[GitHub] [tvm] vicalloy commented on issue #7896: `relay.vm.compile()` get a error `Cannot store int32 to buffer of int64`

vicalloy commented on issue #7896: URL: https://github.com/apache/tvm/issues/7896#issuecomment-824561791 thanks, the output is correct. -- 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] vicalloy commented on issue #7908: `vm.run(data).asnumpy()` and get `TypeError: runtime.ADT is not registered via TVM_REGISTER_NODE_TYPE`

vicalloy commented on issue #7908: URL: https://github.com/apache/tvm/issues/7908#issuecomment-824557421 thanks -- 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

[GitHub] [tvm] vicalloy closed issue #7908: `vm.run(data).asnumpy()` and get `TypeError: runtime.ADT is not registered via TVM_REGISTER_NODE_TYPE`

vicalloy closed issue #7908: URL: https://github.com/apache/tvm/issues/7908 -- 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] junrushao1994 commented on pull request #7809: [Target][Lowering] Update Op Intrinsic Lowering Mechanism And Intrinsic Lowering Pass

junrushao1994 commented on pull request #7809: URL: https://github.com/apache/tvm/pull/7809#issuecomment-824532034 @tqchen would you like to take a second look? If there is other comments, I will get it merged the other day -- This is an automated message from the Apache Git Service. To

[GitHub] [tvm] masahi commented on issue #7896: `relay.vm.compile()` get a error `Cannot store int32 to buffer of int64`

masahi commented on issue #7896: URL: https://github.com/apache/tvm/issues/7896#issuecomment-824528048 See my change above. You need to wrap `x_shape` and `y_shape` too. ``` cond_shape = maybe_expand(inputs[0]) x_shape = maybe_expand(inputs[1]) y_shape =

[GitHub] [tvm] vicalloy commented on issue #7896: `relay.vm.compile()` get a error `Cannot store int32 to buffer of int64`

vicalloy commented on issue #7896: URL: https://github.com/apache/tvm/issues/7896#issuecomment-824524312 1. I got another error: ``` [21:09:04] /home/hu/tvm/src/relay/transforms/let_list.h:54: Warning: letlist not used Traceback (most recent call last): File "run-tvm.py",

[GitHub] [tvm] ChenNie99 commented on issue #7857: pynq z1 fails to make the VTA, caused by the libbacktrace

ChenNie99 commented on issue #7857: URL: https://github.com/apache/tvm/issues/7857#issuecomment-824520834 Yeah, I have already deployed the VTA on my pynq-z1. As for the problem, I'm not sure but you can try this command: chmod +x /path/to/tvm/3rdparty/libbacktrace/configure

[GitHub] [tvm] masahi commented on pull request #7867: [Runtime] Driver version + consistent clock speed units

masahi commented on pull request #7867: URL: https://github.com/apache/tvm/pull/7867#issuecomment-824518257 thanks @Lunderberg @areusch @tmoreau89 -- 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

[tvm] branch main updated (1c71a06 -> 46e0634)

This is an automated email from the ASF dual-hosted git repository. masahi pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 1c71a06 [ONNX][TOPI][RELAY] Resize refactor (#7883) add 46e0634 [Runtime] Driver version + consistent clock speed

[GitHub] [tvm] masahi merged pull request #7867: [Runtime] Driver version + consistent clock speed units

masahi merged pull request #7867: URL: https://github.com/apache/tvm/pull/7867 -- 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] masahi commented on issue #7908: `vm.run(data).asnumpy()` and get `TypeError: runtime.ADT is not registered via TVM_REGISTER_NODE_TYPE`

masahi commented on issue #7908: URL: https://github.com/apache/tvm/issues/7908#issuecomment-824517458 See https://github.com/apache/tvm/blob/f4490708a559930c6554eec26c377d49bfb46d84/tests/python/frontend/pytorch/test_lstm.py#L202-L211 and its use in

