dagger: A Python Framework for Reproducible Machine Learning Experiment Orchestration

TL;DR

Dagger is a Python framework designed to facilitate reproducible and reusable orchestration of complex, multi-stage machine learning experiments, addressing challenges in tracking experimental provenance.

Contribution

It introduces a novel framework that simplifies managing complex experiment workflows and ensures reproducibility in machine learning research.

Findings

Enables reproducible experiment pipelines

Simplifies tracking of experimental provenance

Supports multi-stage machine learning workflows

Abstract

Many research directions in machine learning, particularly in deep learning, involve complex, multi-stage experiments, commonly involving state-mutating operations acting on models along multiple paths of execution. Although machine learning frameworks provide clean interfaces for defining model architectures and unbranched flows, burden is often placed on the researcher to track experimental provenance, that is, the state tree that leads to a final model configuration and result in a multi-stage experiment. Originally motivated by analysis reproducibility in the context of neural network pruning research, where multi-stage experiment pipelines are common, we present dagger, a framework to facilitate reproducible and reusable experiment orchestration. We describe the design principles of the framework and example usage.