## python/tvm/micro/testing/aot_test_utils.py: ## @@ -65,6 +65,17 @@ }, ) +AOT_SPIKE_RUNNER = AOTTestRunner( Review Comment: AFAIK AOTTestRunner and the

[GitHub] [tvm] PhilippvK commented on pull request #12522: [AutoTVM] [TVMC] Visualize tuning progress

PhilippvK commented on PR #12522: URL: https://github.com/apache/tvm/pull/12522#issuecomment-1223432097 > @PhilippvK this looks very cool! the only thing missing is to declare the dependency on matplotlib so that the user is not surprised with and ImportError when trying to use it :). >

[GitHub] [tvm] mehrdadh merged pull request #12539: [microTVM] Rework evaluate_model_accuracy into a more generic helper function

mehrdadh merged PR #12539: URL: https://github.com/apache/tvm/pull/12539 -- 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 (5cef6bf559 -> 58f2139ffd)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 5cef6bf559 [microTVM] Rework evaluate_model_accuracy into a more generic helper function (#12539) add 58f2139ffd

[GitHub] [tvm] mehrdadh merged pull request #12530: [microTVM] Replace static fixtures with parameterization

mehrdadh merged PR #12530: URL: https://github.com/apache/tvm/pull/12530 -- 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 (3bd168194f -> 5cef6bf559)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 3bd168194f check for CMSIS_PATH in project generation (#12547) add 5cef6bf559 [microTVM] Rework

[GitHub] [tvm] mkatanbaf commented on a diff in pull request #12390: [MicroTVM] add heap-size to project options

mkatanbaf commented on code in PR #12390: URL: https://github.com/apache/tvm/pull/12390#discussion_r952067644 ## apps/microtvm/zephyr/template_project/boards.json: ## @@ -54,7 +54,8 @@ "is_qemu": false, "fpu": true, "vid_hex": "0483", -

[GitHub] [tvm] mehrdadh commented on a diff in pull request #12512: [bug fix] fix split filter in conv1d and conv2d to avoid empty search space

mehrdadh commented on code in PR #12512: URL: https://github.com/apache/tvm/pull/12512#discussion_r952065783 ## python/tvm/topi/arm_cpu/mprofile/dsp/conv1d.py: ## @@ -105,7 +106,12 @@ def conv1d_nwc_dsp_compute(cfg, data, kernel, strides, padding, dilation, out_dt

[GitHub] [tvm] github-actions[bot] commented on pull request #12414: [ci] Default to n=2 for test parallelism

github-actions[bot] commented on PR #12414: URL: https://github.com/apache/tvm/pull/12414#issuecomment-1223420755 Built docs for commit ae149a7b3c2e2ca05f7c3731a267faa42b8153cf can be found [here](https://pr-docs.tlcpack.ai/PR-12414/5/docs/index.html). -- This is an automated

[tvm] branch micro/dense_dsp_add_split created (now 894be8029e)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch micro/dense_dsp_add_split in repository https://gitbox.apache.org/repos/asf/tvm.git at 894be8029e lint No new revisions were added by this update.

[tvm] branch main updated (8e95bba697 -> 3bd168194f)

This is an automated email from the ASF dual-hosted git repository. mehrdadh pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 8e95bba697 Remove mutable defaults in mlp_model (#12546) add 3bd168194f check for CMSIS_PATH in project generation

[GitHub] [tvm] mehrdadh merged pull request #12547: [bug fix] verify CMSIS_PATH at project generation

mehrdadh merged PR #12547: URL: https://github.com/apache/tvm/pull/12547 -- 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] mehrdadh commented on pull request #12547: [bug fix] verify CMSIS_PATH at project generation

mehrdadh commented on PR #12547: URL: https://github.com/apache/tvm/pull/12547#issuecomment-1223412512 cc @alanmacd -- 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 commented on a diff in pull request #12544: [MetaSchedule] Add software pipeline in CUDA tensor core auto tensorization

vinx13 commented on code in PR #12544: URL: https://github.com/apache/tvm/pull/12544#discussion_r952057498 ## python/tvm/meta_schedule/schedule_rule/multi_level_tiling.py: ## @@ -172,6 +174,7 @@ def __init__( vector_load_lens: Optional[List[int]] = None,

[GitHub] [tvm] Hzfengsy commented on a diff in pull request #12544: [MetaSchedule] Add software pipeline in CUDA tensor core auto tensorization

