Guilherme Martins / Tomato_leaf_disease_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

Teste 01

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

Run #1 - Teste 01 (Jul 09 2024, 16:52:48, finished)
79%
rgb-mobilenetv2-a09
PERFORMANCE
LATENCY
2632 ms of 100 ms
Exceeds target by 2532 ms
RAM
339 kB of 256 kB
Exceeds target by 83 kB
ROM
648 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 20 | 79%

MobileNetV2 0.35
64 | 0.5

7/9/2024, 5:20:58 PM

76%
rgb-mobilenetv2-7f1
PERFORMANCE
LATENCY
2770 ms of 100 ms
Exceeds target by 2670 ms
RAM
672 kB of 256 kB
Exceeds target by 416 kB
ROM
224 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 76%

MobileNetV2 0.1
16 | 0.1 |

7/9/2024, 5:33:29 PM

73%
rgb-mobilenetv2-55c
PERFORMANCE
LATENCY
5883 ms of 100 ms
Exceeds target by 5783 ms
RAM
726 kB of 256 kB
Exceeds target by 470 kB
ROM
587 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 73%

MobileNetV2 0.35
16 | 0.1

7/9/2024, 6:13:39 PM

68%
grayscale-mobilenetv2-4d7
PERFORMANCE
LATENCY
804 ms of 100 ms
Exceeds target by 704 ms
RAM
275 kB of 256 kB
Exceeds target by 19 kB
ROM
235 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

Grayscale

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

0.0005 | 20 | 68%

MobileNetV2 0.05
64 | 0.1 |

7/9/2024, 5:14:28 PM

59%
rgb-mobilenetv1-f3a
PERFORMANCE
LATENCY
3335 ms of 100 ms
Exceeds target by 3235 ms
RAM
258 kB of 256 kB
Exceeds target by 2 kB
ROM
307 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 59%

MobileNetV1 0.25
16 | 0.5 |

7/9/2024, 5:21:57 PM

58%
grayscale-conv2d-f79
PERFORMANCE
LATENCY
360 ms of 100 ms
Exceeds target by 260 ms
RAM
49 kB of 256 kB
ROM
62 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

64 |
64

IMAGE

Grayscale

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 58%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.5

7/9/2024, 5:02:05 PM

47%
rgb-mobilenetv1-379
PERFORMANCE
LATENCY
426 ms of 100 ms
Exceeds target by 326 ms
RAM
123 kB of 256 kB
ROM
110 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 47%

MobileNetV1 0.1
64 | 0.1

7/9/2024, 5:54:17 PM

45%
grayscale-mobilenetv1-326
PERFORMANCE
LATENCY
3433 ms of 100 ms
Exceeds target by 3333 ms
RAM
258 kB of 256 kB
Exceeds target by 2 kB
ROM
307 kB of 1024 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

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

0.0005 | 20 | 45%

MobileNetV1 0.25
16 | 0.1 |

7/9/2024, 5:20:20 PM

rgb-mobilenetv2-01b
PERFORMANCE
LATENCY
100 ms
RAM
256 kB
ROM
1024 kB
Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20

MobileNetV2 0.35
16 | 0.1 |

rgb-mobilenetv2-a23
PERFORMANCE
LATENCY
100 ms
RAM
256 kB
ROM
1024 kB
Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20

MobileNetV2 160x160 0.35
64 | 0.5 |

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