Open sissini opened 2 years ago
I have the same problem while using ./centerpoint
in the instructions
&&&& RUNNING TensorRT.sample_onnx_centerpoint # ./centerpoint
[11/01/2021-20:05:38] [I] Building and running a GPU inference engine for CenterPoint
[11/01/2021-20:05:43] [I] [TRT] ----------------------------------------------------------------
[11/01/2021-20:05:43] [I] [TRT] Input filename: ../data/centerpoint/pointpillars_trt.onnx
[11/01/2021-20:05:43] [I] [TRT] ONNX IR version: 0.0.7
[11/01/2021-20:05:43] [I] [TRT] Opset version: 11
[11/01/2021-20:05:43] [I] [TRT] Producer name: pytorch
[11/01/2021-20:05:43] [I] [TRT] Producer version: 1.10
[11/01/2021-20:05:43] [I] [TRT] Domain:
[11/01/2021-20:05:43] [I] [TRT] Model version: 0
[11/01/2021-20:05:43] [I] [TRT] Doc string:
[11/01/2021-20:05:43] [I] [TRT] ----------------------------------------------------------------
[11/01/2021-20:05:43] [W] [TRT] /home/jenkins/workspace/OSS/L0_MergeRequest/oss/parsers/onnx/onnx2trt_utils.cpp:226: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[11/01/2021-20:05:43] [I] [TRT] No importer registered for op: ScatterND. Attempting to import as plugin.
[11/01/2021-20:05:43] [I] [TRT] Searching for plugin: ScatterND, plugin_version: 1, plugin_namespace:
[11/01/2021-20:05:43] [I] [TRT] Successfully created plugin: ScatterND
[11/01/2021-20:05:43] [W] [TRT] Tensor DataType is determined at build time for tensors not marked as input or output.
[11/01/2021-20:06:07] [I] [TRT] Some tactics do not have sufficient workspace memory to run. Increasing workspace size may increase performance, please check verbose output.
[11/01/2021-20:06:29] [I] [TRT] Detected 1 inputs and 6 output network tensors.
[11/01/2021-20:06:29] [I] getNbInputs: 2
[11/01/2021-20:06:29] [I] getNbOutputs: 6
[11/01/2021-20:06:29] [I] getNbOutputs Name: 315
[11/01/2021-20:06:29] [E] [TRT] INVALID_ARGUMENT: Cannot find binding of given name: input.1
filePath[idx]: ../data/centerpoint//points/0a0d6b8c2e884134a3b48df43d54c36a.bin
[11/01/2021-20:06:29] [I] [INFO] pointNum : 278272
Segmentation fault (core dumped)
@CarkusL, I have tried your steps on pointpillars with Waymo config. But I faced the error above! Can you please tell me what I am missing!. Thanks!
@CarkusL, I have tried your steps on pointpillars with Waymo config. But I faced the error above! Can you please tell me what I am missing!. Thanks!
[11/01/2021-20:06:29] [E] [TRT] INVALID_ARGUMENT: Cannot find binding of given name: input.1
I think you should check the input name of your onnx model
Thanks, I have already solved this problem. Yeah, the input was not correct.
i follwed your instruction, and run 'python tools/create_data.py waymo_data_prep --root_path=data/Waymo --split train --nsweeps=1', but failed, shows 'Segmentation fault (core dumped)' did u ever meet this problem? hope for help