Hzfengsy commented on code in PR #12544: URL: https://github.com/apache/tvm/pull/12544#discussion_r952056307 ## python/tvm/meta_schedule/schedule_rule/multi_level_tiling.py: ## @@ -172,6 +174,7 @@ def __init__( vector_load_lens: Optional[List[int]] = None,

[GitHub] [tvm] github-actions[bot] commented on pull request #12551: Add instructions on how to add a new CI docker image

github-actions[bot] commented on PR #12551: URL: https://github.com/apache/tvm/pull/12551#issuecomment-1223394723 Built docs for commit c852c6d3c73912e1dd96b16176e87d58431c412c can be found [here](https://pr-docs.tlcpack.ai/PR-12551/1/docs/index.html). -- This is an automated

[GitHub] [tvm] masahi commented on a diff in pull request #12544: [MetaSchedule] Add software pipeline in CUDA tensor core auto tensorization

masahi commented on code in PR #12544: URL: https://github.com/apache/tvm/pull/12544#discussion_r952043594 ## src/meta_schedule/schedule_rule/multi_level_tiling_tensor_core.cc: ## @@ -286,6 +293,54 @@ std::vector MultiLevelTilingTensorCoreNode::AddReadReuseTensorCore(

[tvm] branch main updated (1e399fa451 -> 8e95bba697)

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 1e399fa451 [TIR] Add pass to check for out of bounds memory access (#12352) add 8e95bba697 Remove mutable defaults in

[GitHub] [tvm] zxybazh merged pull request #12546: Remove mutable default arguments in mlp_model

zxybazh merged PR #12546: URL: https://github.com/apache/tvm/pull/12546 -- 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] michalpiszczek commented on pull request #12546: Remove mutable default arguments in mlp_model

michalpiszczek commented on PR #12546: URL: https://github.com/apache/tvm/pull/12546#issuecomment-1223352639 Thank you for the reviews @junrushao @Kathryn-cat ! -- 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] juda commented on a diff in pull request #12485: [Relay][Op] embedding_bag operator implementation

juda commented on code in PR #12485: URL: https://github.com/apache/tvm/pull/12485#discussion_r952026615 ## python/tvm/relay/frontend/pytorch.py: ## @@ -2242,6 +2243,29 @@ def embedding(self, inputs, input_types): return _op.take(weight, indices.astype("int32"),

[GitHub] [tvm] juda commented on a diff in pull request #12485: [Relay][Op] embedding_bag operator implementation

juda commented on code in PR #12485: URL: https://github.com/apache/tvm/pull/12485#discussion_r952026083 ## python/tvm/relay/frontend/pytorch.py: ## @@ -3919,7 +3950,8 @@ def _handel_nested_input(inputs): out_names = _get_output_names(op_node)

[GitHub] [tvm] gromero commented on pull request #12519: [MicroTVM] fix compile error when the compiler implements char as unsigned

gromero commented on PR #12519: URL: https://github.com/apache/tvm/pull/12519#issuecomment-1223347783 @Lucien0 @mehrdadh Could we used `int ch` like in other functions in `load_json.c`, like `JSONReader_NextObjectItem()`? -- This is an automated message from the Apache Git Service. To

[GitHub] [tvm] driazati commented on a diff in pull request #12551: Add instructions on how to add a new CI docker image

driazati commented on code in PR #12551: URL: https://github.com/apache/tvm/pull/12551#discussion_r952024632 ## docs/contribute/ci.rst: ## @@ -174,6 +174,29 @@ The images for these containers are hosted in the `tlcpack Docker Hub

[tvm] branch nightly-docker-update updated (3cb2f17836 -> 7320d09f95)

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 3cb2f17836 [ci][docker] Nightly Docker image update add 262906516a [TVMScript] Printer: add

[GitHub] [tvm] areusch opened a new pull request, #12551: Add instructions on how to add a new CI docker image

areusch opened a new pull request, #12551: URL: https://github.com/apache/tvm/pull/12551 This PR documents the steps to introducing a new CI docker image, which we've been doing a lot lately. @driazati -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] Kathryn-cat commented on pull request #12546: Remove mutable default arguments in mlp_model

Kathryn-cat commented on PR #12546: URL: https://github.com/apache/tvm/pull/12546#issuecomment-122038 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 above to go to the specific comment. To

[tvm] branch last-successful updated (e5e05feee3 -> 902343a817)

