Methods and systems for correlating source code commitments and model result records during model development

Abstract

Methods and systems are described herein for improvements to model development, particularly in terms of source code commitments for these models. For example, methods and systems allow contributors to manage when and where commitments and/or the results of those commitments are stored and/or do so without the use of an additional layer of external software by correlating source code commitments and model result records during model development. Moreover, the methods and systems allow contributors to manage when and where commitments are store automatically without the use of an additional layer of external software through user of native source code commands and libraries.

What is claimed is: 
     
         1 . A system for correlating source code commitments and model result records during model development without use of an additional layer of external software through the use of native source code commands and libraries, comprising:
 cloud-based memory configured to store, for a software development version control system, a plurality of source code commitments for a model development project for a model;   cloud-based control circuitry configured to:
 receive, at a software development version control system, a first subset of user inputs comprising a first source code commitment for a model development project for a model, wherein the first source code commitment comprises a first storage identifier, wherein the first storage identifier comprises a first source code string in the first source code commitment, wherein the first source code commitment comprises a plurality of lines of source code for updating the model, and wherein the first storage identifier is located within the plurality of lines of source code; 
 store, based on the first storage identifier, model data in the software development version control system; 
 receive, at the software development version control system, a second subset of user inputs comprising a second source code commitment for the model development project for the model, wherein the second source code commitment comprises a second storage identifier; 
 store, based on the second storage identifier, additional model data in the software development version control system; and 
   cloud-based I/O circuitry configured to generate for simultaneous display, in a user interface for the software development version control system, a first icon, a second icon, a third icon, and a fourth icon, wherein selection of the first icon causes a human-readable version of the first source code commitment to be displayed in the user interface, wherein selection of the second icon causes a human-readable version the model data to be displayed in the user interface, wherein selection of the third icon causes a human-readable version of the second source code commitment to be displayed in the user interface, and wherein selection of the fourth icon causes a human-readable version of the additional model data to be displayed in the user interface.   
     
     
         2 . A method for correlating source code commitments and model result records during model development without use of an additional layer of external software through the use of native source code commands and libraries, the method comprising:
 receiving, at a software development version control system, a first subset of user inputs comprising a first source code commitment for a model development project for a model, wherein the first source code commitment comprises a first storage identifier, wherein the first storage identifier comprises a first source code string in the first source code commitment, wherein the first source code commitment comprises a plurality of lines of source code for updating the model, and wherein the first storage identifier is located within the plurality of lines of source code;   storing, based on the first storage identifier, model data in the software development version control system; and   generating for simultaneous display, in a user interface for the software development version control system, a first icon and a second icon, wherein selection of the first icon causes a human-readable version of the first source code commitment to be displayed in the user interface, and wherein selection of the second icon causes a human-readable version of the model data to be displayed in the user interface.   
     
     
         3 . The method of  claim 2 , further comprising:
 receiving, at the software development version control system, a second subset of user inputs comprising a second source code commitment for the model development project for the model, wherein the second source code commitment comprises a second storage identifier;   storing, based on the second storage identifier, additional model data in the software development version control system;   generating for simultaneous display, in a user interface for the software development version control system, a third icon and a fourth icon, wherein selection of the third icon causes a human-readable version of the second source code commitment to be displayed in the user interface, and wherein selection of the fourth icon causes a human-readable version of the additional model data to be displayed in the user interface.   
     
     
         4 . The method of  claim 3 , wherein the first source code commitment is executed prior to a validation of the first source code commitment by the software development version control system, and wherein the second source code commitment is executed prior to a validation of the second source code commitment by the software development version control system. 
     
     
         5 . The method of  claim 3 , wherein the second storage identifier comprises a second source code string in the second source code commitment. 
     
     
         6 . The method of  claim 5 , wherein the first source code string is written in Python and indicates whether the model data should be stored on a local device or a remote device, and the second source code string is written in Python and indicates whether the additional model data should be stored on the local device or the remote device. 
     
     
         7 . The method of  claim 2 , wherein storing, based on the first storage identifier, the model data in the software development version control system, comprises:
 retrieving a library comprising a common Pythonic interface with an application programming interface that is uniform across Secure Shell network protocols, Hadoop distributed file system protocols, and Simple Storage Service protocols;   determining, based on the library, a first storage location for the model data based on the first storage identifier; and   transmitting a storage instruction to store the model data at the first storage location.   
     