[GitHub] [tvm] masahi commented on issue #7896: `relay.vm.compile()` get a error `Cannot store int32 to buffer of int64`

masahi commented on issue #7896: URL: https://github.com/apache/tvm/issues/7896#issuecomment-824516302 ok can you try replacing https://github.com/apache/tvm/blob/8e23806d2d522b71979d0a2730b38cc5c3bf6185/python/tvm/relay/op/_transform.py#L1017-L1023 with ```

[GitHub] [tvm] mehrdadh commented on pull request #7838: [RPC] microtvm: fix RPC large transfer size issue

mehrdadh commented on pull request #7838: URL: https://github.com/apache/tvm/pull/7838#issuecomment-824515577 @areusch PTAL. Thanks! -- 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] vicalloy opened a new issue #7908: `vm.run(data).asnumpy()` and get `TypeError: runtime.ADT is not registered via TVM_REGISTER_NODE_TYPE`

vicalloy opened a new issue #7908: URL: https://github.com/apache/tvm/issues/7908 1. The follow code is used to run the model by vm. ```python import onnx import time import tvm import numpy as np import tvm.relay as relay target = 'llvm' # model from

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

shihangfu commented on issue #7857: URL: https://github.com/apache/tvm/issues/7857#issuecomment-824510397 Have you solved this problem？ I have a similar question. [ 0%] Creating directories for 'project_libbacktrace' [ 0%] No download step for 'project_libbacktrace' [

[GitHub] [tvm] leoluopy edited a comment on pull request #7839: [Fix] deployment code fails with USE_FALLBACK_STL_MAP=0

leoluopy edited a comment on pull request #7839: URL: https://github.com/apache/tvm/pull/7839#issuecomment-824507672 > Would you like to add the line below in tvm_runtime_pack.cc? > > ```c++ > #include "../../src/runtime/container.cc" > #include "../../src/runtime/logging.cc"

[GitHub] [tvm] leoluopy commented on pull request #7839: [Fix] deployment code fails with USE_FALLBACK_STL_MAP=0

leoluopy commented on pull request #7839: URL: https://github.com/apache/tvm/pull/7839#issuecomment-824507672 > Would you like to add the line below in tvm_runtime_pack.cc? > > ```c++ > #include "../../src/runtime/container.cc" > #include "../../src/runtime/logging.cc" > ```

[GitHub] [tvm] masahi commented on issue #7896: `relay.vm.compile()` get a error `Cannot store int32 to buffer of int64`

masahi commented on issue #7896: URL: https://github.com/apache/tvm/issues/7896#issuecomment-824502231 This error is fixed by replacing `3` below with `int64(3)` https://github.com/apache/tvm/blob/390b4d1ce32533810fc8d65dcf211f6d147764aa/python/tvm/relay/op/vision/_vision.py#L97 But

[GitHub] [tvm] masahi commented on issue #7896: `relay.vm.compile()` get a error `Cannot store int32 to buffer of int64`

masahi commented on issue #7896: URL: https://github.com/apache/tvm/issues/7896#issuecomment-824497504 ok thanks reproduced. At first sight it is not clear what operator is causing the issue. -- This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] [tvm] vicalloy commented on issue #7896: `relay.vm.compile()` get a error `Cannot store int32 to buffer of int64`

vicalloy commented on issue #7896: URL: https://github.com/apache/tvm/issues/7896#issuecomment-824494392 The model: [face-detection-0200.zip](https://github.com/apache/tvm/files/6355080/face-detection-0200.zip) -- This is an automated message from the Apache Git Service. To respond

[GitHub] [tvm] ghostplant opened a new pull request #7907: Remove unnecessary bracelet around make_int

ghostplant opened a new pull request #7907: URL: https://github.com/apache/tvm/pull/7907 Turning from `((make_int4)(exp))` to `(make_int4(exp))`. The former case is incompatible with "macro-defined function". -- This is an automated message from the Apache Git Service. To respond to the

[GitHub] [tvm] jwfromm opened a new pull request #7906: [Relay][ONNX] 1-D global and adaptive pooling.

