[GitHub] [tvm] anijain2305 merged pull request #7639: [TFLite] Cast operator adapted for MLIR-based convertor

anijain2305 merged pull request #7639: URL: https://github.com/apache/tvm/pull/7639 -- 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. For queries about this service, please

[tvm] branch main updated: [TFLite] Cast operator adapted for MLIR-based convertor (#7639)

This is an automated email from the ASF dual-hosted git repository. anijain2305 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 2ee860e [TFLite] Cast operator adapted for MLIR-b

[GitHub] [tvm] anijain2305 commented on pull request #7639: [TFLite] Cast operator adapted for MLIR-based convertor

anijain2305 commented on pull request #7639: URL: https://github.com/apache/tvm/pull/7639#issuecomment-802598299 Thanks @d-smirnov @u99127 -- 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 sp

[GitHub] [tvm] jwfromm commented on pull request #7698: [TVMC] Python Scripting Init Files

jwfromm commented on pull request #7698: URL: https://github.com/apache/tvm/pull/7698#issuecomment-802552863 @leandron yeah we think TVMC has an excellent set of utilities to make it a simplified python API for new users, kind of like keras. You did an awesome job packing tons of functiona

[tvm] 01/01: update to staging containers

This is an automated email from the ASF dual-hosted git repository. areusch pushed a commit to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git commit 53e3656e77ab3c3bf0da909cec3b50d9e8a11739 Author: Andrew Reusch AuthorDate: Thu Mar 18 20:31:03 2021 -0700

[tvm] branch ci-docker-staging updated (f0796a6 -> 53e3656)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a change to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git. discard f0796a6 add ci-qemu-staging omit 37e0bce Fix path for test. omit 65f0b4e Lint comments. omit

[tvm] 01/01: add ci-qemu-staging

This is an automated email from the ASF dual-hosted git repository. areusch pushed a commit to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git commit f0796a6b126a334607e7242b086be463f6f4def9 Author: Andrew Reusch AuthorDate: Thu Mar 18 17:22:08 2021 -0700

[tvm] branch ci-docker-staging updated (6a77cdd -> f0796a6)

This is an automated email from the ASF dual-hosted git repository. areusch pushed a change to branch ci-docker-staging in repository https://gitbox.apache.org/repos/asf/tvm.git. omit 6a77cdd prefix junit results with FFI type omit f9ff088 Merge remote-tracking branch 'origin/main' int

[GitHub] [tvm] FrozenGene commented on pull request #7670: [CPP_RPC] allow user supplied work dir

FrozenGene commented on pull request #7670: URL: https://github.com/apache/tvm/pull/7670#issuecomment-802512877 Thanks @euntaik -- 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 comm

[tvm] branch main updated (431a7d6 -> e467748)

This is an automated email from the ASF dual-hosted git repository. zhaowu pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 431a7d6 Default value for graph_runtime Init lookup_linked_param_func (#7676) add e467748 [CPP_RPC] allow user sup

[GitHub] [tvm] FrozenGene merged pull request #7670: [CPP_RPC] allow user supplied work dir

FrozenGene merged pull request #7670: URL: https://github.com/apache/tvm/pull/7670 -- 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. For queries about this service, please

[GitHub] [tvm] Hzfengsy commented on a change in pull request #7643: [docs] Getting Started with TVM: AutoTVM and Matrix Multiply

Hzfengsy commented on a change in pull request #7643: URL: https://github.com/apache/tvm/pull/7643#discussion_r597369445 ## File path: tutorials/get_started/autotvm_matmul.py ## @@ -0,0 +1,377 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contribut

[GitHub] [tvm] comaniac opened a new pull request #7703: [AutoScheduler] Add function name in message

comaniac opened a new pull request #7703: URL: https://github.com/apache/tvm/pull/7703 This PR makes auto_scheduler prints the function name when no schedule is found, as function name is usually more informatic than the compute DAG at the first glance. Example: ``` --

[GitHub] [tvm] zhuwenxi commented on issue #7596: [BUG][CodeGenC]“tvm_call_packed” crashes when build with target “c”

zhuwenxi commented on issue #7596: URL: https://github.com/apache/tvm/issues/7596#issuecomment-802453554 > @zhuwenxi Did you build tvm with cblas support? Yes, definitely. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHu

[GitHub] [tvm] junrushao1994 edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

junrushao1994 edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802447025 Yeah @tkonolige if there are implementation details you don't understand, as a reviewer, you can just ask. Then if you documents could help with better understanding, i

[GitHub] [tvm] junrushao1994 edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

junrushao1994 edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802447025 Yeah @tkonolige if there are implementation details you don't understand, as a reviewer, you can just ask. Then if you documents could help with better understanding, i

[GitHub] [tvm] junrushao1994 commented on pull request #7630: [TensorIR] TVMScript Parser/Printer

junrushao1994 commented on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802447025 Yeah @tkonolige if there are implementation details you don't understand, as a review, you can just ask. Then if you documents could help with better understanding, indeed you

[GitHub] [tvm] icemelon9 commented on a change in pull request #7624: [Profiling,VM] Profiling interface for VM and Graph Runtime

icemelon9 commented on a change in pull request #7624: URL: https://github.com/apache/tvm/pull/7624#discussion_r597328855 ## File path: include/tvm/runtime/profiling.h ## @@ -144,6 +149,134 @@ class Timer : public ObjectRef { */ Timer DefaultTimer(TVMContext ctx); +namespa

[GitHub] [tvm] apivovarov closed pull request #7701: Use dmlc-core as a cmake project

apivovarov closed pull request #7701: URL: https://github.com/apache/tvm/pull/7701 -- 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. For queries about this service, please

[tvm-site] branch asf-site updated: Build at Thu Mar 18 20:31:02 EDT 2021

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/tvm-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 41536bf Build at Thu Mar 18 20:31:02 EDT

[tvm-site] branch main updated: Fix link (#27)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/tvm-site.git The following commit(s) were added to refs/heads/main by this push: new 85c7e4e Fix link (#27) 85c7e4e is described below

[GitHub] [tvm] tqchen commented on pull request #7701: Use dmlc-core as a cmake project

tqchen commented on pull request #7701: URL: https://github.com/apache/tvm/pull/7701#issuecomment-802408703 right, the dependency of TVM on dmlc-core is only limited to the header only, so we do not want to introduce a separate build dep step. Hopefully things can be easier after #7153

[GitHub] [tvm] tkonolige commented on pull request #7701: Use dmlc-core as a cmake project

tkonolige commented on pull request #7701: URL: https://github.com/apache/tvm/pull/7701#issuecomment-802402509 I've proposed this change in the past, but it seems like we want to use dmlc-core as a header only library. @tqchen can provide more info. You'll be happy to know that #7153

[GitHub] [tvm] trevor-m commented on pull request #7702: [BYOC][TensorRT] Explicitly free TensorRT engine and context in destructor.

trevor-m commented on pull request #7702: URL: https://github.com/apache/tvm/pull/7702#issuecomment-802397445 @comaniac Could you take a look? -- 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] trevor-m opened a new pull request #7702: [BYOC][TensorRT] Explicitly free TensorRT engine and context in destructor.

trevor-m opened a new pull request #7702: URL: https://github.com/apache/tvm/pull/7702 We weren't freeing the TRT engines when TRT runtime module was destroyed. -- 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] anijain2305 commented on pull request #7676: Default value for graph_runtime Init lookup_linked_param_func

anijain2305 commented on pull request #7676: URL: https://github.com/apache/tvm/pull/7676#issuecomment-802391514 Thanks @apivovarov @d-smirnov @zhiics -- 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

[tvm] branch main updated (45442ed -> 431a7d6)

This is an automated email from the ASF dual-hosted git repository. anijain2305 pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 45442ed [Relay][Training][Pass] Factor out first-order AD to a module pass (#7677) add 431a7d6 Default value

[GitHub] [tvm] anijain2305 merged pull request #7676: Default value for graph_runtime Init lookup_linked_param_func

anijain2305 merged pull request #7676: URL: https://github.com/apache/tvm/pull/7676 -- 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. For queries about this service, please

