a-box detection - Impulses

Experiments (6 / 10)

Name	Input	DSP Blocks	Learn Blocks	Tr_Proc	f32_V_mAP@[IoU=50]	i8_V_mAP@[IoU=50]	f32_T_Acc	i8_T_Acc	f32_T_mAP@[IoU=50]	i8_T_mAP@[IoU=50]
MobileNetV2 cycles 25 rate 0.001	320×320	Image	MobileNetV2 SSD FPN-Lite	GPU	0.62	0.62	-	-	-	-	View impulse
MobileNetV2 cycles 40 rate 0.001	320×320	Image	MobileNetV2 SSD FPN-Lite	GPU	0.70	0.70	-	-	-	-	View impulse
MobileNetV2 cycles 60 rate 0.001	320×320	Image	MobileNetV2 SSD FPN-Lite	GPU	0.73	0.74	-	-	-	-	View impulse
MobileNetV2 cycles 60 rate BACK TO 0.001	320×320	Image	MobileNetV2 SSD FPN-Lite	GPU	0.73	0.74	-	-	-	-	View impulse
MobileNetV2 cycles 60 rate 0.0001	320×320	Image	MobileNetV2 SSD FPN-Lite	GPU	0.04	0.04	-	-	-	-	View impulse
MobileNetV2 cycles 80 rate 0.001	320×320	Image	MobileNetV2 SSD FPN-Lite	GPU	0.73	0.74	31.9%	29.8%	0.71	0.70	View impulse

Impulse information

Tags - Impulse tags

Source - Impulse source (Manual vs EON Tuner)

Created by - The user who originally created this impulse

Key metrics

Tr_Proc - Training processor

f32_Latency - float32 Latency for the complete impulse.

i8_Latency - int8 Latency for the complete impulse.

f32_V_mAP@[IoU=50] - Mean Average Precision (mAP) [IoU=50] (float32, validation set)

i8_V_mAP@[IoU=50] - Mean Average Precision (mAP) [IoU=50] (int8, validation set)

f32_T_Acc - Accuracy (float32, test set)

i8_T_Acc - Accuracy (int8, test set)

f32_T_mAP@[IoU=50] - Mean Average Precision (mAP) [IoU=50] (float32, test set)

i8_T_mAP@[IoU=50] - Mean Average Precision (mAP) [IoU=50] (int8, test set)

Input block config

Img_Height - Image height

Img_Width - Image width

Resize_Mode - Resize mode

Processing block config

Norm - Normalization

Col_dep - Image > Color depth (Processing Block #1)

Learn block config

Model - ML Model architecture (Learn Block #1)

Notc - Number of training cycles (Learn Block #1)

Lea_rat - Learning rate (Learn Block #1)

Dat_aug - Data augmentation (Learn Block #1)

Bat_siz - Batch size (Learn Block #1)

Val_Size - Validation set size (Learn Block #1)

Val_Key - Split train/validation set on metadata key (Learn Block #1)

Mode - Visual / expert mode (Learn Block #1)

Thr_St - Score threshold (Learn Block #1)

Accuracy metrics

f32_V_mAP - float32 mAP (validation set)

f32_V_mAP@[IoU=75] - float32 mAP@[IoU=75] (validation set)

f32_V_mAP@[area=s] - float32 mAP@[area=small] (validation set)

f32_V_mAP@[area=m] - float32 mAP@[area=medium] (validation set)

f32_V_mAP@[area=l] - float32 mAP@[area=large] (validation set)

f32_V_Recall@[max_det=1] - float32 Recall@[max_detections=1] (validation set)

f32_V_Recall@[max_det=10] - float32 Recall@[max_detections=10] (validation set)

f32_V_Recall@[max_det=100] - float32 Recall@[max_detections=100] (validation set)

f32_V_Recall@[area=s] - float32 Recall@[area=small] (validation set)

f32_V_Recall@[area=m] - float32 Recall@[area=medium] (validation set)

f32_V_Recall@[area=l] - float32 Recall@[area=large] (validation set)

