- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
After implementing a C code that creates a neural network by using a header file containing the weights and activations for each layer in the structure, I am trying to see if I can run the network such that I can observe a prediction or two for verification. The header file I have used is 10KB in size. From what I read about the storage space available in the MSP430 - it seems like the model with the weights would be small enough to fit onto the microcontroller. However, I also read about the possibility to use FRAM of the MSP to store data - based on this I have a couple of questions -
Is it absolutely necessary - if my models are in sizes varying from about 10-40 KB - to store these layers and the weights in the FRAM?
Also, If I do need to use the FRAM to store my data - could anyone please point me toward an answer or thread where the data has been intentionally stored on the FRAM - as I could refer the same and try to see how I can use it for my model. Please feel free to let me know if any more clarity is needed to help answer the question.
Thanks!
**Attention** This is a public forum