[GitHub] [tvm] apivovarov opened a new pull request #7701: Use dmlc-core as a cmake project

apivovarov opened a new pull request #7701: URL: https://github.com/apache/tvm/pull/7701 About the problem: `dmlc-core` has cmake generated `build_config.h` which enables `DMLC_LOG_STACK_TRACE` in case `cxxabi.h` and `execinfo.h` files are present on the system. This feature (LOG_STACK_

[GitHub] [tvm] trevor-m commented on pull request #7700: [TOPI][CUDA] Fix 0 valid boxes case for NMS when return_indices=False

trevor-m commented on pull request #7700: URL: https://github.com/apache/tvm/pull/7700#issuecomment-802339326 @anijain2305 @masahi Please take a look -- 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] trevor-m opened a new pull request #7700: [TOPI][CUDA] Fix 0 valid boxes case for NMS when return_indices=False

trevor-m opened a new pull request #7700: URL: https://github.com/apache/tvm/pull/7700 When return_indices is False, the values of all boxes, scores must be set to -1 when they are invalid. The current code was not doing this, resulting in uninitialized memory being returned in the op outp

[GitHub] [tvm] tqchen commented on pull request #7624: [Profiling,VM] Profiling interface for VM and Graph Runtime

tqchen commented on pull request #7624: URL: https://github.com/apache/tvm/pull/7624#issuecomment-802336272 cc @trevor-m @icemelon9 @yzhliu would be great if you can take a look as well -- This is an automated message from the Apache Git Service. To respond to the message, please log on t

[GitHub] [tvm] jwfromm opened a new pull request #7699: [Relay][Frontend][ONNX] Raise error when user provides an input not in the onnx graph.

jwfromm opened a new pull request #7699: URL: https://github.com/apache/tvm/pull/7699 There was recently some discussion about how the onnx importer should handle being passed a shape dict that contains inputs not found in the onnx graph. Currently, the importer will just ignore these inpu

[GitHub] [tvm] masahi commented on pull request #7694: [TORCH] Implement avg_pool1d

masahi commented on pull request #7694: URL: https://github.com/apache/tvm/pull/7694#issuecomment-802319133 No, we do squash merge anyway when the PR is merged. -- 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] cgerum commented on pull request #7694: [TORCH] Implement avg_pool1d

cgerum commented on pull request #7694: URL: https://github.com/apache/tvm/pull/7694#issuecomment-802318347 Lint error should be fixed, should I squash the commits? -- 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] areusch commented on pull request #7653: Rename GraphRuntime to GraphExecutor

areusch commented on pull request #7653: URL: https://github.com/apache/tvm/pull/7653#issuecomment-802318182 @zhiics good call. I added python backwards compatibility for the `create` API. I don't know whether we are sensitive to backwards compatibility for the other languages. since the r

[GitHub] [tvm] masahi commented on pull request #7694: [TORCH] Implement avg_pool1d

masahi commented on pull request #7694: URL: https://github.com/apache/tvm/pull/7694#issuecomment-802312776 Yes thanks, please fix the lint error -- 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] cgerum commented on pull request #7694: [TORCH] Implement avg_pool1d

cgerum commented on pull request #7694: URL: https://github.com/apache/tvm/pull/7694#issuecomment-802309798 Thanks for the review, I hope this is what you had in mind. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use

[GitHub] [tvm] tqchen commented on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen commented on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802304707 @Hzfengsy please do your best effort in applying the actionable comments :) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] zxybazh commented on a change in pull request #7534: [Target] Add support for target object with host field compatible with previous api

zxybazh commented on a change in pull request #7534: URL: https://github.com/apache/tvm/pull/7534#discussion_r597244567 ## File path: python/tvm/autotvm/task/relay_integration.py ## @@ -122,6 +124,9 @@ def extract_from_multiple_program(mods, params, target, target_host=None, o

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] majercakdavid commented on issue #7697: Build error: TypeError: Map is not registered via TVM_REGISTER_NODE_TYPE

majercakdavid commented on issue #7697: URL: https://github.com/apache/tvm/issues/7697#issuecomment-802298169 @comaniac I have tried to update the local repository, build it through conda environment, create a new python environment (venv) and run setup.py install from the build folder but