f32_V_Precision score (legacy) - float32 Precision score (legacy) (validation set)

i8_V_mAP - int8 mAP (validation set)

i8_V_mAP@[IoU=75] - int8 mAP@[IoU=75] (validation set)

i8_V_mAP@[area=s] - int8 mAP@[area=small] (validation set)

i8_V_mAP@[area=m] - int8 mAP@[area=medium] (validation set)

i8_V_mAP@[area=l] - int8 mAP@[area=large] (validation set)

i8_V_Recall@[max_det=1] - int8 Recall@[max_detections=1] (validation set)

i8_V_Recall@[max_det=10] - int8 Recall@[max_detections=10] (validation set)

i8_V_Recall@[max_det=100] - int8 Recall@[max_detections=100] (validation set)

i8_V_Recall@[area=s] - int8 Recall@[area=small] (validation set)

i8_V_Recall@[area=m] - int8 Recall@[area=medium] (validation set)

i8_V_Recall@[area=l] - int8 Recall@[area=large] (validation set)

i8_V_Precision score (legacy) - int8 Precision score (legacy) (validation set)

f32_V_Acc - Legacy "Accuracy" metric (float32, validation set)

i8_V_Acc - Legacy "Accuracy" metric (int8, validation set)

f32_T_mAP - float32 mAP (test set)

f32_T_mAP@[IoU=75] - float32 mAP@[IoU=75] (test set)

f32_T_mAP@[area=s] - float32 mAP@[area=small] (test set)

f32_T_mAP@[area=m] - float32 mAP@[area=medium] (test set)

f32_T_mAP@[area=l] - float32 mAP@[area=large] (test set)

f32_T_Recall@[max_det=1] - float32 Recall@[max_detections=1] (test set)

f32_T_Recall@[max_det=10] - float32 Recall@[max_detections=10] (test set)

f32_T_Recall@[max_det=100] - float32 Recall@[max_detections=100] (test set)

f32_T_Recall@[area=s] - float32 Recall@[area=small] (test set)

f32_T_Recall@[area=m] - float32 Recall@[area=medium] (test set)

f32_T_Recall@[area=l] - float32 Recall@[area=large] (test set)

f32_T_Precision score (legacy) - float32 Precision score (legacy) (test set)

i8_T_mAP - int8 mAP (test set)

i8_T_mAP@[IoU=75] - int8 mAP@[IoU=75] (test set)

i8_T_mAP@[area=s] - int8 mAP@[area=small] (test set)

i8_T_mAP@[area=m] - int8 mAP@[area=medium] (test set)

i8_T_mAP@[area=l] - int8 mAP@[area=large] (test set)

i8_T_Recall@[max_det=1] - int8 Recall@[max_detections=1] (test set)

i8_T_Recall@[max_det=10] - int8 Recall@[max_detections=10] (test set)

i8_T_Recall@[max_det=100] - int8 Recall@[max_detections=100] (test set)

i8_T_Recall@[area=s] - int8 Recall@[area=small] (test set)

i8_T_Recall@[area=m] - int8 Recall@[area=medium] (test set)

i8_T_Recall@[area=l] - int8 Recall@[area=large] (test set)

i8_T_Precision score (legacy) - int8 Precision score (legacy) (test set)

Post-processing block config

Obj_tra - Object tracking enabled

Target device

Define your target device requirements to inform model optimizations and performance calculations. No device yet? Use the default settings which you can change at any time.

Target device

Processor family

Clock rate

| MHz

Max

Custom device name

Application budget

Specify the available RAM and ROM for the model's operation, along with the maximum allowed latency for your specific application. Not sure yet? Start with the defaults and modify them later on.

RAM

| KB

Max

ROM

| KB

Max

Latency

| ms

Max

Experiments (6 / 10)

MobileNetV2 cycles 25 rate 0.001

MobileNetV2 cycles 40 rate 0.001

MobileNetV2 cycles 60 rate 0.001

MobileNetV2 cycles 60 rate BACK TO 0.001

MobileNetV2 cycles 60 rate 0.0001

MobileNetV2 cycles 80 rate 0.001