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 e5e05feee3 [LLVM] Add "cl-opt" attribute to target_kind "llvm" (#12440) add 902343a817 [BugFix][UMA]

[GitHub] [tvm] csullivan commented on a diff in pull request #12411: [Hexagon] Asynchronous DMA support

csullivan commented on code in PR #12411: URL: https://github.com/apache/tvm/pull/12411#discussion_r952001668 ## src/runtime/hexagon/hexagon_user_dma.cc: ## @@ -82,12 +95,13 @@ int HexagonUserDMA::Copy(void* dst, void* src, uint32_t length) { dmstart(dma_desc);

[GitHub] [tvm] csullivan commented on a diff in pull request #12411: [Hexagon] Asynchronous DMA support

csullivan commented on code in PR #12411: URL: https://github.com/apache/tvm/pull/12411#discussion_r951993684 ## tests/cpp-runtime/hexagon/hexagon_user_dma_tests.cc: ## @@ -57,9 +57,9 @@ TEST_F(HexagonUserDMATest, wait) { HexagonUserDMA::Get().Wait(10); } Review Comment:

[GitHub] [tvm] mbaret commented on pull request #12550: [AOT] Add AOTLowerMain pass to lower a Relay main into TIR

mbaret commented on PR #12550: URL: https://github.com/apache/tvm/pull/12550#issuecomment-1223289608 cc @gigiblender @mikepapadim @leandron @ekalda -- 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] mbaret opened a new pull request, #12550: [AOT] Add AOTLowerMain pass to lower a Relay main into TIR

mbaret opened a new pull request, #12550: URL: https://github.com/apache/tvm/pull/12550 Tracking issue: https://github.com/apache/tvm/issues/12548 This is a pass refactored out of the AOTExecutorCodegen. Instead of combining all of the functionality of the AOTExecutorCodegen into a

[GitHub] [tvm] mbaret opened a new pull request, #12549: [DRAFT][AOT] Refactor of AOTCodegenExecutor

mbaret opened a new pull request, #12549: URL: https://github.com/apache/tvm/pull/12549 This is a draft to illustrate the refactor described up to P4 in tracking issue https://github.com/apache/tvm/issues/12548. The pull requests 1-4 will appear in sequence to be merged separately to this

[GitHub] [tvm] adstraw commented on a diff in pull request #12411: [Hexagon] Asynchronous DMA support

adstraw commented on code in PR #12411: URL: https://github.com/apache/tvm/pull/12411#discussion_r951975770 ## src/runtime/hexagon/hexagon_user_dma.h: ## @@ -0,0 +1,90 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[tvm] branch last-successful updated (c1bd0227be -> e5e05feee3)

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 c1bd0227be [FIX] Fix bug in resize2d unittest func name (#12498) add 4d7e7a8b57 [ci] xfail failing

[GitHub] [tvm] mbaret opened a new issue, #12548: [Tracking Issue] Refactor AOT executor codegen into IRModule -> IRModule form

mbaret opened a new issue, #12548: URL: https://github.com/apache/tvm/issues/12548 The AOT executor codegen has accumulated significant technical debt and dead code during its development, resulting in a component that is difficult to understand, test and extend. Refactoring it will

[GitHub] [tvm] mkatanbaf opened a new pull request, #12547: [bug fix] check for CMSIS_PATH in project generation

mkatanbaf opened a new pull request, #12547: URL: https://github.com/apache/tvm/pull/12547 This PR checks if the CMSIS_PATH is defined during project generation and sets the value appropriately in the CMakeLists.txt file. This would prevent a later build error if the CMSIS_PATH is not

[GitHub] [tvm] michalpiszczek commented on pull request #12546: Remove mutable default arguments in mlp_model

michalpiszczek commented on PR #12546: URL: https://github.com/apache/tvm/pull/12546#issuecomment-1223116485 cc @zxybazh @Kathryn-cat PTAL -- 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] michalpiszczek opened a new pull request, #12546: Remove mutable defaults in mlp_model

michalpiszczek opened a new pull request, #12546: URL: https://github.com/apache/tvm/pull/12546 Removes [mutable default arguments](https://docs.python-guide.org/writing/gotchas/#mutable-default-arguments) in `python/tvm/meta_schedule/cost_model/mlp_model.py`. -- This is an automated