jwfromm opened a new pull request #7906: URL: https://github.com/apache/tvm/pull/7906 This PR adds support for 1D global and adaptive pooling operators in relay, specifically `adaptive_max_pool1d`, `adaptive_avg_pool1d`, `global_max_pool1d`, and `global_avg_pool1d`. It also adds

[GitHub] [tvm] masahi commented on issue #7896: `relay.vm.compile()` get a error `Cannot store int32 to buffer of int64`

masahi commented on issue #7896: URL: https://github.com/apache/tvm/issues/7896#issuecomment-824466778 Can you attach the onnx model or tell me how to download it? -- 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] srinidhigoud commented on a change in pull request #7901: [Frontend][Tensorflow] SelectV2 and BroadcastArgs op support for tf2 models

srinidhigoud commented on a change in pull request #7901: URL: https://github.com/apache/tvm/pull/7901#discussion_r617979127 ## File path: python/tvm/relay/frontend/tensorflow.py ## @@ -1765,6 +1765,44 @@ def _impl(inputs, attr, params, mod): return _impl +def

[GitHub] [tvm] rohanmukh commented on a change in pull request #7901: [Frontend][Tensorflow] SelectV2 and BroadcastArgs op support for tf2 models

rohanmukh commented on a change in pull request #7901: URL: https://github.com/apache/tvm/pull/7901#discussion_r617970388 ## File path: python/tvm/relay/frontend/tensorflow.py ## @@ -1765,6 +1765,44 @@ def _impl(inputs, attr, params, mod): return _impl +def

[GitHub] [tvm] rohanmukh commented on a change in pull request #7901: [Frontend][Tensorflow] SelectV2 and BroadcastArgs op support for tf2 models

rohanmukh commented on a change in pull request #7901: URL: https://github.com/apache/tvm/pull/7901#discussion_r617970021 ## File path: python/tvm/relay/frontend/tensorflow.py ## @@ -1765,6 +1765,44 @@ def _impl(inputs, attr, params, mod): return _impl +def

[GitHub] [tvm] zxybazh opened a new pull request #7905: [WIP][TIR][OP] Add New Tir Ops

zxybazh opened a new pull request #7905: URL: https://github.com/apache/tvm/pull/7905 This PR intends to introduce a few new target dependent intrinsic lowering function of tir level ops for better support. The ops are selected from numpy ops and supported in `cmath`. -- This is an

[GitHub] [tvm] mdw-octoml commented on pull request #7869: Update ICHECK error message with link to documentation page.

mdw-octoml commented on pull request #7869: URL: https://github.com/apache/tvm/pull/7869#issuecomment-824425907 I think I have resolved the merge issue, this should be good to land once CI passes. -- This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] srinidhigoud edited a comment on pull request #7901: [Frontend][Tensorflow] SelectV2 and BroadcastArgs op support for tf2 models

srinidhigoud edited a comment on pull request #7901: URL: https://github.com/apache/tvm/pull/7901#issuecomment-824373069 @zhiics@yongwww @comaniac @rohanmukh -- 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] anwang2009 opened a new issue #7904: feature request: RelayVM get_input

anwang2009 opened a new issue #7904: URL: https://github.com/apache/tvm/issues/7904 Graph executor has the `get_input` function defined. Could we add something similar for Relay VM? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to

[GitHub] [tvm] srinidhigoud commented on pull request #7901: [Frontend][Tensorflow] SelectV2 and BroadcastArgs op support for tf2 models

srinidhigoud commented on pull request #7901: URL: https://github.com/apache/tvm/pull/7901#issuecomment-824373069 @zhiics@yongwww @comaniac -- 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 main updated (f28c75f -> 1c71a06)

This is an automated email from the ASF dual-hosted git repository. jwfromm pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from f28c75f [Frontend][Tensorflow] Support SAME padding for dynamic h, w when stride == 1 (#7885) add 1c71a06