     
         8 . The method of  claim 2 , wherein storing, based on the first storage identifier, the model data in the software development version control system, comprises:
 identifying the first storage identifier in the first source code commitment; and   determining, using a command-line interface of the software development version control system, a first storage location for the model data.   
     
     
         9 . The method of  claim 2 , wherein the first storage identifier comprises an extension for pushing project files to the software development version control system in a predefined interval without taking any action by a developer. 
     
     
         10 . The method of  claim 2 , wherein the first storage identifier comprises a file system object that identifies a first storage location for the model data. 
     
     
         11 . The method of  claim 2 , further comprising:
 generating a commit hash based on the first source code commitment;   storing the first source code commitment in the software development version control system; and   linking the first source code commitment to the model data in the software development version control system based on the commit hash.   
     
     
         12 . A non-transitory computer-readable medium for correlating source code commitments and model result records during model development without use of an additional layer of external software through the use of native source code commands and libraries comprising instructions that, when executed by one or more processors, cause operations comprising:
 receiving, at a software development version control system, a first subset of user inputs comprising a first source code commitment for a model development project for a model, wherein the first source code commitment comprises a first storage identifier, wherein the first storage identifier comprises a first source code string in the first source code commitment, wherein the first source code commitment comprises a plurality of lines of source code for updating the model, and wherein the first storage identifier is located within the plurality of lines of source code;   storing, based on the first storage identifier, model data in the software development version control system; and   generating for simultaneous display, in a user interface for the software development version control system, a first icon and a second icon, wherein selection of the first icon causes a human-readable version of the first source code commitment to be displayed in the user interface, and wherein selection of the second icon causes a human-readable version of the model data to be displayed in the user interface.   
     
     
         13 . The non-transitory computer readable medium of  claim 12 , wherein the instructions further cause operations comprising:
 receiving, at the software development version control system, a second subset of user inputs comprising a second source code commitment for the model development project for the model, wherein the second source code commitment comprises a second storage identifier;   storing, based on the second storage identifier, additional model data in the software development version control system;   generating for simultaneous display, in a user interface for the software development version control system, a third icon and a fourth icon, wherein selection of the third icon causes a human-readable version of the second source code commitment to be displayed in the user interface, and wherein selection of the fourth icon causes a human-readable version of the additional model data to be displayed in the user interface.   
     
     
         14 . The non-transitory computer readable medium of  claim 13 , wherein the first source code commitment is executed prior to a validation of the first source code commitment by the software development version control system, and wherein the second source code commitment is executed prior to a validation of the second source code commitment by the software development version control system. 
     
     
         15 . The non-transitory computer readable medium of  claim 13 , wherein the second storage identifier comprises a second source code string in the second source code commitment. 
     
     
         16 . The non-transitory computer readable medium of  claim 15 , wherein the first source code string is written in Python and indicates whether the model data should be stored on a local device or a remote device, and the second source code string is written in Python and indicates whether the additional model data should be stored on the local device or the remote device. 
     
     
         17 . The non-transitory computer readable medium of  claim 12 , wherein storing, based on the first storage identifier, the model data in the software development version control system, comprises:
 retrieving a library comprising a common Pythonic interface with an application programming interface that is uniform across Secure Shell network protocols, Hadoop distributed file system protocols, and Simple Storage Service protocols;   determining, based on the library, a first storage location for the model data based on the first storage identifier; and   transmitting a storage instruction to store the model data at the first storage location.   
     
     
         18 . The non-transitory computer readable medium of  claim 12 , wherein storing, based on the first storage identifier, the model data in the software development version control system, comprises:
 identifying the first storage identifier in the first source code commitment; and   determining, using a command-line interface of the software development version control system, a first storage location for the model data.   
     
     
         19 . The non-transitory computer readable medium of  claim 12 , wherein the first storage identifier comprises an extension for pushing project files to the software development version control system in a predefined interval without taking any action by a developer. 
     
     
         20 . The non-transitory computer readable medium of  claim 12 , wherein the instructions further cause operations comprising:
 generating a commit hash based on the first source code commitment;   storing the first source code commitment in the software development version control system; and   linking the first source code commitment to the model data in the software development version control system based on the commit hash.