How can I reduce the RAM consumption of AER applications?

[Fyouj.1]

Dear ST Engineer,

I use CubeMX. AI debugging the AER applications downloaded from https://github.com/Shahnawax/AER-CNN-KERAS, need RAM 9.31 MiB, no matter use what kind of compression is so big, that is to say there is no that kind of MCU can adapt to this kind of use,What can be done to reduce RAM consumption?Or is there any MCU that has that much RAM?

[Tesla DeLorean]

Is this even designed for an MCU target, or an embedded linux platform?

You might need code the algorithm yourself in more efficient manner, in native code.

For high RAM requirements, perhaps look at the STM32 DISCO boards with SDRAM

[Tesla DeLorean]

Perhaps continue in your existing threads..

https://community.st.com/s/question/0D53W000014TzjMSAS/how-to-train-a-convolutional-neural-network-with-data-collected-by-adc

https://community.st.com/s/question/0D53W000014WA3nSAG/aer

[TDK]

No STM32 MCUs have 9MB of built-in RAM. You could add an external chip such as SDRAM to accommodate this. However, it seems like the application you are running is not intended for an embedded platform.