[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

comaniac commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448093018 ## File path: python/tvm/ansor/compute_dag.py ## @@ -0,0 +1,103 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

comaniac commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448092322 ## File path: python/tvm/ansor/auto_schedule.py ## @@ -0,0 +1,186 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [incubator-tvm] hewumars closed issue #5973: 1d vector dot product

hewumars closed issue #5973: URL: https://github.com/apache/incubator-tvm/issues/5973 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] [incubator-tvm] tqchen commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

tqchen commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448089682 ## File path: python/tvm/ansor/auto_schedule.py ## @@ -0,0 +1,186 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [incubator-tvm] hewumars opened a new issue #5973: 1d vector dot product

hewumars opened a new issue #5973: URL: https://github.com/apache/incubator-tvm/issues/5973 N = 512 ctx = tvm.context("cpu") k = te.reduce_axis((0,N),'k') A0 = te.placeholder((N,),'uint16', name='A0') A1 = te.placeholder((N,),'uint16', name='A1') B = te.compute((1,),

[GitHub] [incubator-tvm] FrozenGene commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

FrozenGene commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448084718 ## File path: tests/python/unittest/test_ansor_measure.py ## @@ -0,0 +1,67 @@ +# Licensed to the Apache Software Foundation (ASF) under one +#

[GitHub] [incubator-tvm] merrymercy commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

merrymercy commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448084658 ## File path: python/tvm/ansor/auto_schedule.py ## @@ -0,0 +1,186 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [incubator-tvm] merrymercy commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

merrymercy commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448084658 ## File path: python/tvm/ansor/auto_schedule.py ## @@ -0,0 +1,186 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [incubator-tvm] FrozenGene commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

FrozenGene commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448084428 ## File path: src/ansor/utils.cc ## @@ -0,0 +1,55 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor

[GitHub] [incubator-tvm] jcf94 commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

jcf94 commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448084431 ## File path: python/tvm/ansor/utils.py ## @@ -0,0 +1,157 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor

[GitHub] [incubator-tvm] jcf94 commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

jcf94 commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r448068982 ## File path: python/tvm/ansor/auto_schedule.py ## @@ -0,0 +1,186 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [incubator-tvm] electriclilies commented on a change in pull request #5932: [Frontend][Relay] Add Parser 2.0

electriclilies commented on a change in pull request #5932: URL: https://github.com/apache/incubator-tvm/pull/5932#discussion_r448050480 ## File path: src/parser/tokenizer.h ## @@ -0,0 +1,460 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more

[GitHub] [incubator-tvm] kevinthesun commented on pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

kevinthesun commented on pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#issuecomment-65201 @jcf94 Thanks for this PR! Will take a look late this week or early next week. This is an automated

[GitHub] [incubator-tvm] mbrookhart commented on a change in pull request #5826: [DYNAMIC] Add Dynamic reshape to a dynamic namespace and add DynamicToStatic Pass

mbrookhart commented on a change in pull request #5826: URL: https://github.com/apache/incubator-tvm/pull/5826#discussion_r448039240 ## File path: python/tvm/relay/op/_tensor_grad.py ## @@ -511,7 +511,7 @@ def batch_matmul_grad(orig, grad): @register_gradient("reshape") def

[GitHub] [incubator-tvm] icemelon9 commented on a change in pull request #5826: [DYNAMIC] Add Dynamic reshape to a dynamic namespace and add DynamicToStatic Pass

icemelon9 commented on a change in pull request #5826: URL: https://github.com/apache/incubator-tvm/pull/5826#discussion_r448021317 ## File path: src/relay/analysis/util.cc ## @@ -448,13 +448,8 @@ bool IsDataDependant(const CallNode* call) { return false; } - if

[GitHub] [incubator-tvm] trevor-m commented on pull request #5967: [MXNet] Add parser for contrib.box_decode

trevor-m commented on pull request #5967: URL: https://github.com/apache/incubator-tvm/pull/5967#issuecomment-652078789 @icemelon9 @zhiics Could you please review? Thanks! This is an automated message from the Apache Git