[GitHub] [tvm] jwfromm merged pull request #7883: [ONNX][TOPI][RELAY] Resize refactor

jwfromm merged pull request #7883: URL: https://github.com/apache/tvm/pull/7883 -- 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] jwfromm commented on pull request #7883: [ONNX][TOPI][RELAY] Resize refactor

jwfromm commented on pull request #7883: URL: https://github.com/apache/tvm/pull/7883#issuecomment-824372527 Thanks @mbrookhart, @masahi. This is now 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] jwfromm commented on a change in pull request #7883: [ONNX][TOPI][RELAY] Resize refactor

jwfromm commented on a change in pull request #7883: URL: https://github.com/apache/tvm/pull/7883#discussion_r617891826 ## File path: python/tvm/relay/op/image/image.py ## @@ -58,6 +61,16 @@ def resize( Refer to the ONNX Resize operator specification for details.

[GitHub] [tvm] mehrdadh opened a new pull request #7903: [microTVM] Graph Executor Debugger: fix parameter dump and introduce two debugging mode

mehrdadh opened a new pull request #7903: URL: https://github.com/apache/tvm/pull/7903 This PR: - changes graph executor debugger to have two modes {timing, accuracy}. These two modes are different in a sense that in timing we measure the performance of each layer, however in accuracy

[GitHub] [tvm] tkonolige opened a new pull request #7902: [FIX,VM] Fix get_outputs on the vm with a single output

tkonolige opened a new pull request #7902: URL: https://github.com/apache/tvm/pull/7902 The VM uses an ADT for multiple outputs and an NDArray for a single output. The single output case was not being handled. @tmoreau89 -- This is an automated message from the Apache Git

[GitHub] [tvm] srinidhigoud opened a new pull request #7901: [Frontend][Tensorflow] SelectV2 and BroadcastArgs op support for tf2 models

srinidhigoud opened a new pull request #7901: URL: https://github.com/apache/tvm/pull/7901 Adding support in tf parser for new ops SelectV2 and BroadcastArgs introduced by tf2. This is focused on compiling and running the tf2 version of the object detection models such as faster rcnn

[GitHub] [tvm] Lunderberg commented on pull request #7890: Remove pointer arithmetic in StorageObj::AllocNDArray

Lunderberg commented on pull request #7890: URL: https://github.com/apache/tvm/pull/7890#issuecomment-824286106 Thank you, updated the commit timestamp to restart the CI. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and

[GitHub] [tvm] AD1024 commented on pull request #7795: [FIX] skip_conv_layers will affect quantization of nn.dense

AD1024 commented on pull request #7795: URL: https://github.com/apache/tvm/pull/7795#issuecomment-824279796 @vinx13 @ZihengJiang Test case added. Please review. Thank you. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub

[tvm] branch main updated (935a7ad -> f28c75f)

This is an automated email from the ASF dual-hosted git repository. kevinthesun pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 935a7ad Add support for the quantized RESIZE_BILINEAR operator to relay TFLite frontend (#7866) add f28c75f

[GitHub] [tvm] junrushao1994 commented on pull request #7809: [Target][Lowering] Update Op Intrinsic Lowering Mechanism And Intrinsic Lowering Pass

junrushao1994 commented on pull request #7809: URL: https://github.com/apache/tvm/pull/7809#issuecomment-824270527 CC: @comaniac @yzhliu @icemelon9 -- 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] kevinthesun merged pull request #7885: [Frontend][Tensorflow] Support SAME padding for dynamic h, w when stride == 1

kevinthesun merged pull request #7885: URL: https://github.com/apache/tvm/pull/7885 -- 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,

[GitHub] [tvm] kevinthesun commented on pull request #7885: [Frontend][Tensorflow] Support SAME padding for dynamic h, w when stride == 1

kevinthesun commented on pull request #7885: URL: https://github.com/apache/tvm/pull/7885#issuecomment-824270592 Thanks @trevor-m -- 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] AndrewZhaoLuo commented on pull request #7893: Update Dev. Doc. on how to add a new relay operator

AndrewZhaoLuo commented on pull request #7893: URL: https://github.com/apache/tvm/pull/7893#issuecomment-824258396 @masahi Thanks for the comments. 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

[GitHub] [tvm] AndrewZhaoLuo commented on a change in pull request #7893: Update Dev. Doc. on how to add a new relay operator

AndrewZhaoLuo commented on a change in pull request #7893: URL: https://github.com/apache/tvm/pull/7893#discussion_r617772065 ## File path: docs/dev/relay_add_op.rst ## @@ -15,75 +15,297 @@ specific language governing permissions and limitations under the License.

