Machine learning model publishing systems and methods
A machine learning (ML) model publisher can, responsive to an indication that a ML model is ready for publication, generate a publication request form or page on a user device. The ML model publisher can be invoked from within a ML modeling application. Responsive to an instruction received through the publication request form or page, the ML model publisher can access a data structure in memory used in training the ML model and populate the publication request form or page with attributes required by the ML model to run. Responsive to activation of a single-click publication actuator, the ML model publisher can publish the ML model directly from the ML modeling application to a target computing system by providing, to the target computing system, a path to a repository location where the ML model is stored and information on the attributes required by the ML model to run.
1. A method, comprising:
receiving an indication that a machine learning (ML) model has been trained and tested and is ready for publication, the receiving performed by a ML model publisher through a user interface of a ML modeling application, the ML modeling application hosted on an artificial intelligence (AI) platform;
responsive to the indication that the ML model has been trained and test and is ready for publication, generating a publication request form or page on a user device through the user interface of the ML modeling application hosted on the AI platform, the generating performed by the ML model publisher, the publication request form or page containing a user interface element and a single-click publication actuator;
responsive to activation of the user interface element on the publication request form or page generated by the ML model publisher, accessing a data structure in memory, the data structure used in training the ML model, the accessing performed by the ML model publisher;
automatically populating, by the ML model publisher, at least a section of the publication request form or page with attributes from the data structure, the attributes required by the ML model to run; and
responsive to activation of the single-click publication actuator on the publication request form or page generated by the ML model publisher in response to the indication that the ML model has been trained and tested and is ready for publication, publishing the ML model to a target computing system on the AI platform, the publishing performed by the ML model publisher, the publishing including providing, to the target computing system, a path to a repository location where the ML model is stored and information on the attributes required by the ML model to run.
2. The method according to claim 1 , wherein the ML model publisher runs within the ML modeling application hosted on the AI platform.
3. The method according to claim 1 , wherein the ML model publisher is operable to examine a ML model design file for the ML model, extract the path from the design ML model design file, and populate the publication request form or page with the path.
4. The method according to claim 1 , wherein the ML model is trained using a Spark-based ML pipeline.
5. The method according to claim 1 , wherein the repository location is in a Hadoop distributed file system.
6. The method according to claim 1 , wherein the ML model publisher is operable to access the data structure in memory by querying a Spark application programming interface with a Dataframe variable for datasets used in training the ML model.
7. The method according to claim 1 , wherein the data structure in memory comprises a Spark Dataframe.
8. A system, comprising:
a processor;
a non-transitory computer-readable medium; and
stored instructions translatable by the processor to implement a ML model publisher for:
receiving an indication that a machine learning (ML) model has been trained and tested and is ready for publication, the receiving performed by the ML model publisher through a user interface of a ML modeling application, the ML modeling application hosted on an artificial intelligence (AI) platform;
responsive to the indication that the ML model has been trained and test and is ready for publication, generating a publication request form or page on a user device through the user interface of the ML modeling application hosted on the AI platform, the publication request form or page containing a user interface element and a single-click publication actuator;
responsive to activation of the user interface element on the publication request form or page generated by the ML model publisher, accessing a data structure in memory, the data structure used in training the ML model;
automatically populating at least a section of the publication request form or page with attributes from the data structure, the attributes required by the ML model to run; and
responsive to activation of the single-click publication actuator on the publication request form or page generated by the ML model publisher in response to the indication that the ML model has been trained and tested and is ready for publication, publishing the ML model to a target computing system on the AI platform, the publishing including providing, to the target computing system, a path to a repository location where the ML model is stored and information on the attributes required by the ML model to run.
9. The system of claim 8 , wherein the ML model publisher runs within the ML modeling application hosted on the AI platform.
10. The system of claim 8 , wherein the ML model publisher is operable to examine a ML model design file for the ML model, extract the path from the design ML model design file, and populate the publication request form or page with the path.
11. The system of claim 8 , wherein the ML model is trained using a Spark-based ML pipeline.
12. The system of claim 8 , wherein the repository location is in a Hadoop distributed file system.
13. The system of claim 8 , wherein the ML model publisher is operable to access the data structure in memory by querying a Spark application programming interface with a Dataframe variable for datasets used in training the ML model.
14. The system of claim 8 , wherein the data structure in memory comprises a Spark Dataframe.
15. A computer program product comprising a non-transitory computer-readable medium storing instructions translatable by a processor to implement a ML model publisher for:
receiving an indication that a machine learning (ML) model has been trained and tested and is ready for publication, the receiving performed by the ML model publisher through a user interface of a ML modeling application, the ML modeling application hosted on an artificial intelligence (AI) platform;
responsive to the indication that the ML model has been trained and test and is ready for publication, generating a publication request form or page on a user device through the user interface of the ML modeling application hosted on the AI platform, the publication request form or page containing a user interface element and a single-click publication actuator;
responsive to activation of the user interface element on the publication request form or page generated by the ML model publisher, accessing a data structure in memory, the data structure used in training the ML model;
automatically populating at least a section of the publication request form or page with attributes from the data structure, the attributes required by the ML model to run; and
responsive to activation of the single-click publication actuator on the publication request form or page generated by the ML model publisher in response to the indication that the ML model has been trained and tested and is ready for publication, publishing the ML model to a target computing system on the AI platform, the publishing including providing, to the target computing system, a path to a repository location where the ML model is stored and information on the attributes required by the ML model to run.
16. The computer program product of claim 15 , wherein the ML model publisher is operable to examine a ML model design file for the ML model, extract the path from the design ML model design file, and populate the publication request form or page with the path.
17. The computer program product of claim 15 , wherein the ML model is trained using a Spark-based ML pipeline.
18. The computer program product of claim 15 , wherein the repository location is in a Hadoop distributed file system.
19. The computer program product of claim 15 , wherein the ML model publisher is operable to access the data structure in memory by querying a Spark application programming interface with a Dataframe variable for datasets used in training the ML model.
20. The computer program product of claim 15 , wherein the data structure in memory comprises a Spark Dataframe.