Edge Impulse Experts / bicycle-street-fomo Public
Target: Sony Spresense (Cortex-M4F 156MHz) 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

Sony Spresense (Cortex-M4F 156MHz)

100 ms

1536 kB

8096 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

78%
rgb-mobilenetv2-5ab
PERFORMANCE
LATENCY
580 ms of 100 ms
Exceeds target by 480 ms
RAM
351 kB of 1536 kB
ROM
586 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 78%

MobileNetV2 0.35
16 | 0.5 |

8/25/2022, 12:23:28 AM

67%
rgb-mobilenetv2-2a1
PERFORMANCE
LATENCY
4279 ms of 100 ms
Exceeds target by 4179 ms
RAM
674 kB of 1536 kB
ROM
173 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 67%

MobileNetV2 0.05
16 | 0.1

8/25/2022, 12:29:55 AM

65%
rgb-mobilenetv1-f73
PERFORMANCE
LATENCY
4982 ms of 100 ms
Exceeds target by 4882 ms
RAM
135 kB of 1536 kB
ROM
324 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 65%

MobileNetV1 0.25
64 | 0.5

8/25/2022, 12:18:42 AM

63%
grayscale-mobilenetv2-c4f
PERFORMANCE
LATENCY
4025 ms of 100 ms
Exceeds target by 3925 ms
RAM
351 kB of 1536 kB
ROM
585 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

Grayscale

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 63%

MobileNetV2 0.35
16 | 0.5 |

8/25/2022, 12:22:52 AM

63%
rgb-mobilenetv2-3ed
PERFORMANCE
LATENCY
956 ms of 100 ms
Exceeds target by 856 ms
RAM
674 kB of 1536 kB
ROM
236 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 63%

MobileNetV2 0.05
64 | 0.5 |

8/25/2022, 12:22:49 AM

58%
grayscale-mobilenetv1-961
PERFORMANCE
LATENCY
1226 ms of 100 ms
Exceeds target by 1126 ms
RAM
264 kB of 1536 kB
ROM
312 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 58%

MobileNetV1 0.25
16 | 0.1 |

8/25/2022, 12:28:57 AM

58%
grayscale-mobilenetv2-2a3
PERFORMANCE
LATENCY
860 ms of 100 ms
Exceeds target by 760 ms
RAM
684 kB of 1536 kB
ROM
223 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 58%

MobileNetV2 0.1
16 | 0.1 |

8/25/2022, 12:20:26 AM

35%
grayscale-mobilenetv1-2e8
PERFORMANCE
LATENCY
853 ms of 100 ms
Exceeds target by 753 ms
RAM
264 kB of 1536 kB
ROM
312 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 35%

MobileNetV1 0.25
16 | 0.5

8/25/2022, 12:25:54 AM

33%
rgb-conv2d-3f4
PERFORMANCE
LATENCY
3976 ms of 100 ms
Exceeds target by 3876 ms
RAM
51 kB of 1536 kB
ROM
42 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

64 |
64

IMAGE

RGB

CLASSIFICATION (KERAS)

0.0005 | 10 | 33%

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

8/25/2022, 12:23:21 AM

6%
rgb-mobilenetv1-6ac
PERFORMANCE
LATENCY
862 ms of 100 ms
Exceeds target by 762 ms
RAM
208 kB of 1536 kB
ROM
228 kB of 8096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

TRANSFER LEARNING (IMAGES)

0.0005 | 20 | 6%

MobileNetV1 0.2
16 | 0.5 |

8/25/2022, 12:19:32 AM

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