RDisrael / cards project data Public
Target: Espressif ESP-EYE (ESP32 240MHz) 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

Espressif ESP-EYE (ESP32 240MHz)

100 ms

4096 kB

4096 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

Run #1 - teste (Jan 30 2024, 17:35:15, finished)
grayscale-conv2d-851
PERFORMANCE
LATENCY
176 ms of 100 ms
Exceeds target by 76 ms
RAM
19 kB of 4096 kB
ROM
57 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

32 |
32

IMAGE

Grayscale

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 0%

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

1/30/2024, 5:48:05 PM

91%
rgb-mobilenetv2-a43
PERFORMANCE
LATENCY
1320 ms of 100 ms
Exceeds target by 1220 ms
RAM
275 kB of 4096 kB
ROM
234 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 20 | 91%

MobileNetV2 0.05
64 | 0.1

1/30/2024, 5:53:42 PM

91%
rgb-mobilenetv2-637
PERFORMANCE
LATENCY
2659 ms of 100 ms
Exceeds target by 2559 ms
RAM
339 kB of 4096 kB
ROM
647 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 20 | 91%

MobileNetV2 0.35
64 | 0.5

1/30/2024, 5:49:08 PM

77%
rgb-mobilenetv2-29a
PERFORMANCE
LATENCY
838 ms of 100 ms
Exceeds target by 738 ms
RAM
285 kB of 4096 kB
ROM
222 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

RGB

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

0.0005 | 20 | 77%

MobileNetV2 0.1
16 | 0.5

1/30/2024, 5:48:55 PM

61%
grayscale-mobilenetv2-976
PERFORMANCE
LATENCY
1275 ms of 100 ms
Exceeds target by 1175 ms
RAM
275 kB of 4096 kB
ROM
234 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

96 |
96

IMAGE

Grayscale

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

0.0005 | 20 | 61%

MobileNetV2 0.05
64 | 0.1 |

1/30/2024, 5:48:29 PM

grayscale-conv2d-beb
PERFORMANCE
LATENCY
352 ms of 100 ms
Exceeds target by 252 ms
RAM
49 kB of 4096 kB
ROM
60 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

64 |
64

IMAGE

Grayscale

ACCURACY (KERAS)
CLASSIFICATION

0.0005 | 10 | 0%

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

1/30/2024, 5:43:32 PM

73%
rgb-mobilenetv2-0d1
PERFORMANCE
LATENCY
4088 ms of 100 ms
Exceeds target by 3988 ms
RAM
672 kB of 4096 kB
ROM
223 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 73%

MobileNetV2 0.1
16 | 0.1 |

1/30/2024, 5:46:26 PM

14%
rgb-mobilenetv1-e48
PERFORMANCE
LATENCY
1020 ms of 100 ms
Exceeds target by 920 ms
RAM
123 kB of 4096 kB
ROM
110 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 14%

MobileNetV1 0.1
64 | 0.1

1/30/2024, 5:50:44 PM

grayscale-mobilenetv1-a8d
PERFORMANCE
LATENCY
4590 ms of 100 ms
Exceeds target by 4490 ms
RAM
258 kB of 4096 kB
ROM
307 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

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

0.0005 | 20 | 0%

MobileNetV1 0.25
16 | 0.1 |

1/30/2024, 5:43:05 PM

2%
rgb-mobilenetv1-f65
PERFORMANCE
LATENCY
4882 ms of 100 ms
Exceeds target by 4782 ms
RAM
258 kB of 4096 kB
ROM
307 kB of 4096 kB
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

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

0.0005 | 20 | 2%

MobileNetV1 0.25
16 | 0.5 |

1/30/2024, 5:45:07 PM

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