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] mdw-octoml commented on a change in pull request #7698: [TVMC] Python Scripting Init Files

mdw-octoml commented on a change in pull request #7698: URL: https://github.com/apache/tvm/pull/7698#discussion_r597240308 ## File path: python/tvm/driver/tvmc/__init__.py ## @@ -21,3 +21,6 @@ from . import autotuner from . import compiler from . import runner +from .fronten

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] tqchen edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timel

[GitHub] [tvm] tqchen commented on pull request #7630: [TensorIR] TVMScript Parser/Printer

tqchen commented on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802294684 Thanks @tkonolige . The level of documentation can be debatabe, while it is always good to add more details in the docs, we also cannot too hung up on things, to ensure timely deliv

[GitHub] [tvm] CircleSpin commented on pull request #7698: [TVMC] Python Scripting Init Files

CircleSpin commented on pull request #7698: URL: https://github.com/apache/tvm/pull/7698#issuecomment-802294361 @jwfromm @leandron @mdw-octoml -- 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 th

[GitHub] [tvm] CircleSpin opened a new pull request #7698: [TVMC] Python Scripting Init Files

CircleSpin opened a new pull request #7698: URL: https://github.com/apache/tvm/pull/7698 This PR adds to two init files to make tvmc cleaner for running python scripts. For example, instead of running tvmc.runner.run_module() you can now run tvmc.run() -- This is an automated message fr

[GitHub] [tvm] tkonolige commented on pull request #7696: [GO] Fix go bindings

tkonolige commented on pull request #7696: URL: https://github.com/apache/tvm/pull/7696#issuecomment-802287680 In this case, the cmake build system is readily available, so we should use it to build. I've re-enabled the test. -- This is an automated message from the Apache Git Ser

[GitHub] [tvm-vta] tqchen merged pull request #23: Use TVMArrayCopy API to replace VTA internal function VTABufferCPUPtr

tqchen merged pull request #23: URL: https://github.com/apache/tvm-vta/pull/23 -- 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. For queries about this service, please cont

[tvm-vta] branch main updated: Use TVMArrayCopyFromBytes API to replace VTA internal function VTABufferCPUPtr (#23)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/tvm-vta.git The following commit(s) were added to refs/heads/main by this push: new f11ca65 Use TVMArrayCopyFromBytes API to replace V

[GitHub] [tvm] tqchen commented on pull request #7696: [GO] Fix go bindings

tqchen commented on pull request #7696: URL: https://github.com/apache/tvm/pull/7696#issuecomment-802283000 The all in one runtime pack approach is valuable in places where a build cmake system is not readily available (e.g. custom compiler command invocation is needed). We should try to

[GitHub] [tvm] mbrookhart closed pull request #7672: [OpenCL] Skip opencl memory operations on empty tensors

mbrookhart closed pull request #7672: URL: https://github.com/apache/tvm/pull/7672 -- 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. For queries about this service, please

[GitHub] [tvm] mbrookhart commented on pull request #7672: [OpenCL] Skip opencl memory operations on empty tensors

mbrookhart commented on pull request #7672: URL: https://github.com/apache/tvm/pull/7672#issuecomment-802274064 Closing in favor of #7691 -- 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 spe

[GitHub] [tvm] masahi commented on pull request #7694: [TORCH] Implement avg_pool1d

masahi commented on pull request #7694: URL: https://github.com/apache/tvm/pull/7694#issuecomment-802268135 Thanks, can you unify the `avg_pool1d` and `2d`? They look identical. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub

[GitHub] [tvm] comaniac closed issue #7697: Build error: TypeError: Map is not registered via TVM_REGISTER_NODE_TYPE

comaniac closed issue #7697: URL: https://github.com/apache/tvm/issues/7697 -- 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. For queries about this service, please contact

[GitHub] [tvm] comaniac commented on issue #7697: Build error: TypeError: Map is not registered via TVM_REGISTER_NODE_TYPE

comaniac commented on issue #7697: URL: https://github.com/apache/tvm/issues/7697#issuecomment-802267750 Hi this usually indicates that your TVM library is not up-to-date. You may check if you `git submodule update --resursive` and rebuild TVM after pulling the last commit. In addit

