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
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
rgb-mobilenetv1-372
PERFORMANCE
LATENCY
912 ms of 100 ms
Exceeds target by 812 ms
RAM
207 kB of 1536 kB
ROM
229 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
160 |
160
IMAGE
RGB
ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)
0.0005 | 20 | 0%
MobileNetV1 0.2
16 | 0.5 |
7/6/2022, 12:54:15 PM
100%
rgb-mobilenetv2-5e9
PERFORMANCE
LATENCY
2260 ms of 100 ms
Exceeds target by 2160 ms
RAM
738 kB of 1536 kB
ROM
653 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
160 |
160
IMAGE
RGB
ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)
0.0005 | 20 | 100%
MobileNetV2 0.35
64 | 0.1
7/6/2022, 12:56:18 PM
rgb-conv2d-812
PERFORMANCE
LATENCY
6216 ms of 100 ms
Exceeds target by 6116 ms
RAM
52 kB of 1536 kB
ROM
50 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
64 |
64
IMAGE
RGB
ACCURACY (KERAS)
CLASSIFICATION (KERAS)
0.0005 | 10 | 0%
| Type | Filters | Kernel | Rate |
|---|---|---|---|
| conv2d | 8 | 3 | - |
| conv2d | 16 | 3 | - |
| conv2d | 32 | 3 | - |
| dropout | - | - | 0.25 |
7/6/2022, 12:52:48 PM
grayscale-conv2d-7ab
PERFORMANCE
LATENCY
615 ms of 100 ms
Exceeds target by 515 ms
RAM
50 kB of 1536 kB
ROM
64 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
64 |
64
IMAGE
Grayscale
ACCURACY (KERAS)
CLASSIFICATION (KERAS)
0.0005 | 10 | 0%
| Type | Filters | Kernel | Rate |
|---|---|---|---|
| conv2d | 8 | 3 | - |
| conv2d | 16 | 3 | - |
| conv2d | 32 | 3 | - |
| conv2d | 64 | 3 | - |
| dropout | - | - | 0.5 |
7/6/2022, 12:54:06 PM
100%
rgb-mobilenetv2-5ad
PERFORMANCE
LATENCY
2418 ms of 100 ms
Exceeds target by 2318 ms
RAM
738 kB of 1536 kB
ROM
587 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
160 |
160
IMAGE
RGB
ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)
0.0005 | 20 | 100%
MobileNetV2 0.35
16 | 0.1 |
7/6/2022, 12:58:56 PM
100%
rgb-mobilenetv2-e4c
PERFORMANCE
LATENCY
15714 ms of 100 ms
Exceeds target by 15614 ms
RAM
738 kB of 1536 kB
ROM
587 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
160 |
160
IMAGE
RGB
ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)
0.0005 | 20 | 100%
MobileNetV2 160x160 0.35
16 | 0.5
7/6/2022, 1:00:17 PM
100%
rgb-mobilenetv2-0e8
PERFORMANCE
LATENCY
962 ms of 100 ms
Exceeds target by 862 ms
RAM
674 kB of 1536 kB
ROM
241 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
160 |
160
IMAGE
RGB
ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)
0.0005 | 20 | 100%
MobileNetV2 0.05
64 | 0.5 |
7/6/2022, 1:00:23 PM
50%
rgb-mobilenetv1-fd3
PERFORMANCE
LATENCY
1082 ms of 100 ms
Exceeds target by 982 ms
RAM
264 kB of 1536 kB
ROM
325 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
160 |
160
IMAGE
RGB
ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)
0.0005 | 20 | 50%
MobileNetV1 0.25
64 | 0.1 |
7/6/2022, 1:03:27 PM
grayscale-mobilenetv1-f4c
PERFORMANCE
LATENCY
1181 ms of 100 ms
Exceeds target by 1081 ms
RAM
264 kB of 1536 kB
ROM
312 kB of 8096 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 | 0.5
7/6/2022, 1:03:04 PM
100%
rgb-mobilenetv2-694
PERFORMANCE
LATENCY
5763 ms of 100 ms
Exceeds target by 5663 ms
RAM
674 kB of 1536 kB
ROM
175 kB of 8096 kB
DSP NN Unused
IMAGE INPUT
160 |
160
IMAGE
RGB
ACCURACY (KERAS-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES)
0.0005 | 20 | 100%
MobileNetV2 0.05
16 | 0.1
7/6/2022, 1:04:09 PM
Training output
Hide
Click 'Run EON Tuner' to begin