[GitHub] [tvm] giuseros commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

giuseros commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r617752937 ## File path: src/relay/backend/build_module.cc ## @@ -473,23 +517,25 @@ class RelayBuildModule : public runtime::ModuleNode { // Relay IRModule ->

[GitHub] [tvm] mbrookhart opened a new pull request #7900: [ONNX] Support NMS Center Box

mbrookhart opened a new pull request #7900: URL: https://github.com/apache/tvm/pull/7900 Thanks for contributing to TVM! Please refer to guideline https://tvm.apache.org/docs/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews

[GitHub] [tvm] giuseros commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

giuseros commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r617747883 ## File path: src/relay/backend/build_module.cc ## @@ -473,23 +517,25 @@ class RelayBuildModule : public runtime::ModuleNode { // Relay IRModule ->

[GitHub] [tvm] mbrookhart opened a new pull request #7899: [ONNX] Support importing Conv with missing attributes

mbrookhart opened a new pull request #7899: URL: https://github.com/apache/tvm/pull/7899 -- 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,

[GitHub] [tvm] comaniac commented on a change in pull request #7807: [Relay][Pass] Update SimplifyTranspose to correctly simplify rank changing layout transforms

comaniac commented on a change in pull request #7807: URL: https://github.com/apache/tvm/pull/7807#discussion_r617736990 ## File path: tests/python/relay/test_pass_simplify_expr.py ## @@ -106,10 +106,112 @@ def expected3(): y = relay.transpose(y, axes=[0, 2, 3, 1])

[GitHub] [tvm] domin1985 commented on pull request #7775: [Relay][Parser] support dict attributes parsing

domin1985 commented on pull request #7775: URL: https://github.com/apache/tvm/pull/7775#issuecomment-824210250 > cc @domin1985 > I think this is an okay work around for now, ideally we could make the printing a little cleaner but I think the challenge is detecting between

[GitHub] [tvm] areusch commented on a change in pull request #7838: [RPC] microtvm: fix RPC large transfer size issue

areusch commented on a change in pull request #7838: URL: https://github.com/apache/tvm/pull/7838#discussion_r617704348 ## File path: src/runtime/rpc/rpc_endpoint.cc ## @@ -980,7 +980,9 @@ class RPCClientSession : public RPCSession, public DeviceAPI { void

[GitHub] [tvm] Lunderberg commented on pull request #7867: [Runtime] Driver version + consistent clock speed units

Lunderberg commented on pull request #7867: URL: https://github.com/apache/tvm/pull/7867#issuecomment-824194789 @areusch Good point, and that is useful feedback. Changes made in each docstring to indicate that they can return `None`. The one exception is `.exist`, which returns a

[GitHub] [tvm] ANSHUMAN87 commented on a change in pull request #7807: [Relay][Pass] Update SimplifyTranspose to correctly simplify rank changing layout transforms

ANSHUMAN87 commented on a change in pull request #7807: URL: https://github.com/apache/tvm/pull/7807#discussion_r617700835 ## File path: src/relay/transforms/simplify_expr.cc ## @@ -163,6 +141,69 @@ class SimplifyTranspose : public DFPatternRewrite { return x; } +

[GitHub] [tvm] tristan-arm commented on a change in pull request #7858: [ETHOSN] Removed support for 20.08 version of the driver stack.

tristan-arm commented on a change in pull request #7858: URL: https://github.com/apache/tvm/pull/7858#discussion_r617699191 ## File path: src/relay/backend/contrib/ethosn/codegen_ethosn.h ## @@ -240,18 +241,14 @@ struct EthosnCompilerConfigNode : public tvm::AttrsNode

