Justin Lutz / Rower Public
The EON Tuner helps you find the most optimal architecture for your embedded machine-learning application. Clone this project to use the EON Tuner.

Target

No name set

Arduino Nano 33 BLE Sense (Cortex-M4F 64MHz)

100 ms

256 kB

1024 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

92%
spectr-dense-081
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
2 kB of 256 kB
ROM
21 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:31:26 AM

80%
spectr-dense-65a
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
24 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

2000 ms |
2000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:25:24 AM

70%
spectr-dense-5bf
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
2 kB of 256 kB
ROM
19 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
2000 ms |
Enabled

SPECTRAL-ANALYSIS

128

ACCURACY (KERAS)

4/15/2022, 1:30:00 AM

64%
spectr-dense-aad
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
2 kB of 256 kB
ROM
22 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

2000 ms |
2000 ms |
Enabled

SPECTRAL-ANALYSIS

128

ACCURACY (KERAS)

4/15/2022, 1:22:45 AM

62%
spectr-dense-a95
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
2 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
2000 ms |
Enabled

SPECTRAL-ANALYSIS

128

ACCURACY (KERAS)

4/15/2022, 1:24:24 AM

58%
spectr-dense-4cf
PERFORMANCE
LATENCY
6 ms of 100 ms
RAM
2 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

2000 ms |
1000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:27:23 AM

53%
spectr-dense-dab
PERFORMANCE
LATENCY
5 ms of 100 ms
RAM
22 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:31:49 AM

50%
spectr-dense-406
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
22 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:19:55 AM

40%
spectr-dense-440
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
2 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:28:41 AM

40%
spectr-dense-431
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
2 kB of 256 kB
ROM
19 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
2000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:29:04 AM

39%
spectr-dense-fff
PERFORMANCE
LATENCY
5 ms of 100 ms
RAM
23 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

2000 ms |
2000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:30:29 AM

39%
spectr-dense-6df
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
6 kB of 256 kB
ROM
22 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTRAL-ANALYSIS

128

ACCURACY (KERAS)

4/15/2022, 1:22:22 AM

38%
spectr-dense-c3d
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
22 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:31:00 AM

37%
spectr-dense-f99
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
2 kB of 256 kB
ROM
21 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
2000 ms |
Enabled

SPECTRAL-ANALYSIS

1024

ACCURACY (KERAS)

4/15/2022, 1:23:44 AM

31%
spectr-dense-0f4
PERFORMANCE
LATENCY
5 ms of 100 ms
RAM
2 kB of 256 kB
ROM
25 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

2000 ms |
2000 ms |
Enabled

SPECTRAL-ANALYSIS

128

ACCURACY (KERAS)

4/15/2022, 1:26:36 AM