Exception in torch.jit.script doesn't indicate where in the code the problem lies.

[BrettRyland]

🐛 Describe the bug

In the following exception caused by

RuntimeError: Attempted to use Dict without contained types. Please add contained type, e.g. Dict[int, int]

there's no indication of where in the model/code the problem lies, only that there's a Dict without contained types somewhere:

In [6]: scripted_model = run_trace(S,args)
---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
Input In [6], in <cell line: 1>()
----> 1 scripted_model = run_trace(S,args)

File ~/repos/Autosensor/NN/deploy/trace_model.py:35, in run_trace(S, args)
     33 with torch.inference_mode():
     34         if args.script:
---> 35                 traced_module = torch.jit.script(S.model)
     36         else:
     37                 example = torch.rand(1, 3, *S.model.input_size).to(S.device)

File ~/.local/lib/python3.10/site-packages/torch/jit/_script.py:1265, in script(obj, optimize, _frames_up, _rcb, example_inputs)
   1263 if isinstance(obj, torch.nn.Module):
   1264     obj = call_prepare_scriptable_func(obj)
-> 1265     return torch.jit._recursive.create_script_module(
   1266         obj, torch.jit._recursive.infer_methods_to_compile
   1267     )
   1269 if isinstance(obj, dict):
   1270     return create_script_dict(obj)

File ~/.local/lib/python3.10/site-packages/torch/jit/_recursive.py:454, in create_script_module(nn_module, stubs_fn, share_types, is_tracing)
    452 if not is_tracing:
    453     AttributeTypeIsSupportedChecker().check(nn_module)
--> 454 return create_script_module_impl(nn_module, concrete_type, stubs_fn)

File ~/.local/lib/python3.10/site-packages/torch/jit/_recursive.py:520, in create_script_module_impl(nn_module, concrete_type, stubs_fn)
    518 # Compile methods if necessary
    519 if concrete_type not in concrete_type_store.methods_compiled:
--> 520     create_methods_and_properties_from_stubs(concrete_type, method_stubs, property_stubs)
    521     # Create hooks after methods to ensure no name collisions between hooks and methods.
    522     # If done before, hooks can overshadow methods that aren't exported.
    523     create_hooks_from_stubs(concrete_type, hook_stubs, pre_hook_stubs)

File ~/.local/lib/python3.10/site-packages/torch/jit/_recursive.py:371, in create_methods_and_properties_from_stubs(concrete_type, method_stubs, property_stubs)
    368 property_defs = [p.def_ for p in property_stubs]
    369 property_rcbs = [p.resolution_callback for p in property_stubs]
--> 371 concrete_type._create_methods_and_properties(property_defs, property_rcbs, method_defs, method_rcbs, method_defaults)

File ~/.local/lib/python3.10/site-packages/torch/jit/annotations.py:350, in try_ann_to_type(ann, loc)
    348 if a is None:
    349     inner.append(NoneType.get())
--> 350 maybe_type = try_ann_to_type(a, loc)
    351 msg = "Unsupported annotation {} could not be resolved because {} could not be resolved."
    352 assert maybe_type, msg.format(repr(ann), repr(maybe_type))

File ~/.local/lib/python3.10/site-packages/torch/jit/annotations.py:321, in try_ann_to_type(ann, loc)
    319     if elem_type:
    320         return ListType(elem_type)
--> 321 if is_dict(ann):
    322     key = try_ann_to_type(ann.__args__[0], loc)
    323     value = try_ann_to_type(ann.__args__[1], loc)

File ~/.local/lib/python3.10/site-packages/torch/_jit_internal.py:880, in is_dict(ann)
    878 def is_dict(ann) -> bool:
    879     if ann is Dict:
--> 880         raise_error_container_parameter_missing("Dict")
    882     if not hasattr(ann, '__module__'):
    883         return False

File ~/.local/lib/python3.10/site-packages/torch/_jit_internal.py:1101, in raise_error_container_parameter_missing(target_type)
   1099 def raise_error_container_parameter_missing(target_type) -> None:
   1100     if target_type == 'Dict':
-> 1101         raise RuntimeError(
   1102             "Attempted to use Dict without "
   1103             "contained types. Please add contained type, e.g. "
   1104             "Dict[int, int]"
   1105         )
   1106     raise RuntimeError(
   1107         f"Attempted to use {target_type} without a "
   1108         "contained type. Please add a contained type, e.g. "
   1109         f"{target_type}[int]"
   1110     )

RuntimeError: Attempted to use Dict without contained types. Please add contained type, e.g. Dict[int, int]

It would be really helpful if the exception could indicate the location of the problematic dictionary.

Versions

Collecting environment information...
PyTorch version: 1.11.0a0+gitbc2c6ed
Is debug build: False
CUDA used to build PyTorch: 11.7
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04 LTS (x86_64)
GCC version: (Ubuntu 11.2.0-19ubuntu1) 11.2.0
Clang version: Could not collect
CMake version: version 3.22.1
Libc version: glibc-2.35

Python version: 3.10.4 (main, Apr  2 2022, 09:04:19) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.15.0-33-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 11.7.64
GPU models and configuration: GPU 0: NVIDIA GeForce GTX 1080 Ti
Nvidia driver version: 515.43.04
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.7.6.5
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.3.2
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.3.2
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.3.2
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.3.2
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.3.2
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.3.2
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.3.2
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

Versions of relevant libraries:
[pip3] mypy-extensions==0.4.3
[pip3] numpy==1.21.6
[pip3] numpy-quaternion==2022.4.1
[pip3] torch==1.11.0
[pip3] torch-tensorrt==1.2.0a0+e9e824c0
[pip3] torchsummary==1.5.1
[pip3] torchvision==0.12.0
[conda] Could not collect

[BrettRyland]

Note: I eventually found the Dict, but wasted half a day in doing so.

[Gamrix]

Can you include a sample repro of this error?