[GitHub] [tvm] ANSHUMAN87 commented on a change in pull request #7807: [Relay][Pass] Update SimplifyTranspose to correctly simplify rank changing layout transforms

ANSHUMAN87 commented on a change in pull request #7807: URL: https://github.com/apache/tvm/pull/7807#discussion_r617697768 ## File path: tests/python/relay/test_pass_simplify_expr.py ## @@ -106,10 +106,28 @@ def expected3(): y = relay.transpose(y, axes=[0, 2, 3, 1])

[GitHub] [tvm] ANSHUMAN87 commented on a change in pull request #7807: [Relay][Pass] Update SimplifyTranspose to correctly simplify rank changing layout transforms

ANSHUMAN87 commented on a change in pull request #7807: URL: https://github.com/apache/tvm/pull/7807#discussion_r617695429 ## File path: tests/python/relay/test_pass_simplify_expr.py ## @@ -106,10 +106,112 @@ def expected3(): y = relay.transpose(y, axes=[0, 2, 3, 1])

[GitHub] [tvm] ANSHUMAN87 commented on a change in pull request #7807: [Relay][Pass] Update SimplifyTranspose to correctly simplify rank changing layout transforms

ANSHUMAN87 commented on a change in pull request #7807: URL: https://github.com/apache/tvm/pull/7807#discussion_r617695429 ## File path: tests/python/relay/test_pass_simplify_expr.py ## @@ -106,10 +106,112 @@ def expected3(): y = relay.transpose(y, axes=[0, 2, 3, 1])

[GitHub] [tvm] ANSHUMAN87 commented on a change in pull request #7807: [Relay][Pass] Update SimplifyTranspose to correctly simplify rank changing layout transforms

ANSHUMAN87 commented on a change in pull request #7807: URL: https://github.com/apache/tvm/pull/7807#discussion_r617694296 ## File path: tests/python/relay/test_pass_simplify_expr.py ## @@ -106,10 +106,112 @@ def expected3(): y = relay.transpose(y, axes=[0, 2, 3, 1])

[GitHub] [tvm] mbaret commented on a change in pull request #7858: [ETHOSN] Removed support for 20.08 version of the driver stack.

mbaret commented on a change in pull request #7858: URL: https://github.com/apache/tvm/pull/7858#discussion_r617692088 ## File path: src/relay/backend/contrib/ethosn/codegen_ethosn.h ## @@ -240,18 +241,14 @@ struct EthosnCompilerConfigNode : public tvm::AttrsNode

[GitHub] [tvm] ANSHUMAN87 commented on a change in pull request #7807: [Relay][Pass] Update SimplifyTranspose to correctly simplify rank changing layout transforms

ANSHUMAN87 commented on a change in pull request #7807: URL: https://github.com/apache/tvm/pull/7807#discussion_r617687443 ## File path: src/relay/transforms/simplify_expr.cc ## @@ -163,6 +141,69 @@ class SimplifyTranspose : public DFPatternRewrite { return x; } +

[GitHub] [tvm] manupa-arm commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

manupa-arm commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r617672408 ## File path: src/relay/backend/build_module.cc ## @@ -473,23 +517,25 @@ class RelayBuildModule : public runtime::ModuleNode { // Relay IRModule ->

[GitHub] [tvm] areusch commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

areusch commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r617667974 ## File path: src/relay/backend/build_module.cc ## @@ -473,23 +517,25 @@ class RelayBuildModule : public runtime::ModuleNode { // Relay IRModule ->

[GitHub] [tvm] mbaret commented on pull request #7866: Quantized RESIZE_BILINEAR operator support in TF Lite Frontend

mbaret commented on pull request #7866: URL: https://github.com/apache/tvm/pull/7866#issuecomment-824162064 This is merged, thanks @NicolaLancellotti ! -- 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 (8bd857d -> 935a7ad)

This is an automated email from the ASF dual-hosted git repository. mbaret pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from 8bd857d [RPC][REFACTOR] Use PopenWorker to handle RPC Server. (#7889) add 935a7ad Add support for the quantized