[GitHub] [tvm] majercakdavid opened a new issue #7697: Build error: TypeError: Map is not registered via TVM_REGISTER_NODE_TYPE

majercakdavid opened a new issue #7697: URL: https://github.com/apache/tvm/issues/7697 Hello, while building the ONNX model (originally Pytorch) in TVM I got the following error: `TypeError: Map is not registered via TVM_REGISTER_NODE_TYPE` I have quite hard time understanding th

[GitHub] [tvm] zxybazh commented on a change in pull request #7534: [Target] Add support for target object with host field compatible with previous api

zxybazh commented on a change in pull request #7534: URL: https://github.com/apache/tvm/pull/7534#discussion_r597215602 ## File path: python/tvm/relay/build_module.py ## @@ -205,7 +208,7 @@ def _build_module_no_factory(mod, target=None, target_host=None, params=None, mo T

[GitHub] [tvm] electriclilies commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

electriclilies commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597212799 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +

[GitHub] [tvm] zxybazh commented on a change in pull request #7534: [Target] Add support for target object with host field compatible with previous api

zxybazh commented on a change in pull request #7534: URL: https://github.com/apache/tvm/pull/7534#discussion_r597212646 ## File path: src/auto_scheduler/measure_record.cc ## @@ -163,8 +163,10 @@ struct Handler<::tvm::auto_scheduler::SearchTaskNode> { writer->WriteArrayItem

[GitHub] [tvm] zxybazh commented on a change in pull request #7534: [Target] Add support for target object with host field compatible with previous api

zxybazh commented on a change in pull request #7534: URL: https://github.com/apache/tvm/pull/7534#discussion_r597209297 ## File path: python/tvm/target/target.py ## @@ -494,3 +498,32 @@ def _load_config_dict(config_dict_str): if not isinstance(key, str): r

[GitHub] [tvm] zxybazh commented on a change in pull request #7534: [Target] Add support for target object with host field compatible with previous api

zxybazh commented on a change in pull request #7534: URL: https://github.com/apache/tvm/pull/7534#discussion_r597207631 ## File path: python/tvm/autotvm/task/relay_integration.py ## @@ -122,6 +124,9 @@ def extract_from_multiple_program(mods, params, target, target_host=None, o

[GitHub] [tvm] tkonolige opened a new pull request #7696: [GO] Fix go bindings

tkonolige opened a new pull request #7696: URL: https://github.com/apache/tvm/pull/7696 Fixes #7683. The go bindings are using a **very** brittle approach to vendoring TVM. They build their own version of the tvm_runtime by just `#include`ing all the runtime files. This does not cor

[GitHub] [tvm-vta] huajsj commented on pull request #23: Use TVMArrayCopy API to replace VTA internal function VTABufferCPUPtr

huajsj commented on pull request #23: URL: https://github.com/apache/tvm-vta/pull/23#issuecomment-802244733 @tqchen @tmoreau89 @liangfu , could you help for a review ? thanks. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub a

[GitHub] [tvm] areusch commented on pull request #7557: Clean up uTVM demo runtime, add ONNX model test and tutorial

areusch commented on pull request #7557: URL: https://github.com/apache/tvm/pull/7557#issuecomment-802243546 @mehrdadh ah, to ci-qemu more precisely. I think we need to update `docker/Dockerfile.ci_qemu` to include installing onnx, then rerun `docker/build.sh` and push the image. I can do

[GitHub] [tvm] mbrookhart commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

mbrookhart commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597194395 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +# On

[GitHub] [tvm] mbrookhart commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

mbrookhart commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597194246 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +# On

[GitHub] [tvm] mehrdadh edited a comment on pull request #7557: Clean up uTVM demo runtime, add ONNX model test and tutorial

mehrdadh edited a comment on pull request #7557: URL: https://github.com/apache/tvm/pull/7557#issuecomment-802230829 @areusch I think we need to add onnx to reference VM to fix the current issue. -- This is an automated message from the Apache Git Service. To respond to the message, plea

