Saurabh Datta / simple v1 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

L:2000, KWS

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 - L:2000, KWS (Mar 28 2024, 18:55:25, finished)
61%
mfe-conv1d-605
PERFORMANCE
LATENCY
980 ms of 100 ms
Exceeds target by 880 ms
RAM
58 kB of 256 kB
ROM
79 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

5000 ms |
5000 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 61%

Type Filters Kernel Rate
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
conv1d 128 3 -
dropout - - 0.25

3/28/2024, 7:05:08 PM

57%
mfe-conv2d-c4a
PERFORMANCE
LATENCY
971 ms of 100 ms
Exceeds target by 871 ms
RAM
95 kB of 256 kB
ROM
63 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 57%

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

3/28/2024, 7:10:25 PM

52%
mfe-conv1d-778
PERFORMANCE
LATENCY
956 ms of 100 ms
Exceeds target by 856 ms
RAM
55 kB of 256 kB
ROM
61 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

5000 ms |
5000 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 52%

Type Filters Kernel Rate
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.25

3/28/2024, 7:09:00 PM

52%
mfe-conv1d-398
PERFORMANCE
LATENCY
890 ms of 100 ms
Exceeds target by 790 ms
RAM
56 kB of 256 kB
ROM
52 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

5000 ms |
5000 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 52%

Type Filters Kernel Rate
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.5

3/28/2024, 7:07:36 PM

43%
mfcc-conv2d-81d
PERFORMANCE
LATENCY
1277 ms of 100 ms
Exceeds target by 1177 ms
RAM
68 kB of 256 kB
ROM
65 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFCC

0.032 | 0.032 | 40

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 43%

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

3/28/2024, 7:06:42 PM

43%
mfe-conv1d-17b
PERFORMANCE
LATENCY
1291 ms of 100 ms
Exceeds target by 1191 ms
RAM
84 kB of 256 kB
ROM
42 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFE

0.02 | 0.01 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 43%

Type Filters Kernel Rate
conv1d 8 3 -
conv1d 16 3 -
conv1d 32 3 -
dropout - - 0.25

3/28/2024, 7:02:23 PM

39%
mfcc-conv1d-d4e
PERFORMANCE
LATENCY
1674 ms of 100 ms
Exceeds target by 1574 ms
RAM
60 kB of 256 kB
ROM
76 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFCC

0.02 | 0.02 | 40

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 39%

Type Filters Kernel Rate
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
conv1d 128 3 -
dropout - - 0.5

3/28/2024, 7:11:04 PM

35%
mfcc-conv2d-051
PERFORMANCE
LATENCY
2125 ms of 100 ms
Exceeds target by 2025 ms
RAM
93 kB of 256 kB
ROM
51 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFCC

0.032 | 0.016 | 40

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 35%

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

3/28/2024, 7:09:08 PM

26%
mfcc-conv2d-7cd
PERFORMANCE
LATENCY
1598 ms of 100 ms
Exceeds target by 1498 ms
RAM
81 kB of 256 kB
ROM
60 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

5000 ms |
5000 ms |
Enabled

MFCC

0.032 | 0.032 | 40

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 26%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.5

3/28/2024, 7:11:32 PM

26%
mfe-conv1d-937
PERFORMANCE
LATENCY
1880 ms of 100 ms
Exceeds target by 1780 ms
RAM
75 kB of 256 kB
ROM
194 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

5000 ms |
5000 ms |
Enabled

MFE

0.032 | 0.016 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 26%

Type Filters Kernel Rate
Data augmentation
conv1d 32 3 -
conv1d 64 3 -
conv1d 128 3 -
conv1d 256 3 -
dropout - - 0.5

3/28/2024, 7:07:19 PM

26%
mfcc-conv1d-8c5
PERFORMANCE
LATENCY
2059 ms of 100 ms
Exceeds target by 1959 ms
RAM
69 kB of 256 kB
ROM
184 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFCC

0.02 | 0.02 | 40

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 26%

Type Filters Kernel Rate
Data augmentation
conv1d 32 3 -
conv1d 64 3 -
conv1d 128 3 -
conv1d 256 3 -
dropout - - 0.25

3/28/2024, 7:06:50 PM

22%
mfcc-conv2d-684
PERFORMANCE
LATENCY
1253 ms of 100 ms
Exceeds target by 1153 ms
RAM
69 kB of 256 kB
ROM
47 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFCC

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 22%

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

3/28/2024, 7:10:35 PM

22%
mfcc-conv2d-069
PERFORMANCE
LATENCY
1945 ms of 100 ms
Exceeds target by 1845 ms
RAM
88 kB of 256 kB
ROM
61 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFCC

0.032 | 0.016 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 22%

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

3/28/2024, 7:04:59 PM

22%
mfe-conv2d-855
PERFORMANCE
LATENCY
1027 ms of 100 ms
Exceeds target by 927 ms
RAM
71 kB of 256 kB
ROM
67 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

4000 ms |
4000 ms |
Enabled

MFE

0.05 | 0.05 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 22%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

3/28/2024, 7:01:09 PM

17%
mfcc-conv2d-f86
PERFORMANCE
LATENCY
1046 ms of 100 ms
Exceeds target by 946 ms
RAM
55 kB of 256 kB
ROM
63 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

5000 ms |
5000 ms |
Enabled

MFCC

0.05 | 0.05 | 32

ACCURACY (KERAS)
CLASSIFICATION

0.005 | 100 | 17%

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

3/28/2024, 7:12:35 PM

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