Kosmas Deligkaris / FrogAI-FirstTest Public
The EON Tuner helps you quickly run hyper-parameter sweeps that explore different pre-processing + model architectures optimized for your defined objectives. Clone this project to use the EON Tuner.

Target

Run #5

ST STM32N6 (Cortex-M55 800MHz + ST Neural-ART accelerator)

100 ms

512 kB

32768 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

5%
spectr-conv1d-4c9
PERFORMANCE
LATENCY
10876 ms of 100 ms
Exceeds target by 10776 ms
RAM
70 kB of 512 kB
ROM
34 kB of 32768 kB
DSP NN Unused
TIME-SERIES INPUT

2000 ms |
2000 ms |
Enabled

SPECTROGRAM

0.025 | 0.0125 | -52

ACCURACY (OVERALL)
CLASSIFICATION

0.005 | 100 | 5%

Type Filters Kernel Rate
conv1d 8 3 -
conv1d 16 3 -
dropout - - 0.5

0 1/27/2026, 3:36:01 AM

0%
spectr-conv2d-fe3
PERFORMANCE
LATENCY
6627 ms of 100 ms
Exceeds target by 6527 ms
RAM
133 kB of 512 kB
ROM
77 kB of 32768 kB
DSP NN Unused
TIME-SERIES INPUT

2000 ms |
2000 ms |
Enabled

SPECTROGRAM

0.05 | 0.025 | -52

ACCURACY (OVERALL)
CLASSIFICATION

0.005 | 100 | 0%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.25

0 1/27/2026, 3:36:34 AM

0%
spectr-conv2d-29d
PERFORMANCE
LATENCY
2501 ms of 100 ms
Exceeds target by 2401 ms
RAM
58 kB of 512 kB
ROM
62 kB of 32768 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
250 ms |
Enabled

SPECTROGRAM

0.075 | 0.0375 | -52

ACCURACY (OVERALL)
CLASSIFICATION

0.005 | 100 | 0%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

0 1/27/2026, 3:37:19 AM