[GitHub] [tvm-rfcs] slyubomirsky commented on a diff in pull request #89: [RFC] Relax Upstreaming

slyubomirsky commented on code in PR #89: URL: https://github.com/apache/tvm-rfcs/pull/89#discussion_r951902969 ## rfcs/0089-relax-upstreaming.md: ## @@ -0,0 +1,701 @@ +- Feature Name: Relax Upstreaming +- Start Date: 2022-08-17 +- RFC PR: 

[GitHub] [tvm] shingjan commented on a diff in pull request #12485: [Relay][Op] embedding_bag operator implementation

shingjan commented on code in PR #12485: URL: https://github.com/apache/tvm/pull/12485#discussion_r951886852 ## python/tvm/relay/frontend/pytorch.py: ## @@ -2242,6 +2243,29 @@ def embedding(self, inputs, input_types): return _op.take(weight, indices.astype("int32"),

[GitHub] [tvm] shingjan commented on a diff in pull request #12485: [Relay][Op] embedding_bag operator implementation

shingjan commented on code in PR #12485: URL: https://github.com/apache/tvm/pull/12485#discussion_r951889739 ## python/tvm/relay/frontend/pytorch.py: ## @@ -2242,6 +2243,29 @@ def embedding(self, inputs, input_types): return _op.take(weight, indices.astype("int32"),

[GitHub] [tvm] shingjan commented on a diff in pull request #12485: [Relay][Op] embedding_bag operator implementation

shingjan commented on code in PR #12485: URL: https://github.com/apache/tvm/pull/12485#discussion_r951888906 ## python/tvm/relay/frontend/pytorch.py: ## @@ -3919,7 +3950,8 @@ def _handel_nested_input(inputs): out_names = _get_output_names(op_node)

[tvm] branch main updated (902343a817 -> 1e399fa451)

This is an automated email from the ASF dual-hosted git repository. andrewzhaoluo pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 902343a817 [BugFix][UMA] Fix order issue in uma_lower (#12447) add 1e399fa451 [TIR] Add pass to check for out

[GitHub] [tvm] AndrewZhaoLuo commented on a diff in pull request #12352: [TIR] Add pass to check for out of bounds memory access

AndrewZhaoLuo commented on code in PR #12352: URL: https://github.com/apache/tvm/pull/12352#discussion_r951067578 ## src/tir/analysis/oob_checker.cc: ## @@ -0,0 +1,130 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] shingjan commented on a diff in pull request #12485: [Relay][Op] embedding_bag operator implementation

shingjan commented on code in PR #12485: URL: https://github.com/apache/tvm/pull/12485#discussion_r951887727 ## python/tvm/relay/frontend/pytorch.py: ## @@ -2485,6 +2509,12 @@ def interpolate(self, inputs, input_types): ) def numel(self, inputs, input_types): +

[GitHub] [tvm] AndrewZhaoLuo merged pull request #12352: [TIR] Add pass to check for out of bounds memory access

AndrewZhaoLuo merged PR #12352: URL: https://github.com/apache/tvm/pull/12352 -- 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 #12390: [MicroTVM] add heap-size to project options

areusch commented on code in PR #12390: URL: https://github.com/apache/tvm/pull/12390#discussion_r951887026 ## apps/microtvm/zephyr/template_project/microtvm_api_server.py: ## @@ -195,6 +195,33 @@ def _get_device_args(options): ) +def _get_board_mem_size(options): +

[GitHub] [tvm] shingjan commented on a diff in pull request #12485: [Relay][Op] embedding_bag operator implementation

shingjan commented on code in PR #12485: URL: https://github.com/apache/tvm/pull/12485#discussion_r951886852 ## python/tvm/relay/frontend/pytorch.py: ## @@ -2242,6 +2243,29 @@ def embedding(self, inputs, input_types): return _op.take(weight, indices.astype("int32"),

[GitHub] [tvm] areusch commented on issue #12511: [Flaky Test] `tests/python/contrib/test_ethosu/test_codegen.py::test_binary_add_with_non_4d_shapes[ifm_shape0-ifm2_shape0-ethos-u55-32]`, `tests/pytho

areusch commented on issue #12511: URL: https://github.com/apache/tvm/issues/12511#issuecomment-1222965187 oh whooops. that was not intentional! -- 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] areusch commented on a diff in pull request #12521: [AutoTVM] Add support for text buffers to ApplyHistoryBest

areusch commented on code in PR #12521: URL: https://github.com/apache/tvm/pull/12521#discussion_r951879892 ## python/tvm/autotvm/task/dispatcher.py: ## @@ -238,12 +246,12 @@ class ApplyHistoryBest(DispatchContext): Parameters -- -records : str, list of

