You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2021/09/01 09:24:06 UTC
[GitHub] [tvm] luyaor opened a new issue #8890: [Bug] Error when compiling a ONNX model with Squeeze operator
luyaor opened a new issue #8890:
URL: https://github.com/apache/tvm/issues/8890
## Description
When compiling following model with TVM, it will error on parsing. I think the reason is similar to https://github.com/apache/tvm/issues/8889 . Because Squeeze operation in ONNX specification make some changes from previous version to current version (move split from attributes to inputs).
<img width="744" alt="image" src="https://user-images.githubusercontent.com/7541296/131554297-e22bff9f-9c3d-44da-9abd-edc3e0fe844f.png">
The model(with ONNX as frontend) with error is as follows, check bug.onnx in
[bug10.zip](https://github.com/apache/tvm/files/7085956/bug10.zip)
## Error Log
```
Traceback (most recent call last):
File "doublecheck.py", line 40, in <module>
mod, params = relay.frontend.from_onnx(onnx_model, shape_dict)
File "/Users/luyaor/Documents/tvm/python/tvm/relay/frontend/onnx.py", line 2806, in from_onnx
mod, params = g.from_onnx(graph, opset, freeze_params)
File "/Users/luyaor/Documents/tvm/python/tvm/relay/frontend/onnx.py", line 2613, in from_onnx
op = self._convert_operator(op_name, inputs, attr, opset)
File "/Users/luyaor/Documents/tvm/python/tvm/relay/frontend/onnx.py", line 2721, in _convert_operator
sym = convert_map[op_name](inputs, attrs, self._params)
File "/Users/luyaor/Documents/tvm/python/tvm/relay/frontend/common.py", line 430, in __call__
return get_relay_op(op_name)(*inputs, **new_attrs)
File "/Users/luyaor/Documents/tvm/python/tvm/relay/op/transform.py", line 153, in squeeze
return _make.squeeze(data, axis)
File "/Users/luyaor/Documents/tvm/python/tvm/_ffi/_ctypes/packed_func.py", line 237, in __call__
raise get_last_ffi_error()
tvm._ffi.base.TVMError: Traceback (most recent call last):
[bt] (6) 7 ??? 0x00007ffee09b9e70 0x0 + 140732666715760
[bt] (5) 6 libffi.7.dylib 0x000000010fd58ead ffi_call_unix64 + 85
[bt] (4) 5 libtvm.dylib 0x00000001373950b6 TVMFuncCall + 70
[bt] (3) 4 libtvm.dylib 0x0000000136f716d7 void tvm::runtime::TypedPackedFunc<tvm::RelayExpr (tvm::RelayExpr, tvm::runtime::Array<tvm::Integer, void>)>::AssignTypedLambda<tvm::RelayExpr (*)(tvm::RelayExpr, tvm::runtime::Array<tvm::Integer, void>)>(tvm::RelayExpr (*)(tvm::RelayExpr, tvm::runtime::Array<tvm::Integer, void>))::'lambda'(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)::operator()(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*) const + 487
[bt] (2) 3 libtvm.dylib 0x00000001366ad36e tvm::runtime::TVMMovableArgValue_::operator tvm::runtime::Array<tvm::Integer, void><tvm::runtime::Array<tvm::Integer, void>, void>() const + 158
[bt] (1) 2 libtvm.dylib 0x00000001366ad8c6 tvm::runtime::Array<tvm::Integer, void> tvm::runtime::TVMPODValue_::AsObjectRef<tvm::runtime::Array<tvm::Integer, void> >() const + 1222
[bt] (0) 1 libtvm.dylib 0x0000000136525c6f dmlc::LogMessageFatal::~LogMessageFatal() + 111
File "/Users/luyaor/Documents/tvm/include/tvm/runtime/packed_func.h", line 1405
TVMError:
---------------------------------------------------------------
An internal invariant was violated during the execution of TVM.
Please read TVM's error reporting guidelines.
More details can be found here: https://discuss.tvm.ai/t/error-reporting/7793.
---------------------------------------------------------------
Check failed: ObjectTypeChecker<TObjectRef>::Check(ptr) == false: Expect Array[IntImm] but get relay.Var
```
## How to reproduce
```
python check.py
```
### Environment
Python3, with tvm, onnx
tvm version: [`c31e338`](https://github.com/apache/tvm/commit/c31e338d5f98a8e8c97286c5b93b20caee8be602) Wed Dec 9 14:52:58 2020 +0900
1. Download [bug10.zip](https://github.com/apache/tvm/files/7085956/bug10.zip)
2. Run `python check.py`.
Hi [@jwfromm](https://github.com/jwfromm) , please check it out! 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.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] AndrewZhaoLuo commented on issue #8890: [Bug] Error when compiling a ONNX model with Squeeze operator
Posted by GitBox <gi...@apache.org>.
AndrewZhaoLuo commented on issue #8890:
URL: https://github.com/apache/tvm/issues/8890#issuecomment-911967024
This is due to the fact that the onnx squeeze is not dynamic.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org