Version management for machine learning pipeline building

A computer system, the computer system comprising:

• one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage medium, and program instructions stored on at least one of the one or more computer-readable tangible storage medium for execution by at least one of the one or more processors via at least one of the one or more computer-readable memories, wherein the computer system is capable of performing a method comprising:
• a computer-based method of automated version management for machine learning pipeline development, the method comprising:
• computing a quality value of a target machine learning pipeline at a predetermined regular interval, and automatically saving updated versions of the target machine learning pipeline based on the computed quality value;
• extracting a series of key features from detected versions of the target machine learning pipeline;
• clustering the detected versions of the target machine learning model pipeline by the extracted series of key features;
• identifying a highest-quality version within each of a series of generated clusters;
• computing similarity scores for subsets of the detected versions within each of the series of generated clusters for each version of the target machine learning pipeline being developed, wherein different versions of the series are analyzed to determine when a set of unique pairs of one or more individual versions exist that are both contained within a same cluster and encoding each branch into a machine learning operation sequence;
• computing the similarity scores for the subsets of the detected versions within each of the series of generated clusters by automatically encoding a series of operations associated with branches corresponding to the subsets of versions;
• computing a final similarity score for each version of the target machine learning pipeline by averaging values from each set of unique pairs;
• generating and outputting to a user, using the final similarity score, a visual representation of the series of generated clusters including version quality data and version similarity data, wherein the visual representation includes a quality evolution visual depicting the highest-quality versions from each of the series of generated clusters in combination with the version quality data;
• and
• automatically saving at least one unique version of a given machine learning pipeline stream anytime a preselected amount of changes or modifications have been made for a later use in management of the given machine learning pipeline stream.

A computer program product, the computer program product comprising:

• one or more non-transitory computer-readable tangible storage medium and program instructions stored on at least one of the one or more computer-readable tangible storage medium, the program instructions executable by a processor capable of performing a method, the method comprising:
• a computer-based method of automated version management for machine learning pipeline development, the method comprising:
• computing a quality value of a target machine learning pipeline at a predetermined regular interval, and automatically saving updated versions of the target machine learning pipeline based on the computed quality value;
• extracting a series of key features from detected versions of the target machine learning pipeline;
• clustering the detected versions of the target machine learning model pipeline by the extracted series of key features;
• identifying a highest-quality version within each of a series of generated clusters;
• computing similarity scores for subsets of the detected versions within each of the series of generated clusters for each version of the target machine learning pipeline being developed, wherein different versions of the series are analyzed to determine when a set of unique pairs of one or more individual versions exist that are both contained within a same cluster and encoding each branch into a machine learning operation sequence;
• computing the similarity scores for the subsets of the detected versions within each of the series of generated clusters by automatically encoding a series of operations associated with branches corresponding to the subsets of versions;
• computing a final similarity score for each version of the target machine learning pipeline by averaging values from each set of unique pairs;
• generating and outputting to a user, using the final similarity score, a visual representation of the series of generated clusters including version quality data and version similarity data, wherein the visual representation includes a quality evolution visual depicting the highest-quality versions from each of the series of generated clusters in combination with the version quality data;
• and
• automatically saving at least one unique version of a given machine learning pipeline stream anytime a preselected amount of changes or modifications have been made for a later use in management of the given machine learning pipeline stream.