Flood Monitor +24Hr - Impulses

Experiments

Name	Tags	Input	Win_Incr	DSP Blocks	Learn Blocks	f32_V_Acc	f32_T_Acc	Tr_Proc	f32_Latency	f32_RAM	f32_Flash	i8_Latency	i8_RAM	i8_Flash
Impulse #1		21,600,000ms.	10,800,000ms.	Raw data	Regression		72.0%	CPU	12 ms.	1.6K	19.8K	1 ms.	1.3K	13.3K	View impulse
Short window		5,400,000ms.	900,000ms.	Raw data	Regression	-	-	CPU	-	-	-	-	-	-	View impulse
5400000 #2	Local Version	5,400,000ms.	900,000ms.	Raw data	Regression		51.4%	CPU	2 ms.	1.8K	32.8K	1 ms.	1.8K	32.8K	View impulse
raw-dense-b17		5,400,000ms.	900,000ms.	Raw Data	Regression		-	CPU	26 ms.	2.3K	112.6K	2 ms.	1.5K	37.3K	View impulse
Impulse #35		5,400,000ms.	900,000ms.	Spectral features	LGBM Random Forest Regressor		72.2%	CPU	10 ms.	3.3K	21.6K	-	-	-	View impulse
raw-dense-04d		5,400,000ms.	900,000ms.	Raw Data	Regression		72.5%	CPU	19 ms.	1.7K	18.8K	1 ms.	1.6K	13.4K	View impulse
No Rainfall (split by year)		5,400,000ms.	900,000ms.	Raw data	Regression		78.7%	CPU	9 ms.	1.3K	12.8K	2 ms.	1.3K	12.8K	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

f32_V_Acc - Accuracy (float32, validation set)

f32_T_Acc - Accuracy (float32, test set)

Tr_Proc - Training processor

f32_Latency - float32 Latency for the complete impulse.

f32_RAM - float32 RAM for the complete impulse.

f32_Flash - float32 Flash for the complete impulse.

i8_Latency - int8 Latency for the complete impulse.

i8_RAM - int8 RAM for the complete impulse.

i8_Flash - int8 Flash for the complete impulse.

Input block config

Win_Size - Window size

Win_Incr - Window increase

Freq - Frequency

Zero_Pad - Zero pad data

Processing block config

Norm - Normalization

Sca_axe - Scaling > Scale axes (Processing Block #1)

Idr - Filter > Input decimation ratio (Processing Block #1)

Typ - Filter > Type (Processing Block #1)

Cof - Filter > Cut-off frequency (Processing Block #1)

Ord - Filter > Order (Processing Block #1)

Typ_ATPB1 - Analysis > Type (Processing Block #1)

FFT_len - Analysis > FFT length (Processing Block #1)

Tlos - Analysis > Take log of spectrum? (Processing Block #1)

OFf - Analysis > Overlap FFT frames? (Processing Block #1)

WDL - Analysis > Wavelet Decomposition Level (Processing Block #1)

Wav - Analysis > Wavelet (Processing Block #1)

Ilfr - Analysis > Improve low frequency resolution? (Processing Block #1)

Learn block config

Model - ML Model architecture (Learn Block #1)

Tc - Training cycles (Learn Block #1)

Lr - Learning rate (Learn Block #1)

Lo - Use learned optimizer? (Learn Block #1)

Acw - Auto-weight classes (Learn Block #1)

Data_Aug - Data augmentation (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)

Lea_rat - Learning rate (Learn Block #1)

Nobi - Number of boosting iterations (Learn Block #1)

Max_dep - Max. depth (Learn Block #1)

L2 - L2 (Learn Block #1)

L1 - L1 (Learn Block #1)

Num_lea - Num. leaves (Learn Block #1)

Mcw - Min. child weight (Learn Block #1)

Fea_fra - Feature fraction (Learn Block #1)

Accuracy metrics

f32_V_MSE - float32 Mean squared error (validation set)

f32_V_MAE - float32 Mean absolute error (validation set)

f32_V_EVS - float32 Explained variance score (validation set)

f32_T_MSE - float32 Mean squared error (test set)

f32_T_MAE - float32 Mean absolute error (test set)

f32_T_EVS - float32 Explained variance score (test set)

i8_V_MSE - int8 Mean squared error (validation set)

i8_V_MAE - int8 Mean absolute error (validation set)

i8_V_EVS - int8 Explained variance score (validation set)

i8_T_MSE - int8 Mean squared error (test set)

i8_T_MAE - int8 Mean absolute error (test set)

i8_T_EVS - int8 Explained variance score (test set)

Post-processing block config

No columns

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

Impulse #1

Short window

5400000 #2

raw-dense-b17

Impulse #35

raw-dense-04d

No Rainfall (split by year)