[tvm] branch last-successful updated (40fd43e24b -> c1bd0227be)

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 40fd43e24b [MicroTVM] expose project options in autotuning (#12479) add 3f56851264 [TIR][Schedule]

[GitHub] [tvm] manupak commented on pull request #12447: [BugFix][UMA] Fix order issue in uma_lower

manupak commented on PR #12447: URL: https://github.com/apache/tvm/pull/12447#issuecomment-1222917809 Thanks @MichaelJKlaiber! -- 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

[tvm] branch main updated (e5e05feee3 -> 902343a817)

This is an automated email from the ASF dual-hosted git repository. manupa pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from e5e05feee3 [LLVM] Add "cl-opt" attribute to target_kind "llvm" (#12440) add 902343a817 [BugFix][UMA] Fix order issue

[GitHub] [tvm] manupak merged pull request #12447: [BugFix][UMA] Fix order issue in uma_lower

manupak merged PR #12447: URL: https://github.com/apache/tvm/pull/12447 -- 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] Icemist opened a new pull request, #12545: Introducing multi_filter into ConfigSpace autotvm.

Icemist opened a new pull request, #12545: URL: https://github.com/apache/tvm/pull/12545 Using multi_filter allows to set additional restrictions on ConfigEntity indexes used by avtotvm tuning and exclude the unsuitable in advance. Co-authored-by: Andrey Malyshev Co-authored-by:

[tvm] branch last-successful updated (389675668a -> 40fd43e24b)

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 389675668a [CI] Set test dependency on "transformers" package with pytest.importorskip (#12528) add

[tvm] branch main updated (66a31e97a7 -> e5e05feee3)

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 66a31e97a7 [CI] Add alexnet and googlenet caffe model to request hook (#12510) add e5e05feee3 [LLVM] Add "cl-opt"

[GitHub] [tvm] kparzysz-quic merged pull request #12440: [LLVM] Add "cl-opt" attribute to target_kind "llvm"

kparzysz-quic merged PR #12440: URL: https://github.com/apache/tvm/pull/12440 -- 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] mkatanbaf commented on a diff in pull request #12390: [MicroTVM] add heap-size to project options

mkatanbaf commented on code in PR #12390: URL: https://github.com/apache/tvm/pull/12390#discussion_r951814458 ## apps/microtvm/zephyr/template_project/microtvm_api_server.py: ## @@ -370,6 +397,12 @@ def _get_nrf_device_args(options): type="bool", help="Run on

[tvm] branch main updated (4d7e7a8b57 -> 66a31e97a7)

This is an automated email from the ASF dual-hosted git repository. driazati pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git from 4d7e7a8b57 [ci] xfail failing ethosu codegen tests (#12508) add 66a31e97a7 [CI] Add alexnet and googlenet caffe

[GitHub] [tvm] driazati merged pull request #12510: [CI] Add alexnet and googlenet caffe model to request hook

driazati merged PR #12510: URL: https://github.com/apache/tvm/pull/12510 -- 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] driazati closed issue #12465: [Flaky Test] `cython.py::tests.python.frontend.darknet.test_forward`

driazati closed issue #12465: [Flaky Test] `cython.py::tests.python.frontend.darknet.test_forward` URL: https://github.com/apache/tvm/issues/12465 -- 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 last-successful updated (8146a9bf2c -> 389675668a)

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 8146a9bf2c [TIR] Support AllocateConst nodes in TensorIR scheduling flow (#12489) add 48a8cbd57d

[GitHub] [tvm] shingjan commented on issue #12511: [Flaky Test] `tests/python/contrib/test_ethosu/test_codegen.py::test_binary_add_with_non_4d_shapes[ifm_shape0-ifm2_shape0-ethos-u55-32]`, `tests/pyth

shingjan commented on issue #12511: URL: https://github.com/apache/tvm/issues/12511#issuecomment-1222791397 Hi @areusch would that be possible that we keep this open until Luke can take a look? I believe David's PR only mitigate the issue instead of fixing it. Thanks! -- This is an

