a11y / Sign Language Detection A&B Public
Target: Arduino Nano 33 BLE Sense (Cortex-M4F 64MHz) Clone this project
Experiments EON Tuner
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

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%
grayscale-conv2d-66f
PERFORMANCE
LATENCY
103 ms of 100 ms
Exceeds target by 3 ms
RAM
17 kB of 256 kB
ROM
37 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

32 |
32

IMAGE

Grayscale

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 98%

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

12/9/2023, 3:06:56 PM

97%
grayscale-conv2d-814
PERFORMANCE
LATENCY
103 ms of 100 ms
Exceeds target by 3 ms
RAM
17 kB of 256 kB
ROM
37 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

32 |
32

IMAGE

Grayscale

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 97%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.5

12/9/2023, 3:09:39 PM

96%
rgb-conv2d-d30
PERFORMANCE
LATENCY
87 ms of 100 ms
RAM
17 kB of 256 kB
ROM
33 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

32 |
32

IMAGE

RGB

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 96%

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

12/9/2023, 3:10:47 PM

96%
rgb-mobilenetv1-1c0
PERFORMANCE
LATENCY
292 ms of 100 ms
Exceeds target by 192 ms
RAM
63 kB of 256 kB
ROM
110 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 96%

MobileNetV1 0.1
64 | 0.5

12/9/2023, 3:09:28 PM

95%
grayscale-mobilenetv1-5bc
PERFORMANCE
LATENCY
268 ms of 100 ms
Exceeds target by 168 ms
RAM
56 kB of 256 kB
ROM
110 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

Grayscale

ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 95%

MobileNetV1 0.1
64 | 0.5

12/9/2023, 3:09:33 PM

91%
rgb-mobilenetv1-998
PERFORMANCE
LATENCY
293 ms of 100 ms
Exceeds target by 193 ms
RAM
63 kB of 256 kB
ROM
110 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 91%

MobileNetV1 0.1
64 | 0.5 |

12/9/2023, 3:14:14 PM

91%
grayscale-mobilenetv1-6b3
PERFORMANCE
LATENCY
269 ms of 100 ms
Exceeds target by 169 ms
RAM
56 kB of 256 kB
ROM
110 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

Grayscale

ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 91%

MobileNetV1 0.1
64 | 0.5 |

12/9/2023, 3:09:41 PM

90%
rgb-mobilenetv1-4d1
PERFORMANCE
LATENCY
294 ms of 100 ms
Exceeds target by 194 ms
RAM
63 kB of 256 kB
ROM
105 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 90%

MobileNetV1 0.1
16 | 0.1 |

12/9/2023, 3:09:30 PM

89%
rgb-mobilenetv1-b62
PERFORMANCE
LATENCY
294 ms of 100 ms
Exceeds target by 194 ms
RAM
63 kB of 256 kB
ROM
105 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 89%

MobileNetV1 0.1
16 | 0.1 |

12/9/2023, 3:13:23 PM

88%
grayscale-conv2d-139
PERFORMANCE
LATENCY
44 ms of 100 ms
RAM
16 kB of 256 kB
ROM
32 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

32 |
32

IMAGE

Grayscale

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 88%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
dropout - - 0.5

12/9/2023, 3:06:59 PM

Help & documentation Get help on the forum Send feedback Contact sales
Join us
Terms of Service Privacy policy