[GitHub] [tvm] mehrdadh commented on pull request #7557: Clean up uTVM demo runtime, add ONNX model test and tutorial

mehrdadh commented on pull request #7557: URL: https://github.com/apache/tvm/pull/7557#issuecomment-802230829 @areusch I think we need to add onnx to reference VM fix the current issue. -- This is an automated message from the Apache Git Service. To respond to the message, please log on t

[GitHub] [tvm] masahi commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

masahi commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597177115 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +# Only r

[GitHub] [tvm] masahi commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

masahi commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597177115 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +# Only r

[GitHub] [tvm] masahi commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

masahi commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597177115 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +# Only r

[GitHub] [tvm] areusch commented on a change in pull request #7653: Rename GraphRuntime to GraphExecutor

areusch commented on a change in pull request #7653: URL: https://github.com/apache/tvm/pull/7653#discussion_r597154342 ## File path: docs/install/from_source.rst ## @@ -88,7 +88,7 @@ The configuration of TVM can be modified by `config.cmake`. - On macOS, for some versions o

[GitHub] [tvm] areusch commented on a change in pull request #7653: Rename GraphRuntime to GraphExecutor

areusch commented on a change in pull request #7653: URL: https://github.com/apache/tvm/pull/7653#discussion_r597152991 ## File path: CMakeLists.txt ## @@ -326,7 +326,7 @@ if(USE_PROFILER) file(GLOB RUNTIME_GRAPH_DEBUG_SRCS src/runtime/graph/debug/*.cc) Review comment:

[GitHub] [tvm] tkonolige commented on pull request #7153: [RUNTIME] Add libbacktrace for backtraces with line numbers

tkonolige commented on pull request #7153: URL: https://github.com/apache/tvm/pull/7153#issuecomment-802202635 Thanks everyone for putting the time in to review this. I know it was a big PR. -- This is an automated message from the Apache Git Service. To respond to the message, please lo

[GitHub] [tvm] tkonolige edited a comment on pull request #7630: [TensorIR] TVMScript Parser/Printer

tkonolige edited a comment on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802200082 I'd like my latest round of reviews addressed. We need documentation in `context_maintainer.py` and `special_stmt.py` so that other developers can understand and work on th

[GitHub] [tvm] tkonolige commented on pull request #7630: [TensorIR] TVMScript Parser/Printer

tkonolige commented on pull request #7630: URL: https://github.com/apache/tvm/pull/7630#issuecomment-802200082 I'd like my latest round of reviews addressed. We need documentation in `context_maintainer.py` and `special_stmt.py` so that other developers can understand and work on this code

[GitHub] [tvm] anijain2305 commented on pull request #7639: [TFLite] Cast operator adapted for MLIR-based convertor

anijain2305 commented on pull request #7639: URL: https://github.com/apache/tvm/pull/7639#issuecomment-802193123 @d-smirnov Can you please check the CI error? I will merge when it passes. Assinging to myself. -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] areusch commented on a change in pull request #7537: [PYTHON][Target] add stm32mp1 target helper

areusch commented on a change in pull request #7537: URL: https://github.com/apache/tvm/pull/7537#discussion_r597102389 ## File path: python/tvm/target/target.py ## @@ -324,6 +324,31 @@ def arm_cpu(model="unknown", options=None): return Target(" ".join(["llvm"] + opts))

[GitHub] [tvm] areusch commented on pull request #7557: Clean up uTVM demo runtime, add ONNX model test and tutorial

areusch commented on pull request #7557: URL: https://github.com/apache/tvm/pull/7557#issuecomment-802130443 @mdw-octoml it's not particularly user-friendly right now, but you can at least `docker/bash.sh ci-cpu tests/scripts/task_.sh` -- This is an automated message from the Apache Git

[GitHub] [tvm] Wheest commented on a change in pull request #6137: Better grouped convolution for CPU targets

Wheest commented on a change in pull request #6137: URL: https://github.com/apache/tvm/pull/6137#discussion_r597001134 ## File path: topi/python/topi/arm_cpu/group_conv2d.py ## @@ -0,0 +1,310 @@ +import tvm +from tvm import autotvm +from tvm import te +from ..util import get_co

