Created by: zjutlzt

复现步骤 使用Padlle-TRT-FP32的模型 没有保存成其他模式-直接是__parma__和 model 1.docker包在 1080ti的 服务器上 编译好 可以运行 2.docker包导入到 2080ti的服务器上运行预测时出现如下错误

I0311 08:51:40.127867 8785 tensorrt_subgraph_pass.cc:285] Prepare TRT engine (Optimize model structure, Select OP kernel etc). This process may cost a lot of time. E0311 08:51:42.247627 8785 helper.h:65] ../rtSafe/cuda/cudaConvolutionRunner.cpp (303) - Cudnn Error in execute: 8 (CUDNN_STATUS_EXECUTION_FAILED) E0311 08:51:42.248762 8785 helper.h:65] ../rtSafe/cuda/cudaConvolutionRunner.cpp (303) - Cudnn Error in execute: 8 (CUDNN_STATUS_EXECUTION_FAILED) terminate called after throwing an instance of 'paddle::platform::EnforceNotMet' what():