Process flow for model-based applications
A process flow for model-based applications, including: receiving data from one or more data sources; applying at least one first transformation on at least a portion of the data to generate transformed input data encoded according to a predefined format; providing the transformed input data to an executed instance of a model facilitating a prediction associated with the data; and exposing access to application data based on an output associated with the model.
1. A method for a process flow for model-based applications, comprising:
receiving data from one or more data sources;
applying at least one first transformation on at least a portion of the data to generate transformed input data encoded according to a predefined format;
providing the transformed input data to an executed instance of a model facilitating a prediction associated with the data;
applying at least one second transformation to the output associated with the model to generate a plurality of versions of application data, wherein each version of the application data is encoded in a different format corresponding to a different application;
exposing access to the plurality of versions of the application data based on an output associated with the model; and
updating an execution metadata store storing metadata facilitating replay of one or more model executions, wherein the metadata is stored separate from data described by the metadata.
2. The method of claim 1 , wherein the one or more data sources comprise a data store or a streaming data source.
3. The method of claim 1 , wherein the data comprises structured data and unstructured data.
4. The method of claim 1 , wherein applying the at least one transformation on at least a portion of the data comprises transforming unstructured data to generate structured data, wherein the transformed input data comprises the generated structured data.
5. The method of claim 1 , wherein applying the at least one transformation on at least a portion of the data comprises transforming first structured data to generate second structured data, wherein the transformed input data comprises the generated second structured data.
6. The method of claim 1 , further comprising storing the transformed input data.
7. The method of claim 6 , wherein the transformed input data is stored independent of the plurality of versions of application data.
8. The method of claim 1 , wherein the predefined format is associated with the model.
9. The method of claim 1 , wherein exposing access to the plurality of versions of the application data comprises exposing access to each version of the application data via a corresponding application program interface (API) of a plurality of APIs.
10. The method of claim 1 , wherein each entry of the metadata comprises, for a given model execution, a model identifier.
11. The method of claim 10 , wherein each entry of the metadata identifies, for the given model execution, input to the model execution and output of the model execution.
12. An apparatus for a process flow for model-based applications, the apparatus comprising one or more processors and memory storing instructions that, when executed, cause the one or more processors to perform steps comprising:
receiving data from one or more data sources;
applying at least one first transformation on at least a portion of the data to generate transformed input data encoded according to a predefined format;
providing the transformed input data to an executed instance of a model facilitating a prediction associated with the data;
applying at least one second transformation to the output associated with the model to generate a plurality of versions of application data, wherein each version of the application data is encoded in a different format corresponding to a different application;
exposing access to the plurality of versions of the application data based on an output associated with the model; and
updating an execution metadata store storing metadata facilitating replay of one or more model executions, wherein the metadata is stored separate from data described by the metadata.
13. The apparatus of claim 12 , wherein the one or more data sources comprise a data store or a streaming data source.
14. The apparatus of claim 12 , wherein the data comprises structured data and unstructured data.
15. The apparatus of claim 12 , wherein applying the at least one transformation on at least a portion of the data comprises transforming unstructured data to generate structured data, wherein the transformed input data comprises the generated structured data.
16. The apparatus of claim 12 , wherein applying the at least one transformation on at least a portion of the data comprises transforming first structured data to generate second structured data, wherein the transformed input data comprises the generated second structured data.
17. The apparatus of claim 12 , wherein the steps further comprise storing the transformed input data.
18. The apparatus of claim 17 , wherein the transformed input data is stored independent of the plurality of versions of application data.
19. The apparatus of claim 12 , wherein the predefined format is associated with the model.
20. The apparatus of claim 12 , wherein exposing access to the plurality of versions of the application data comprises exposing access to each version of the application data via a corresponding application program interface (API) of a plurality of APIs.