Description
when I ran: traced_model = torch.jit.trace(model, (input_ids,))
I got:
/home/jhy/py3.6/lib/python3.6/site-packages/torch/tensor.py:389: RuntimeWarning: Iterating over a tensor might cause the trace to be incorrect. Passing a tensor of different shape won't change the number of iterations executed (and might lead to errors or silently give incorrect results).
'incorrect results).', category=RuntimeWarning)
Traceback (most recent call last):
File "/home/jhy/project/xlnet/src/xlnet_test.py", line 13, in
traced_model = torch.jit.trace(model, (input_ids,))
File "/home/jhy/py3.6/lib/python3.6/site-packages/torch/jit/init.py", line 772, in trace
check_tolerance, _force_outplace, _module_class)
File "/home/jhy/py3.6/lib/python3.6/site-packages/torch/jit/init.py", line 904, in trace_module
module._c._create_method_from_trace(method_name, func, example_inputs, var_lookup_fn, _force_outplace)
RuntimeError: Tracer cannot infer type of (tensor([[[-0.9993, 0.2632, -0.6305, ..., -0.3520, -1.2041, -1.5944],
[ 4.5358, 2.6032, -1.4790, ..., 2.1211, 1.6621, -0.9913],
[ 2.0586, 2.1398, 0.6811, ..., 1.9191, 0.0836, -1.2848],
...,
[-1.4818, 0.5329, 0.5212, ..., 0.6176, 1.7843, -1.8773],
[-2.8784, 1.9871, 0.5379, ..., 1.3778, 1.0554, -1.3039],
[-4.1723, 1.3071, 0.6565, ..., 1.2515, 1.6618, -0.8640]]],
grad_fn=), (None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None))
:Cannot infer type of a None value (toTraceableIValue at /pytorch/torch/csrc/jit/pybind_utils.h:268)
frame #0: c10::Error::Error(c10::SourceLocation, std::string const&) + 0x33 (0x7f8ea599c273 in /home/jhy/py3.6/lib/python3.6/site-packages/torch/lib/libc10.so)
frame #1: + 0x44e288 (0x7f8ea69db288 in /home/jhy/py3.6/lib/python3.6/site-packages/torch/lib/libtorch_python.so)
frame #2: + 0x4bdda2 (0x7f8ea6a4ada2 in /home/jhy/py3.6/lib/python3.6/site-packages/torch/lib/libtorch_python.so)
frame #3: + 0x4d1d81 (0x7f8ea6a5ed81 in /home/jhy/py3.6/lib/python3.6/site-packages/torch/lib/libtorch_python.so)
frame #4: + 0x1d3ef4 (0x7f8ea6760ef4 in /home/jhy/py3.6/lib/python3.6/site-packages/torch/lib/libtorch_python.so)
frame #5: _PyCFunction_FastCallDict + 0x288 (0x566ad8 in /home/jhy/py3.6/bin/python)
frame #6: /home/jhy/py3.6/bin/python() [0x5067b0]
frame #7: _PyEval_EvalFrameDefault + 0x4de (0x50729e in /home/jhy/py3.6/bin/python)
frame #8: /home/jhy/py3.6/bin/python() [0x504232]
frame #9: /home/jhy/py3.6/bin/python() [0x505e83]
frame #10: /home/jhy/py3.6/bin/python() [0x5066f0]
frame #11: _PyEval_EvalFrameDefault + 0x4de (0x50729e in /home/jhy/py3.6/bin/python)
frame #12: /home/jhy/py3.6/bin/python() [0x504232]
frame #13: /home/jhy/py3.6/bin/python() [0x505e83]
frame #14: /home/jhy/py3.6/bin/python() [0x5066f0]
frame #15: _PyEval_EvalFrameDefault + 0x4de (0x50729e in /home/jhy/py3.6/bin/python)
frame #16: /home/jhy/py3.6/bin/python() [0x504232]
frame #17: PyEval_EvalCode + 0x23 (0x6022e3 in /home/jhy/py3.6/bin/python)
frame #18: /home/jhy/py3.6/bin/python() [0x647fa2]
frame #19: PyRun_FileExFlags + 0x9a (0x64806a in /home/jhy/py3.6/bin/python)
frame #20: PyRun_SimpleFileExFlags + 0x197 (0x649d97 in /home/jhy/py3.6/bin/python)
frame #21: Py_Main + 0x5c2 (0x63c352 in /home/jhy/py3.6/bin/python)
frame #22: main + 0xe9 (0x4dbcb9 in /home/jhy/py3.6/bin/python)
frame #23: __libc_start_main + 0xf0 (0x7f8eabcff830 in /lib/x86_64-linux-gnu/libc.so.6)
frame #24: _start + 0x29 (0x5cb639 in /home/jhy/py3.6/bin/python)