: If the engine fails to read the tuned parameters, ensure the file path in mace_run.config matches the actual location on the device.
: Note that this binary is device-specific ; a file generated for a Snapdragon 660 will not work optimally on a different chipset.
OpenCL programs are normally written as .cl source files and compiled at runtime by the GPU driver (just-in-time compilation). That compilation can take tens to hundreds of milliseconds. By saving the compiled binary, MACE can load it directly (no recompilation), cutting model initialization time dramatically – critical for real-time apps like camera filters or voice assistants. mace-cl-compiled-program.bin
: Using a pre-compiled .bin file significantly reduces the "first-run" latency of an AI application by skipping the kernel compilation step at startup.
: Execute the converter.py tool. When the gpu runtime and target SOCs are correctly specified, MACE automatically compiles the OpenCL kernels for that hardware. : If the engine fails to read the
The file is typically generated during the model deployment or tuning phase. How to build - MACE documentation - Read the Docs
The file will likely be automatically recreated by the system. That compilation can take tens to hundreds of milliseconds
: It contains compiled OpenCL code designed to accelerate neural network inference on mobile GPUs. Tuning and Deployment