[GitHub] [tvm] vinx13 opened a new pull request, #12544: [MetaSchedule] Add software pipeline in CUDA tensor core auto tensorization

vinx13 opened a new pull request, #12544: URL: https://github.com/apache/tvm/pull/12544 cc @spectrometerHBH @masahi @junrushao -- 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-rfcs] junrushao commented on pull request #89: [RFC] Relax Upstreaming

junrushao commented on PR #89: URL: https://github.com/apache/tvm-rfcs/pull/89#issuecomment-1222788202 Thank you @leandron @ekalda for the questions, and @zhiics, @slyubomirsky, @Hzfengsy, @sunggg for the discussion! As a long-term contributor since 2018, the pre-Relay era, and the

[GitHub] [tvm] quic-sanirudh commented on a diff in pull request #12204: [Hexagon] Add Hand written HVX conv2d

quic-sanirudh commented on code in PR #12204: URL: https://github.com/apache/tvm/pull/12204#discussion_r951798525 ## include/tvm/runtime/hexagon/ops/conv2d.h: ## @@ -0,0 +1,133 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] driazati commented on pull request #12533: [CI] Remove Vela from ci_cpu

driazati commented on PR #12533: URL: https://github.com/apache/tvm/pull/12533#issuecomment-1222784277 adding a new image name requires a small infra change: https://github.com/tlc-pack/ci/commit/71b059493f78376af6bd9a6ddeb720f8aaeeeff2, that is in place now and should fix the issue, I

[GitHub] [tvm] quic-sanirudh commented on a diff in pull request #12204: [Hexagon] Add Hand written HVX conv2d

quic-sanirudh commented on code in PR #12204: URL: https://github.com/apache/tvm/pull/12204#discussion_r951797734 ## src/runtime/hexagon/ops/conv2d_hvx.cc: ## @@ -0,0 +1,473 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] quic-sanirudh commented on a diff in pull request #12204: [Hexagon] Add Hand written HVX conv2d

quic-sanirudh commented on code in PR #12204: URL: https://github.com/apache/tvm/pull/12204#discussion_r951797972 ## include/tvm/runtime/hexagon/ops/conv2d.h: ## @@ -0,0 +1,133 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] quic-sanirudh commented on a diff in pull request #12204: [Hexagon] Add Hand written HVX conv2d

quic-sanirudh commented on code in PR #12204: URL: https://github.com/apache/tvm/pull/12204#discussion_r951796717 ## include/tvm/runtime/hexagon/ops/conv2d.h: ## @@ -0,0 +1,133 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] quic-sanirudh commented on a diff in pull request #12204: [Hexagon] Add Hand written HVX conv2d

quic-sanirudh commented on code in PR #12204: URL: https://github.com/apache/tvm/pull/12204#discussion_r951795171 ## src/runtime/hexagon/ops/conv_utils.cc: ## @@ -0,0 +1,243 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license

[GitHub] [tvm] driazati commented on a diff in pull request #12436: [CI] Assert some unittests are not skipped in CI

driazati commented on code in PR #12436: URL: https://github.com/apache/tvm/pull/12436#discussion_r951790460 ## tests/scripts/github_skipped_tests_comment.py: ## @@ -0,0 +1,206 @@ +#!/usr/bin/env python3 +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [tvm] trahman-quic commented on pull request #12406: [TOPI][Hexagon] Implement quantized elementwise

trahman-quic commented on PR #12406: URL: https://github.com/apache/tvm/pull/12406#issuecomment-1222774093 Hi @mehrdadh CI is green now. Please let me know if you have any feedback on the PR -- This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] [tvm] mehrdadh commented on a diff in pull request #12530: [microTVM] Replace static fixtures with parameterization

mehrdadh commented on code in PR #12530: URL: https://github.com/apache/tvm/pull/12530#discussion_r951769878 ## python/tvm/micro/testing/pytest_plugin.py: ## @@ -26,14 +26,18 @@ from tvm.contrib.utils import tempdir -from .utils import get_supported_boards +from .utils

[tvm] branch last-successful updated (7c318d7196 -> 8146a9bf2c)

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 7c318d7196 [ETHOSN] Remove support for older versions of the driver stack (#12347) add 8146a9bf2c

[GitHub] [tvm] leandron commented on pull request #12541: [ACL] Adjust mobilenet test for Keras 2.9

leandron commented on PR #12541: URL: https://github.com/apache/tvm/pull/12541#issuecomment-1222749028 For reference, this is the change on Keras side: ``` https://github.com/keras-team/keras/commit/b6abfaed1326e3c5e1d233e35c97bde177b562fa ``` -- This is an automated message

  1   2   3   >