This describes how to enable and use the MLTK command-line interface.


This assumes the MLTK has been installed and is available on the command prompt.

Command Basics

Enable Python Virtual Environment

If you’re using a Python virtual environment as described in the installation guide, ensure you ‘activate’ it to make the mltk command accessible on the command prompt:

source ./mltk_pyvenv/bin/activate

Command Format

All MLTK commands are accessible via the mltk command-line command.
The mltk command expects arguments with the format:

mltk <operation> [<arguments>] [<options> ...]


  • <operation> - The specific operation to perform (e.g. profile, train, etc.)

  • <arguments> - Operation-specific arguments (e.g. The name of an ML model)

  • <options> - Additional flags & arguments to give to the operation


All MLTK commands provide details about their supported arguments/options by appending the --help option, e.g.:

mltk --help
mltk profile --help
mltk train --help

Supported Operations

The following operations are supported by the mltk command:

Name Description
profile Profile a model to determine how efficiently is may run on hardware
train Train a model and generate a archive containing a .tflite model file
ssh Train a model on a remote cloud server via SSH
evaluate Evaluate a trained model to determine how accurate it is
quantize Quantize a trained model to reduce its memory footprint
summarize Generate a text summary of a model
view View a model's graph in an interactive visualizer
update_params Update the parameters embedded into a generated .tflite model file
view_audio Visualize the spectrograms generated by the Audio Feature Generator
classify_audio Classify real-time audio from a development board's or PC's microphone
classify_image Classify images from an RGB camera connected to a development board
fingerprint_reader View fingerprint images from a fingerprint module connected to a development board
commander Run the Silicon Lab's Simplicity Commander utility


To get more information about a specific operation, issue the command:

mltk <operation> --help