7.7.20. Washing Machine Regression

Predict washing machine load weight from sensor data.

7.7.20.1. Overview

This example demonstrates load weight estimation for washing machines using motor current and vibration data. Accurate load estimation enables optimal water and detergent dosing, improving efficiency and wash quality.

Application: Smart appliances, home automation, energy efficiency

Task Type: Time Series Regression

Data Type: Multivariate (motor and sensor data)

7.7.20.2. Configuration

common:
  target_module: 'timeseries'
  task_type: 'generic_timeseries_regression'
  target_device: 'MSPM0G5187'

dataset:
  dataset_name: 'reg_washing_machine'

training:
  model_name: 'REGR_2k_NPU'
  training_epochs: 50
  batch_size: 32

testing: {}
compilation: {}

7.7.20.3. Running the Example

cd tinyml-modelzoo
./run_tinyml_modelzoo.sh examples/reg_washing_machine/config.yaml

7.7.20.4. Dataset Details

Input Variables:

  • Motor current

  • Drum speed

  • Vibration sensor data

Output:

  • Load weight (continuous value in kg)

7.7.20.5. Results

Float Model Predictions:

Float Model Predictions

Actual vs predicted load weight using float model

Quantized Model Predictions:

Quantized Model Predictions

Actual vs predicted load weight using quantized model

7.7.20.6. See Also