mshr-h opened a new issue, #16126:
URL: https://github.com/apache/tvm/issues/16126
Seems like the aten::new_zeros support is broken. #14747
### Steps to reproduce
```python
from tvm import relay
import torch
torch.set_grad_enabled(False)
class NewZeros1(torch.nn.Module):
def forward(self, x):
return x.new_zeros((2, 3))
input_data = torch.tensor((), dtype=torch.float)
module = NewZeros1().float().eval()
scripted_module = torch.jit.trace(module, input_data)
input_infos = [('x', (input_data.shape, 'float32'))]
mod, params = relay.frontend.from_pytorch(scripted_module, input_infos)
```
### Actual behavior
```
Traceback (most recent call last):
File "/home/ubuntu/workspace/sandbox/tvm_/frontend/new_zeros.py", line 17,
in <module>
mod, params = relay.frontend.from_pytorch(scripted_module, input_infos)
File
"/home/ubuntu/workspace/sandbox/.dep/tvm/python/tvm/relay/frontend/pytorch.py",
line 5183, in from_pytorch
outputs = converter.convert_operators(operator_nodes, outputs, ret_name)
File
"/home/ubuntu/workspace/sandbox/.dep/tvm/python/tvm/relay/frontend/pytorch.py",
line 4402, in convert_operators
relay_out = relay_op(
File
"/home/ubuntu/workspace/sandbox/.dep/tvm/python/tvm/relay/frontend/pytorch.py",
line 851, in new_zeros
return self.full_impl(data, 0, dtype)
File
"/home/ubuntu/workspace/sandbox/.dep/tvm/python/tvm/relay/frontend/pytorch.py",
line 755, in full_impl
out = _op.full(fill_value, size, dtype=dtype)
File
"/home/ubuntu/workspace/sandbox/.dep/tvm/python/tvm/relay/op/transform.py",
line 535, in full
return _make.full(fill_value, shape, dtype)
File
"/home/ubuntu/workspace/sandbox/.dep/tvm/python/tvm/_ffi/_ctypes/packed_func.py",
line 239, in __call__
raise_last_ffi_error()
File "/home/ubuntu/workspace/sandbox/.dep/tvm/python/tvm/_ffi/base.py",
line 481, in raise_last_ffi_error
raise py_err
tvm._ffi.base.TVMError: Traceback (most recent call last):
2: _ZN3tvm7runtime13PackedFuncObj
1: tvm::runtime::TypedPackedFunc<tvm::RelayExpr (tvm::RelayExpr,
tvm::runtime::Array<tvm::Integer, void>,
tvm::runtime::DataType)>::AssignTypedLambda<tvm::RelayExpr (*)(tvm::RelayExpr,
tvm::runtime::Array<tvm::Integer, void>,
tvm::runtime::DataType)>(tvm::RelayExpr (*)(tvm::RelayExpr,
tvm::runtime::Array<tvm::Integer, void>, tvm::runtime::DataType),
std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>
>)::{lambda(tvm::runtime::TVMArgs const&,
tvm::runtime::TVMRetValue*)#1}::operator()(tvm::runtime::TVMArgs const&,
tvm::runtime::TVMRetValue*) const
0: tvm::runtime::TVMMovableArgValueWithContext_::operator
tvm::runtime::DataType<tvm::runtime::DataType>() const
3: _ZN3tvm7runtime13PackedFuncObj
2: tvm::runtime::TypedPackedFunc<tvm::RelayExpr (tvm::RelayExpr,
tvm::runtime::Array<tvm::Integer, void>,
tvm::runtime::DataType)>::AssignTypedLambda<tvm::RelayExpr (*)(tvm::RelayExpr,
tvm::runtime::Array<tvm::Integer, void>,
tvm::runtime::DataType)>(tvm::RelayExpr (*)(tvm::RelayExpr,
tvm::runtime::Array<tvm::Integer, void>, tvm::runtime::DataType),
std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>
>)::{lambda(tvm::runtime::TVMArgs const&,
tvm::runtime::TVMRetValue*)#1}::operator()(tvm::runtime::TVMArgs const&,
tvm::runtime::TVMRetValue*) const
1: tvm::runtime::TVMMovableArgValueWithContext_::operator
tvm::runtime::DataType<tvm::runtime::DataType>() const
0: tvm::runtime::TVMArgValue::operator DLDataType() const
File
"/home/ubuntu/workspace/sandbox/.dep/tvm/include/tvm/runtime/packed_func.h",
line 777
TVMError: In function relay.op._make.full(0: RelayExpr, 1: Array<IntImm>, 2:
DataType) -> RelayExpr: error while converting argument 2: [19:47:16]
/home/ubuntu/workspace/sandbox/.dep/tvm/include/tvm/runtime/packed_func.h:2210:
InternalError: Check failed: type_code_ == kTVMDataType (8 vs. 5) : expected
DLDataType but got Object
```
### Environment
TVM: c8ef902a752e71dc92fb7fdfd7c3fcc4178a3455
LLVM=ON
CUDA=OFF
### Triage
Please refer to the list of label tags
[here](https://github.com/apache/tvm/wiki/Issue-Triage-Labels) to find the
relevant tags and add them below in a bullet format (example below).
* frontend:pytorch
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
