Closed JiaqinKang closed 2 years ago
This has no content
What do you mean this has no content, This is the error code I received with tensorrt.
Doing what? Vanilla code? What versions of tookit are you using? What version of cuDNN? What version of Tensorrt?
This is an internal error to tensorrt, not our code. So it could be a bunch of different issues. The error says you are missing num_bindings, why are they missing? Can't answer that unless I know the prior content. You posting an error is extremely vague.
Additionally, tensorrt version is for ML pros only since it can be pretty cancer to work with.
If you want more personal help, you can join our discord.
There's no need to be so aggressive, I have already joined the discord but no one offered any helps. if you require more information, you can kindly ask me to provide more details. There's no need to be a ass. Was I being rude first? wtf.
On Tue, 9 Aug 2022 at 16:27, Elijah Harmon @.***> wrote:
Doing what? Vanilla code? What versions of tookit are you using? What version of cuDNN? What version of Tensorrt?
This is an internal error to tensorrt, not our code. So it could be a bunch of different issues. The error says you are missing num_bindings, why are they missing? Can't answer that unless I know the prior content. You posting an error is extremely vague.
Additionally, tensorrt version is for ML pros only since it can be pretty cancer to work with.
If you want more personal help, you can join our discord.
— Reply to this email directly, view it on GitHub https://github.com/RootKit-Org/AI-Aimbot/issues/21#issuecomment-1209530884, or unsubscribe https://github.com/notifications/unsubscribe-auth/AI4OGGUHK6VHMQQS5QGTYH3VYJ2FRANCNFSM55JFRPKA . You are receiving this because you authored the thread.Message ID: @.***>
Loading yolov5s320Half.engine for TensorRT inference... [08/02/2022-00:09:49] [TRT] [I] [MemUsageChange] Init CUDA: CPU +412, GPU +0, now: CPU 12438, GPU 1245 (MiB) [08/02/2022-00:09:49] [TRT] [I] Loaded engine size: 15 MiB [08/02/2022-00:09:49] [TRT] [E] 6: The engine plan file is generated on an incompatible device, expecting compute 8.6 got compute 7.5, please rebuild. [08/02/2022-00:09:49] [TRT] [E] 4: [runtime.cpp::nvinfer1::Runtime::deserializeCudaEngine::50] Error Code 4: Internal Error (Engine deserialization failed.) Traceback (most recent call last): File "C:\ai\main_tensorrt_gpu.py", line 187, in
main()
File "C:\ai\main_tensorrt_gpu.py", line 83, in main
model = DetectMultiBackend('yolov5s320Half.engine', device=torch.device('cuda'), dnn=False, data='', fp16=True)
File "C:\ai\models\common.py", line 389, in init
for index in range(model.num_bindings):
AttributeError: 'NoneType' object has no attribute 'num_bindings'