Closed hoveychen closed 1 month ago
[07/22/2024-10:21:17] [TRT] [I] [MemUsageChange] Init CUDA: CPU +14, GPU +0, now: CPU 24, GPU 17382 (MiB) [07/22/2024-10:21:19] [TRT] [I] [MemUsageChange] Init builder kernel library: CPU +2052, GPU +384, now: CPU 2229, GPU 17766 (MiB) Traceback (most recent call last): File "/root/FasterLivePortrait/scripts/onnx2trt.py", line 157, in <module> main(args) File "/root/FasterLivePortrait/scripts/onnx2trt.py", line 134, in main builder = EngineBuilder(args.verbose) File "/root/FasterLivePortrait/scripts/onnx2trt.py", line 57, in __init__ self.config.max_workspace_size = 12 * (2 ** 30) # 12 GB AttributeError: 'tensorrt_bindings.tensorrt.IBuilderConfig' object has no attribute 'max_workspace_size'
I searched some threads, and it seems to be related to tensorrt version. https://github.com/NVIDIA-AI-IOT/torch2trt/issues/557
https://github.com/NVIDIA-AI-IOT/torch2trt/issues/557
So what's the expected tensorrt version to use?
only test on 8.6.1
I searched some threads, and it seems to be related to tensorrt version.
https://github.com/NVIDIA-AI-IOT/torch2trt/issues/557
So what's the expected tensorrt version to use?