[GitHub] [tvm] jwfromm commented on pull request #7006: [frontend][keras] Add support for TimeDistributed

jwfromm commented on pull request #7006: URL: https://github.com/apache/tvm/pull/7006#issuecomment-802116173 @slyubomirsky can you resolve the conflicts this PR has with the current main branch? Once you do I think we can merge this as it LGTM. -- This is an automated message from the Ap

[GitHub] [tvm] electriclilies commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

electriclilies commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597071262 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +

[GitHub] [tvm] Wheest commented on a change in pull request #6137: Better grouped convolution for CPU targets

Wheest commented on a change in pull request #6137: URL: https://github.com/apache/tvm/pull/6137#discussion_r596994641 ## File path: python/tvm/topi/x86/group_conv2d.py ## @@ -0,0 +1,371 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor lic

[GitHub] [tvm] electriclilies commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

electriclilies commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597072538 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +

[GitHub] [tvm] electriclilies commented on a change in pull request #7438: [ONNX] Enable GPU in ONNX importer tests

electriclilies commented on a change in pull request #7438: URL: https://github.com/apache/tvm/pull/7438#discussion_r597071262 ## File path: tests/scripts/task_python_frontend.sh ## @@ -31,6 +31,9 @@ find . -type f -path "*.pyc" | xargs rm -f # Rebuild cython make cython3 +

[GitHub] [tvm] Hzfengsy commented on a change in pull request #7642: [docs] Getting Started With TVM: Tensor Expressions

Hzfengsy commented on a change in pull request #7642: URL: https://github.com/apache/tvm/pull/7642#discussion_r596989768 ## File path: tutorials/get_started/tensor_expr_get_started.py ## @@ -17,16 +17,39 @@ """ .. _tutorial-tensor-expr-get-started: -Get Started with Tensor

[GitHub] [tvm] mdw-octoml commented on pull request #7557: Clean up uTVM demo runtime, add ONNX model test and tutorial

mdw-octoml commented on pull request #7557: URL: https://github.com/apache/tvm/pull/7557#issuecomment-802048221 Yep, thanks. I really need to figure out how to run the CI stuff locally. On Thu, Mar 18, 2021 at 8:57 AM Andrew Reusch ***@***.***> wrote: > @mdw-octoml

[GitHub] [tvm] areusch commented on pull request #7557: Clean up uTVM demo runtime, add ONNX model test and tutorial

areusch commented on pull request #7557: URL: https://github.com/apache/tvm/pull/7557#issuecomment-802043638 @mdw-octoml looks like you need to adjust the path in `tests/scripts/task_python_microtvm.sh` -- This is an automated message from the Apache Git Service. To respond to the messag

[GitHub] [tvm] Wheest commented on a change in pull request #6137: Better grouped convolution for CPU targets

Wheest commented on a change in pull request #6137: URL: https://github.com/apache/tvm/pull/6137#discussion_r597001134 ## File path: topi/python/topi/arm_cpu/group_conv2d.py ## @@ -0,0 +1,310 @@ +import tvm +from tvm import autotvm +from tvm import te +from ..util import get_co

[GitHub] [tvm] Wheest commented on a change in pull request #6137: Better grouped convolution for CPU targets

Wheest commented on a change in pull request #6137: URL: https://github.com/apache/tvm/pull/6137#discussion_r596998966 ## File path: python/tvm/relay/op/strategy/x86.py ## @@ -205,22 +205,20 @@ def conv2d_strategy_cpu(attrs, inputs, out_type, target): else: # group_conv2d

[GitHub] [tvm] Wheest commented on a change in pull request #6137: Better grouped convolution for CPU targets

Wheest commented on a change in pull request #6137: URL: https://github.com/apache/tvm/pull/6137#discussion_r596997928 ## File path: python/tvm/relay/op/strategy/x86.py ## @@ -175,21 +175,10 @@ def conv2d_strategy_cpu(attrs, inputs, out_type, target): else: # group_conv2d

[GitHub] [tvm] Wheest commented on a change in pull request #6137: Better grouped convolution for CPU targets

Wheest commented on a change in pull request #6137: URL: https://github.com/apache/tvm/pull/6137#discussion_r596994641 ## File path: python/tvm/topi/x86/group_conv2d.py ## @@ -0,0 +1,371 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor lic

  1   2   >