【Mindspore内部，紧急】Onnx转Mindspore问题

[NoOneUST]

很抱歉叨扰您了~可是真的很紧急。
我先将一个Pytorch模型转成了Onnx（OpSet 11）并且验证过了等价性，但是现在我在无法用Mindconveter将Onnx转为Mindspore模型和权重。

我用于将Pytorch模型转为Onnx的代码：

直接将生成的Onnx转为Mindspore遇到Simplifying问题：

尝试将Onnx先进行优化：

优化后再用Mindconverter转换提示Operator不支持：

输出文件夹里没有权重和模型：

操作符扫描报告内容：
operator_name, count, status
onnx::Reshape, 261, Supported
onnx::Gather, 174, Supported
onnx::Mul, 99, Supported
onnx::Conv, 277, Supported
onnx::Transpose, 96, Supported
onnx::Squeeze, 2, Supported
onnx::Relu, 155, Supported
onnx::Div, 179, Supported
onnx::Slice, 341, Supported
onnx::Expand, 2, Unsupported
onnx::MatMul, 42, Supported
onnx::Add, 206, Supported
onnx::CumSum, 2, Supported
onnx::Tile, 4, Unsupported
onnx::Neg, 44, Supported
onnx::Concat, 70, Supported
onnx::Greater, 45, Supported
onnx::Not, 45, Supported
onnx::Where, 126, Supported
onnx::Sub, 90, Supported
onnx::ScatterND, 84, Unsupported
onnx::ConvTranspose, 26, Supported
onnx::Unsqueeze, 27, Supported
onnx::Pow, 27, Supported
onnx::GridSample, 42, Unsupported
onnx::ReduceSum, 72, Supported
onnx::5598, 1, Unsupported
onnx::5601, 1, Unsupported
onnx::5604, 1, Unsupported
onnx::5607, 1, Unsupported
onnx::5610, 1, Unsupported
onnx::5613, 1, Unsupported
onnx::5616, 1, Unsupported
onnx::Softmax, 24, Supported
onnx::Clip, 23, Supported
onnx::Log, 21, Supported
onnx::Exp, 21, Supported
onnx::5619, 1, Unsupported
onnx::Resize, 2, Supported
onnx::Abs, 3, Supported
onnx::Cast, 3, Supported
onnx::ReduceMean, 3, Supported
onnx::4244, 1, Unsupported
onnx::4247, 1, Unsupported
onnx::4250, 1, Unsupported
onnx::4253, 1, Unsupported
onnx::4256, 1, Unsupported
onnx::4259, 1, Unsupported
onnx::4262, 1, Unsupported
onnx::4265, 1, Unsupported
onnx::4268, 1, Unsupported
onnx::4271, 1, Unsupported
onnx::4274, 1, Unsupported
onnx::4277, 1, Unsupported
onnx::4280, 1, Unsupported
onnx::4283, 1, Unsupported
onnx::4286, 1, Unsupported
onnx::4289, 1, Unsupported

如果我在Pytorch文件里直接转：

报错4000000：
ldd: warning: you do not have execution permission for /export/data/lwangcg/anaconda3/lib/python3.7/site-packages/mindspore/run_check/../_c_expression.cpython-37m-x86_64-linux-gnu.so' ldd: warning: you do not have execution permission for /export/data/lwangcg/anaconda3/lib/python3.7/site-packages/mindspore/run_check/../_c_expression.cpython-37m-x86_64-linux-gnu.so'
[INFO] MI(190315:140408814913344,MainProcess):2022-07-11-03:14:47.696.105 [MINDCONVERTER] Pytorch model loading begins.
[ERROR] MI(190315:140408814913344,MainProcess):2022-07-11-03:14:47.696.692 [MINDCONVERTER] Failed to trace pytorch model to graph.
[ERROR] MI(190315:140408814913344,MainProcess):2022-07-11-03:14:47.699.808 [MINDCONVERTER] [SubGraphSearchingError] code: 4000000, msg: module 'torch.onnx.utils' has no attribute '_unique_state_dict'
/export/data/lwangcg/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint8 = np.dtype([("qint8", np.int8, 1)])
/export/data/lwangcg/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/export/data/lwangcg/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint16 = np.dtype([("qint16", np.int16, 1)])
/export/data/lwangcg/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/export/data/lwangcg/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint32 = np.dtype([("qint32", np.int32, 1)])
/export/data/lwangcg/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
np_resource = np.dtype([("resource", np.ubyte, 1)])
[WARNING] MI(190315:140408814913344,MainProcess):2022-07-11-03:14:51.074.040 [MINDCONVERTER] For command line, libraries mindspore(>=1.2.0), onnx(>=1.8.0), onnxruntime(>=1.5.2) and onnxoptimizer(>=0.1.2) are required by converter, tensorflow(>=1.15.0) and tf2onnx(>=1.7.1) are required when converted from TF(.pb). For API, libraries mindspore(>=1.2.0) and torch(==1.8.2) are required. Current versions are: mindspore(==1.7.0), onnx(==1.12.0), onnxruntime(==1.11.1), onnxoptimizer(==0.2.7), tensorflow(==1.14.0), tf2onnx(==NotFound), mindspore(==1.7.0) and torch(==1.12.0+cu113).
Process finished with exit code 255

这就是我现在面对的问题的情况了，请问我应该如何操作以将Onnx或者我的Pytorch模型转为Mindspore？非常感谢！