[ERROR] Unsupported numpy data type: i64. Cannot implicitly convert to tensorrt.Weights.

[zzlnys]

[06/17/2023-20:06:23] [TRT] [I] [MemUsageChange] Init CUDA: CPU +522, GPU +0, now: CPU 10870, GPU 2466 (MiB) [06/17/2023-20:06:23] [TRT] [I] [MemUsageSnapshot] Begin constructing builder kernel library: CPU 10923 MiB, GPU 2466 MiB [06/17/2023-20:06:23] [TRT] [I] [MemUsageSnapshot] End constructing builder kernel library: CPU 11103 MiB, GPU 2510 MiB Warning: Encountered known unsupported method torch.empty Warning: Encountered known unsupported method torch.is_grad_enabled Warning: Encountered known unsupported method torch.is_grad_enabled Warning: Encountered known unsupported method torch.Tensor.uniform_ Warning: Encountered known unsupported method torch.is_grad_enabled Warning: Encountered known unsupported method torch.empty Warning: Encountered known unsupported method torch.empty Warning: Encountered known unsupported method torch.zeros Warning: Encountered known unsupported method torch.ones [ERROR] Unsupported numpy data type: i64. Cannot implicitly convert to tensorrt.Weights.

How should I fix the error reported above after using TensorRT?

[itachi1232gg]

try convert you model to eval mode by model.eval(), hope it can help