According to the project documentation, the model is supposed to run at 50Hz during inference. However, after local deployment on RTX 5090D hardware, I can only achieve a steady 30Hz.
Could you share relevant optimization methods to reach the expected 50Hz inference frequency?
According to the project documentation, the model is supposed to run at 50Hz during inference. However, after local deployment on RTX 5090D hardware, I can only achieve a steady 30Hz.
Could you share relevant optimization methods to reach the expected 50Hz inference frequency?