Capicú Edge ML Inference

Capicú Edge ML Inference (CEMI) is an Edge AI/TinyML experiment workspace managing and analyzing runs, comparisons, and explicit action-event streams across local and extensible testing of model compression and evaluation workflows.

cemi-cli is the packaged command-line distribution for Capicu Edge ML Inference (CEMI).

It provides the local-first workflow for:

• serving the CEMI workspace locally
• opening saved runs in the browser
• launching instrumented commands with CEMI environment wiring
• reading and serving run data from a shared local save directory

What This Package Includes

Install

From PyPI

pip install cemi-cli

From source

From the repository root:

pip install -e ./cli

For local development and tests:

pip install -e "./cli[dev]"
pytest cli/tests/ -q

Quickstart

Minimal example

from cemi.writer import create_writer

writer = create_writer(project="demo", log_dir=".cemi")
writer.start_run(name="baseline")
writer.log_parameter(key="learning_rate", value=0.001)
writer.log_metric(name="loss", value=0.42, step=1)
writer.log_summary_metrics({"final_accuracy": 0.95})
writer.emit_run_record()
writer.end_run(status="succeeded")
writer.emit_run_record()

Start the local gateway:

cemi gateway

Open the workspace:

cemi view

Or visit:

http://127.0.0.1:3141/workspace

Command Summary

Documentation

Visit https://docs.capicu.ai for more information.