[incubator-tvm] branch master updated (44194d0 -> 36a0bf9)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git. from 44194d0 Raise an exception when extern function does not return Stmt (#5964) add 36a0bf9 Print right

[GitHub] [incubator-tvm] tqchen merged pull request #5965: Print right number of parentheses for LoadNode

tqchen merged pull request #5965: URL: https://github.com/apache/incubator-tvm/pull/5965 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

[GitHub] [incubator-tvm] trevor-m commented on a change in pull request #5961: [Relay/TOPI][OP] Add meshgrid op in Relay, TOPI, Pytorch frontend

trevor-m commented on a change in pull request #5961: URL: https://github.com/apache/incubator-tvm/pull/5961#discussion_r448013103 ## File path: src/relay/op/tensor/transform.cc ## @@ -1269,6 +1269,93 @@ RELAY_REGISTER_OP("repeat") .set_attr("FTVMCompute", RepeatCompute)

[GitHub] [incubator-tvm] yzhliu closed issue #5947: [VOTE] Release Apache TVM (incubating) v0.6.1.rc1

yzhliu closed issue #5947: URL: https://github.com/apache/incubator-tvm/issues/5947 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] [incubator-tvm] yzhliu commented on issue #5947: [VOTE] Release Apache TVM (incubating) v0.6.1.rc1

yzhliu commented on issue #5947: URL: https://github.com/apache/incubator-tvm/issues/5947#issuecomment-652063557 Thanks everyone for voting. The voting result has been sent out. This is an automated message from the Apache

[GitHub] [incubator-tvm] yzhliu opened a new issue #5972: [RESULT][VOTE] Release Apache TVM (incubating) v0.6.1.rc1

yzhliu opened a new issue #5972: URL: https://github.com/apache/incubator-tvm/issues/5972 Dear TVM community, I'm glad to announce the results of the vote. This vote passes with 12 +1 votes (9 binding), no 0 votes, and 0 -1 vote. +1 votes * Tianqi Chen (binding)

[GitHub] [incubator-tvm] zhiics commented on a change in pull request #5961: [Relay/TOPI][OP] Add meshgrid op in Relay, TOPI, Pytorch frontend

zhiics commented on a change in pull request #5961: URL: https://github.com/apache/incubator-tvm/pull/5961#discussion_r447994741 ## File path: src/relay/op/tensor/transform.cc ## @@ -1269,6 +1269,93 @@ RELAY_REGISTER_OP("repeat") .set_attr("FTVMCompute", RepeatCompute)

[GitHub] [incubator-tvm] jroesch opened a new issue #5971: [Relay][InferType] Standardize the behavior between IRModule::Add and IRModule::IRModule

jroesch opened a new issue #5971: URL: https://github.com/apache/incubator-tvm/issues/5971 Currently you get different behavior if you pass a set of functions to IRModule directly, vs iteratively calling Add. This behavior can be problematic when comparing modules for equality in testing

[GitHub] [incubator-tvm] MarisaKirisame commented on a change in pull request #5932: [Frontend][Relay] Add Parser 2.0

MarisaKirisame commented on a change in pull request #5932: URL: https://github.com/apache/incubator-tvm/pull/5932#discussion_r447985013 ## File path: tests/python/relay/test_ir_parser.py ## @@ -76,6 +76,7 @@ def graph_equal(lhs, rhs): def roundtrip(expr): x =

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

comaniac commented on a change in pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#discussion_r447916858 ## File path: python/tvm/ansor/measure.py ## @@ -0,0 +1,379 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more

[GitHub] [incubator-tvm] zhiics commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

zhiics commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447951976 ## File path: src/relay/backend/contrib/codegen_json/codegen_json.h ## @@ -0,0 +1,353 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] [incubator-tvm] zhiics commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

zhiics commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447949386 ## File path: src/relay/backend/contrib/codegen_json/codegen_json.h ## @@ -0,0 +1,353 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] [incubator-tvm] comaniac commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

comaniac commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447942670 ## File path: src/relay/backend/contrib/codegen_json/codegen_json.h ## @@ -0,0 +1,353 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] [incubator-tvm] areusch opened a new pull request #5970: Improve docker/bash script to handle git worktrees

areusch opened a new pull request #5970: URL: https://github.com/apache/incubator-tvm/pull/5970 The lint tool doesn't mount the correct .git directory into the docker image when using git-worktree. Adding support for that.

[GitHub] [incubator-tvm] jroesch opened a new issue #5969: [Relay][Parser] Add support for annotating error messages to add negative tests.

jroesch opened a new issue #5969: URL: https://github.com/apache/incubator-tvm/issues/5969 The parser should support getting annotated error comments and indexing them so we can use same parsing infrastructure to write negative test cases. This is common in other compilers and will help

[GitHub] [incubator-tvm] jroesch opened a new issue #5968: [Relay][Parser] Add type checking of all successful parses.

jroesch opened a new issue #5968: URL: https://github.com/apache/incubator-tvm/issues/5968 The Relay parser tests should type check all produced fragments of code, currently they just ensure they parse. This causes all kinds of issues as all type information must be identical between the

[incubator-tvm] branch master updated (9efe119 -> 44194d0)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git. from 9efe119 Fix small typo in nn.conv2d_gemm_weight_transform (#5925) add 44194d0 Raise an exception when

[incubator-tvm] branch master updated (7176483 -> 9efe119)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git. from 7176483 Make first order gradient graphs more efficient (#5959) add 9efe119 Fix small typo in

[incubator-tvm] branch master updated (bc22fb9 -> 7176483)

This is an automated email from the ASF dual-hosted git repository. tqchen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git. from bc22fb9 Fix the meaning of conv{1,2}d_transpose output_padding parameter. (#5758) add 7176483 Make

[GitHub] [incubator-tvm] t-vi commented on pull request #5959: Make first order gradient graphs more efficient

t-vi commented on pull request #5959: URL: https://github.com/apache/incubator-tvm/pull/5959#issuecomment-651857225 @MarisaKirisame Thank you! I rebased and now the CI is all happy again. This is an automated message from

[GitHub] [incubator-tvm] kparzysz-quic opened a new pull request #5965: Print right number of parentheses for LoadNode

kparzysz-quic opened a new pull request #5965: URL: https://github.com/apache/incubator-tvm/pull/5965 Stop printing the unnecessary `)` after each `LoadNode` that didn't have a matching `(`. This is an automated message

[GitHub] [incubator-tvm] FrozenGene edited a comment on pull request #5914: [clflush] Enable x86 cpu cache flush

FrozenGene edited a comment on pull request #5914: URL: https://github.com/apache/incubator-tvm/pull/5914#issuecomment-651708824 > How about we go with A1 for now, @FrozenGene can you update this PR to A1? @tqchen If we use ```python # cache flush packed is a packed func that

[GitHub] [incubator-tvm] FrozenGene edited a comment on pull request #5914: [clflush] Enable x86 cpu cache flush

FrozenGene edited a comment on pull request #5914: URL: https://github.com/apache/incubator-tvm/pull/5914#issuecomment-651708824 > How about we go with A1 for now, @FrozenGene can you update this PR to A1? @tqchen If we use ```python # cache flush packed is a packed func that

[GitHub] [incubator-tvm] kparzysz-quic opened a new pull request #5964: Raise an exception when extern function does not return Stmt

kparzysz-quic opened a new pull request #5964: URL: https://github.com/apache/incubator-tvm/pull/5964 The function for `tvm.te.extern` should return either `PrimExpr` or `Stmt`, however there is no check if it actually does so. If it does not, the result may be a segmentation fault later

[GitHub] [incubator-tvm] giuseros commented on pull request #5925: Fix small typo in nn.conv2d_gemm_weight_transform

giuseros commented on pull request #5925: URL: https://github.com/apache/incubator-tvm/pull/5925#issuecomment-651718265 Hi @tqchen , Any update on this? Thanks a lot This is an automated message from the Apache

[GitHub] [incubator-tvm] FrozenGene edited a comment on pull request #5914: [clflush] Enable x86 cpu cache flush

FrozenGene edited a comment on pull request #5914: URL: https://github.com/apache/incubator-tvm/pull/5914#issuecomment-651708824 > How about we go with A1 for now, @FrozenGene can you update this PR to A1? @tqchen If we use ```python # cache flush packed is a packed func that

