Unified parameter and feature access in machine learning models

1. A method, comprising:

obtaining a function call for a function that calculates an attribute associated with a machine learning model;

for each argument of the function call, identifying, by a computer system, a parameter type of the argument, wherein the parameter type represents a type of data used with the machine learning model;

obtaining a value accessor for retrieving features specific to the parameter type;

obtaining, by the computer system, a value represented by the argument using the value accessor;

calculating, by the computer system, the attribute by applying the function to the value; and

using the attribute to execute the machine learning model.

2. The method of claim 1 , further comprising:

verifying a match between a data type of the value and an expected type from a function declaration for the function prior to applying the function to the value.

3. The method of claim 1 , further comprising:

obtaining the argument through a universal access interface for accessing multiple parameter types associated with the machine learning model.

4. The method of claim 3 , wherein identifying the parameter type of the argument comprises:

obtaining the parameter type from a context associated with a model definition for the machine learning model.

5. The method of claim 4 , wherein using the value accessor for the parameter type to obtain the value represented by the argument further comprises:

obtaining the value accessor from an implementation of the universal access interface.

6. The method of claim 4 , wherein the context is obtained from an internal representation of the machine learning model.

7. The method of claim 1 , wherein using the value accessor for the parameter type to obtain the value represented by the argument comprises:

calling the value accessor from an object representing the argument.

8. The method of claim 1 , wherein the value accessor is further associated with a data type of the argument.

9. The method of claim 1 , wherein the attribute comprises at least one of:

a model parameter;

a request parameter;

a feature; or

a set of documents used to execute the machine learning model.

10. The method of claim 1 , wherein the parameter type is at least one of:

a model parameter;

a request parameter;

a feature; or

a constant.

11. A system, comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the system to:

obtain a function call for a function that calculates an attribute associated with a machine learning model;

for each argument of the function call, identify a parameter type of the argument, wherein the parameter type represents a type of data used with the machine learning model;

obtain a value accessor for retrieving features specific to the parameter type;

obtain a value represented by the argument using the value accessor;

calculate the attribute by applying the function to the value; and

use the attribute to execute the machine learning model.

12. The system of claim 11 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the system to:

verify a match between a data type of the value and an expected type from a function declaration for the function prior to applying the function to the value.

13. The system of claim 11 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the system to:

obtain the argument through a universal access interface for accessing multiple parameter types associated with the machine learning model.

14. The system of claim 11 , wherein identifying the parameter type of the argument comprises:

obtaining the parameter type from a context associated with a model definition for the machine learning model.

15. The system of claim 14 , wherein using the value accessor for the parameter type to obtain the value represented by the argument further comprises:

obtaining the value accessor from an implementation of the universal access interface.

16. The system of claim 11 , wherein the value accessor is further associated with a data type of the argument.

17. The system of claim 11 , wherein the attribute comprises at least one of:

a model parameter;

a request parameter;

a feature; or

a set of documents used to execute the machine learning model.

18. The system of claim 11 , wherein the parameter type is at least one of:

a model parameter;

a request parameter;

a feature; or

a constant.

19. A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method, the method comprising:

obtaining a function call for a function that calculates an attribute associated with a machine learning model;

for each argument of the function call, identifying a parameter type of the argument, wherein the parameter type represents a type of data used with the machine learning model;

obtaining a value accessor for retrieving features specific to the parameter type;

obtaining a value represented by the argument using the value accessor;

calculating the attribute by applying the function to the value; and

using the attribute to execute the machine learning model.

20. The non-transitory computer readable storage medium of claim 19 , wherein the method further comprises:

verifying a match between a data type of the value and an expected type from a function declaration for the function prior to applying the function to the value.