I get the following warning message when running inference via tvm for bert-base-cased
[20:53:25] /neo-ai-dlr/3rdparty/tvm/src/runtime/contrib/tensorrt/tensorrt_logger.h:54: Warning: WARNING: The implicit batch dimension mode has been deprecated. Please create the network with NetworkDefinitionCreationFlag::kEXPLICIT_BATCH flag whenever possible.
If I set it to False like this tvm.target.Target("tensorrt -use_implicit_batch=False")
the code fails with the following
RuntimeError: Failed to run model. Error: TVMError: cannot create std::vector larger than max_size()