[GitHub] [incubator-tvm] FrozenGene commented on pull request #5914: [clflush] Enable x86 cpu cache flush

FrozenGene commented on pull request #5914: URL: https://github.com/apache/incubator-tvm/pull/5914#issuecomment-651708824 > How about we go with A1 for now, @FrozenGene can you update this PR to A1? @tqchen If we use ``` # cache flush packed is a packed func that performs the

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

masahi commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447569351 ## File path: src/relay/backend/contrib/codegen_json/codegen_json.h ## @@ -0,0 +1,353 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

masahi commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447565441 ## File path: src/runtime/contrib/dnnl/dnnl_json_runtime.cc ## @@ -0,0 +1,455 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

masahi commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447564439 ## File path: src/relay/backend/contrib/codegen_json/codegen_json.h ## @@ -0,0 +1,353 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

masahi commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447564223 ## File path: src/relay/backend/contrib/codegen_json/codegen_json.h ## @@ -0,0 +1,353 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

masahi commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447564607 ## File path: src/relay/backend/contrib/codegen_json/codegen_json.h ## @@ -0,0 +1,353 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

masahi commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447564786 ## File path: src/relay/backend/contrib/dnnl/codegen.cc ## @@ -417,13 +424,87 @@ class DNNLModuleCodegen : public CSourceModuleCodegenBase {

[GitHub] [incubator-tvm] masahi commented on a change in pull request #5919: [BYOC] JSON Runtime with DNNL End-to-End Flow

masahi commented on a change in pull request #5919: URL: https://github.com/apache/incubator-tvm/pull/5919#discussion_r447563925 ## File path: src/relay/backend/contrib/codegen_json/codegen_json.h ## @@ -0,0 +1,353 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] [incubator-tvm] siju-samuel commented on a change in pull request #5961: [Relay/TOPI][OP] Add meshgrid op in Relay, TOPI, Pytorch frontend

siju-samuel commented on a change in pull request #5961: URL: https://github.com/apache/incubator-tvm/pull/5961#discussion_r447550719 ## File path: tests/python/frontend/pytorch/test_forward.py ## @@ -466,6 +466,17 @@ def forward(self, *args):

[GitHub] [incubator-tvm] lixiaoquan opened a new pull request #5963: Refine LSTMBlockCell to support dynamic rnn

lixiaoquan opened a new pull request #5963: URL: https://github.com/apache/incubator-tvm/pull/5963 1. Refine conversion of `LSTMBlockCell` 1) Make its output follows definition in TensorFlow 2) Avoid introducing variables which doesn't match any placeholder nodes in

[GitHub] [incubator-tvm] comaniac commented on pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

comaniac commented on pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962#issuecomment-651593570 @icemelon9 @yzhliu @zhiics @kevinthesun @anijain2305 it would be great if you guys could also help review the core data structure and flow to make sure everything is

[GitHub] [incubator-tvm] vinx13 merged pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

vinx13 merged pull request #5758: URL: https://github.com/apache/incubator-tvm/pull/5758 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

[incubator-tvm] branch master updated: Fix the meaning of conv{1, 2}d_transpose output_padding parameter. (#5758)

This is an automated email from the ASF dual-hosted git repository. wuwei pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-tvm.git The following commit(s) were added to refs/heads/master by this push: new bc22fb9 Fix the meaning of

[GitHub] [incubator-tvm] siju-samuel commented on pull request #5758: Fix the meaning of conv{1,2}d_transpose output_padding parameter.

siju-samuel commented on pull request #5758: URL: https://github.com/apache/incubator-tvm/pull/5758#issuecomment-651576650 @abergeron Thanks for bringing it up and the PR. I followed the existing conv2d_transpose implementation. You can add another PR to fix for fixing this issue.

[GitHub] [incubator-tvm] jcf94 opened a new pull request #5962: [Ansor][AutoTVM v2.0] Part 0: Ansor minimum system for auto schedule generating

jcf94 opened a new pull request #5962: URL: https://github.com/apache/incubator-tvm/pull/5962 Hi all, The last PR of Ansor #5883 is not clear enough for reviewers to fully understand our design. After some discussion, we changed our upstream plan to propose a minimal version of Ansor