Christopher Mendez / AI Meter
This is your Edge Impulse project. From here you acquire new training data, design impulses and train models.
About this project
This project consists on a "Smart Energy Meter" that is capable to identify and segregate loads and appliances connected to our house grid by analysing their harmonics behaviour extracted from their current and voltage raw signals, also, with some custom code, the microcontroller shares the inferences results and energy measurements to a server in the cloud using a WiFi Notecard
- YHDC CT 100A:50mA
- YHDC PT 230-9 VAC
- ST Nucleo-F767ZI Dev Board
- Custom PCB
Hackster project guide
Creating your first impulse (100% complete)
Every Machine Learning project starts with data. You can capture data from a development board or your phone, or import data you already collected.
Design an impulse
Teach the model to interpret previously unseen data, based on historical data. Use this to categorize new data, or to find anomalies in sensor readings.
Package the complete impulse up, from signal processing code to trained model, and deploy it on your device. This ensures that the impulse runs with low latency and without requiring a network connection.
Download block output
|Spectral features training data||NPY file||133760 windows|
|Spectral features training labels||NPY file||133760 windows|
|Spectral features testing data||NPY file||3415 windows|
|Spectral features testing labels||NPY file||3415 windows|
|Raw data training data||NPY file||133760 windows|
|Raw data training labels||NPY file||133760 windows|
|Raw data testing data||NPY file||3415 windows|
|Raw data testing labels||NPY file||3415 windows|
|Flatten training data||NPY file||133760 windows|
|Flatten training labels||NPY file||133760 windows|
|Flatten testing data||NPY file||3415 windows|
|Flatten testing labels||NPY file||3415 windows|
|NN Classifier model||TensorFlow Lite (float32)||18 KB|
|NN Classifier model||TensorFlow Lite (int8 quantized)||7 KB|
|NN Classifier model||TensorFlow Lite (int8 quantized with float32 input and output)||7 KB|
|NN Classifier model||TensorFlow SavedModel||28 KB|
|Anomaly detection model||JSON||144 KB|
Data collected2h 12m 9s