pailab the integrated machine learning workbench
Project description
pailab
pailab is an integrated machine learning workbench to version, analyze and automatize the machine learning model building processes and deployments. It keeps track of changes in your machine learning pipeline (code, data, parameters) similar to classical version control systems considering special characteristics of AI model building processes.
It provides:
Versioning of all objects of the ML development cycle
Full transparency over the whole ML development cycle
Support to work in a team on the same AI projects, sharing data, algorithms and results
Consistency checks across the whole pipeline
Distributed execution of parallel jobs, e.g. for parameter studies
Standardized analysis plots
Jupyter widgets to administrate and control the ML repo
Al objects added to the repository are split into a part containing big data and a part with the remaining data handling them separately with different technqiues. For example one may use git to administrate the remaining data part. It also adds information to each object such as the version, the author, date etc. and each object is labeled with a category defining the role in the ml process.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.