-
I just wanted to ask on how I should write the python code for oneAPI back-end, it only produces a emulation file, what I want is a fpga hardware file so that I can synthesize it on my FPGA board.currently I have this:
|
Beta Was this translation helpful? Give feedback.
Replies: 2 comments
-
The command to use, after compiling the model, is build, (This is not the function you call directly, but |
Beta Was this translation helpful? Give feedback.
-
By the way, as an aside, it's also usually a good idea to pass |
Beta Was this translation helpful? Give feedback.
The command to use, after compiling the model, is build,
hls_model.build(build_type='fpga')
. You can look at the options at:https://fastmachinelearning.org/hls4ml/autodoc/hls4ml.backends.oneapi.html#hls4ml.backends.oneapi.oneapi_backend.OneAPIBackend.build
(This is not the function you call directly, but
hls_model.build()
calls the backend-specific function with the same arguments.) Note that the building is based on what is given in oneAPI-samples, so you can also manually work with the cmake file.