Quantcast
Channel: Raspberry Pi Forums
Viewing all articles
Browse latest Browse all 8621

General • Re: Workarounds for RAM overflow error - TFlite Micro

$
0
0
I dug this up from a README somewhere:

Code:

### Reference **MCU** memory footprint based on ESC-10 dataset| Model             | Format | Resolution | Series  | Activation RAM (kB) | Runtime RAM (kB) | Weights Flash (kB) | Code Flash (kB) | Total RAM (kB)  | Total Flash (kB) | STM32Cube.AI version  ||-------------------|--------|------------|---------|---------------------|------------------|--------------------|-----------------|-----------------|------------------|-----------------------||[Yamnet 256]       | Int8   | 64x96x1    | ...     | 109.57              |   7.61           |   135.91           |   57.74         | 117.18          | 193.65           | 10.0.0                |
If this is the one, it should fit... But then there is this:

https://github.com/raspberrypi/pico-tflmicro

while this:

https://github.com/tensorflow/tflite-micro

does not list RPi Pico support. Not a tflite-micro user, so I'm just talking generalities, filling in some blanks to move this forward, if at all possible.

Question is, when running micro_speech with its original model, how much free memory is left?

Statistics: Posted by katak255 — Wed Oct 08, 2025 11:50 pm



Viewing all articles
Browse latest Browse all 8621

Trending Articles