Edge Impulse Experts / Bangle Activities 2 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

Cortex-M4F 80MHz

100 ms

128 kB

1024 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

77%
raw-dense-d16
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
20 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

3000 ms |
3000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 77%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dropout - - 0.25

10/31/2024, 6:27:38 PM

72%
raw-dense-64b
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
22 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 72%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dropout - - 0.25

10/31/2024, 6:38:59 PM

60%
raw-dense-05e
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
22 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 60%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dropout - - 0.5

10/31/2024, 6:35:47 PM

53%
raw-dense-a8b
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
1 kB of 128 kB
ROM
20 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

3000 ms |
3000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 53%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dropout - - 0.5

10/31/2024, 6:31:03 PM

51%
raw-dense-7f3
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
22 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 51%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dense 10 - -
dropout - - 0.25

10/31/2024, 6:31:53 PM

50%
raw-dense-e50
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
18 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 50%

Type Filters Kernel Rate
dense 20 - -
dense 10 - -
dropout - - 0.25

10/31/2024, 6:31:48 PM

48%
raw-dense-c76
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
21 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

3000 ms |
3000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 48%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dense 10 - -
dropout - - 0.25

10/31/2024, 6:35:49 PM

43%
raw-dense-0c0
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
1 kB of 128 kB
ROM
17 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

3000 ms |
3000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 43%

Type Filters Kernel Rate
dense 20 - -
dense 10 - -
dropout - - 0.25

10/31/2024, 6:31:51 PM

40%
raw-dense-c63
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
22 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

RAW

1

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 40%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dense 10 - -
dropout - - 0.5

10/31/2024, 6:35:11 PM

39%
spectr-dense-6b7
PERFORMANCE
LATENCY
3 ms of 100 ms
RAM
3 kB of 128 kB
ROM
20 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

SPECTRAL-ANALYSIS

64

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 39%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dropout - - 0.25

10/31/2024, 6:29:23 PM

25%
spectr-dense-abf
PERFORMANCE
LATENCY
3 ms of 100 ms
RAM
3 kB of 128 kB
ROM
20 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

3000 ms |
3000 ms |
Enabled

SPECTRAL-ANALYSIS

64

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 25%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dropout - - 0.25

10/31/2024, 6:27:46 PM

18%
spectr-dense-90c
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
18 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

3000 ms |
3000 ms |
Enabled

SPECTRAL-ANALYSIS

16

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 18%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dense 10 - -
dropout - - 0.25

10/31/2024, 6:30:00 PM

15%
spectr-dense-392
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
2 kB of 128 kB
ROM
18 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

SPECTRAL-ANALYSIS

16

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 15%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dense 10 - -
dropout - - 0.25

10/31/2024, 6:34:06 PM

15%
spectr-dense-3fb
PERFORMANCE
LATENCY
6 ms of 100 ms
RAM
3 kB of 128 kB
ROM
18 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

3000 ms |
3000 ms |
Enabled

SPECTRAL-ANALYSIS

16

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 15%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dropout - - 0.25

10/31/2024, 6:28:39 PM

13%
spectr-dense-c5b
PERFORMANCE
LATENCY
1 ms of 100 ms
RAM
1 kB of 128 kB
ROM
20 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

SPECTRAL-ANALYSIS

64

ACCURACY (OVERALL)
CLASSIFICATION

0.0005 | 30 | 13%

Type Filters Kernel Rate
dense 40 - -
dense 20 - -
dropout - - 0.5

10/31/2024, 6:37:18 PM