Model integration tool
Certain aspects involve models for generating code executed on data-processing platforms. One example involves receiving an electronic data-processing model, which generates an analytical output from input attributes weighted with respective modeling coefficients. A target data-processing platform is identified that requires bin ranges for the modeling coefficients and reason codes for the input attributes. Modeling code is generated that implements the electronic data-processing model with the bin ranges and the reason codes. The processor outputs executable code that implements the electronic data-processing model.
1. A computing system comprising:
a processor communicatively coupled via a data network to a client computing system; and
a non-transitory computer-readable medium storing program code that is executable by the processor;
wherein the processor is configured for executing the program code to perform operations comprising:
receiving an electronic data-processing model that generates an analytical output by applying a function to input attributes weighted with modeling coefficients, respectively,
identifying a target data-processing platform that requires bin ranges respectively associated with the modeling coefficients and that also requires reason codes respectively associated with at least some of the bin ranges, wherein a bin range identifies a range of input attribute values and one or more modeling coefficients are valid for input attribute values within the range of input attribute values, wherein each reason code indicates a respective impact of a respective input attribute on the analytical output,
identifying the bin ranges and the reason codes,
generating executable code that implements the electronic data-processing model with the bin ranges and the reason codes, wherein the executable code includes code for implementing:
(a) mapping at least one of the input attributes to a production attribute from the target data-processing platform, the production attribute having a normalized value that is compliant with the electronic data-processing model,
(b) a first analytical function,
(c) a second analytical function, and
(d) segmentation logic comprising a decision tree that (e) selects the first analytical function based on a specific input attribute having a first value and (f) selects the second analytical function based on the specific input attribute having a second value; and
verifying, from a test of the executable code, that a test analytical output from the test matches an expected analytical output from an audit file and that a test reason code from the test matches an expected reason code from the audit file.
2. The computing system of claim 1 , wherein the electronic data-processing model comprises a regression model and wherein the modeling coefficients comprise regression coefficients, respectively.
3. The computing system of claim 1 , wherein the first analytical function comprises a first decision tree function and the second analytical function comprises a second decision tree function.
4. The computing system of claim 1 , wherein the executable code includes additional code for generating the analytical output as a combination of decision tree functions in the electronic data-processing model.
5. The computing system of claim 1 , wherein the executable code includes additional code for:
transforming respective values of the production attribute into respective values that are compliant with the electronic data-processing model.
6. The computing system of claim 5 , wherein transforming the respective values of the production attribute comprises generating the normalized value of the production attribute.
7. The computing system of claim 1 , wherein the operations further comprise generating, from the executable code, source code that implements the electronic data-processing model on the target data-processing platform.
8. A method in which processing hardware of a computing system performs operations comprising:
receiving an electronic data-processing model that generates an analytical output by applying a function to input attributes weighted with modeling coefficients, respectively;
identifying a target data-processing platform that requires bin ranges respectively associated with the modeling coefficients and that also requires reason codes respectively associated with at least some of the bin ranges, wherein a bin range identifies a range of input attribute values and one or more modeling coefficients are valid for input attribute values within the range of input attribute values, wherein each reason code indicates a respective impact of a respective input attribute on the analytical output;
identifying the bin ranges and the reason codes;
generating executable code that implements the electronic data-processing model with the bin ranges and the reason codes, wherein the executable code includes code for implementing:
(a) mapping at least one of the input attributes to a production attribute from the target data-processing platform, the production attribute having a normalized value that is compliant with the electronic data-processing model,
(b) a first analytical function,
(c) a second analytical function, and
(d) segmentation logic comprising a decision tree that (e) selects the first analytical function based on a specific input attribute having a first value and (f) selects the second analytical function based on the specific input attribute having a second value; and
verifying, from a test of the executable code, that a test analytical output from the test matches an expected analytical output from an audit file and that a test reason code from the test matches an expected reason code from the audit file.
9. The method of claim 8 , wherein the electronic data-processing model comprises a regression model and wherein the modeling coefficients comprise regression coefficients, respectively.
10. The method of claim 8 , wherein the first analytical function comprises a first decision tree function and the second analytical function comprises a second decision tree function.
11. The method of claim 8 , wherein the executable code includes additional code for generating the analytical output as a combination of decision tree functions in the electronic data-processing model.
12. The method of claim 8 , wherein the executable code includes additional code for:
transforming respective values of the production attribute into respective values that are compliant with the electronic data-processing model.
13. The method of claim 12 , wherein transforming the respective values of the production attribute comprises generating the normalized value of the production attribute.
14. The method of claim 8 , wherein the operations further comprise generating, from the executable code, source code that implements the electronic data-processing model on the target data-processing platform.
15. A non-transitory computer-readable medium storing program code that, when executed by processing hardware of a computing system, performs operations comprising:
receiving an electronic data-processing model that generates an analytical output by applying a function to input attributes weighted with modeling coefficients, respectively;
identifying a target data-processing platform that requires bin ranges respectively associated with the modeling coefficients and that also requires reason codes respectively associated with at least some of the bin ranges, wherein a bin range identifies a range of input attribute values and one or more modeling coefficients are valid for input attribute values within the range of input attribute values, wherein each reason code indicates a respective impact of a respective input attribute on the analytical output;
identifying the bin ranges and the reason codes;
generating executable code that implements the electronic data-processing model with the bin ranges and the reason codes, wherein the executable code includes code for implementing:
(a) mapping at least one of the input attributes to a production attribute from the target data-processing platform, the production attribute having a normalized value that is compliant with the electronic data-processing model,
(b) a first analytical function,
(c) a second analytical function, and
(d) segmentation logic comprising a decision tree that (e) selects the first analytical function based on a specific input attribute having a first value and (f) selects the second analytical function based on the specific input attribute having a second value; and
verifying, from a test of the executable code, that a test analytical output from the test matches an expected analytical output from an audit file and that a test reason code from the test matches an expected reason code from the audit file.
16. The non-transitory computer-readable medium of claim 15 , wherein the electronic data-processing model comprises a regression model and wherein the modeling coefficients comprise regression coefficients, respectively.
17. The non-transitory computer-readable medium of claim 15 , wherein the first analytical function comprises a first decision tree function and the second analytical function comprises a second decision tree function.
18. The non-transitory computer-readable medium of claim 15 , wherein the executable code includes additional code for generating the analytical output as a combination of decision tree functions in the electronic data-processing model.
19. The non-transitory computer-readable medium of claim 15 , wherein the executable code includes additional code for:
transforming respective values of the production attribute into respective values that are compliant with the electronic data-processing model, wherein the transforming comprises generating the normalized value of the production attribute.
20. The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise generating, from the executable code, source code that implements the electronic data-processing model on the target data-processing platform.