goodseed 0.2.5

ML experiment tracker

Goodseed

ML experiment tracker. Logs metrics and configs to local SQLite files, serves them via a built-in HTTP server, and visualizes them in the browser.

Install

cd goodseed
pip install -e .

For development:

pip install -e ".[dev]"

Quick Start

import goodseed

with goodseed.Run(experiment_name="my-experiment") as run:
    run.log_configs({"learning_rate": 0.001, "batch_size": 32})

    for step in range(100):
        loss = train_step()
        run.log_metrics({"loss": loss}, step=step)

Then view your runs:

goodseed serve

How It Works

Each Run() creates a SQLite file at ~/.goodseed/projects/<project>/runs/<run_name>.sqlite. Metrics and configs are written there during training. After the run closes, the WAL is checkpointed so the result is a single .sqlite file.

The goodseed serve command starts a local HTTP server that reads these files and exposes a JSON API. The frontend at goodseed.ai connects to this server to display your runs.

Configuration

Variable	Description
GOODSEED_HOME	Data directory (default: ~/.goodseed)
GOODSEED_PROJECT	Default project name (default: default)

CLI

goodseed                   # Start the server (default command)
goodseed serve [dir]       # Start the server, optionally from a specific directory
goodseed serve --port 9000 # Use a custom port
goodseed list              # List projects
goodseed list -p default   # List runs in a project

Tests

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

See DOCS.md for architecture details and API reference.