RuntimeError: Tracer 无法推断 Seq2SeqLMOutput 的类型

Traceback (most recent call last): File "E:\Python\project\Chinese_Chat_T5_Base-main\convertModel.py", line 37, in <module> traced_model = torch.jit.trace(model, (input_ids, attention_mask, decoder_input_ids), strict=False) File "D:\Program Files\Python310\lib\site-packages\torch\jit\_trace.py", line 759, in trace return trace_module( File "D:\Program Files\Python310\lib\site-packages\torch\jit\_trace.py", line 976, in trace_module module._c._create_method_from_trace( RuntimeError: Tracer cannot infer type of Seq2SeqLMOutput(loss=None, logits=tensor([[[-8.0331, -0.6127, 1.7029, ..., -6.0205, -4.9355, -7.5521]]], grad_fn=<UnsafeViewBackward0>), past_key_values=((tensor([[[[-4.1845e-01, -3.1748e+00, 3.5584e-01, 1.3317e-01, -4.8382e-01, 4.9041e-01, 1.2883e+00, 5.5251e-01, 2.3777e+00, 3.6629e-01, -2.3793e-01, 1.6337e+00, 9.4133e-01, -1.0904e+00, -2.8644e+00, -5.2565e-02, 2.9996e-01, -4.1858e-01, -7.8744e-01, -1.7734e+00, -1.0728e+00, 5.5014e-01, -1.5405e+00, 2.7343e+00, 3.5340e+00, -1.5999e-02, -7.7990e-01, 4.5489e-01, -2.4964e-01, -2.9343e-01, 7.0564e-01, 9.1929e-01, 3.4561e+00, -6.6381e-01, 8.5702e-01, 6.3156e-01, -7.5711e-01, 1.6548e+00, -8.5602e-01, -9.3094e-01, 9.1188e-02, -8.6472e-01, 6.4054e-01, 4.7034e-01, 3.4763e+00, -1.0079e+00, 1.2279e-01, 1.5227e+00, 1.6583e-01, 9.4017e-01, 1.5735e+00, 3.4655e-01, -8.0972e-01, 9.2279e-01, 3.1652e-01, -2.3178e+00, 5.2484e-02,