[GitHub] [tvm] mbaret merged pull request #7866: Quantized RESIZE_BILINEAR operator support in TF Lite Frontend

mbaret merged pull request #7866: URL: https://github.com/apache/tvm/pull/7866 -- 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] stoa commented on pull request #7742: Contributing the STM32 port

stoa commented on pull request #7742: URL: https://github.com/apache/tvm/pull/7742#issuecomment-824143000 @areusch I have moved all ai_runner code from tvm/contrib/stm32 to the project,application directory apps/stm32. Only emitter.py remains in the contrib/stm32 directory. Arthur

[GitHub] [tvm] manupa-arm commented on a change in pull request #7785: [AOT] Introducing AOT in TVM

manupa-arm commented on a change in pull request #7785: URL: https://github.com/apache/tvm/pull/7785#discussion_r617637024 ## File path: src/relay/backend/build_module.cc ## @@ -473,23 +517,25 @@ class RelayBuildModule : public runtime::ModuleNode { // Relay IRModule ->

[GitHub] [tvm] mbrookhart commented on pull request #7883: [ONNX][TOPI][RELAY] Resize refactor

mbrookhart commented on pull request #7883: URL: https://github.com/apache/tvm/pull/7883#issuecomment-824136011 I changed the order of operations on align_corner from out = a / b * x to out = x * a / b. That seems to have caused some rounding errors on GPU, if I revert the change the

[GitHub] [tvm] mbrookhart commented on pull request #7883: [ONNX][TOPI][RELAY] Resize refactor

mbrookhart commented on pull request #7883: URL: https://github.com/apache/tvm/pull/7883#issuecomment-824129779 I'm hitting a small and intermittent atol issue with the TF tests on GPU. Before bumping it from 1e-5 to 2e-5, I'm trying to figure out if I actually changed the behavior for

[GitHub] [tvm] tqchen commented on issue #7870: [TEST][FLAKY] test_vm_rpc

tqchen commented on issue #7870: URL: https://github.com/apache/tvm/issues/7870#issuecomment-824098515 should be fixed by https://github.com/apache/tvm/pull/7889 -- 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] tqchen closed issue #7870: [TEST][FLAKY] test_vm_rpc

tqchen closed issue #7870: URL: https://github.com/apache/tvm/issues/7870 -- 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] leeexyz opened a new pull request #7898: [TE] Fix bug if find a loop in compute_at attach path

leeexyz opened a new pull request #7898: URL: https://github.com/apache/tvm/pull/7898 Fix the bug reported at https://discuss.tvm.apache.org/t/incorrect-compute-at-schedule-makes-tvm-stuck/9780/2 -- This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] K1504296 commented on issue #7470: [TEST][FLAKY] tests/python/relay/test_op_level5.py::test_crop_and_resize

K1504296 commented on issue #7470: URL: https://github.com/apache/tvm/issues/7470#issuecomment-824036721 Hello, When running with img shape 224 x 224 on aarch64 this error occurs. This doesn't occur with other image sizes (244, 488 255, 128, 488 are all okay.) and it doesn't matter if

[GitHub] [tvm] PENGUINLIONG opened a new pull request #7897: Enable StackVM in AutoTVM

PENGUINLIONG opened a new pull request #7897: URL: https://github.com/apache/tvm/pull/7897 It turns out in my edge use case StackVM is rather useful because it's minimum. I think it can help if tuning via StackVM can be officially supported. -- This is an automated message from the

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

vinceab commented on a change in pull request #7537: URL: https://github.com/apache/tvm/pull/7537#discussion_r617479382 ## 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] masahi commented on pull request #7893: Update Dev. Doc. on how to add a new relay operator

masahi commented on pull request #7893: URL: https://github.com/apache/tvm/pull/7893#issuecomment-823977647 @hogepodge -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment.

