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
first_EON_run_grove_vision_ai_v2
Seeed Vision AI Module (HX6537-A ARC DSP 400MHz)
100 ms
2048 kB
2048 kB
Filters
Status
DSP type
Model type
View
Data set
Variant
Sort
General
F1-score
Precision
Recall
Run #1 - first_EON_run_grove_vision_ai_v2 (Apr 09 2024, 02:21:18, finished)
rgb-organization-2588-364
PERFORMANCE
LATENCY
5 ms of
100 ms
RAM
2649 kB of
2048 kB
Exceeds target by 601 kB
ROM
14982 kB of
2048 kB
Exceeds target by 12934 kB
DSP
NN
Unused
IMAGE INPUT
224
|
224
IMAGE
RGB
ORGANIZATION BLOCK 2588
| Parameter | Value |
|---|---|
| hue | 0.1 |
| epochs | 60 |
| backbone | mobilenet_v2 |
| contrast | 0.5 |
| batch-size | 16 |
| brightness | 32 |
| saturation | 1.5 |
| random-flip | 0.5 |
| iou-threshold | 0.5 |
| max-learning-rate | 0.015 |
| min-learning-rate | 0.0005 |
| random-crop-max-ar | 2.0 |
| random-crop-min-ar | 0.5 |
| zoom-out-max-scale | 4.0 |
| zoom-out-min-scale | 1.0 |
| freeze-blocks-vgg16 | [0,1,2,3,4,5] |
| freeze-blocks-vgg19 | [0,1,2,3,4,5] |
| confidence-threshold | 0.01 |
| random-crop-max-scale | 1.0 |
| random-crop-min-scale | 0.3 |
| freeze-blocks-darknet19 | [0,1,2,3,4,5] |
| freeze-blocks-darknet53 | [0,1,2,3,4,5] |
| freeze-blocks-googlenet | [0,1,2,3,4,5,6,7] |
| freeze-blocks-resnet-10 | [0,1,2,3] |
| freeze-blocks-resnet-18 | [0,1,2,3] |
| freeze-blocks-resnet-34 | [0,1,2,3] |
| freeze-blocks-resnet-50 | [0,1,2,3] |
| freeze-blocks-squeezenet | [0,1,2,3] |
| freeze-blocks-mobilenetv1 | [0,1,2,3,4,5,6,7,8,9,10,11] |
| freeze-blocks-mobilenetv2 | [0,1,2,3,4,5,6,7,8,9,10,11,12,13] |
| quantization-aware-training | true |
| freeze-blocks-mobilenetv1-a0.1-96 | [0,1,2,3,4,5,6,7,8,9,10,11] |
| freeze-blocks-mobilenetv1-a0.5-96 | [0,1,2,3,4,5,6,7,8,9,10,11] |
| freeze-blocks-mobilenetv2-a0.1-96 | [0,1,2,3,4,5,6,7,8,9,10,11,12,13] |
| freeze-blocks-mobilenetv1-a0.1-160 | [0,1,2,3,4,5,6,7,8,9,10,11] |
| freeze-blocks-mobilenetv1-a0.25-96 | [0,1,2,3,4,5,6,7,8,9,10,11] |
| freeze-blocks-mobilenetv1-a0.5-160 | [0,1,2,3,4,5,6,7,8,9,10,11] |
| freeze-blocks-mobilenetv2-a0.1-160 | [0,1,2,3,4,5,6,7,8,9,10,11,12,13] |
| freeze-blocks-mobilenetv2-a0.25-96 | [0,1,2,3,4,5,6,7,8,9,10,11,12,13] |
| freeze-blocks-mobilenetv1-a0.25-160 | [0,1,2,3,4,5,6,7,8,9,10,11] |
| freeze-blocks-mobilenetv2-a0.25-160 | [0,1,2,3,4,5,6,7,8,9,10,11,12,13] |
4/9/2024, 3:31:42 AM
Training output
Hide
Click 'Run EON Tuner' to begin
