Brainchip / Akida Image Classification Public
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

BrainChip AKD1000 MINI PCIe Board

100 ms

524288 kB

16384 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

91%
rgb-other-d47
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
829 kB of 16384 kB
Allocated NPs
33
Activity sparsity
69.47%
MACs
357642988
DSP NN Unused
IMAGE INPUT

224 |
224

IMAGE

RGB

ACCURACY (KERAS-AKIDA-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES) - BRAINCHIP AKIDA™

0.0005 | 20 | 91%

Type Filters Kernel Rate
transfer_akidanet_imagenet_160_a50 64 - 0.5

8/8/2023, 1:26:44 PM

91%
rgb-other-d4e
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
810 kB of 16384 kB
Allocated NPs
33
Activity sparsity
69.39%
MACs
357556228
DSP NN Unused
IMAGE INPUT

224 |
224

IMAGE

RGB

ACCURACY (KERAS-AKIDA-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES) - BRAINCHIP AKIDA™

0.0005 | 20 | 91%

Type Filters Kernel Rate
transfer_akidanet_imagenet_160_a50 16 - 0.1

8/8/2023, 2:00:39 PM

88%
rgb-conv2d-b3a
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
62 kB of 16384 kB
Allocated NPs
4
Activity sparsity
81.08%
MACs
3766672
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

ACCURACY (KERAS-AKIDA)
CLASSIFICATION - BRAINCHIP AKIDA™

0.0005 | 10 | 88%

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

8/8/2023, 1:11:52 PM

72%
grayscale-conv2d-606
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
42 kB of 16384 kB
Allocated NPs
6
Activity sparsity
95.28%
MACs
4230409
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

ACCURACY (KERAS-AKIDA)
CLASSIFICATION - BRAINCHIP AKIDA™

0.0005 | 10 | 72%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.25

8/8/2023, 1:12:41 PM

71%
rgb-other-184
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
829 kB of 16384 kB
Allocated NPs
33
Activity sparsity
68.19%
MACs
350037225
DSP NN Unused
IMAGE INPUT

224 |
224

IMAGE

RGB

ACCURACY (KERAS-AKIDA-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES) - BRAINCHIP AKIDA™

0.0005 | 20 | 71%

Type Filters Kernel Rate
transfer_akidanet_imagenet_160_a50 64 - 0.1

8/8/2023, 1:42:20 PM

67%
grayscale-other-480
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
715 kB of 16384 kB
Allocated NPs
25
Activity sparsity
73.13%
MACs
176764137
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

ACCURACY (KERAS-AKIDA-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES) - BRAINCHIP AKIDA™

0.0005 | 20 | 67%

Type Filters Kernel Rate
transfer_akidanet_imagenet_224_a50 64 - 0.1

8/8/2023, 1:19:56 PM

62%
grayscale-other-c36
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
828 kB of 16384 kB
Allocated NPs
33
Activity sparsity
76.98%
MACs
346424553
DSP NN Unused
IMAGE INPUT

224 |
224

IMAGE

Grayscale

ACCURACY (KERAS-AKIDA-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES) - BRAINCHIP AKIDA™

0.0005 | 20 | 62%

Type Filters Kernel Rate
transfer_akidanet_imagenet_160_a50 64 - 0.5

8/8/2023, 1:23:33 PM

57%
grayscale-other-f2c
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
715 kB of 16384 kB
Allocated NPs
25
Activity sparsity
74.60%
MACs
176764137
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

Grayscale

ACCURACY (KERAS-AKIDA-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES) - BRAINCHIP AKIDA™

0.0005 | 20 | 57%

Type Filters Kernel Rate
transfer_akidanet_imagenet_224_a50 64 - 0.5

8/8/2023, 1:29:37 PM

24%
rgb-other-0cb
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
697 kB of 16384 kB
Allocated NPs
25
Activity sparsity
68.56%
MACs
178582593
DSP NN Unused
IMAGE INPUT

160 |
160

IMAGE

RGB

ACCURACY (KERAS-AKIDA-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES) - BRAINCHIP AKIDA™

0.0005 | 20 | 24%

Type Filters Kernel Rate
transfer_akidanet_imagenet_160_a50 16 - 0.1

8/8/2023, 1:22:28 PM

rgb-other-d54
PERFORMANCE
LATENCY
2 ms of 100 ms
RAM
4 kB of 524288 kB
ROM
810 kB of 16384 kB
Allocated NPs
33
Activity sparsity
69.29%
MACs
350012481
DSP NN Unused
IMAGE INPUT

224 |
224

IMAGE

RGB

ACCURACY (KERAS-AKIDA-TRANSFER-IMAGE)
TRANSFER LEARNING (IMAGES) - BRAINCHIP AKIDA™

0.0005 | 20 | 0%

Type Filters Kernel Rate
transfer_akidanet_imagenet_160_a50 16 - 0.5

8/8/2023, 1:40:38 PM