2025-06-30 9:14 AM
Hello, I'm writing my theorical part of my Bachelor Thesis. I'm interesed in whether two output ports from two Processing Unit (for example) the Convolution accelerators can be connected to one input port of the POOL, is it possible to do this through the Stream switch? So that we can have two CAxs to work together parallel for the pooling. From the documentation I cant find a specific answer.
And what is the reason here to divide the convolution accelerators as CA0 and CA1?????????????????????????????????????????????????? is that because the can work parallel for one pooling as I mentioned before ????????????????????????
Thanks a lot in advance, I really need an answer
Solved! Go to Solution.
2025-07-01 1:08 AM
Hello @Einstein_rookie_version,
In the reference manual are listed some examples of typical operation that the NPU can do. It does not describe all of them. Depending on your model, it could use different processing stream.
In any case, you do not have any control on that. You can only use the NPU through the ST Edge AI core, which will decide what to do.
Combining the two conv to pass through a single pool may not be faster.
For your second question, yes, it is indeed to work in parallel.
Have a good day,
Julian
2025-07-01 1:08 AM
Hello @Einstein_rookie_version,
In the reference manual are listed some examples of typical operation that the NPU can do. It does not describe all of them. Depending on your model, it could use different processing stream.
In any case, you do not have any control on that. You can only use the NPU through the ST Edge AI core, which will decide what to do.
Combining the two conv to pass through a single pool may not be faster.
For your second question, yes, it is indeed to work in parallel.
Have a good day,
Julian