Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU) :- dGPU
• DeepStream Version :- 8.0
serialize cuda engine to file: /some_path/model_b64_gpu2_fp32.engine successfully
how can I change name and/or path of the TRT engine that gets created ?
Check this out: Feature Contribution - Smart Engine File Caching for nvdsinfer
If the parameters for generating the engine (batch_size/gpu/datatype) are exactly the same as those in nvinfer, then nvinfer will not attempt to regenerate the engine file (even with any path and name).
You can also refer to @Levi_Pereira’s reply.
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks.