7. Examples & Applications
This section provides ready-to-run examples demonstrating various AI applications for TI microcontrollers. Each example includes complete configuration files and step-by-step instructions.
7.1. Running an Example
All examples are located in tinyml-modelzoo/examples/. To run an example:
cd tinyml-modelzoo
./run_tinyml_modelzoo.sh examples/<example_name>/config.yaml
cd tinyml-modelzoo
run_tinyml_modelzoo.bat examples\<example_name>\config.yaml
Output artifacts will be saved to ../tinyml-modelmaker/data/projects/<project_name>/.
You can choose to save the output artifacts in your own custom directory by specifying
in the respective config.yaml under the common section:
common:
projects_path: './your/choice' # or absolute path
# ... other settings
7.2. Generic Examples
If you do not find an application that matches your use case, use these generic examples as starting points:
Task Type |
Example |
Description |
|---|---|---|
Time Series Classification |
Classify sine/square/sawtooth waveforms. Start here to learn the toolchain. |
|
Time Series Regression |
Generic regression example for continuous value prediction. |
|
Time Series Forecasting |
Generic forecasting example for time series prediction. |
|
Time Series Anomaly Detection |
Generic anomaly detection example using autoencoders. |
7.3. Classification Examples
No. |
Example |
Data Type |
Description |
|---|---|---|---|
1 |
Univariate |
Detect DC arc faults from current waveforms for electrical safety. |
|
2 |
Univariate |
Detect AC arc faults in electrical systems. |
|
3 |
Multivariate |
Classify 5 bearing fault types + normal operation from vibration data. |
|
4 |
Multivariate |
Detect blade imbalance in HVAC blowers using 3-phase motor currents. |
|
5 |
Multivariate |
Detect faults in BLDC fans from accelerometer data. |
|
6 |
Multivariate |
Classify transmission line faults using voltage and current. |
|
7 |
Multivariate |
Predict power grid stability from node parameters. |
|
8 |
Multivariate |
Identify gas type and concentration from sensor array data. |
|
9 |
Multivariate |
Human Activity Recognition from accelerometer/gyroscope data. |
|
10 |
Multivariate |
Classify normal vs anomalous heartbeats from ECG signals. |
|
11 |
Multivariate |
Non-Intrusive Load Monitoring - identify active appliances. |
|
12 |
Multivariate |
Detect presence/motion using PIR sensor data. |
|
13 |
Univariate |
Detect AC grid faults in EV on-board chargers using current measurements. |
7.4. Regression Examples
No. |
Example |
Data Type |
Description |
|---|---|---|---|
1 |
Multivariate |
Predict PMSM motor torque from current measurements. |
|
2 |
Multivariate |
Predict induction motor speed from electrical signals. |
|
3 |
Multivariate |
Predict washing machine load weight. |
|
4 |
Multivariate |
Predict MOSFET junction temperature for thermal management in power converters. |
7.5. Forecasting Examples
No. |
Example |
Data Type |
Description |
|---|---|---|---|
1 |
Multivariate |
Forecast PMSM rotor winding temperature. |
|
2 |
Multivariate |
Predict indoor temperature for HVAC control. |
7.6. Anomaly Detection Examples
No. |
Example |
Data Type |
Description |
|---|---|---|---|
1 |
Univariate |
Detect anomalous DC arc patterns using autoencoder. |
|
2 |
Multivariate |
Detect anomalous heartbeat patterns from ECG signals. |
|
3 |
Multivariate |
Detect anomalous fan blade behavior from accelerometer data. |
|
4 |
Multivariate |
Detect anomalous bearing behavior from vibration data. |
7.7. Image Classification Examples
No. |
Example |
Data Type |
Description |
|---|---|---|---|
1 |
Image |
Handwritten digit recognition (MNIST dataset). |