Edge Impulse Experts / xG24 Card colour sorting Public
Target: SiLabs EFR32MG24 (Cortex-M33 78MHz) Clone this project
Experiments EON Tuner
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

SiLabs EFR32MG24 (Cortex-M33 78MHz)

100 ms

256 kB

1536 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

99%
rgb-mobilenetv1-9c0
Show logs
PERFORMANCE
LATENCY
371 ms of 100 ms
Exceeds target by 271 ms
RAM
135 kB of 256 kB
ROM
311 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 40 | 99%

MobileNetV1 0.25
16 | 0.1 |

4/10/2023, 12:42:04 PM

98%
rgb-mobilenetv1-065
Show logs
PERFORMANCE
LATENCY
1156 ms of 100 ms
Exceeds target by 1056 ms
RAM
264 kB of 256 kB
Exceeds target by 8 kB
ROM
311 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 40 | 98%

MobileNetV1 0.25
16 | 0.05 |

4/10/2023, 12:41:29 PM

97%
rgb-mobilenetv1-489
Show logs
PERFORMANCE
LATENCY
427 ms of 100 ms
Exceeds target by 327 ms
RAM
135 kB of 256 kB
ROM
311 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 40 | 97%

MobileNetV1 0.25
16 | 0.1

4/10/2023, 12:38:27 PM

97%
rgb-mobilenetv1-41d
Show logs
PERFORMANCE
LATENCY
737 ms of 100 ms
Exceeds target by 637 ms
RAM
208 kB of 256 kB
ROM
227 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 40 | 97%

MobileNetV1 0.2
16 | 0.1

4/10/2023, 12:47:16 PM

95%
grayscale-mobilenetv1-3c4
Show logs
PERFORMANCE
LATENCY
1235 ms of 100 ms
Exceeds target by 1135 ms
RAM
264 kB of 256 kB
Exceeds target by 8 kB
ROM
323 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

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

0.0005 | 40 | 95%

MobileNetV1 0.25
64 | 0.05 |

4/10/2023, 12:40:49 PM

95%
rgb-mobilenetv1-95f
Show logs
PERFORMANCE
LATENCY
276 ms of 100 ms
Exceeds target by 176 ms
RAM
131 kB of 256 kB
ROM
108 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 40 | 95%

MobileNetV1 0.1
16 | 0.05 |

4/10/2023, 12:45:53 PM

95%
rgb-mobilenetv1-b35
Show logs
PERFORMANCE
LATENCY
108 ms of 100 ms
Exceeds target by 8 ms
RAM
71 kB of 256 kB
ROM
108 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 40 | 95%

MobileNetV1 0.1
16 | 0.1

4/10/2023, 12:37:49 PM

94%
rgb-mobilenetv1-70f
Show logs
PERFORMANCE
LATENCY
274 ms of 100 ms
Exceeds target by 174 ms
RAM
131 kB of 256 kB
ROM
108 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 40 | 94%

MobileNetV1 0.1
16 | 0.1

4/10/2023, 12:39:48 PM

89%
grayscale-mobilenetv1-fd7
Show logs
PERFORMANCE
LATENCY
1326 ms of 100 ms
Exceeds target by 1226 ms
RAM
264 kB of 256 kB
Exceeds target by 8 kB
ROM
311 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

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

0.0005 | 40 | 89%

MobileNetV1 0.25
16 | 0.05 |

4/10/2023, 12:40:43 PM

75%
rgb-conv2d-dd1
Show logs
PERFORMANCE
LATENCY
31 ms of 100 ms
RAM
18 kB of 256 kB
ROM
30 kB of 1536 kB
DSP NN Unused
IMAGE INPUT

32 |
32

IMAGE

RGB

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 20 | 75%

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

4/10/2023, 12:37:39 PM

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