Emitters

HiveWatch keeps the runtime API small by pushing output behavior into emitter classes. You create one or more emitters and pass them to hivewatch.init().

SSE Emitter

Use SSEEmitter when you want a local dashboard, replayable run artifacts, and the geographic map view without depending on an external service.

import hivewatch as hw
from hivewatch.emitters import SSEEmitter

hw.init(
    algorithm="FedAvg",
    emitters=[SSEEmitter(port=7070, serve_map=False)],
)

This emitter writes:

• runs/<run_id>.jsonl with the full event stream

• runs/<run_id>.map.json with map-ready metadata

Weights & Biases Emitter

Use WandbEmitter when you want hosted experiment tracking, round metrics, per-client metrics, and alert-style events inside W&B.

import hivewatch as hw
from hivewatch.emitters import WandbEmitter

hw.init(
    algorithm="FedAvg",
    emitters=[WandbEmitter(project="my-fl-project")],
)

If a W&B run already exists in the current process, HiveWatch adopts it instead of creating a duplicate run.

MLflow Emitter

Use MLflowEmitter when you want self-hosted experiment tracking and artifact logging.

import hivewatch as hw
from hivewatch.emitters import MLflowEmitter

hw.init(
    algorithm="FedAvg",
    emitters=[MLflowEmitter(
        tracking_uri="http://localhost:5000",
        experiment="my-fl-project",
    )],
)

Start a local MLflow server if needed:

mlflow server --host 0.0.0.0 --port 5000

Combined workflow

HiveWatch can log to all three emitters in the same run:

import hivewatch as hw
from hivewatch.emitters import MLflowEmitter, SSEEmitter, WandbEmitter

hw.init(
    emitters=[
        SSEEmitter(serve_map=False),
        WandbEmitter(project="my-fl-project"),
        MLflowEmitter(experiment="my-fl-project"),
    ],
)