Samuel Alexander / Edge AI Recycle Bin 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

98%
spectr-conv1d-e57
PERFORMANCE
LATENCY
115 ms of 100 ms
Exceeds target by 15 ms
RAM
24 kB of 256 kB
ROM
40 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
250 ms |
Enabled

SPECTROGRAM

0.05 | 0.025 | -72

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 98%

Type Filters Kernel Rate
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.5

6/26/2023, 6:26:55 AM

98%
mfe-conv1d-fce
PERFORMANCE
LATENCY
222 ms of 100 ms
Exceeds target by 122 ms
RAM
20 kB of 256 kB
ROM
66 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
250 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 98%

Type Filters Kernel Rate
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
conv1d 128 3 -
dropout - - 0.25

6/26/2023, 6:32:47 AM

98%
mfe-conv1d-b96
PERFORMANCE
LATENCY
227 ms of 100 ms
Exceeds target by 127 ms
RAM
20 kB of 256 kB
ROM
31 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
250 ms |
Enabled

MFE

0.032 | 0.016 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 98%

Type Filters Kernel Rate
Data augmentation
conv1d 8 3 -
conv1d 16 3 -
conv1d 32 3 -
dropout - - 0.5

6/26/2023, 6:33:54 AM

96%
spectr-conv1d-769
PERFORMANCE
LATENCY
130 ms of 100 ms
Exceeds target by 30 ms
RAM
23 kB of 256 kB
ROM
42 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
250 ms |
Enabled

SPECTROGRAM

0.05 | 0.025 | -32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
Data augmentation
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.25

6/26/2023, 6:23:09 AM

96%
spectr-conv1d-b47
PERFORMANCE
LATENCY
231 ms of 100 ms
Exceeds target by 131 ms
RAM
39 kB of 256 kB
ROM
72 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.025 | 0.0125 | -32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
conv1d 32 3 -
conv1d 64 3 -
conv1d 128 3 -
dropout - - 0.5

6/26/2023, 6:23:33 AM

96%
spectr-conv1d-ea6
PERFORMANCE
LATENCY
51 ms of 100 ms
RAM
15 kB of 256 kB
ROM
40 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
500 ms |
Enabled

SPECTROGRAM

0.075 | 0.075 | -52

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
Data augmentation
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.25

6/26/2023, 6:26:44 AM

96%
mfe-conv1d-60e
PERFORMANCE
LATENCY
241 ms of 100 ms
Exceeds target by 141 ms
RAM
19 kB of 256 kB
ROM
67 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
250 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
Data augmentation
conv1d 32 3 -
conv1d 64 3 -
conv1d 128 3 -
dropout - - 0.25

6/26/2023, 6:31:53 AM

96%
spectr-conv1d-1c1
PERFORMANCE
LATENCY
66 ms of 100 ms
RAM
16 kB of 256 kB
ROM
41 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
500 ms |
Enabled

SPECTROGRAM

0.05 | 0.05 | -72

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.5

6/26/2023, 6:27:13 AM

96%
spectr-conv1d-fd0
PERFORMANCE
LATENCY
63 ms of 100 ms
RAM
16 kB of 256 kB
ROM
41 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
500 ms |
Enabled

SPECTROGRAM

0.05 | 0.05 | -32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.25

6/26/2023, 6:29:16 AM

96%
spectr-conv2d-9c7
PERFORMANCE
LATENCY
115 ms of 100 ms
Exceeds target by 15 ms
RAM
21 kB of 256 kB
ROM
34 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.075 | 0.075 | -52

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.25

6/26/2023, 6:31:41 AM

96%
spectr-conv1d-2f2
PERFORMANCE
LATENCY
59 ms of 100 ms
RAM
17 kB of 256 kB
ROM
39 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.05 | 0.05 | -72

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
Data augmentation
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.25

6/26/2023, 6:32:12 AM

96%
spectr-conv2d-702
PERFORMANCE
LATENCY
155 ms of 100 ms
Exceeds target by 55 ms
RAM
21 kB of 256 kB
ROM
34 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
250 ms |
Enabled

SPECTROGRAM

0.075 | 0.075 | -52

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.25

6/26/2023, 6:36:07 AM

96%
spectr-conv1d-fcb
PERFORMANCE
LATENCY
71 ms of 100 ms
RAM
19 kB of 256 kB
ROM
31 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.075 | 0.0375 | -32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

Type Filters Kernel Rate
Data augmentation
conv1d 8 3 -
conv1d 16 3 -
conv1d 32 3 -
dropout - - 0.5

6/26/2023, 6:36:05 AM

96%
spectr-conv1d-c3b
PERFORMANCE
LATENCY
80 ms of 100 ms
RAM
21 kB of 256 kB
ROM
28 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.025 | 0.025 | -72

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 96%

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

6/26/2023, 6:36:03 AM

94%
mfe-conv2d-6b0
PERFORMANCE
LATENCY
253 ms of 100 ms
Exceeds target by 153 ms
RAM
28 kB of 256 kB
ROM
34 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
250 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 94%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
dropout - - 0.25

6/26/2023, 6:25:08 AM