[GitHub] [tvm] masahi commented on a change in pull request #7893: Update Dev. Doc. on how to add a new relay operator

masahi commented on a change in pull request #7893: URL: https://github.com/apache/tvm/pull/7893#discussion_r617435627 ## File path: docs/dev/relay_add_op.rst ## @@ -15,75 +15,297 @@ specific language governing permissions and limitations under the License. -..

[GitHub] [tvm] masahi commented on a change in pull request #7893: Update Dev. Doc. on how to add a new relay operator

masahi commented on a change in pull request #7893: URL: https://github.com/apache/tvm/pull/7893#discussion_r617435051 ## File path: docs/dev/relay_add_op.rst ## @@ -15,75 +15,297 @@ specific language governing permissions and limitations under the License. -..

[GitHub] [tvm] masahi commented on a change in pull request #7893: Update Dev. Doc. on how to add a new relay operator

masahi commented on a change in pull request #7893: URL: https://github.com/apache/tvm/pull/7893#discussion_r617434021 ## File path: docs/dev/relay_add_op.rst ## @@ -15,75 +15,297 @@ specific language governing permissions and limitations under the License. -..

[GitHub] [tvm] masahi commented on a change in pull request #7893: Update Dev. Doc. on how to add a new relay operator

masahi commented on a change in pull request #7893: URL: https://github.com/apache/tvm/pull/7893#discussion_r617433178 ## File path: docs/dev/relay_add_op.rst ## @@ -15,75 +15,297 @@ specific language governing permissions and limitations under the License. -..

[GitHub] [tvm] llehtahw closed pull request #7879: [Tensorize]Fix tensorize error while reusing compute

llehtahw closed pull request #7879: URL: https://github.com/apache/tvm/pull/7879 -- 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] llehtahw commented on pull request #7879: [Tensorize]Fix tensorize error while reusing compute

llehtahw commented on pull request #7879: URL: https://github.com/apache/tvm/pull/7879#issuecomment-823957245 > why not use call_extern, etc. to achieve the same goal? In fact, my example was just drafted for showing the `Check failed`, so is't not important whether to use

[GitHub] [tvm] u99127 commented on pull request #7858: [ETHOSN] Removed support for 20.08 version of the driver stack.

u99127 commented on pull request #7858: URL: https://github.com/apache/tvm/pull/7858#issuecomment-823909705 @mbaret -- 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

[GitHub] [tvm] tristan-arm commented on pull request #7858: [ETHOSN] Removed support for 20.08 version of the driver stack.

tristan-arm commented on pull request #7858: URL: https://github.com/apache/tvm/pull/7858#issuecomment-823908805 cc @leandron @Leo-arm @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

[GitHub] [tvm] leeexyz commented on pull request #7879: [Tensorize]Fix tensorize error while reusing compute

leeexyz commented on pull request #7879: URL: https://github.com/apache/tvm/pull/7879#issuecomment-823822447 What I need to explain more is in our case, we use compute to define the HW intrinsic, it is may as the same as Operators, such as element-wise. But it is not the same compute (I

[GitHub] [tvm] leeexyz commented on pull request #7879: [Tensorize]Fix tensorize error while reusing compute

leeexyz commented on pull request #7879: URL: https://github.com/apache/tvm/pull/7879#issuecomment-823814290 @llehtahw Thanks for pointing this out. :) One more thing, why not use call_extern, etc. to achieve the same goal? -- This is an automated message from the Apache Git Service. To

[GitHub] [tvm] leeexyz removed a comment on pull request #7879: [Tensorize]Fix tensorize error while reusing compute

leeexyz removed a comment on pull request #7879: URL: https://github.com/apache/tvm/pull/7879#issuecomment-823809171 @llehtahw Thanks for pointing this out. :) Let me update it and add a new case. -- This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [tvm] leeexyz commented on pull request #7879: [Tensorize]Fix tensorize error while reusing compute

leeexyz commented on pull request #7879: URL: https://github.com/apache/tvm/pull/7879#issuecomment-823809171 @llehtahw Thanks for pointing this out. :) Let me update it and add a new case. -- This is an automated message from the Apache Git Service. To respond to the message, please