Thaís / face_mask_detection 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

97%
rgb-mobilenetv2-55f
PERFORMANCE
LATENCY
1974 ms of 100 ms
Exceeds target by 1874 ms
RAM
351 kB of 256 kB
Exceeds target by 95 kB
ROM
643 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 20 | 97%

MobileNetV2 0.35
64 | 0.5

11/23/2022, 10:15:09 PM

97%
rgb-mobilenetv1-451
PERFORMANCE
LATENCY
3215 ms of 100 ms
Exceeds target by 3115 ms
RAM
264 kB of 256 kB
Exceeds target by 8 kB
ROM
324 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 97%

MobileNetV1 0.25
64 | 0.1 |

11/23/2022, 9:59:12 PM

96%
rgb-mobilenetv2-3c4
PERFORMANCE
LATENCY
3066 ms of 100 ms
Exceeds target by 2966 ms
RAM
386 kB of 256 kB
Exceeds target by 130 kB
ROM
906 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 20 | 96%

MobileNetV2 160x160 0.5
16 | 0.1

11/23/2022, 9:51:02 PM

96%
rgb-mobilenetv2-c56
PERFORMANCE
LATENCY
1959 ms of 100 ms
Exceeds target by 1859 ms
RAM
351 kB of 256 kB
Exceeds target by 95 kB
ROM
643 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 20 | 96%

MobileNetV2 160x160 0.35
64 | 0.5

11/23/2022, 10:22:57 PM

94%
rgb-conv2d-0db
PERFORMANCE
LATENCY
488 ms of 100 ms
Exceeds target by 388 ms
RAM
34 kB of 256 kB
ROM
128 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

32 |
32

IMAGE

RGB

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 94%

Type Filters Kernel Rate
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
conv2d 128 3 -
dropout - - 0.25

11/23/2022, 9:24:55 PM

92%
rgb-conv2d-a48
PERFORMANCE
LATENCY
582 ms of 100 ms
Exceeds target by 482 ms
RAM
55 kB of 256 kB
ROM
55 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

64 |
64

IMAGE

RGB

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 92%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

11/23/2022, 9:22:13 PM

87%
rgb-mobilenetv1-73b
PERFORMANCE
LATENCY
2208 ms of 100 ms
Exceeds target by 2108 ms
RAM
208 kB of 256 kB
ROM
238 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 87%

MobileNetV1 0.2
64 | 0.5 |

11/23/2022, 9:49:51 PM

79%
grayscale-conv2d-414
PERFORMANCE
LATENCY
322 ms of 100 ms
Exceeds target by 222 ms
RAM
48 kB of 256 kB
ROM
55 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

32 |
32

IMAGE

Grayscale

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 79%

Type Filters Kernel Rate
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.5

11/23/2022, 9:19:48 PM

78%
grayscale-conv2d-de5
PERFORMANCE
LATENCY
687 ms of 100 ms
Exceeds target by 587 ms
RAM
87 kB of 256 kB
ROM
48 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

64 |
64

IMAGE

Grayscale

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 78%

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

11/23/2022, 9:57:40 PM

grayscale-mobilenetv2-229
PERFORMANCE
LATENCY
100 ms
RAM
256 kB
ROM
1024 kB
Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

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

0.0005 | 20

MobileNetV2 0.35
64 | 0.5

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